-mm1 --- diff/CREDITS 2004-02-18 08:54:06.000000000 +0000 +++ source/CREDITS 2004-03-01 15:42:14.000000000 +0000 @@ -289,6 +289,15 @@ S: Via Delle Palme, 9 S: Terni 05100 S: Italy +N: Krzysztof Benedyczak +E: golbi@mat.uni.torun.pl +W: http://www.mat.uni.torun.pl/~golbi +D: POSIX message queues fs (with M. Wronski) +S: ul. Podmiejska 52 +S: Radunica +S: 83-000 Pruszcz Gdanski +S: Poland + N: Randolph Bentson E: bentson@grieg.seaslug.org W: http://www.aa.net/~bentson/ @@ -3489,6 +3498,14 @@ S: 12725 SW Millikan Way, Suite 400 S: Beaverton, OR 97005 S: USA +N: Michal Wronski +E: wrona@mat.uni.torun.pl +W: http://www.mat.uni.torun.pl/~wrona +D: POSIX message queues fs (with K. Benedyczak) +S: ul. Teczowa 23/12 +S: 80-680 Gdansk-Sobieszewo +S: Poland + N: Frank Xia E: qx@math.columbia.edu D: Xiafs filesystem [defunct] --- diff/Documentation/00-INDEX 2003-10-09 09:47:16.000000000 +0100 +++ source/Documentation/00-INDEX 2004-03-01 15:42:14.000000000 +0000 @@ -1,7 +1,7 @@ This is a brief list of all the files in ./linux/Documentation and what -they contain. If you add a documentation file, please list it here in -alphabetical order as well, or risk being hunted down like a rabid dog. +they contain. If you add a documentation file, please list it here in +alphabetical order as well, or risk being hunted down like a rabid dog. Please try and keep the descriptions small enough to fit on one line. Thanks -- Paul G. @@ -26,10 +26,14 @@ DocBook/ - directory with DocBook templates etc. for kernel documentation. IO-mapping.txt - how to access I/O mapped memory from within device drivers. +IPMI.txt + - info on Linux Intelligent Platform Management Interface (IPMI) Driver. IRQ-affinity.txt - how to select which CPU(s) handle which interrupt events on SMP. +MSI-HOWTO.txt + - the Message Signaled Interrupts (MSI) Driver Guide HOWTO and FAQ. README.DAC960 - - info on Mylex DAC960/DAC1100 PCI RAID Controller Driver for Linux + - info on Mylex DAC960/DAC1100 PCI RAID Controller Driver for Linux. README.moxa - release notes for Moxa mutiport serial card. SAK.txt @@ -42,6 +46,10 @@ VGA-softcursor.txt - how to change your VGA cursor from a blinking underscore. arm/ - directory with info about Linux on the ARM architecture. +as-iosched.txt + - info on anticipatory IO scheduler. +basic_profiling.txt + - basic instructions for those who wants to profile Linux kernel. binfmt_misc.txt - info on the kernel support for extra binary formats. block/ @@ -53,25 +61,31 @@ cciss.txt cdrom/ - directory with information on the CD-ROM drivers that Linux has. computone.txt - - info on Computone Intelliport II/Plus Multiport Serial Driver + - info on Computone Intelliport II/Plus Multiport Serial Driver. cpqarray.txt - info on using Compaq's SMART2 Intelligent Disk Array Controllers. cpufreq/ - - info on CPU frequency and voltage scaling + - info on CPU frequency and voltage scaling. cris/ - directory with info about Linux on CRIS architecture. +debugging-modules.txt + - some notes on debugging modules after Linux 2.6.3. devices.txt - - plain ASCII listing of all the nodes in /dev/ with major minor #'s + - plain ASCII listing of all the nodes in /dev/ with major minor #'s. digiboard.txt - info on the Digiboard PC/X{i,e,eve} multiport boards. digiepca.txt - info on Digi Intl. {PC,PCI,EISA}Xx and Xem series cards. dnotify.txt - info about directory notification in Linux. -driver-model.txt - - info about Linux driver model. +driver-model/ + - directory with info about Linux driver model. +dvb/ + - info on Linux Digital Video Broadcast (DVB) subsystem. early-userspace/ - info about initramfs, klibc, and userspace early during boot. +eisa.txt + - info on EISA bus support. exception.txt - how Linux v2.2 handles exceptions without verify_area etc. fb/ @@ -81,47 +95,51 @@ filesystems/ floppy.txt - notes and driver options for the floppy disk driver. ftape.txt - - notes about the floppy tape device driver + - notes about the floppy tape device driver. hayes-esp.txt - info on using the Hayes ESP serial driver. highuid.txt - notes on the change from 16 bit to 32 bit user/group IDs. +hw_random.txt + - info on Linux support for random number generator in i8xx chipsets. i2c/ - - directory with info about the I2C bus/protocol (2 wire, kHz speed) + - directory with info about the I2C bus/protocol (2 wire, kHz speed). i386/ - - directory with info about Linux on intel 32 bit architecture. -i810_rng.txt - - info on Linux support for random number generator in i8xx chipsets. + - directory with info about Linux on Intel 32 bit architecture. ia64/ - - directory with info about Linux on intel 64 bit architecture. + - directory with info about Linux on Intel 64 bit architecture. ide.txt - - important info for users of ATA devices (IDE/EIDE disks and CD-ROMS) + - important info for users of ATA devices (IDE/EIDE disks and CD-ROMS). initrd.txt - how to use the RAM disk as an initial/temporary root filesystem. input/ - info on Linux input device support. ioctl-number.txt - how to implement and register device/driver ioctl calls. +iostats.txt + - info on I/O statistics Linux kernel provides. isapnp.txt - - info on Linux ISA Plug & Play support + - info on Linux ISA Plug & Play support. isdn/ - directory with info on the Linux ISDN support, and supported cards. java.txt - - info on the in-kernel binary support for Java(tm) + - info on the in-kernel binary support for Java(tm). kbuild/ - - directory with info about the kernel build process + - directory with info about the kernel build process. kernel-doc-nano-HOWTO.txt - mini HowTo on generation and location of kernel documentation files. kernel-docs.txt - listing of various WWW + books that document kernel internals. kernel-parameters.txt - summary listing of command line / boot prompt args for the kernel. +kobject.txt + - info of the kobject infrastructure of the Linux kernel. ldm.txt - a brief description of LDM (Windows Dynamic Disks). locks.txt - info on file locking implementations, flock() vs. fcntl(), etc. logo.gif - - Full colour GIF image of Linux logo (penguin) + - Full colour GIF image of Linux logo (penguin). logo.txt - Info on creator of above logo & site to get additional images from. m68k/ @@ -133,27 +151,27 @@ mandatory.txt mca.txt - info on supporting Micro Channel Architecture (e.g. PS/2) systems. md.txt - - info on boot arguments for the multiple devices driver + - info on boot arguments for the multiple devices driver. memory.txt - info on typical Linux memory problems. mips/ - directory with info about Linux on MIPS architecture. mkdev.cciss - - script to make /dev entries for SMART controllers (see cciss.txt) + - script to make /dev entries for SMART controllers (see cciss.txt). mkdev.ida - script to make /dev entries for Intelligent Disk Array Controllers. moxa-smartio - info on installing/using Moxa multiport serial driver. mtrr.txt - - how to use PPro Memory Type Range Registers to increase performance + - how to use PPro Memory Type Range Registers to increase performance. nbd.txt - info on a TCP implementation of a network block device. networking/ - directory with info on various aspects of networking with Linux. nfsroot.txt - - short guide on setting up a diskless box with NFS root filesystem + - short guide on setting up a diskless box with NFS root filesystem. nmi_watchdog.txt - - info on NMI watchdog for SMP systems + - info on NMI watchdog for SMP systems. oops-tracing.txt - how to decode those nasty internal kernel error dump messages. paride.txt @@ -165,11 +183,11 @@ parport.txt parport-lowlevel.txt - description and usage of the low level parallel port functions. pci.txt - - info on the PCI subsystem for device driver authors -pcwd-watchdog.txt - - info and sample code for using with the PC Watchdog reset card. + - info on the PCI subsystem for device driver authors. pm.txt - info on Linux power management support. +pnp.txt + - Linux Plug and Play documentation. power/ - directory with info on Linux PCI power management. powerpc/ @@ -181,29 +199,31 @@ ramdisk.txt riscom8.txt - notes on using the RISCom/8 multi-port serial driver. rocket.txt - - info on installing/using the Comtrol RocketPort multiport serial driver + - info on installing/using the Comtrol RocketPort multiport serial driver. +rpc-cache.txt + - introduction to the caching mechanisms in the sunrpc layer. rtc.txt - notes on how to use the Real Time Clock (aka CMOS clock) driver. s390/ - directory with info on using Linux on the IBM S390. -sh/ - - directory with info on porting Linux to a new architecture. +sched-design.txt + - goals, design and implementation of the Linux O(1) scheduler. scsi/ - directory with info on Linux scsi support. serial-console.txt - how to set up Linux with a serial line console as the default. sgi-visws.txt - short blurb on the SGI Visual Workstations. +sh/ + - directory with info on porting Linux to a new architecture. smart-config.txt - description of the Smart Config makefile feature. -smp.tex - - LaTeX document describing implementation of Multiprocessor Linux smp.txt - - a few more notes on symmetric multi-processing + - a few notes on symmetric multi-processing. sonypi.txt - info on Linux Sony Programmable I/O Device support. sound/ - - directory with info on sound card support + - directory with info on sound card support. sparc/ - directory with info on using Linux on Sparc architecture. specialix.txt @@ -217,9 +237,9 @@ svga.txt sx.txt - info on the Specialix SX/SI multiport serial driver. sysctl/ - - directory with info on the /proc/sys/* files + - directory with info on the /proc/sys/* files. sysrq.txt - - info on the magic SysRq key + - info on the magic SysRq key. telephony/ - directory with info on telephony (e.g. voice over IP) support. unicode.txt @@ -230,7 +250,7 @@ video4linux/ - directory with info regarding video/TV/radio cards and linux. vm/ - directory with info on the Linux vm code. -watchdog.txt +watchdog/ - how to auto-reboot Linux if it has "fallen and can't get up". ;-) x86_64/ - directory with info on Linux support for AMD x86-64 (Hammer) machines. @@ -238,4 +258,3 @@ xterm-linux.xpm - XPM image of penguin logo (see logo.txt) sitting on an xterm. zorro.txt - info on writing drivers for Zorro bus devices found on Amigas. - --- diff/Documentation/DMA-mapping.txt 2003-08-26 10:00:51.000000000 +0100 +++ source/Documentation/DMA-mapping.txt 2004-03-01 15:42:14.000000000 +0000 @@ -283,7 +283,7 @@ There are two types of DMA mappings: in order to get correct behavior on all platforms. - Streaming DMA mappings which are usually mapped for one DMA transfer, - unmapped right after it (unless you use pci_dma_sync below) and for which + unmapped right after it (unless you use pci_dma_sync_* below) and for which hardware can optimize for sequential accesses. This of "streaming" as "asynchronous" or "outside the coherency @@ -543,14 +543,30 @@ same bus address space) and you could re all bus addresses. If you need to use the same streaming DMA region multiple times and touch -the data in between the DMA transfers, just map it with -pci_map_{single,sg}, and after each DMA transfer call either: +the data in between the DMA transfers, the buffer needs to be synced +properly in order for the cpu and device to see the most uptodate and +correct copy of the DMA buffer. - pci_dma_sync_single(dev, dma_handle, size, direction); +So, firstly, just map it with pci_map_{single,sg}, and after each DMA +transfer call either: + + pci_dma_sync_single_for_cpu(dev, dma_handle, size, direction); or: - pci_dma_sync_sg(dev, sglist, nents, direction); + pci_dma_sync_sg_for_cpu(dev, sglist, nents, direction); + +as appropriate. + +Then, if you wish to let the device get at the DMA area again, +finish accessing the data with the cpu, and then before actually +giving the buffer to the hardware call either: + + pci_dma_sync_single_for_device(dev, dma_handle, size, direction); + +or: + + pci_dma_sync_sg_for_device(dev, sglist, nents, direction); as appropriate. @@ -590,8 +606,9 @@ to use the pci_dma_sync_*() interfaces. * the DMA transfer with the CPU first * so that we see updated contents. */ - pci_dma_sync_single(cp->pdev, cp->rx_dma, cp->rx_len, - PCI_DMA_FROMDEVICE); + pci_dma_sync_single_for_cpu(cp->pdev, cp->rx_dma, + cp->rx_len, + PCI_DMA_FROMDEVICE); /* Now it is safe to examine the buffer. */ hp = (struct my_card_header *) cp->rx_buf; @@ -601,7 +618,13 @@ to use the pci_dma_sync_*() interfaces. pass_to_upper_layers(cp->rx_buf); make_and_setup_new_rx_buf(cp); } else { - /* Just give the buffer back to the card. */ + /* Just sync the buffer and give it back + * to the card. + */ + pci_dma_sync_single_for_device(cp->pdev, + cp->rx_dma, + cp->rx_len, + PCI_DMA_FROMDEVICE); give_rx_buf_to_card(cp); } } @@ -709,12 +732,21 @@ interfaces. To reiterate: When the DMA transfer is complete, invoke: - void pci_dac_dma_sync_single(struct pci_dev *pdev, - dma64_addr_t dma_addr, - size_t len, int direction); + void pci_dac_dma_sync_single_for_cpu(struct pci_dev *pdev, + dma64_addr_t dma_addr, + size_t len, int direction); This must be done before the CPU looks at the buffer again. -This interface behaves identically to pci_dma_sync_{single,sg}(). +This interface behaves identically to pci_dma_sync_{single,sg}_for_cpu(). + +And likewise, if you wish to let the device get back at the buffer after +the cpu has read/written it, invoke: + + void pci_dac_dma_sync_single_for_device(struct pci_dev *pdev, + dma64_addr_t dma_addr, + size_t len, int direction); + +before letting the device access the DMA area again. If you need to get back to the PAGE/OFFSET tuple from a dma64_addr_t the following interfaces are provided: --- diff/Documentation/DocBook/gadget.tmpl 2003-09-30 15:46:10.000000000 +0100 +++ source/Documentation/DocBook/gadget.tmpl 2004-03-01 15:42:14.000000000 +0000 @@ -454,6 +454,7 @@ but some optional utilities are provided !Edrivers/usb/gadget/usbstring.c +!Edrivers/usb/gadget/config.c --- diff/Documentation/binfmt_misc.txt 2003-10-09 09:47:33.000000000 +0100 +++ source/Documentation/binfmt_misc.txt 2004-03-01 15:42:14.000000000 +0000 @@ -15,7 +15,7 @@ First you must mount binfmt_misc: mount binfmt_misc -t binfmt_misc /proc/sys/fs/binfmt_misc To actually register a new binary type, you have to set up a string looking like -:name:type:offset:magic:mask:interpreter: (where you can choose the ':' upon +:name:type:offset:magic:mask:interpreter:flags (where you can choose the ':' upon your needs) and echo it to /proc/sys/fs/binfmt_misc/register. Here is what the fields mean: - 'name' is an identifier string. A new /proc file will be created with this @@ -34,6 +34,28 @@ Here is what the fields mean: The mask is anded with the byte sequence of the file. - 'interpreter' is the program that should be invoked with the binary as first argument (specify the full path) + - 'flags' is an optional field that controls several aspects of the invocation + of the interpreter. It is a string of capital letters, each controls a certain + aspect. The following flags are supported - + 'P' - preserve-argv[0]. Legacy behavior of binfmt_misc is to overwrite the + original argv[0] with the full path to the binary. When this flag is + included, binfmt_misc will add an argument to the argument vector for + this purpose, thus preserving the original argv[0]. + 'O' - open-binary. Legacy behavior of binfmt_misc is to pass the full path + of the binary to the interpreter as an argument. When this flag is + included, binfmt_misc will open the file for reading and pass its + descriptor as an argument, instead of the full path, thus allowing + the interpreter to execute non-readable binaries. This feature should + be used with care - the interpreter has to be trusted not to emit + the contents of the non-readable binary. + 'C' - credentials. Currently, the behavior of binfmt_misc is to calculate + the credentials and security token of the new process according to + the interpreter. When this flag is included, these attributes are + calculated according to the binary. It also implies the 'O' flag. + This feature should be used with care as the interpreter + will run with root permissions when a setuid binary owned by root + is run with binfmt_misc. + There are some restrictions: - the whole register string may not exceed 255 characters @@ -83,9 +105,9 @@ If you want to pass special arguments to write a wrapper script for it. See Documentation/java.txt for an example. -Your interpreter should NOT look in the PATH for the filename; the -kernel passes it the full filename to use. Using the PATH can cause -unexpected behaviour and be a security hazard. +Your interpreter should NOT look in the PATH for the filename; the kernel +passes it the full filename (or the file descriptor) to use. Using $PATH can +cause unexpected behaviour and can be a security hazard. There is a web page about binfmt_misc at --- diff/Documentation/early-userspace/README 2003-08-26 10:00:51.000000000 +0100 +++ source/Documentation/early-userspace/README 2004-03-01 15:42:14.000000000 +0000 @@ -71,5 +71,31 @@ custom initramfs images that meet your n For questions and help, you can sign up for the early userspace mailing list at http://www.zytor.com/mailman/listinfo/klibc +How does it work? +================= + +The kernel has currently 3 ways to mount the root filesystem: + +a) all required device and filesystem drivers compiled into the kernel, no + initrd. init/main.c:init() will call prepare_namespace() to mount the + final root filesystem, based on the root= option and optional init= to run + some other init binary than listed at the end of init/main.c:init(). + +b) some device and filesystem drivers built as modules and stored in an + initrd. The initrd must contain a binary '/linuxrc' which is supposed to + load these driver modules. It is also possible to mount the final root + filesystem via linuxrc and use the pivot_root syscall. The initrd is + mounted and executed via prepare_namespace(). + +c) using initramfs. The call to prepare_namespace() must be skipped. + This means that a binary must do all the work. Said binary can be stored + into initramfs either via modifying usr/gen_init_cpio.c or via the new + initrd format, an cpio archive. It must be called "/init". This binary + is responsible to do all the things prepare_namespace() would do. + + To remain backwards compatibility, the /init binary will only run if it + comes via an initramfs cpio archive. If this is not the case, + init/main.c:init() will run prepare_namespace() to mount the final root + and exec one of the predefined init binaries. Bryan O'Sullivan --- diff/Documentation/filesystems/proc.txt 2004-03-01 14:36:04.000000000 +0000 +++ source/Documentation/filesystems/proc.txt 2004-03-01 15:42:14.000000000 +0000 @@ -38,6 +38,7 @@ Table of Contents 2.8 /proc/sys/net/ipv4 - IPV4 settings 2.9 Appletalk 2.10 IPX + 2.11 /proc/sys/fs/mqueue - POSIX message queues filesystem ------------------------------------------------------------------------------ Preface @@ -1814,6 +1815,30 @@ The /proc/net/ipx_route table holds a gives the destination network, the router node (or Directly) and the network address of the router (or Connected) for internal networks. +2.11 /proc/sys/fs/mqueue - POSIX message queues filesystem +---------------------------------------------------------- + +The "mqueue" filesystem provides the necessary kernel features to enable the +creation of a user space library that implements the POSIX message queues +API (as noted by the MSG tag in the POSIX 1003.1-2001 version of the System +Interfaces specification.) + +The "mqueue" filesystem contains values for determining/setting the amount of +resources used by the file system. + +/proc/sys/fs/mqueue/queues_max is a read/write file for setting/getting the +maximum number of message queues allowed on the system. + +/proc/sys/fs/mqueue/msg_max is a read/write file for setting/getting the +maximum number of messages in a queue value. In fact it is the limiting value +for another (user) limit which is set in mq_open invocation. This attribute of +a queue must be less or equal then msg_max. + +/proc/sys/fs/mqueue/msgsize_max is a read/write file for setting/getting the +maximum message size value (it is every message queue's attribute set during +its creation). + + ------------------------------------------------------------------------------ Summary ------------------------------------------------------------------------------ --- diff/Documentation/filesystems/ufs.txt 2002-10-16 04:28:31.000000000 +0100 +++ source/Documentation/filesystems/ufs.txt 2004-03-01 15:42:14.000000000 +0000 @@ -20,6 +20,9 @@ ufstype=type_of_ufs 44bsd used in FreeBSD, NetBSD, OpenBSD supported os read-write + ufs2 used in FreeBSD 5.x + supported os read-only + sun used in SunOS (Solaris) supported as read-write --- diff/Documentation/kernel-parameters.txt 2004-03-01 14:36:04.000000000 +0000 +++ source/Documentation/kernel-parameters.txt 2004-03-01 15:42:14.000000000 +0000 @@ -90,10 +90,13 @@ running once the system is up. Format: , default is 13 acpi= [HW,ACPI] Advanced Configuration and Power Interface - Format: { force | off | ht } + Format: { force | off | ht | strict } force -- enables ACPI for systems with default off off -- disabled ACPI for systems with default on ht -- run only enough ACPI to enable Hyper Threading + strict -- Be less tolerant of platforms that are not + strictly ACPI specification compliant. + See also Documentation/pm.txt. acpi_pic_sci= [HW,ACPI] ACPI System Control Interrupt trigger mode --- diff/Documentation/networking/arcnet.txt 2003-08-20 14:16:23.000000000 +0100 +++ source/Documentation/networking/arcnet.txt 2004-03-01 15:42:14.000000000 +0000 @@ -186,7 +186,6 @@ loadable module. You can also say 'y' to to the chipset support if you wish. make config - make dep make clean make zImage make modules --- diff/Documentation/sound/alsa/ALSA-Configuration.txt 2004-02-18 08:54:06.000000000 +0000 +++ source/Documentation/sound/alsa/ALSA-Configuration.txt 2004-03-01 15:42:14.000000000 +0000 @@ -692,6 +692,15 @@ Module parameters The power-management is supported. + Module snd-mixart + ----------------- + + Module for Digigram miXart8 soundcards. + + Module supports multiple cards. + Note: One miXart8 board will be represented as 4 alsa cards. + See MIXART.txt for details. + Module snd-mpu401 ----------------- @@ -1170,6 +1179,16 @@ Module parameters The power-management is supported. + Module snd-pdaudiocf + -------------------- + + Module for Sound Core PDAudioCF soundcard. + + irq_mask - IRQ mask (PCMCIA type) + irq_list - List of available interrupts for this soundcard + + Note: the driver is build only when CONFIG_ISA is set. + Configuring Non-ISAPNP Cards ============================ --- diff/Documentation/sound/alsa/DocBook/writing-an-alsa-driver.tmpl 2004-02-18 08:54:06.000000000 +0000 +++ source/Documentation/sound/alsa/DocBook/writing-an-alsa-driver.tmpl 2004-03-01 15:42:14.000000000 +0000 @@ -3769,14 +3769,20 @@ struct _snd_pcm_runtime { value.enumerated.items = 4; - if (uinfo->value.enumerated.item > 3) - uinfo->value.enumerated.item = 3; - strcpy(uinfo->value.enumerated.name, - texts[uinfo->value.enumerated.item]); + snd_ctl_elem_info_t *uinfo) + { + static char *texts[4] = { + "First", "Second", "Third", "Fourth" + }; + uinfo->type = SNDRV_CTL_ELEM_TYPE_ENUMERATED; + uinfo->count = 1; + uinfo->value.enumerated.items = 4; + if (uinfo->value.enumerated.item > 3) + uinfo->value.enumerated.item = 3; + strcpy(uinfo->value.enumerated.name, + texts[uinfo->value.enumerated.item]); + return 0; + } ]]> --- diff/Documentation/sound/alsa/Joystick.txt 2004-02-18 08:54:06.000000000 +0000 +++ source/Documentation/sound/alsa/Joystick.txt 2004-03-01 15:42:14.000000000 +0000 @@ -43,7 +43,8 @@ The following PCI drivers support the jo ens1370 joystick 0 = disable (default), 1 = enable ens1371 joystick_port 0 = disable (default), 1 = auto-detect, manual: 0x200, 0x208, 0x210, 0x218 - cmipci joystick 0 = disable (default), 1 = enable + cmipci joystick_port 0 = disable (default), 1 = auto-detect, + manual: any address (e.g. 0x200) cs4281 N/A N/A cs46xx N/A N/A es1938 N/A N/A --- diff/Documentation/usb/error-codes.txt 2002-10-16 04:27:08.000000000 +0100 +++ source/Documentation/usb/error-codes.txt 2004-03-01 15:42:14.000000000 +0000 @@ -70,7 +70,9 @@ one or more packets could finish before (That is, if drivers see this it's a bug.) -EPROTO (*) a) bitstuff error - b) unknown USB error + b) no response packet received within the + prescribed bus turn-around time + c) unknown USB error -EILSEQ (*) CRC mismatch --- diff/MAINTAINERS 2004-03-01 14:36:04.000000000 +0000 +++ source/MAINTAINERS 2004-03-01 15:42:14.000000000 +0000 @@ -172,7 +172,7 @@ ACPI P: Len Brown M: len.brown@intel.com L: acpi-devel@lists.sourceforge.net -W: http://sf.net/projects/acpi/ +W: http://acpi.sourceforge.net/ S: Maintained AD1816 SOUND DRIVER @@ -1186,6 +1186,12 @@ W: http://sf.net/projects/kernel-janitor W: http://developer.osdl.org/rddunlap/kj-patches/ S: Maintained +KGDB FOR I386 PLATFORM +P: George Anzinger +M: george@mvista.com +L: linux-net@vger.kernel.org +S: Supported + KERNEL NFSD P: Neil Brown M: neilb@cse.unsw.edu.au @@ -2395,7 +2401,7 @@ S: Maintained ZR36067 VIDEO FOR LINUX DRIVER P: Ronald Bultje -M: R.S.Bultje@pharm.uu.nl +M: rbultje@ronald.bitfreak.net L: mjpeg-users@lists.sourceforge.net W: http://mjpeg.sourceforge.net/driver-zoran/ S: Maintained --- diff/Makefile 2004-03-01 14:36:04.000000000 +0000 +++ source/Makefile 2004-03-01 15:42:14.000000000 +0000 @@ -1,7 +1,7 @@ VERSION = 2 PATCHLEVEL = 6 SUBLEVEL = 4 -EXTRAVERSION =-rc1 +EXTRAVERSION =-rc1-mm1 NAME=Feisty Dunnart # *DOCUMENTATION* @@ -444,6 +444,7 @@ endif ifdef CONFIG_DEBUG_INFO CFLAGS += -g +AFLAGS += -g endif # warn about C99 declaration after statement --- diff/README 2004-01-19 10:22:54.000000000 +0000 +++ source/README 2004-03-01 15:42:14.000000000 +0000 @@ -172,24 +172,15 @@ COMPILING the kernel: Please note that you can still run a.out user programs with this kernel. - - Do a "make bzImage" to create a compressed kernel image. If you want - to make a boot disk (without root filesystem or LILO), insert a floppy - in your A: drive, and do a "make bzdisk". It is also possible to do - "make install" if you have lilo installed to suit the kernel makefiles, - but you may want to check your particular lilo setup first. + - Do a "make" to create a compressed kernel image. It is also + possible to do "make install" if you have lilo installed to suit the + kernel makefiles, but you may want to check your particular lilo setup first. To do the actual install you have to be root, but none of the normal build should require that. Don't take the name of root in vain. - - In the unlikely event that your system cannot boot bzImage kernels you - can still compile your kernel as zImage. However, since zImage support - will be removed at some point in the future in favor of bzImage we - encourage people having problems with booting bzImage kernels to report - these, with detailed hardware configuration information, to the - linux-kernel mailing list and to H. Peter Anvin . - - If you configured any of the parts of the kernel as `modules', you - will have to do "make modules" followed by "make modules_install". + will also have to do "make modules_install". - Keep a backup kernel handy in case something goes wrong. This is especially true for the development releases, since each new release @@ -200,11 +191,11 @@ COMPILING the kernel: do a "make modules_install". - In order to boot your new kernel, you'll need to copy the kernel - image (found in .../linux/arch/i386/boot/bzImage after compilation) + image (e.g. .../linux/arch/i386/boot/bzImage after compilation) to the place where your regular bootable kernel is found. - For some, this is on a floppy disk, in which case you can copy the - kernel bzImage file to /dev/fd0 to make a bootable floppy. + - Booting a kernel directly from a floppy without the assistance of a + bootloader such as LILO, is no longer supported. If you boot Linux from the hard drive, chances are you use LILO which uses the kernel image as specified in the file /etc/lilo.conf. The @@ -262,8 +253,9 @@ IF SOMETHING GOES WRONG: the above example it's due to a bad kernel pointer). More information on making sense of the dump is in Documentation/oops-tracing.txt - - You can use the "ksymoops" program to make sense of the dump. This - utility can be downloaded from + - If you compiled the kernel with CONFIG_KALLSYMS you can send the dump + as is, otherwise you will have to use the "ksymoops" program to make + sense of the dump. This utility can be downloaded from ftp://ftp..kernel.org/pub/linux/utils/kernel/ksymoops. Alternately you can do the dump lookup by hand: --- diff/arch/alpha/kernel/alpha_ksyms.c 2004-03-01 14:36:04.000000000 +0000 +++ source/arch/alpha/kernel/alpha_ksyms.c 2004-03-01 15:42:14.000000000 +0000 @@ -35,9 +35,6 @@ #include #include -#define __KERNEL_SYSCALLS__ -#include - extern struct hwrpb_struct *hwrpb; extern void dump_thread(struct pt_regs *, struct user *); extern spinlock_t rtc_lock; --- diff/arch/alpha/kernel/smp.c 2003-10-09 09:47:33.000000000 +0100 +++ source/arch/alpha/kernel/smp.c 2004-03-01 15:42:14.000000000 +0000 @@ -39,9 +39,6 @@ #include #include -#define __KERNEL_SYSCALLS__ -#include - #include "proto.h" #include "irq_impl.h" --- diff/arch/arm/common/sa1111-pcibuf.c 2004-03-01 14:36:04.000000000 +0000 +++ source/arch/arm/common/sa1111-pcibuf.c 2004-03-01 15:42:14.000000000 +0000 @@ -457,8 +457,8 @@ void sa1111_unmap_sg(struct device *dev, local_irq_restore(flags); } -void sa1111_dma_sync_single(struct device *dev, dma_addr_t dma_addr, - size_t size, enum dma_data_direction dir) +void sa1111_dma_sync_single_for_cpu(struct device *dev, dma_addr_t dma_addr, + size_t size, enum dma_data_direction dir) { unsigned long flags; @@ -472,8 +472,44 @@ void sa1111_dma_sync_single(struct devic local_irq_restore(flags); } -void sa1111_dma_sync_sg(struct device *dev, struct scatterlist *sg, - int nents, enum dma_data_direction dir) +void sa1111_dma_sync_single_for_device(struct device *dev, dma_addr_t dma_addr, + size_t size, enum dma_data_direction dir) +{ + unsigned long flags; + + dev_dbg(dev, "%s(ptr=%08lx,size=%d,dir=%x)\n", + __func__, dma_addr, size, dir); + + local_irq_save(flags); + + sync_single(dev, dma_addr, size, dir); + + local_irq_restore(flags); +} + +void sa1111_dma_sync_sg_for_cpu(struct device *dev, struct scatterlist *sg, + int nents, enum dma_data_direction dir) +{ + unsigned long flags; + int i; + + dev_dbg(dev, "%s(sg=%p,nents=%d,dir=%x)\n", + __func__, sg, nents, dir); + + local_irq_save(flags); + + for (i = 0; i < nents; i++, sg++) { + dma_addr_t dma_addr = sg->dma_address; + unsigned int length = sg->length; + + sync_single(dev, dma_addr, length, dir); + } + + local_irq_restore(flags); +} + +void sa1111_dma_sync_sg_for_device(struct device *dev, struct scatterlist *sg, + int nents, enum dma_data_direction dir) { unsigned long flags; int i; @@ -497,8 +533,10 @@ EXPORT_SYMBOL(sa1111_map_single); EXPORT_SYMBOL(sa1111_unmap_single); EXPORT_SYMBOL(sa1111_map_sg); EXPORT_SYMBOL(sa1111_unmap_sg); -EXPORT_SYMBOL(sa1111_dma_sync_single); -EXPORT_SYMBOL(sa1111_dma_sync_sg); +EXPORT_SYMBOL(sa1111_dma_sync_single_for_cpu); +EXPORT_SYMBOL(sa1111_dma_sync_single_for_device); +EXPORT_SYMBOL(sa1111_dma_sync_sg_for_cpu); +EXPORT_SYMBOL(sa1111_dma_sync_sg_for_device); /* **************************************** */ --- diff/arch/arm/kernel/Makefile 2004-03-01 14:36:04.000000000 +0000 +++ source/arch/arm/kernel/Makefile 2004-03-01 15:42:14.000000000 +0000 @@ -29,7 +29,7 @@ obj-$(CONFIG_DEBUG_LL) += debug.o extra-y := $(head-y) init_task.o vmlinux.lds.s -# Spell out some dependencies that `make dep' doesn't spot +# Spell out some dependencies that aren't automatically figured out $(obj)/entry-armv.o: $(obj)/entry-header.S include/asm-arm/constants.h $(obj)/entry-common.o: $(obj)/entry-header.S include/asm-arm/constants.h \ $(obj)/calls.S --- diff/arch/arm/kernel/time.c 2004-03-01 14:36:04.000000000 +0000 +++ source/arch/arm/kernel/time.c 2004-03-01 15:42:14.000000000 +0000 @@ -178,7 +178,7 @@ static int __init leds_init(void) int ret; ret = sysdev_class_register(&leds_sysclass); if (ret == 0) - ret = sys_device_register(&leds_device); + ret = sysdev_register(&leds_device); return ret; } --- diff/arch/arm/mach-integrator/integrator_ap.c 2004-02-18 08:54:06.000000000 +0000 +++ source/arch/arm/mach-integrator/integrator_ap.c 2004-03-01 15:42:14.000000000 +0000 @@ -173,7 +173,7 @@ static int __init irq_init_sysfs(void) { int ret = sysdev_class_register(&irq_class); if (ret == 0) - ret = sys_device_register(&irq_device); + ret = sysdev_register(&irq_device); return ret; } --- diff/arch/arm/mach-sa1100/irq.c 2003-06-30 10:07:32.000000000 +0100 +++ source/arch/arm/mach-sa1100/irq.c 2004-03-01 15:42:14.000000000 +0000 @@ -278,7 +278,7 @@ static struct sys_device sa1100irq_devic static int __init sa1100irq_init_devicefs(void) { sysdev_class_register(&sa1100irq_sysclass); - return sys_device_register(&sa1100irq_device); + return sysdev_register(&sa1100irq_device); } device_initcall(sa1100irq_init_devicefs); --- diff/arch/arm26/mm/Makefile 2003-06-30 10:07:18.000000000 +0100 +++ source/arch/arm26/mm/Makefile 2004-03-01 15:42:14.000000000 +0000 @@ -1,12 +1,5 @@ # # Makefile for the linux arm26-specific parts of the memory manager. # -# Note! Dependencies are done automagically by 'make dep', which also -# removes any old dependencies. DON'T put your own dependencies here -# unless it's something special (ie not a .c file). -# -# Note 2! The CFLAGS definition is now in the main makefile... - -# Object file lists. obj-y := init.o extable.o proc-funcs.o mm-memc.o fault.o --- diff/arch/cris/kernel/process.c 2003-10-09 09:47:33.000000000 +0100 +++ source/arch/cris/kernel/process.c 2004-03-01 15:42:14.000000000 +0000 @@ -91,8 +91,6 @@ * This file handles the architecture-dependent parts of process handling.. */ -#define __KERNEL_SYSCALLS__ - #include #include #include --- diff/arch/h8300/kernel/sys_h8300.c 2004-03-01 14:36:04.000000000 +0000 +++ source/arch/h8300/kernel/sys_h8300.c 2004-03-01 15:42:14.000000000 +0000 @@ -260,11 +260,6 @@ asmlinkage int sys_ipc (uint call, int f return -EINVAL; } -asmlinkage long sys_ioperm(unsigned long from, unsigned long num, int on) -{ - return -ENOSYS; -} - /* sys_cacheflush -- no support. */ asmlinkage int sys_cacheflush (unsigned long addr, int scope, int cache, unsigned long len) --- diff/arch/h8300/kernel/syscalls.S 2003-08-26 10:00:51.000000000 +0100 +++ source/arch/h8300/kernel/syscalls.S 2004-03-01 15:42:14.000000000 +0000 @@ -116,7 +116,7 @@ SYMBOL_NAME_LABEL(sys_call_table) .long SYMBOL_NAME(sys_ni_syscall) /* old profil syscall holder */ .long SYMBOL_NAME(sys_statfs) .long SYMBOL_NAME(sys_fstatfs) /* 100 */ - .long SYMBOL_NAME(sys_ioperm) + .long SYMBOL_NAME(sys_ni_syscall) /* ioperm for i386 */ .long SYMBOL_NAME(sys_socketcall) .long SYMBOL_NAME(sys_syslog) .long SYMBOL_NAME(sys_setitimer) --- diff/arch/h8300/mm/Makefile 2004-02-09 10:36:07.000000000 +0000 +++ source/arch/h8300/mm/Makefile 2004-03-01 15:42:14.000000000 +0000 @@ -1,10 +1,5 @@ # # Makefile for the linux m68k-specific parts of the memory manager. # -# Note! Dependencies are done automagically by 'make dep', which also -# removes any old dependencies. DON'T put your own dependencies here -# unless it's something special (ie not a .c file). -# -# Note 2! The CFLAGS definition is now in the main makefile... obj-y := init.o fault.o memory.o kmap.o --- diff/arch/h8300/platform/h8300h/Makefile 2003-05-21 11:50:08.000000000 +0100 +++ source/arch/h8300/platform/h8300h/Makefile 2004-03-01 15:42:14.000000000 +0000 @@ -6,10 +6,6 @@ #VPATH := $(VPATH):$(BOARD) -# Note! Dependencies are done automagically by 'make dep', which also -# removes any old dependencies. DON'T put your own dependencies here -# unless it's something special (ie not a .c file). -# .S.o: $(CC) -D__ASSEMBLY__ $(AFLAGS) -I. -c $< -o $*.o --- diff/arch/h8300/platform/h8300h/aki3068net/Makefile 2003-08-20 14:16:24.000000000 +0100 +++ source/arch/h8300/platform/h8300h/aki3068net/Makefile 2004-03-01 15:42:14.000000000 +0000 @@ -1,10 +1,6 @@ # # Makefile for the linux kernel. # -# Note! Dependencies are done automagically by 'make dep', which also -# removes any old dependencies. DON'T put your own dependencies here -# unless it's something special (ie not a .c file). -# extra-y := crt0_ram.o obj-y := timer.o --- diff/arch/h8300/platform/h8300h/generic/Makefile 2003-08-20 14:16:24.000000000 +0100 +++ source/arch/h8300/platform/h8300h/generic/Makefile 2004-03-01 15:42:14.000000000 +0000 @@ -1,10 +1,6 @@ # # Makefile for the linux kernel. # -# Note! Dependencies are done automagically by 'make dep', which also -# removes any old dependencies. DON'T put your own dependencies here -# unless it's something special (ie not a .c file). -# obj-y := timer.o extra-y = crt0_$(MODEL).o --- diff/arch/h8300/platform/h8300h/h8max/Makefile 2003-08-20 14:16:24.000000000 +0100 +++ source/arch/h8300/platform/h8300h/h8max/Makefile 2004-03-01 15:42:14.000000000 +0000 @@ -1,10 +1,6 @@ # # Makefile for the linux kernel. # -# Note! Dependencies are done automagically by 'make dep', which also -# removes any old dependencies. DON'T put your own dependencies here -# unless it's something special (ie not a .c file). -# extra-y := crt0_ram.o obj-y := timer.o --- diff/arch/h8300/platform/h8s/Makefile 2003-08-20 14:16:36.000000000 +0100 +++ source/arch/h8300/platform/h8s/Makefile 2004-03-01 15:42:14.000000000 +0000 @@ -6,10 +6,6 @@ #VPATH := $(VPATH):$(BOARD) -# Note! Dependencies are done automagically by 'make dep', which also -# removes any old dependencies. DON'T put your own dependencies here -# unless it's something special (ie not a .c file). -# .S.o: $(CC) -D__ASSEMBLY__ $(AFLAGS) -I. -c $< -o $*.o --- diff/arch/h8300/platform/h8s/edosk2674/Makefile 2003-08-20 14:16:36.000000000 +0100 +++ source/arch/h8300/platform/h8s/edosk2674/Makefile 2004-03-01 15:42:14.000000000 +0000 @@ -1,10 +1,6 @@ # # Makefile for the linux kernel. # -# Note! Dependencies are done automagically by 'make dep', which also -# removes any old dependencies. DON'T put your own dependencies here -# unless it's something special (ie not a .c file). -# extra-y := crt0_ram.o obj-y := timer.o --- diff/arch/h8300/platform/h8s/generic/Makefile 2003-08-20 14:16:36.000000000 +0100 +++ source/arch/h8300/platform/h8s/generic/Makefile 2004-03-01 15:42:14.000000000 +0000 @@ -1,10 +1,6 @@ # # Makefile for the linux kernel. # -# Note! Dependencies are done automagically by 'make dep', which also -# removes any old dependencies. DON'T put your own dependencies here -# unless it's something special (ie not a .c file). -# extra-y = crt0_$(MODEL).o obj-y := timer.o --- diff/arch/i386/Kconfig 2004-03-01 14:36:04.000000000 +0000 +++ source/arch/i386/Kconfig 2004-03-01 15:42:14.000000000 +0000 @@ -421,6 +421,54 @@ config X86_OOSTORE depends on (MWINCHIP3D || MWINCHIP2 || MWINCHIPC6) && MTRR default y +config X86_4G + bool "4 GB kernel-space and 4 GB user-space virtual memory support" + help + This option is only useful for systems that have more than 1 GB + of RAM. + + The default kernel VM layout leaves 1 GB of virtual memory for + kernel-space mappings, and 3 GB of VM for user-space applications. + This option ups both the kernel-space VM and the user-space VM to + 4 GB. + + The cost of this option is additional TLB flushes done at + system-entry points that transition from user-mode into kernel-mode. + I.e. system calls and page faults, and IRQs that interrupt user-mode + code. There's also additional overhead to kernel operations that copy + memory to/from user-space. The overhead from this is hard to tell and + depends on the workload - it can be anything from no visible overhead + to 20-30% overhead. A good rule of thumb is to count with a runtime + overhead of 20%. + + The upside is the much increased kernel-space VM, which more than + quadruples the maximum amount of RAM supported. Kernels compiled with + this option boot on 64GB of RAM and still have more than 3.1 GB of + 'lowmem' left. Another bonus is that highmem IO bouncing decreases, + if used with drivers that still use bounce-buffers. + + There's also a 33% increase in user-space VM size - database + applications might see a boost from this. + + But the cost of the TLB flushes and the runtime overhead has to be + weighed against the bonuses offered by the larger VM spaces. The + dividing line depends on the actual workload - there might be 4 GB + systems that benefit from this option. Systems with less than 4 GB + of RAM will rarely see a benefit from this option - but it's not + out of question, the exact circumstances have to be considered. + +config X86_SWITCH_PAGETABLES + def_bool X86_4G + +config X86_4G_VM_LAYOUT + def_bool X86_4G + +config X86_UACCESS_INDIRECT + def_bool X86_4G + +config X86_HIGH_ENTRY + def_bool X86_4G + config HPET_TIMER bool "HPET Timer Support" help @@ -478,6 +526,16 @@ config NR_CPUS This is purely to save memory - each supported CPU adds approximately eight kilobytes to the kernel image. +config SCHED_SMT + bool "SMT (Hyperthreading) scheduler support" + depends on SMP + default off + help + SMT scheduler support improves the CPU scheduler's decision making + when dealing with Intel Pentium 4 chips with HyperThreading at a + cost of slightly increased overhead in some places. If unsure say + N here. + config PREEMPT bool "Preemptible Kernel" help @@ -552,7 +610,7 @@ config X86_MCE the 386 and 486, so nearly everyone can say Y here. config X86_MCE_NONFATAL - bool "Check for non-fatal errors on AMD Athlon/Duron / Intel Pentium 4" + tristate "Check for non-fatal errors on AMD Athlon/Duron / Intel Pentium 4" depends on X86_MCE help Enabling this feature starts a timer that triggers every 5 seconds which @@ -1068,12 +1126,16 @@ config PCI_GOBIOS PCI-based systems don't have any BIOS at all. Linux can also try to detect the PCI hardware directly without using the BIOS. - With this option, you can specify how Linux should detect the PCI - devices. If you choose "BIOS", the BIOS will be used, if you choose - "Direct", the BIOS won't be used, and if you choose "Any", the - kernel will try the direct access method and falls back to the BIOS - if that doesn't work. If unsure, go with the default, which is - "Any". + With this option, you can specify how Linux should detect the + PCI devices. If you choose "BIOS", the BIOS will be used, + if you choose "Direct", the BIOS won't be used, and if you + choose "MMConfig", then PCI Express MMCONFIG will be used. + If you choose "Any", the kernel will try MMCONFIG, then the + direct access method and falls back to the BIOS if that doesn't + work. If unsure, go with the default, which is "Any". + +config PCI_GOMMCONFIG + bool "MMConfig" config PCI_GODIRECT bool "Direct" @@ -1093,6 +1155,12 @@ config PCI_DIRECT depends on PCI && ((PCI_GODIRECT || PCI_GOANY) || X86_VISWS) default y +config PCI_MMCONFIG + bool + depends on PCI && (PCI_GOMMCONFIG || PCI_GOANY) + select ACPI_BOOT + default y + config PCI_USE_VECTOR bool "Vector-based interrupt indexing (MSI)" depends on X86_LOCAL_APIC && X86_IO_APIC @@ -1231,17 +1299,6 @@ config DEBUG_SLAB allocation as well as poisoning memory on free to catch use of freed memory. -config DEBUG_IOVIRT - bool "Memory mapped I/O debugging" - depends on DEBUG_KERNEL - help - Say Y here to get warned whenever an attempt is made to do I/O on - obviously invalid addresses such as those generated when ioremap() - calls are forgotten. Memory mapped I/O will go through an extra - check to catch access to unmapped ISA addresses, an access method - that can still be used by old drivers that are being ported from - 2.0/2.2. - config MAGIC_SYSRQ bool "Magic SysRq key" depends on DEBUG_KERNEL @@ -1273,6 +1330,15 @@ config DEBUG_PAGEALLOC This results in a large slowdown, but helps to find certain types of memory corruptions. +config SPINLINE + bool "Spinlock inlining" + depends on DEBUG_KERNEL + help + This will change spinlocks from out of line to inline, making them + account cost to the callers in readprofile, rather than the lock + itself (as ".text.lock.filename"). This can be helpful for finding + the callers of locks. + config DEBUG_HIGHMEM bool "Highmem debugging" depends on DEBUG_KERNEL && HIGHMEM @@ -1289,20 +1355,208 @@ config DEBUG_INFO Say Y here only if you plan to use gdb to debug the kernel. If you don't debug the kernel, you can say N. +config LOCKMETER + bool "Kernel lock metering" + depends on SMP + help + Say Y to enable kernel lock metering, which adds overhead to SMP locks, + but allows you to see various statistics using the lockstat command. + config DEBUG_SPINLOCK_SLEEP bool "Sleep-inside-spinlock checking" help If you say Y here, various routines which may sleep will become very noisy if they are called with a spinlock held. +config KGDB + bool "Include kgdb kernel debugger" + depends on DEBUG_KERNEL + help + If you say Y here, the system will be compiled with the debug + option (-g) and a debugging stub will be included in the + kernel. This stub communicates with gdb on another (host) + computer via a serial port. The host computer should have + access to the kernel binary file (vmlinux) and a serial port + that is connected to the target machine. Gdb can be made to + configure the serial port or you can use stty and setserial to + do this. See the 'target' command in gdb. This option also + configures in the ability to request a breakpoint early in the + boot process. To request the breakpoint just include 'kgdb' + as a boot option when booting the target machine. The system + will then break as soon as it looks at the boot options. This + option also installs a breakpoint in panic and sends any + kernel faults to the debugger. For more information see the + Documentation/i386/kgdb/kgdb.txt file. + +choice + depends on KGDB + prompt "Debug serial port BAUD" + default KGDB_115200BAUD + help + Gdb and the kernel stub need to agree on the baud rate to be + used. Some systems (x86 family at this writing) allow this to + be configured. + +config KGDB_9600BAUD + bool "9600" + +config KGDB_19200BAUD + bool "19200" + +config KGDB_38400BAUD + bool "38400" + +config KGDB_57600BAUD + bool "57600" + +config KGDB_115200BAUD + bool "115200" +endchoice + +config KGDB_PORT + hex "hex I/O port address of the debug serial port" + depends on KGDB + default 3f8 + help + Some systems (x86 family at this writing) allow the port + address to be configured. The number entered is assumed to be + hex, don't put 0x in front of it. The standard address are: + COM1 3f8 , irq 4 and COM2 2f8 irq 3. Setserial /dev/ttySx + will tell you what you have. It is good to test the serial + connection with a live system before trying to debug. + +config KGDB_IRQ + int "IRQ of the debug serial port" + depends on KGDB + default 4 + help + This is the irq for the debug port. If everything is working + correctly and the kernel has interrupts on a control C to the + port should cause a break into the kernel debug stub. + +config DEBUG_INFO + bool + depends on KGDB + default y + +config KGDB_MORE + bool "Add any additional compile options" + depends on KGDB + default n + help + Saying yes here turns on the ability to enter additional + compile options. + + +config KGDB_OPTIONS + depends on KGDB_MORE + string "Additional compile arguments" + default "-O1" + help + This option allows you enter additional compile options for + the whole kernel compile. Each platform will have a default + that seems right for it. For example on PPC "-ggdb -O1", and + for i386 "-O1". Note that by configuring KGDB "-g" is already + turned on. In addition, on i386 platforms + "-fomit-frame-pointer" is deleted from the standard compile + options. + +config NO_KGDB_CPUS + int "Number of CPUs" + depends on KGDB && SMP + default NR_CPUS + help + + This option sets the number of cpus for kgdb ONLY. It is used + to prune some internal structures so they look "nice" when + displayed with gdb. This is to overcome possibly larger + numbers that may have been entered above. Enter the real + number to get nice clean kgdb_info displays. + +config KGDB_TS + bool "Enable kgdb time stamp macros?" + depends on KGDB + default n + help + Kgdb event macros allow you to instrument your code with calls + to the kgdb event recording function. The event log may be + examined with gdb at a break point. Turning on this + capability also allows you to choose how many events to + keep. Kgdb always keeps the lastest events. + +choice + depends on KGDB_TS + prompt "Max number of time stamps to save?" + default KGDB_TS_128 + +config KGDB_TS_64 + bool "64" + +config KGDB_TS_128 + bool "128" + +config KGDB_TS_256 + bool "256" + +config KGDB_TS_512 + bool "512" + +config KGDB_TS_1024 + bool "1024" + +endchoice + +config STACK_OVERFLOW_TEST + bool "Turn on kernel stack overflow testing?" + depends on KGDB + default n + help + This option enables code in the front line interrupt handlers + to check for kernel stack overflow on interrupts and system + calls. This is part of the kgdb code on x86 systems. + +config KGDB_CONSOLE + bool "Enable serial console thru kgdb port" + depends on KGDB + default n + help + This option enables the command line "console=kgdb" option. + When the system is booted with this option in the command line + all kernel printk output is sent to gdb (as well as to other + consoles). For this to work gdb must be connected. For this + reason, this command line option will generate a breakpoint if + gdb has not yet connected. After the gdb continue command is + given all pent up console output will be printed by gdb on the + host machine. Neither this option, nor KGDB require the + serial driver to be configured. + +config KGDB_SYSRQ + bool "Turn on SysRq 'G' command to do a break?" + depends on KGDB + default y + help + This option includes an option in the SysRq code that allows + you to enter SysRq G which generates a breakpoint to the KGDB + stub. This will work if the keyboard is alive and can + interrupt the system. Because of constraints on when the + serial port interrupt can be enabled, this code may allow you + to interrupt the system before the serial port control C is + available. Just say yes here. + config FRAME_POINTER bool "Compile the kernel with frame pointers" + default KGDB help If you say Y here the resulting kernel image will be slightly larger and slower, but it will give very useful debugging information. If you don't debug the kernel, you can say N, but we may not be able to solve problems without frame pointers. +config MAGIC_SYSRQ + bool + depends on KGDB_SYSRQ + default y + config X86_FIND_SMP_CONFIG bool depends on X86_LOCAL_APIC || X86_VOYAGER @@ -1336,11 +1590,6 @@ config X86_BIOS_REBOOT depends on !(X86_VISWS || X86_VOYAGER) default y -config X86_TRAMPOLINE - bool - depends on SMP || X86_VISWS - default y - config PC bool depends on X86 && !EMBEDDED --- diff/arch/i386/Makefile 2004-03-01 14:36:04.000000000 +0000 +++ source/arch/i386/Makefile 2004-03-01 15:42:14.000000000 +0000 @@ -19,7 +19,7 @@ LDFLAGS := -m elf_i386 OBJCOPYFLAGS := -O binary -R .note -R .comment -S LDFLAGS_vmlinux := -CFLAGS += -pipe +CFLAGS += -pipe -msoft-float # prevent gcc from keeping the stack 16 byte aligned CFLAGS += $(call check_gcc,-mpreferred-stack-boundary=2,) @@ -97,6 +97,9 @@ mcore-$(CONFIG_X86_ES7000) := mach-es700 # default subarch .h files mflags-y += -Iinclude/asm-i386/mach-default +mflags-$(CONFIG_KGDB) += -gdwarf-2 +mflags-$(CONFIG_KGDB_MORE) += $(shell echo $(CONFIG_KGDB_OPTIONS) | sed -e 's/"//g') + head-y := arch/i386/kernel/head.o arch/i386/kernel/init_task.o libs-y += arch/i386/lib/ --- diff/arch/i386/boot/setup.S 2004-03-01 14:36:04.000000000 +0000 +++ source/arch/i386/boot/setup.S 2004-03-01 15:42:14.000000000 +0000 @@ -164,7 +164,7 @@ cmd_line_ptr: .long 0 # (Header versio # can be located anywhere in # low memory 0x10000 or higher. -ramdisk_max: .long MAXMEM-1 # (Header version 0x0203 or later) +ramdisk_max: .long __MAXMEM-1 # (Header version 0x0203 or later) # The highest safe address for # the contents of an initrd --- diff/arch/i386/boot/tools/build.c 2003-05-21 11:49:49.000000000 +0100 +++ source/arch/i386/boot/tools/build.c 2004-03-01 15:42:14.000000000 +0000 @@ -150,10 +150,8 @@ int main(int argc, char ** argv) sz = sb.st_size; fprintf (stderr, "System is %d kB\n", sz/1024); sys_size = (sz + 15) / 16; - /* 0x40000*16 = 4.0 MB, reasonable estimate for the current maximum */ - if (sys_size > (is_big_kernel ? 0x40000 : DEF_SYSSIZE)) - die("System is too big. Try using %smodules.", - is_big_kernel ? "" : "bzImage or "); + if (!is_big_kernel && sys_size > DEF_SYSSIZE) + die("System is too big. Try using bzImage or modules."); while (sz > 0) { int l, n; --- diff/arch/i386/defconfig 2004-02-18 08:54:07.000000000 +0000 +++ source/arch/i386/defconfig 2004-03-01 15:42:14.000000000 +0000 @@ -1195,5 +1195,4 @@ CONFIG_CRC32=y CONFIG_X86_SMP=y CONFIG_X86_HT=y CONFIG_X86_BIOS_REBOOT=y -CONFIG_X86_TRAMPOLINE=y CONFIG_PC=y --- diff/arch/i386/kernel/Makefile 2004-03-01 14:36:04.000000000 +0000 +++ source/arch/i386/kernel/Makefile 2004-03-01 15:42:14.000000000 +0000 @@ -7,19 +7,19 @@ extra-y := head.o init_task.o vmlinux.ld obj-y := process.o semaphore.o signal.o entry.o traps.o irq.o vm86.o \ ptrace.o i8259.o ioport.o ldt.o setup.o time.o sys_i386.o \ pci-dma.o i386_ksyms.o i387.o dmi_scan.o bootflag.o \ - doublefault.o + doublefault.o entry_trampoline.o obj-y += cpu/ obj-y += timers/ obj-$(CONFIG_ACPI_BOOT) += acpi/ obj-$(CONFIG_X86_BIOS_REBOOT) += reboot.o obj-$(CONFIG_MCA) += mca.o +obj-$(CONFIG_KGDB) += kgdb_stub.o obj-$(CONFIG_X86_MSR) += msr.o obj-$(CONFIG_X86_CPUID) += cpuid.o obj-$(CONFIG_MICROCODE) += microcode.o obj-$(CONFIG_APM) += apm.o -obj-$(CONFIG_X86_SMP) += smp.o smpboot.o -obj-$(CONFIG_X86_TRAMPOLINE) += trampoline.o +obj-$(CONFIG_X86_SMP) += smp.o smpboot.o trampoline.o obj-$(CONFIG_X86_MPPARSE) += mpparse.o obj-$(CONFIG_X86_LOCAL_APIC) += apic.o nmi.o obj-$(CONFIG_X86_IO_APIC) += io_apic.o --- diff/arch/i386/kernel/acpi/boot.c 2004-03-01 14:36:04.000000000 +0000 +++ source/arch/i386/kernel/acpi/boot.c 2004-03-01 15:42:15.000000000 +0000 @@ -43,11 +43,12 @@ #define PREFIX "ACPI: " -int acpi_noirq __initdata = 0; /* skip ACPI IRQ initialization */ +int acpi_noirq __initdata; /* skip ACPI IRQ initialization */ int acpi_ht __initdata = 1; /* enable HT */ int acpi_lapic; int acpi_ioapic; +int acpi_strict; /* -------------------------------------------------------------------------- Boot-time Configuration @@ -96,6 +97,31 @@ char *__acpi_map_table(unsigned long phy } +#ifdef CONFIG_PCI_MMCONFIG +static int __init acpi_parse_mcfg(unsigned long phys_addr, unsigned long size) +{ + struct acpi_table_mcfg *mcfg; + + if (!phys_addr || !size) + return -EINVAL; + + mcfg = (struct acpi_table_mcfg *) __acpi_map_table(phys_addr, size); + if (!mcfg) { + printk(KERN_WARNING PREFIX "Unable to map MCFG\n"); + return -ENODEV; + } + + if (mcfg->base_reserved) { + printk(KERN_ERR PREFIX "MMCONFIG not in low 4GB of memory\n"); + return -ENODEV; + } + + pci_mmcfg_base_addr = mcfg->base_address; + + return 0; +} +#endif /* CONFIG_PCI_MMCONFIG */ + #ifdef CONFIG_X86_LOCAL_APIC static u64 acpi_lapic_addr __initdata = APIC_DEFAULT_PHYS_BASE; @@ -259,7 +285,7 @@ acpi_parse_nmi_src ( * programs the PIC-mode SCI to Level Trigger. * (NO-OP if the BIOS set Level Trigger already) * - * If a PIC-mode SCI is not recogznied or gives spurious IRQ7's + * If a PIC-mode SCI is not recognized or gives spurious IRQ7's * it may require Edge Trigger -- use "acpi_pic_sci=edge" * (NO-OP if the BIOS set Edge Trigger already) * @@ -339,7 +365,7 @@ acpi_scan_rsdp ( * RSDP signature. */ for (offset = 0; offset < length; offset += 16) { - if (strncmp((char *) (start + offset), "RSD PTR ", sig_len)) + if (strncmp((char *) __va(start + offset), "RSD PTR ", sig_len)) continue; return (start + offset); } @@ -429,55 +455,10 @@ acpi_find_rsdp (void) return rsdp_phys; } -/* - * acpi_boot_init() - * called from setup_arch(), always. - * 1. maps ACPI tables for later use - * 2. enumerates lapics - * 3. enumerates io-apics - * - * side effects: - * acpi_lapic = 1 if LAPIC found - * acpi_ioapic = 1 if IOAPIC found - * if (acpi_lapic && acpi_ioapic) smp_found_config = 1; - * if acpi_blacklisted() acpi_disabled = 1; - * acpi_irq_model=... - * ... - * - * return value: (currently ignored) - * 0: success - * !0: failure - */ -int __init -acpi_boot_init (void) +static int acpi_apic_setup(void) { - int result = 0; - - if (acpi_disabled && !acpi_ht) - return 1; - - /* - * The default interrupt routing model is PIC (8259). This gets - * overriden if IOAPICs are enumerated (below). - */ - acpi_irq_model = ACPI_IRQ_MODEL_PIC; - - /* - * Initialize the ACPI boot-time table parser. - */ - result = acpi_table_init(); - if (result) { - acpi_disabled = 1; - return result; - } - - result = acpi_blacklisted(); - if (result) { - printk(KERN_WARNING PREFIX "BIOS listed in blacklist, disabling ACPI support\n"); - acpi_disabled = 1; - return result; - } + int result; #ifdef CONFIG_X86_PM_TIMER acpi_table_parse(ACPI_FADT, acpi_parse_fadt); @@ -541,24 +522,17 @@ acpi_boot_init (void) acpi_lapic = 1; -#endif /*CONFIG_X86_LOCAL_APIC*/ +#endif /* CONFIG_X86_LOCAL_APIC */ #if defined(CONFIG_X86_IO_APIC) && defined(CONFIG_ACPI_INTERPRETER) /* * I/O APIC - * -------- */ - /* - * ACPI interpreter is required to complete interrupt setup, - * so if it is off, don't enumerate the io-apics with ACPI. - * If MPS is present, it will handle them, - * otherwise the system will stay in PIC mode - */ - if (acpi_disabled || acpi_noirq) { + if (acpi_noirq) { return 1; - } + } /* * if "noapic" boot option, don't look for IO-APICs @@ -573,8 +547,7 @@ acpi_boot_init (void) if (!result) { printk(KERN_ERR PREFIX "No IOAPIC entries present\n"); return -ENODEV; - } - else if (result < 0) { + } else if (result < 0) { printk(KERN_ERR PREFIX "Error parsing IOAPIC entry\n"); return result; } @@ -611,9 +584,82 @@ acpi_boot_init (void) } #endif + return 0; +} + +/* + * acpi_boot_init() + * called from setup_arch(), always. + * 1. maps ACPI tables for later use + * 2. enumerates lapics + * 3. enumerates io-apics + * + * side effects: + * acpi_lapic = 1 if LAPIC found + * acpi_ioapic = 1 if IOAPIC found + * if (acpi_lapic && acpi_ioapic) smp_found_config = 1; + * if acpi_blacklisted() acpi_disabled = 1; + * acpi_irq_model=... + * ... + * + * return value: (currently ignored) + * 0: success + * !0: failure + */ + +int __init +acpi_boot_init (void) +{ + int result, error; + + if (acpi_disabled && !acpi_ht) + return 1; + + /* + * The default interrupt routing model is PIC (8259). This gets + * overriden if IOAPICs are enumerated (below). + */ + acpi_irq_model = ACPI_IRQ_MODEL_PIC; + + /* + * Initialize the ACPI boot-time table parser. + */ + result = acpi_table_init(); + if (result) { + acpi_disabled = 1; + return result; + } + + result = acpi_blacklisted(); + if (result) { + printk(KERN_WARNING PREFIX "BIOS listed in blacklist, disabling ACPI support\n"); + acpi_disabled = 1; + return result; + } + + error = acpi_apic_setup(); + +#ifdef CONFIG_PCI_MMCONFIG + result = acpi_table_parse(ACPI_MCFG, acpi_parse_mcfg); + if (result < 0) { + printk(KERN_ERR PREFIX "Error %d parsing MCFG\n", result); + if (!error) + error = result; + } else if (result > 1) { + printk(KERN_WARNING PREFIX "Multiple MCFG tables exist\n"); + } +#endif /* CONFIG_PCI_MMCONFIG */ + #ifdef CONFIG_HPET_TIMER - acpi_table_parse(ACPI_HPET, acpi_parse_hpet); + result = acpi_table_parse(ACPI_HPET, acpi_parse_hpet); + if (result < 0) { + printk(KERN_ERR PREFIX "Error %d parsing HPET\n", result); + if (!error) + error = result; + } else if (result > 1) { + printk(KERN_WARNING PREFIX "Multiple HPET tables exist\n"); + } #endif - return 0; + return error; } --- diff/arch/i386/kernel/apic.c 2004-02-09 10:36:07.000000000 +0000 +++ source/arch/i386/kernel/apic.c 2004-03-01 15:42:15.000000000 +0000 @@ -595,7 +595,7 @@ static int __init init_lapic_sysfs(void) error = sysdev_class_register(&lapic_sysclass); if (!error) - error = sys_device_register(&device_lapic); + error = sysdev_register(&device_lapic); return error; } device_initcall(init_lapic_sysfs); --- diff/arch/i386/kernel/asm-offsets.c 2003-10-09 09:47:16.000000000 +0100 +++ source/arch/i386/kernel/asm-offsets.c 2004-03-01 15:42:15.000000000 +0000 @@ -4,9 +4,11 @@ * to extract and format the required data. */ +#include #include #include #include "sigframe.h" +#include #define DEFINE(sym, val) \ asm volatile("\n->" #sym " %0 " #val : : "i" (val)) @@ -28,4 +30,20 @@ void foo(void) DEFINE(RT_SIGFRAME_sigcontext, offsetof (struct rt_sigframe, uc.uc_mcontext)); + + DEFINE(TI_task, offsetof (struct thread_info, task)); + DEFINE(TI_exec_domain, offsetof (struct thread_info, exec_domain)); + DEFINE(TI_flags, offsetof (struct thread_info, flags)); + DEFINE(TI_preempt_count, offsetof (struct thread_info, preempt_count)); + DEFINE(TI_addr_limit, offsetof (struct thread_info, addr_limit)); + DEFINE(TI_real_stack, offsetof (struct thread_info, real_stack)); + DEFINE(TI_virtual_stack, offsetof (struct thread_info, virtual_stack)); + DEFINE(TI_user_pgd, offsetof (struct thread_info, user_pgd)); + + DEFINE(FIX_ENTRY_TRAMPOLINE_0_addr, + __fix_to_virt(FIX_ENTRY_TRAMPOLINE_0)); + DEFINE(FIX_VSYSCALL_addr, __fix_to_virt(FIX_VSYSCALL)); + DEFINE(PAGE_SIZE_asm, PAGE_SIZE); + DEFINE(task_thread_db7, + offsetof (struct task_struct, thread.debugreg[7])); } --- diff/arch/i386/kernel/cpu/centaur.c 2003-05-21 11:49:49.000000000 +0100 +++ source/arch/i386/kernel/cpu/centaur.c 2004-03-01 15:42:15.000000000 +0000 @@ -246,7 +246,15 @@ static void __init winchip2_protect_mcr( lo&=~0x1C0; /* blank bits 8-6 */ wrmsr(MSR_IDT_MCR_CTRL, lo, hi); } -#endif +#endif /* CONFIG_X86_OOSTORE */ + +#define ACE_PRESENT (1 << 6) +#define ACE_ENABLED (1 << 7) +#define ACE_FCR (1 << 28) /* MSR_VIA_FCR */ + +#define RNG_PRESENT (1 << 2) +#define RNG_ENABLED (1 << 3) +#define RNG_ENABLE (1 << 6) /* MSR_VIA_RNG */ static void __init init_c3(struct cpuinfo_x86 *c) { @@ -254,6 +262,24 @@ static void __init init_c3(struct cpuinf /* Test for Centaur Extended Feature Flags presence */ if (cpuid_eax(0xC0000000) >= 0xC0000001) { + u32 tmp = cpuid_edx(0xC0000001); + + /* enable ACE unit, if present and disabled */ + if ((tmp & (ACE_PRESENT | ACE_ENABLED)) == ACE_PRESENT) { + rdmsr (MSR_VIA_FCR, lo, hi); + lo |= ACE_FCR; /* enable ACE unit */ + wrmsr (MSR_VIA_FCR, lo, hi); + printk(KERN_INFO "CPU: Enabled ACE h/w crypto\n"); + } + + /* enable RNG unit, if present and disabled */ + if ((tmp & (RNG_PRESENT | RNG_ENABLED)) == RNG_PRESENT) { + rdmsr (MSR_VIA_RNG, lo, hi); + lo |= RNG_ENABLE; /* enable RNG unit */ + wrmsr (MSR_VIA_RNG, lo, hi); + printk(KERN_INFO "CPU: Enabled h/w RNG\n"); + } + /* store Centaur Extended Feature Flags as * word 5 of the CPU capability bit array */ --- diff/arch/i386/kernel/cpu/common.c 2004-01-19 10:22:55.000000000 +0000 +++ source/arch/i386/kernel/cpu/common.c 2004-03-01 15:42:15.000000000 +0000 @@ -514,12 +514,16 @@ void __init cpu_init (void) set_tss_desc(cpu,t); cpu_gdt_table[cpu][GDT_ENTRY_TSS].b &= 0xfffffdff; load_TR_desc(); - load_LDT(&init_mm.context); + if (cpu) + load_LDT(&init_mm.context); /* Set up doublefault TSS pointer in the GDT */ __set_tss_desc(cpu, GDT_ENTRY_DOUBLEFAULT_TSS, &doublefault_tss); cpu_gdt_table[cpu][GDT_ENTRY_DOUBLEFAULT_TSS].b &= 0xfffffdff; + if (cpu) + trap_init_virtual_GDT(); + /* Clear %fs and %gs. */ asm volatile ("xorl %eax, %eax; movl %eax, %fs; movl %eax, %gs"); --- diff/arch/i386/kernel/cpu/cpufreq/p4-clockmod.c 2004-02-18 08:54:07.000000000 +0000 +++ source/arch/i386/kernel/cpu/cpufreq/p4-clockmod.c 2004-03-01 15:42:15.000000000 +0000 @@ -57,8 +57,7 @@ static int cpufreq_p4_setdc(unsigned int u32 l, h; cpumask_t cpus_allowed, affected_cpu_map; struct cpufreq_freqs freqs; - int hyperthreading = 0; - int sibling = 0; + int j; if (!cpu_online(cpu) || (newstate > DC_DISABLE) || (newstate == DC_RESV)) @@ -68,13 +67,10 @@ static int cpufreq_p4_setdc(unsigned int cpus_allowed = current->cpus_allowed; /* only run on CPU to be set, or on its sibling */ - affected_cpu_map = cpumask_of_cpu(cpu); -#ifdef CONFIG_X86_HT - hyperthreading = ((cpu_has_ht) && (smp_num_siblings == 2)); - if (hyperthreading) { - sibling = cpu_sibling_map[cpu]; - cpu_set(sibling, affected_cpu_map); - } +#ifdef CONFIG_SMP + affected_cpu_map = cpu_sibling_map[cpu]; +#else + affected_cpu_map = cpumask_of_cpu(cpu); #endif set_cpus_allowed(current, affected_cpu_map); BUG_ON(!cpu_isset(smp_processor_id(), affected_cpu_map)); @@ -97,11 +93,11 @@ static int cpufreq_p4_setdc(unsigned int /* notifiers */ freqs.old = stock_freq * l / 8; freqs.new = stock_freq * newstate / 8; - freqs.cpu = cpu; - cpufreq_notify_transition(&freqs, CPUFREQ_PRECHANGE); - if (hyperthreading) { - freqs.cpu = sibling; - cpufreq_notify_transition(&freqs, CPUFREQ_PRECHANGE); + for_each_cpu(j) { + if (cpu_isset(j, affected_cpu_map)) { + freqs.cpu = j; + cpufreq_notify_transition(&freqs, CPUFREQ_PRECHANGE); + } } rdmsr(MSR_IA32_THERM_STATUS, l, h); @@ -132,10 +128,11 @@ static int cpufreq_p4_setdc(unsigned int set_cpus_allowed(current, cpus_allowed); /* notifiers */ - cpufreq_notify_transition(&freqs, CPUFREQ_POSTCHANGE); - if (hyperthreading) { - freqs.cpu = cpu; - cpufreq_notify_transition(&freqs, CPUFREQ_POSTCHANGE); + for_each_cpu(j) { + if (cpu_isset(j, affected_cpu_map)) { + freqs.cpu = j; + cpufreq_notify_transition(&freqs, CPUFREQ_POSTCHANGE); + } } return 0; --- diff/arch/i386/kernel/cpu/cpufreq/powernow-k8.c 2004-03-01 14:36:04.000000000 +0000 +++ source/arch/i386/kernel/cpu/cpufreq/powernow-k8.c 2004-03-01 15:42:15.000000000 +0000 @@ -8,6 +8,8 @@ * * Based on the powernow-k7.c module written by Dave Jones. * (C) 2003 Dave Jones on behalf of SuSE Labs + * (C) 2004 Dominik Brodowski + * (C) 2004 Pavel Machek * Licensed under the terms of the GNU GPL License version 2. * Based upon datasheets & sample CPUs kindly provided by AMD. * @@ -34,10 +36,6 @@ #define VERSION "version 1.00.08a" #include "powernow-k8.h" -#ifdef CONFIG_PREEMPT -#warning this driver has not been tested on a preempt system -#endif - static u32 vstable; /* voltage stabalization time, from PSB, units 20 us */ static u32 plllock; /* pll lock time, from PSB, units 1 us */ static u32 numps; /* number of p-states, from PSB */ @@ -636,13 +634,22 @@ find_psb_table(void) return -ENOMEM; } - for (j = 0; j < numps; j++) { - printk(KERN_INFO PFX " %d : fid 0x%x (%d MHz), vid 0x%x\n", j, - pst[j].fid, find_freq_from_fid(pst[j].fid), pst[j].vid); + for (j = 0; j < psb->numpstates; j++) { powernow_table[j].index = pst[j].fid; /* lower 8 bits */ powernow_table[j].index |= (pst[j].vid << 8); /* upper 8 bits */ - powernow_table[j].frequency = find_freq_from_fid(pst[j].fid); } + + /* If you want to override your frequency tables, this + is right place. */ + + for (j = 0; j < numps; j++) { + powernow_table[j].frequency = find_freq_from_fid(powernow_table[j].index & 0xff)*1000; + printk(KERN_INFO PFX " %d : fid 0x%x (%d MHz), vid 0x%x\n", j, + powernow_table[j].index & 0xff, + powernow_table[j].frequency/1000, + powernow_table[j].index >> 8); + } + powernow_table[numps].frequency = CPUFREQ_TABLE_END; powernow_table[numps].index = 0; --- diff/arch/i386/kernel/cpu/intel.c 2004-03-01 14:36:04.000000000 +0000 +++ source/arch/i386/kernel/cpu/intel.c 2004-03-01 15:42:15.000000000 +0000 @@ -10,6 +10,7 @@ #include #include #include +#include #include "cpu.h" @@ -19,8 +20,6 @@ #include #endif -extern int trap_init_f00f_bug(void); - #ifdef CONFIG_X86_INTEL_USERCOPY /* * Alignment at which movsl is preferred for bulk memory copies. @@ -165,7 +164,7 @@ static void __init init_intel(struct cpu c->f00f_bug = 1; if ( !f00f_workaround_enabled ) { - trap_init_f00f_bug(); + trap_init_virtual_IDT(); printk(KERN_NOTICE "Intel Pentium with F0 0F bug - workaround enabled.\n"); f00f_workaround_enabled = 1; } @@ -250,6 +249,12 @@ static void __init init_intel(struct cpu /* SEP CPUID bug: Pentium Pro reports SEP but doesn't have it until model 3 mask 3 */ if ((c->x86<<8 | c->x86_model<<4 | c->x86_mask) < 0x633) clear_bit(X86_FEATURE_SEP, c->x86_capability); + /* + * FIXME: SEP is disabled for 4G/4G for now: + */ +#ifdef CONFIG_X86_HIGH_ENTRY + clear_bit(X86_FEATURE_SEP, c->x86_capability); +#endif /* Names for the Pentium II/Celeron processors detectable only by also checking the cache size. --- diff/arch/i386/kernel/cpu/proc.c 2003-08-26 10:00:51.000000000 +0100 +++ source/arch/i386/kernel/cpu/proc.c 2004-03-01 15:42:15.000000000 +0000 @@ -50,7 +50,7 @@ static int show_cpuinfo(struct seq_file NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, /* VIA/Cyrix/Centaur-defined */ - NULL, NULL, "xstore", NULL, NULL, NULL, NULL, NULL, + NULL, NULL, "rng", "rng_en", NULL, NULL, "ace", "ace_en", NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, --- diff/arch/i386/kernel/doublefault.c 2003-10-09 09:47:16.000000000 +0100 +++ source/arch/i386/kernel/doublefault.c 2004-03-01 15:42:15.000000000 +0000 @@ -7,12 +7,13 @@ #include #include #include +#include #define DOUBLEFAULT_STACKSIZE (1024) static unsigned long doublefault_stack[DOUBLEFAULT_STACKSIZE]; #define STACK_START (unsigned long)(doublefault_stack+DOUBLEFAULT_STACKSIZE) -#define ptr_ok(x) ((x) > 0xc0000000 && (x) < 0xc1000000) +#define ptr_ok(x) (((x) > __PAGE_OFFSET && (x) < (__PAGE_OFFSET + 0x01000000)) || ((x) >= FIXADDR_START)) static void doublefault_fn(void) { @@ -38,8 +39,8 @@ static void doublefault_fn(void) printk("eax = %08lx, ebx = %08lx, ecx = %08lx, edx = %08lx\n", t->eax, t->ebx, t->ecx, t->edx); - printk("esi = %08lx, edi = %08lx\n", - t->esi, t->edi); + printk("esi = %08lx, edi = %08lx, ebp = %08lx\n", + t->esi, t->edi, t->ebp); } } --- diff/arch/i386/kernel/entry.S 2004-03-01 14:36:04.000000000 +0000 +++ source/arch/i386/kernel/entry.S 2004-03-01 15:42:15.000000000 +0000 @@ -43,11 +43,25 @@ #include #include #include +#include #include #include +#include #include #include #include "irq_vectors.h" + /* We do not recover from a stack overflow, but at least + * we know it happened and should be able to track it down. + */ +#ifdef CONFIG_STACK_OVERFLOW_TEST +#define STACK_OVERFLOW_TEST \ + testl $7680,%esp; \ + jnz 10f; \ + call stack_overflow; \ +10: +#else +#define STACK_OVERFLOW_TEST +#endif #define nr_syscalls ((syscall_table_size)/4) @@ -87,7 +101,102 @@ TSS_ESP0_OFFSET = (4 - 0x200) #define resume_kernel restore_all #endif -#define SAVE_ALL \ +#ifdef CONFIG_X86_HIGH_ENTRY + +#ifdef CONFIG_X86_SWITCH_PAGETABLES + +#if defined(CONFIG_PREEMPT) && defined(CONFIG_SMP) +/* + * If task is preempted in __SWITCH_KERNELSPACE, and moved to another cpu, + * __switch_to repoints %esp to the appropriate virtual stack; but %ebp is + * left stale, so we must check whether to repeat the real stack calculation. + */ +#define repeat_if_esp_changed \ + xorl %esp, %ebp; \ + testl $-THREAD_SIZE, %ebp; \ + jnz 0b +#else +#define repeat_if_esp_changed +#endif + +/* clobbers ebx, edx and ebp */ + +#define __SWITCH_KERNELSPACE \ + cmpl $0xff000000, %esp; \ + jb 1f; \ + \ + /* \ + * switch pagetables and load the real stack, \ + * keep the stack offset: \ + */ \ + \ + movl $swapper_pg_dir-__PAGE_OFFSET, %edx; \ + \ + /* GET_THREAD_INFO(%ebp) intermixed */ \ +0: \ + movl %esp, %ebp; \ + movl %esp, %ebx; \ + andl $(-THREAD_SIZE), %ebp; \ + andl $(THREAD_SIZE-1), %ebx; \ + orl TI_real_stack(%ebp), %ebx; \ + repeat_if_esp_changed; \ + \ + movl %edx, %cr3; \ + movl %ebx, %esp; \ +1: + +#endif + + +#define __SWITCH_USERSPACE \ + /* interrupted any of the user return paths? */ \ + \ + movl EIP(%esp), %eax; \ + \ + cmpl $int80_ret_start_marker, %eax; \ + jb 33f; /* nope - continue with sysexit check */\ + cmpl $int80_ret_end_marker, %eax; \ + jb 22f; /* yes - switch to virtual stack */ \ +33: \ + cmpl $sysexit_ret_start_marker, %eax; \ + jb 44f; /* nope - continue with user check */ \ + cmpl $sysexit_ret_end_marker, %eax; \ + jb 22f; /* yes - switch to virtual stack */ \ + /* return to userspace? */ \ +44: \ + movl EFLAGS(%esp),%ecx; \ + movb CS(%esp),%cl; \ + testl $(VM_MASK | 3),%ecx; \ + jz 2f; \ +22: \ + /* \ + * switch to the virtual stack, then switch to \ + * the userspace pagetables. \ + */ \ + \ + GET_THREAD_INFO(%ebp); \ + movl TI_virtual_stack(%ebp), %edx; \ + movl TI_user_pgd(%ebp), %ecx; \ + \ + movl %esp, %ebx; \ + andl $(THREAD_SIZE-1), %ebx; \ + orl %ebx, %edx; \ +int80_ret_start_marker: \ + movl %edx, %esp; \ + movl %ecx, %cr3; \ + \ + __RESTORE_ALL; \ +int80_ret_end_marker: \ +2: + +#else /* !CONFIG_X86_HIGH_ENTRY */ + +#define __SWITCH_KERNELSPACE +#define __SWITCH_USERSPACE + +#endif + +#define __SAVE_ALL \ cld; \ pushl %es; \ pushl %ds; \ @@ -102,7 +211,7 @@ TSS_ESP0_OFFSET = (4 - 0x200) movl %edx, %ds; \ movl %edx, %es; -#define RESTORE_INT_REGS \ +#define __RESTORE_INT_REGS \ popl %ebx; \ popl %ecx; \ popl %edx; \ @@ -111,29 +220,28 @@ TSS_ESP0_OFFSET = (4 - 0x200) popl %ebp; \ popl %eax -#define RESTORE_REGS \ - RESTORE_INT_REGS; \ -1: popl %ds; \ -2: popl %es; \ +#define __RESTORE_REGS \ + __RESTORE_INT_REGS; \ +111: popl %ds; \ +222: popl %es; \ .section .fixup,"ax"; \ -3: movl $0,(%esp); \ - jmp 1b; \ -4: movl $0,(%esp); \ - jmp 2b; \ +444: movl $0,(%esp); \ + jmp 111b; \ +555: movl $0,(%esp); \ + jmp 222b; \ .previous; \ .section __ex_table,"a";\ .align 4; \ - .long 1b,3b; \ - .long 2b,4b; \ + .long 111b,444b;\ + .long 222b,555b;\ .previous - -#define RESTORE_ALL \ - RESTORE_REGS \ +#define __RESTORE_ALL \ + __RESTORE_REGS \ addl $4, %esp; \ -1: iret; \ +333: iret; \ .section .fixup,"ax"; \ -2: sti; \ +666: sti; \ movl $(__USER_DS), %edx; \ movl %edx, %ds; \ movl %edx, %es; \ @@ -142,10 +250,19 @@ TSS_ESP0_OFFSET = (4 - 0x200) .previous; \ .section __ex_table,"a";\ .align 4; \ - .long 1b,2b; \ + .long 333b,666b;\ .previous +#define SAVE_ALL \ + __SAVE_ALL; \ + __SWITCH_KERNELSPACE; \ + STACK_OVERFLOW_TEST; + +#define RESTORE_ALL \ + __SWITCH_USERSPACE; \ + __RESTORE_ALL; +.section .entry.text,"ax" ENTRY(lcall7) pushfl # We get a different stack layout with call @@ -163,7 +280,7 @@ do_lcall: movl %edx,EIP(%ebp) # Now we move them to their "normal" places movl %ecx,CS(%ebp) # GET_THREAD_INFO_WITH_ESP(%ebp) # GET_THREAD_INFO - movl TI_EXEC_DOMAIN(%ebp), %edx # Get the execution domain + movl TI_exec_domain(%ebp), %edx # Get the execution domain call *4(%edx) # Call the lcall7 handler for the domain addl $4, %esp popl %eax @@ -208,7 +325,7 @@ ENTRY(resume_userspace) cli # make sure we don't miss an interrupt # setting need_resched or sigpending # between sampling and the iret - movl TI_FLAGS(%ebp), %ecx + movl TI_flags(%ebp), %ecx andl $_TIF_WORK_MASK, %ecx # is there any work to be done on # int/exception return? jne work_pending @@ -216,18 +333,18 @@ ENTRY(resume_userspace) #ifdef CONFIG_PREEMPT ENTRY(resume_kernel) - cmpl $0,TI_PRE_COUNT(%ebp) # non-zero preempt_count ? + cmpl $0,TI_preempt_count(%ebp) # non-zero preempt_count ? jnz restore_all need_resched: - movl TI_FLAGS(%ebp), %ecx # need_resched set ? + movl TI_flags(%ebp), %ecx # need_resched set ? testb $_TIF_NEED_RESCHED, %cl jz restore_all testl $IF_MASK,EFLAGS(%esp) # interrupts off (exception path) ? jz restore_all - movl $PREEMPT_ACTIVE,TI_PRE_COUNT(%ebp) + movl $PREEMPT_ACTIVE,TI_preempt_count(%ebp) sti call schedule - movl $0,TI_PRE_COUNT(%ebp) + movl $0,TI_preempt_count(%ebp) cli jmp need_resched #endif @@ -246,37 +363,50 @@ sysenter_past_esp: pushl $(__USER_CS) pushl $SYSENTER_RETURN -/* - * Load the potential sixth argument from user stack. - * Careful about security. - */ - cmpl $__PAGE_OFFSET-3,%ebp - jae syscall_fault -1: movl (%ebp),%ebp -.section __ex_table,"a" - .align 4 - .long 1b,syscall_fault -.previous - pushl %eax SAVE_ALL GET_THREAD_INFO(%ebp) cmpl $(nr_syscalls), %eax jae syscall_badsys - testb $_TIF_SYSCALL_TRACE,TI_FLAGS(%ebp) + testb $_TIF_SYSCALL_TRACE,TI_flags(%ebp) jnz syscall_trace_entry call *sys_call_table(,%eax,4) movl %eax,EAX(%esp) cli - movl TI_FLAGS(%ebp), %ecx + movl TI_flags(%ebp), %ecx testw $_TIF_ALLWORK_MASK, %cx jne syscall_exit_work + +#ifdef CONFIG_X86_SWITCH_PAGETABLES + + GET_THREAD_INFO(%ebp) + movl TI_virtual_stack(%ebp), %edx + movl TI_user_pgd(%ebp), %ecx + movl %esp, %ebx + andl $(THREAD_SIZE-1), %ebx + orl %ebx, %edx +sysexit_ret_start_marker: + movl %edx, %esp + movl %ecx, %cr3 +#endif + /* + * only ebx is not restored by the userspace sysenter vsyscall + * code, it assumes it to be callee-saved. + */ + movl EBX(%esp), %ebx + /* if something modifies registers it must also disable sysexit */ + movl EIP(%esp), %edx movl OLDESP(%esp), %ecx + sti sysexit +#ifdef CONFIG_X86_SWITCH_PAGETABLES +sysexit_ret_end_marker: + nop +#endif # system call handler stub @@ -287,7 +417,7 @@ ENTRY(system_call) cmpl $(nr_syscalls), %eax jae syscall_badsys # system call tracing in operation - testb $_TIF_SYSCALL_TRACE,TI_FLAGS(%ebp) + testb $_TIF_SYSCALL_TRACE,TI_flags(%ebp) jnz syscall_trace_entry syscall_call: call *sys_call_table(,%eax,4) @@ -296,10 +426,23 @@ syscall_exit: cli # make sure we don't miss an interrupt # setting need_resched or sigpending # between sampling and the iret - movl TI_FLAGS(%ebp), %ecx + movl TI_flags(%ebp), %ecx testw $_TIF_ALLWORK_MASK, %cx # current->work jne syscall_exit_work restore_all: +#ifdef CONFIG_TRAP_BAD_SYSCALL_EXITS + movl EFLAGS(%esp), %eax # mix EFLAGS and CS + movb CS(%esp), %al + testl $(VM_MASK | 3), %eax + jz resume_kernelX # returning to kernel or vm86-space + + cmpl $0,TI_preempt_count(%ebp) # non-zero preempt_count ? + jz resume_kernelX + + int $3 + +resume_kernelX: +#endif RESTORE_ALL # perform work that needs to be done immediately before resumption @@ -312,7 +455,7 @@ work_resched: cli # make sure we don't miss an interrupt # setting need_resched or sigpending # between sampling and the iret - movl TI_FLAGS(%ebp), %ecx + movl TI_flags(%ebp), %ecx andl $_TIF_WORK_MASK, %ecx # is there any work to be done other # than syscall tracing? jz restore_all @@ -327,6 +470,22 @@ work_notifysig: # deal with pending s # vm86-space xorl %edx, %edx call do_notify_resume + +#if CONFIG_X86_HIGH_ENTRY + /* + * Reload db7 if necessary: + */ + movl TI_flags(%ebp), %ecx + testb $_TIF_DB7, %cl + jnz work_db7 + + jmp restore_all + +work_db7: + movl TI_task(%ebp), %edx; + movl task_thread_db7(%edx), %edx; + movl %edx, %db7; +#endif jmp restore_all ALIGN @@ -382,7 +541,7 @@ syscall_badsys: */ .data ENTRY(interrupt) -.text +.previous vector=0 ENTRY(irq_entries_start) @@ -392,7 +551,7 @@ ENTRY(irq_entries_start) jmp common_interrupt .data .long 1b -.text +.previous vector=vector+1 .endr @@ -433,12 +592,17 @@ error_code: movl ES(%esp), %edi # get the function address movl %eax, ORIG_EAX(%esp) movl %ecx, ES(%esp) - movl %esp, %edx pushl %esi # push the error code - pushl %edx # push the pt_regs pointer movl $(__USER_DS), %edx movl %edx, %ds movl %edx, %es + +/* clobbers edx, ebx and ebp */ + __SWITCH_KERNELSPACE + + leal 4(%esp), %edx # prepare pt_regs + pushl %edx # push pt_regs + call *%edi addl $8, %esp jmp ret_from_exception @@ -529,7 +693,7 @@ nmi_stack_correct: pushl %edx call do_nmi addl $8, %esp - RESTORE_ALL + jmp restore_all nmi_stack_fixup: FIX_STACK(12,nmi_stack_correct, 1) @@ -606,6 +770,8 @@ ENTRY(spurious_interrupt_bug) pushl $do_spurious_interrupt_bug jmp error_code +.previous + .data ENTRY(sys_call_table) .long sys_restart_syscall /* 0 - old "setup()" system call, used for restarting */ @@ -882,5 +1048,11 @@ ENTRY(sys_call_table) .long sys_utimes .long sys_fadvise64_64 .long sys_ni_syscall /* sys_vserver */ + .long sys_mq_open + .long sys_mq_unlink /* 275 */ + .long sys_mq_timedsend + .long sys_mq_timedreceive + .long sys_mq_notify + .long sys_mq_getsetattr syscall_table_size=(.-sys_call_table) --- diff/arch/i386/kernel/head.S 2004-03-01 14:36:04.000000000 +0000 +++ source/arch/i386/kernel/head.S 2004-03-01 15:42:15.000000000 +0000 @@ -17,7 +17,7 @@ #include #include #include - +#include #define OLD_CL_MAGIC_ADDR 0x90020 #define OLD_CL_MAGIC 0xA33F @@ -40,49 +40,89 @@ #define X86_VENDOR_ID CPU_PARAMS+36 /* offset dependent on NCAPINTS */ /* - * Initialize page tables + * This is how much memory *in addition to the memory covered up to + * and including _end* we need mapped initially. We need one bit for + * each possible page, but only in low memory, which means + * 2^32/4096/8 = 128K worst case (4G/4G split.) + * + * Modulo rounding, each megabyte assigned here requires a kilobyte of + * memory, which is currently unreclaimed. + * + * This should be a multiple of a page. */ -#define INIT_PAGE_TABLES \ - movl $pg0 - __PAGE_OFFSET, %edi; \ - /* "007" doesn't mean with license to kill, but PRESENT+RW+USER */ \ - movl $007, %eax; \ -2: stosl; \ - add $0x1000, %eax; \ - cmp $empty_zero_page - __PAGE_OFFSET, %edi; \ - jne 2b; +#define INIT_MAP_BEYOND_END (128*1024) + /* - * swapper_pg_dir is the main page directory, address 0x00101000 - * - * On entry, %esi points to the real-mode code as a 32-bit pointer. + * 32-bit kernel entrypoint; only used by the boot CPU. On entry, + * %esi points to the real-mode code as a 32-bit pointer. + * CS and DS must be 4 GB flat segments, but we don't depend on + * any particular GDT layout, because we load our own as soon as we + * can. */ ENTRY(startup_32) -#ifdef CONFIG_X86_VISWS /* - * On SGI Visual Workstations boot CPU starts in protected mode. + * Set segments to known values. */ - orw %bx, %bx - jnz 1f - INIT_PAGE_TABLES - movl $swapper_pg_dir - __PAGE_OFFSET, %eax - movl %eax, %cr3 - lgdt boot_gdt -1: -#endif + cld + lgdt boot_gdt_descr - __PAGE_OFFSET + movl $(__BOOT_DS),%eax + movl %eax,%ds + movl %eax,%es + movl %eax,%fs + movl %eax,%gs /* - * Set segments to known values + * Initialize page tables. This creates a PDE and a set of page + * tables, which are located immediately beyond _end. The variable + * init_pg_tables_end is set up to point to the first "safe" location. + * + * Warning: don't use %esi or the stack in this code. However, %esp + * can be used as a GPR if you really need it... */ +page_pde_offset = (__PAGE_OFFSET >> 20); + + movl $(pg0 - __PAGE_OFFSET), %edi + movl $(swapper_pg_dir - __PAGE_OFFSET), %edx + movl $0x007, %eax /* 0x007 = PRESENT+RW+USER */ +10: + leal 0x007(%edi),%ecx /* Create PDE entry */ + movl %ecx,(%edx) /* Store identity PDE entry */ + movl %ecx,page_pde_offset(%edx) /* Store kernel PDE entry */ + addl $4,%edx + movl $1024, %ecx +11: + stosl + addl $0x1000,%eax + loop 11b + /* End condition: we must map up to and including INIT_MAP_BEYOND_END */ + /* bytes beyond the end of our own page tables; the +0x007 is the attribute bits */ + leal (INIT_MAP_BEYOND_END+0x007)(%edi),%ebp + cmpl %ebp,%eax + jb 10b + movl %edi,(init_pg_tables_end - __PAGE_OFFSET) + +#ifdef CONFIG_SMP + xorl %ebx,%ebx /* This is the boot CPU (BSP) */ + jmp 3f + +/* + * Non-boot CPU entry point; entered from trampoline.S + * We can't lgdt here, because lgdt itself uses a data segment, but + * we know the trampoline has already loaded the boot_gdt_table GDT + * for us. + */ +ENTRY(startup_32_smp) cld movl $(__BOOT_DS),%eax movl %eax,%ds movl %eax,%es movl %eax,%fs movl %eax,%gs -#ifdef CONFIG_SMP - orw %bx,%bx - jz 1f + + xorl %ebx,%ebx + incl %ebx /* This is a secondary processor (AP) */ /* * New page tables may be in 4Mbyte page mode and may @@ -99,37 +139,40 @@ ENTRY(startup_32) * not yet offset PAGE_OFFSET.. */ #define cr4_bits mmu_cr4_features-__PAGE_OFFSET - cmpl $0,cr4_bits - je 3f + movl cr4_bits,%edx + andl %edx,%edx + jz 3f movl %cr4,%eax # Turn on paging options (PSE,PAE,..) - orl cr4_bits,%eax + orl %edx,%eax movl %eax,%cr4 - jmp 3f -1: -#endif - INIT_PAGE_TABLES + +3: +#endif /* CONFIG_SMP */ + /* * Enable paging */ -3: movl $swapper_pg_dir-__PAGE_OFFSET,%eax movl %eax,%cr3 /* set the page table pointer.. */ movl %cr0,%eax orl $0x80000000,%eax movl %eax,%cr0 /* ..and set paging (PG) bit */ - jmp 1f /* flush the prefetch-queue */ -1: - movl $1f,%eax - jmp *%eax /* make sure eip is relocated */ + ljmp $__BOOT_CS,$1f /* Clear prefetch and normalize %eip */ 1: /* Set up the stack pointer */ lss stack_start,%esp -#ifdef CONFIG_SMP - orw %bx,%bx - jz 1f /* Initial CPU cleans BSS */ +/* + * Initialize eflags. Some BIOS's leave bits like NT set. This would + * confuse the debugger if this code is traced. + * XXX - best to initialize before switching to protected mode. + */ pushl $0 popfl + +#ifdef CONFIG_SMP + andl %ebx,%ebx + jz 1f /* Initial CPU cleans BSS */ jmp checkCPUtype 1: #endif /* CONFIG_SMP */ @@ -142,21 +185,15 @@ ENTRY(startup_32) movl $__bss_start,%edi movl $__bss_stop,%ecx subl %edi,%ecx - rep - stosb + shrl $2,%ecx + rep ; stosl /* * start system 32-bit setup. We need to re-do some of the things done * in 16-bit mode for the "real" operations. */ call setup_idt -/* - * Initialize eflags. Some BIOS's leave bits like NT set. This would - * confuse the debugger if this code is traced. - * XXX - best to initialize before switching to protected mode. - */ - pushl $0 - popfl + /* * Copy bootup parameters out of the way. First 2kB of * _empty_zero_page is for boot parameters, second 2kB @@ -273,7 +310,7 @@ is386: movl $2,%ecx # set MP call initialize_secondary jmp L6 1: -#endif +#endif /* CONFIG_SMP */ call start_kernel L6: jmp L6 # main should never return here, but @@ -309,6 +346,8 @@ check_x87: * and the kernel moved to PAGE_OFFSET. Interrupts * are enabled elsewhere, when we can be relatively * sure everything is ok. + * + * Warning: %esi is live across this function. */ setup_idt: lea ignore_int,%edx @@ -332,7 +371,7 @@ ENTRY(stack_start) /* This is the default interrupt "handler" :-) */ int_msg: - .asciz "Unknown interrupt\n" + .asciz "Unknown interrupt or fault at EIP %p %p %p\n" ALIGN ignore_int: cld @@ -344,9 +383,13 @@ ignore_int: movl $(__KERNEL_DS),%eax movl %eax,%ds movl %eax,%es + pushl 16(%esp) + pushl 24(%esp) + pushl 32(%esp) + pushl 40(%esp) pushl $int_msg call printk - popl %eax + addl $(5*4),%esp popl %ds popl %es popl %edx @@ -361,10 +404,17 @@ ignore_int: * segment size, and 32-bit linear address value: */ +.globl boot_gdt_descr .globl idt_descr .globl cpu_gdt_descr ALIGN +# early boot GDT descriptor (must use 1:1 address mapping) + .word 0 # 32 bit align gdt_desc.address +boot_gdt_descr: + .word __BOOT_DS+7 + .long boot_gdt_table - __PAGE_OFFSET + .word 0 # 32-bit align idt_desc.address idt_descr: .word IDT_ENTRIES*8-1 # idt contains 256 entries @@ -379,41 +429,25 @@ cpu_gdt_descr: .fill NR_CPUS-1,8,0 # space for the other GDT descriptors /* - * This is initialized to create an identity-mapping at 0-8M (for bootup - * purposes) and another mapping of the 0-8M area at virtual address - * PAGE_OFFSET. + * swapper_pg_dir is the main page directory, address 0x00101000 + * + * This is initialized to create an identity-mapping at 0 (for bootup + * purposes) and another mapping at virtual address PAGE_OFFSET. The + * values put here should be all invalid (zero); the valid + * entries are created dynamically at boot time. + * + * The code creates enough page tables to map 0-_end, the page tables + * themselves, plus INIT_MAP_BEYOND_END bytes; see comment at beginning. */ .org 0x1000 ENTRY(swapper_pg_dir) - .long 0x00102007 - .long 0x00103007 - .fill BOOT_USER_PGD_PTRS-2,4,0 - /* default: 766 entries */ - .long 0x00102007 - .long 0x00103007 - /* default: 254 entries */ - .fill BOOT_KERNEL_PGD_PTRS-2,4,0 + .fill 1024,4,0 -/* - * The page tables are initialized to only 8MB here - the final page - * tables are set up later depending on memory size. - */ .org 0x2000 -ENTRY(pg0) - -.org 0x3000 -ENTRY(pg1) - -/* - * empty_zero_page must immediately follow the page tables ! (The - * initialization loop counts until empty_zero_page) - */ - -.org 0x4000 ENTRY(empty_zero_page) + .fill 4096,1,0 -.org 0x5000 - +.org 0x3000 /* * Real beginning of normal "text" segment */ @@ -428,20 +462,19 @@ ENTRY(_stext) .data /* - * The Global Descriptor Table contains 28 quadwords, per-CPU. - */ -#if defined(CONFIG_SMP) || defined(CONFIG_X86_VISWS) -/* * The boot_gdt_table must mirror the equivalent in setup.S and is - * used only by the trampoline for booting other CPUs + * used only for booting. */ .align L1_CACHE_BYTES ENTRY(boot_gdt_table) .fill GDT_ENTRY_BOOT_CS,8,0 .quad 0x00cf9a000000ffff /* kernel 4GB code at 0x00000000 */ .quad 0x00cf92000000ffff /* kernel 4GB data at 0x00000000 */ -#endif - .align L1_CACHE_BYTES + +/* + * The Global Descriptor Table contains 28 quadwords, per-CPU. + */ + .align PAGE_SIZE_asm ENTRY(cpu_gdt_table) .quad 0x0000000000000000 /* NULL descriptor */ .quad 0x0000000000000000 /* 0x0b reserved */ @@ -488,4 +521,3 @@ ENTRY(cpu_gdt_table) #ifdef CONFIG_SMP .fill (NR_CPUS-1)*GDT_ENTRIES,8,0 /* other CPU's GDT */ #endif - --- diff/arch/i386/kernel/i386_ksyms.c 2004-02-18 08:54:07.000000000 +0000 +++ source/arch/i386/kernel/i386_ksyms.c 2004-03-01 15:42:15.000000000 +0000 @@ -88,16 +88,11 @@ EXPORT_SYMBOL(get_cmos_time); EXPORT_SYMBOL(cpu_khz); EXPORT_SYMBOL(apm_info); -#ifdef CONFIG_DEBUG_IOVIRT -EXPORT_SYMBOL(__io_virt_debug); -#endif - EXPORT_SYMBOL_NOVERS(__down_failed); EXPORT_SYMBOL_NOVERS(__down_failed_interruptible); EXPORT_SYMBOL_NOVERS(__down_failed_trylock); EXPORT_SYMBOL_NOVERS(__up_wakeup); /* Networking helper routines. */ -EXPORT_SYMBOL(csum_partial_copy_generic); /* Delay loops */ EXPORT_SYMBOL(__ndelay); EXPORT_SYMBOL(__udelay); @@ -111,13 +106,17 @@ EXPORT_SYMBOL_NOVERS(__get_user_4); EXPORT_SYMBOL(strpbrk); EXPORT_SYMBOL(strstr); +#if !defined(CONFIG_X86_UACCESS_INDIRECT) EXPORT_SYMBOL(strncpy_from_user); -EXPORT_SYMBOL(__strncpy_from_user); +EXPORT_SYMBOL(__direct_strncpy_from_user); EXPORT_SYMBOL(clear_user); EXPORT_SYMBOL(__clear_user); EXPORT_SYMBOL(__copy_from_user_ll); EXPORT_SYMBOL(__copy_to_user_ll); EXPORT_SYMBOL(strnlen_user); +#else /* CONFIG_X86_UACCESS_INDIRECT */ +EXPORT_SYMBOL(direct_csum_partial_copy_generic); +#endif EXPORT_SYMBOL(dma_alloc_coherent); EXPORT_SYMBOL(dma_free_coherent); --- diff/arch/i386/kernel/i387.c 2004-02-18 08:54:07.000000000 +0000 +++ source/arch/i386/kernel/i387.c 2004-03-01 15:42:15.000000000 +0000 @@ -218,6 +218,7 @@ void set_fpu_mxcsr( struct task_struct * static int convert_fxsr_to_user( struct _fpstate __user *buf, struct i387_fxsave_struct *fxsave ) { + struct _fpreg tmp[8]; /* 80 bytes scratch area */ unsigned long env[7]; struct _fpreg __user *to; struct _fpxreg *from; @@ -234,23 +235,25 @@ static int convert_fxsr_to_user( struct if ( __copy_to_user( buf, env, 7 * sizeof(unsigned long) ) ) return 1; - to = &buf->_st[0]; + to = tmp; from = (struct _fpxreg *) &fxsave->st_space[0]; for ( i = 0 ; i < 8 ; i++, to++, from++ ) { unsigned long *t = (unsigned long *)to; unsigned long *f = (unsigned long *)from; - if (__put_user(*f, t) || - __put_user(*(f + 1), t + 1) || - __put_user(from->exponent, &to->exponent)) - return 1; + *t = *f; + *(t + 1) = *(f+1); + to->exponent = from->exponent; } + if (copy_to_user(buf->_st, tmp, sizeof(struct _fpreg [8]))) + return 1; return 0; } static int convert_fxsr_from_user( struct i387_fxsave_struct *fxsave, struct _fpstate __user *buf ) { + struct _fpreg tmp[8]; /* 80 bytes scratch area */ unsigned long env[7]; struct _fpxreg *to; struct _fpreg __user *from; @@ -258,6 +261,8 @@ static int convert_fxsr_from_user( struc if ( __copy_from_user( env, buf, 7 * sizeof(long) ) ) return 1; + if (copy_from_user(tmp, buf->_st, sizeof(struct _fpreg [8]))) + return 1; fxsave->cwd = (unsigned short)(env[0] & 0xffff); fxsave->swd = (unsigned short)(env[1] & 0xffff); @@ -269,15 +274,14 @@ static int convert_fxsr_from_user( struc fxsave->fos = env[6]; to = (struct _fpxreg *) &fxsave->st_space[0]; - from = &buf->_st[0]; + from = tmp; for ( i = 0 ; i < 8 ; i++, to++, from++ ) { unsigned long *t = (unsigned long *)to; unsigned long *f = (unsigned long *)from; - if (__get_user(*t, f) || - __get_user(*(t + 1), f + 1) || - __get_user(to->exponent, &from->exponent)) - return 1; + *t = *f; + *(t + 1) = *(f + 1); + to->exponent = from->exponent; } return 0; } --- diff/arch/i386/kernel/i8259.c 2004-01-19 10:22:55.000000000 +0000 +++ source/arch/i386/kernel/i8259.c 2004-03-01 15:42:15.000000000 +0000 @@ -258,7 +258,7 @@ static int __init i8259A_init_sysfs(void { int error = sysdev_class_register(&i8259_sysdev_class); if (!error) - error = sys_device_register(&device_i8259A); + error = sysdev_register(&device_i8259A); return error; } @@ -401,7 +401,7 @@ static int __init init_timer_sysfs(void) { int error = sysdev_class_register(&timer_sysclass); if (!error) - error = sys_device_register(&device_timer); + error = sysdev_register(&device_timer); return error; } --- diff/arch/i386/kernel/init_task.c 2003-10-09 09:47:33.000000000 +0100 +++ source/arch/i386/kernel/init_task.c 2004-03-01 15:42:15.000000000 +0000 @@ -26,7 +26,7 @@ EXPORT_SYMBOL(init_mm); */ union thread_union init_thread_union __attribute__((__section__(".data.init_task"))) = - { INIT_THREAD_INFO(init_task) }; + { INIT_THREAD_INFO(init_task, init_thread_union) }; /* * Initial task structure. @@ -44,5 +44,5 @@ EXPORT_SYMBOL(init_task); * section. Since TSS's are completely CPU-local, we want them * on exact cacheline boundaries, to eliminate cacheline ping-pong. */ -struct tss_struct init_tss[NR_CPUS] __cacheline_aligned = { [0 ... NR_CPUS-1] = INIT_TSS }; +struct tss_struct init_tss[NR_CPUS] __attribute__((__section__(".data.tss"))) = { [0 ... NR_CPUS-1] = INIT_TSS }; --- diff/arch/i386/kernel/io_apic.c 2004-03-01 14:36:04.000000000 +0000 +++ source/arch/i386/kernel/io_apic.c 2004-03-01 15:42:15.000000000 +0000 @@ -317,8 +317,7 @@ struct irq_cpu_info { #define IRQ_ALLOWED(cpu, allowed_mask) cpu_isset(cpu, allowed_mask) -#define CPU_TO_PACKAGEINDEX(i) \ - ((physical_balance && i > cpu_sibling_map[i]) ? cpu_sibling_map[i] : i) +#define CPU_TO_PACKAGEINDEX(i) (first_cpu(cpu_sibling_map[i])) #define MAX_BALANCED_IRQ_INTERVAL (5*HZ) #define MIN_BALANCED_IRQ_INTERVAL (HZ/2) @@ -401,6 +400,7 @@ static void do_irq_balance(void) unsigned long max_cpu_irq = 0, min_cpu_irq = (~0); unsigned long move_this_load = 0; int max_loaded = 0, min_loaded = 0; + int load; unsigned long useful_load_threshold = balanced_irq_interval + 10; int selected_irq; int tmp_loaded, first_attempt = 1; @@ -452,7 +452,7 @@ static void do_irq_balance(void) for (i = 0; i < NR_CPUS; i++) { if (!cpu_online(i)) continue; - if (physical_balance && i > cpu_sibling_map[i]) + if (i != CPU_TO_PACKAGEINDEX(i)) continue; if (min_cpu_irq > CPU_IRQ(i)) { min_cpu_irq = CPU_IRQ(i); @@ -471,7 +471,7 @@ tryanothercpu: for (i = 0; i < NR_CPUS; i++) { if (!cpu_online(i)) continue; - if (physical_balance && i > cpu_sibling_map[i]) + if (i != CPU_TO_PACKAGEINDEX(i)) continue; if (max_cpu_irq <= CPU_IRQ(i)) continue; @@ -551,9 +551,14 @@ tryanotherirq: * We seek the least loaded sibling by making the comparison * (A+B)/2 vs B */ - if (physical_balance && (CPU_IRQ(min_loaded) >> 1) > - CPU_IRQ(cpu_sibling_map[min_loaded])) - min_loaded = cpu_sibling_map[min_loaded]; + load = CPU_IRQ(min_loaded) >> 1; + for_each_cpu_mask(j, cpu_sibling_map[min_loaded]) { + if (load > CPU_IRQ(j)) { + /* This won't change cpu_sibling_map[min_loaded] */ + load = CPU_IRQ(j); + min_loaded = j; + } + } cpus_and(allowed_mask, cpu_online_map, irq_affinity[selected_irq]); target_cpu_mask = cpumask_of_cpu(min_loaded); --- diff/arch/i386/kernel/irq.c 2004-03-01 14:36:04.000000000 +0000 +++ source/arch/i386/kernel/irq.c 2004-03-01 15:42:15.000000000 +0000 @@ -508,6 +508,8 @@ out: irq_exit(); + kgdb_process_breakpoint(); + return 1; } --- diff/arch/i386/kernel/ldt.c 2003-10-09 09:47:16.000000000 +0100 +++ source/arch/i386/kernel/ldt.c 2004-03-01 15:42:15.000000000 +0000 @@ -2,7 +2,7 @@ * linux/kernel/ldt.c * * Copyright (C) 1992 Krishna Balasubramanian and Linus Torvalds - * Copyright (C) 1999 Ingo Molnar + * Copyright (C) 1999, 2003 Ingo Molnar */ #include @@ -18,6 +18,8 @@ #include #include #include +#include +#include #ifdef CONFIG_SMP /* avoids "defined but not used" warnig */ static void flush_ldt(void *null) @@ -29,34 +31,31 @@ static void flush_ldt(void *null) static int alloc_ldt(mm_context_t *pc, int mincount, int reload) { - void *oldldt; - void *newldt; - int oldsize; + int oldsize, newsize, i; if (mincount <= pc->size) return 0; + /* + * LDT got larger - reallocate if necessary. + */ oldsize = pc->size; mincount = (mincount+511)&(~511); - if (mincount*LDT_ENTRY_SIZE > PAGE_SIZE) - newldt = vmalloc(mincount*LDT_ENTRY_SIZE); - else - newldt = kmalloc(mincount*LDT_ENTRY_SIZE, GFP_KERNEL); - - if (!newldt) - return -ENOMEM; - - if (oldsize) - memcpy(newldt, pc->ldt, oldsize*LDT_ENTRY_SIZE); - oldldt = pc->ldt; - memset(newldt+oldsize*LDT_ENTRY_SIZE, 0, (mincount-oldsize)*LDT_ENTRY_SIZE); - pc->ldt = newldt; - wmb(); + newsize = mincount*LDT_ENTRY_SIZE; + for (i = 0; i < newsize; i += PAGE_SIZE) { + int nr = i/PAGE_SIZE; + BUG_ON(i >= 64*1024); + if (!pc->ldt_pages[nr]) { + pc->ldt_pages[nr] = alloc_page(GFP_HIGHUSER); + if (!pc->ldt_pages[nr]) + return -ENOMEM; + clear_highpage(pc->ldt_pages[nr]); + } + } pc->size = mincount; - wmb(); - if (reload) { #ifdef CONFIG_SMP cpumask_t mask; + preempt_disable(); load_LDT(pc); mask = cpumask_of_cpu(smp_processor_id()); @@ -67,21 +66,20 @@ static int alloc_ldt(mm_context_t *pc, i load_LDT(pc); #endif } - if (oldsize) { - if (oldsize*LDT_ENTRY_SIZE > PAGE_SIZE) - vfree(oldldt); - else - kfree(oldldt); - } return 0; } static inline int copy_ldt(mm_context_t *new, mm_context_t *old) { - int err = alloc_ldt(new, old->size, 0); - if (err < 0) + int i, err, size = old->size, nr_pages = (size*LDT_ENTRY_SIZE + PAGE_SIZE-1)/PAGE_SIZE; + + err = alloc_ldt(new, size, 0); + if (err < 0) { + new->size = 0; return err; - memcpy(new->ldt, old->ldt, old->size*LDT_ENTRY_SIZE); + } + for (i = 0; i < nr_pages; i++) + copy_user_highpage(new->ldt_pages[i], old->ldt_pages[i], 0); return 0; } @@ -96,6 +94,7 @@ int init_new_context(struct task_struct init_MUTEX(&mm->context.sem); mm->context.size = 0; + memset(mm->context.ldt_pages, 0, sizeof(struct page *) * MAX_LDT_PAGES); old_mm = current->mm; if (old_mm && old_mm->context.size > 0) { down(&old_mm->context.sem); @@ -107,23 +106,21 @@ int init_new_context(struct task_struct /* * No need to lock the MM as we are the last user + * Do not touch the ldt register, we are already + * in the next thread. */ void destroy_context(struct mm_struct *mm) { - if (mm->context.size) { - if (mm == current->active_mm) - clear_LDT(); - if (mm->context.size*LDT_ENTRY_SIZE > PAGE_SIZE) - vfree(mm->context.ldt); - else - kfree(mm->context.ldt); - mm->context.size = 0; - } + int i, nr_pages = (mm->context.size*LDT_ENTRY_SIZE + PAGE_SIZE-1) / PAGE_SIZE; + + for (i = 0; i < nr_pages; i++) + __free_page(mm->context.ldt_pages[i]); + mm->context.size = 0; } static int read_ldt(void __user * ptr, unsigned long bytecount) { - int err; + int err, i; unsigned long size; struct mm_struct * mm = current->mm; @@ -138,8 +135,25 @@ static int read_ldt(void __user * ptr, u size = bytecount; err = 0; - if (copy_to_user(ptr, mm->context.ldt, size)) - err = -EFAULT; + /* + * This is necessary just in case we got here straight from a + * context-switch where the ptes were set but no tlb flush + * was done yet. We rather avoid doing a TLB flush in the + * context-switch path and do it here instead. + */ + __flush_tlb_global(); + + for (i = 0; i < size; i += PAGE_SIZE) { + int nr = i / PAGE_SIZE, bytes; + char *kaddr = kmap(mm->context.ldt_pages[nr]); + + bytes = size - i; + if (bytes > PAGE_SIZE) + bytes = PAGE_SIZE; + if (copy_to_user(ptr + i, kaddr, size - i)) + err = -EFAULT; + kunmap(mm->context.ldt_pages[nr]); + } up(&mm->context.sem); if (err < 0) return err; @@ -158,7 +172,7 @@ static int read_default_ldt(void __user err = 0; address = &default_ldt[0]; - size = 5*sizeof(struct desc_struct); + size = 5*LDT_ENTRY_SIZE; if (size > bytecount) size = bytecount; @@ -200,7 +214,15 @@ static int write_ldt(void __user * ptr, goto out_unlock; } - lp = (__u32 *) ((ldt_info.entry_number << 3) + (char *) mm->context.ldt); + /* + * No rescheduling allowed from this point to the install. + * + * We do a TLB flush for the same reason as in the read_ldt() path. + */ + preempt_disable(); + __flush_tlb_global(); + lp = (__u32 *) ((ldt_info.entry_number << 3) + + (char *) __kmap_atomic_vaddr(KM_LDT_PAGE0)); /* Allow LDTs to be cleared by the user. */ if (ldt_info.base_addr == 0 && ldt_info.limit == 0) { @@ -221,6 +243,7 @@ install: *lp = entry_1; *(lp+1) = entry_2; error = 0; + preempt_enable(); out_unlock: up(&mm->context.sem); @@ -248,3 +271,26 @@ asmlinkage int sys_modify_ldt(int func, } return ret; } + +/* + * load one particular LDT into the current CPU + */ +void load_LDT_nolock(mm_context_t *pc, int cpu) +{ + struct page **pages = pc->ldt_pages; + int count = pc->size; + int nr_pages, i; + + if (likely(!count)) { + pages = &default_ldt_page; + count = 5; + } + nr_pages = (count*LDT_ENTRY_SIZE + PAGE_SIZE-1) / PAGE_SIZE; + + for (i = 0; i < nr_pages; i++) { + __kunmap_atomic_type(KM_LDT_PAGE0 - i); + __kmap_atomic(pages[i], KM_LDT_PAGE0 - i); + } + set_ldt_desc(cpu, (void *)__kmap_atomic_vaddr(KM_LDT_PAGE0), count); + load_LDT_desc(); +} --- diff/arch/i386/kernel/mpparse.c 2004-02-18 08:54:07.000000000 +0000 +++ source/arch/i386/kernel/mpparse.c 2004-03-01 15:42:15.000000000 +0000 @@ -668,7 +668,7 @@ void __init get_smp_config (void) * Read the physical hardware table. Anything here will * override the defaults. */ - if (!smp_read_mpc((void *)mpf->mpf_physptr)) { + if (!smp_read_mpc((void *)phys_to_virt(mpf->mpf_physptr))) { smp_found_config = 0; printk(KERN_ERR "BIOS bug, MP table errors detected!...\n"); printk(KERN_ERR "... disabling SMP support. (tell your hw vendor)\n"); @@ -1156,7 +1156,7 @@ void __init mp_parse_prt (void) continue; } if ((1<irq = acpi_irq_to_vector(irq); continue; --- diff/arch/i386/kernel/nmi.c 2004-03-01 14:36:04.000000000 +0000 +++ source/arch/i386/kernel/nmi.c 2004-03-01 15:42:15.000000000 +0000 @@ -31,7 +31,16 @@ #include #include +#ifdef CONFIG_KGDB +#include +#ifdef CONFIG_SMP +unsigned int nmi_watchdog = NMI_IO_APIC; +#else +unsigned int nmi_watchdog = NMI_LOCAL_APIC; +#endif +#else unsigned int nmi_watchdog = NMI_NONE; +#endif static unsigned int nmi_hz = HZ; unsigned int nmi_perfctr_msr; /* the MSR to reset in NMI handler */ extern void show_registers(struct pt_regs *regs); @@ -248,7 +257,7 @@ static int __init init_lapic_nmi_sysfs(v error = sysdev_class_register(&nmi_sysclass); if (!error) - error = sys_device_register(&device_lapic_nmi); + error = sysdev_register(&device_lapic_nmi); return error; } /* must come after the local APIC's device_initcall() */ @@ -408,6 +417,9 @@ void touch_nmi_watchdog (void) for (i = 0; i < NR_CPUS; i++) alert_counter[i] = 0; } +#ifdef CONFIG_KGDB +int tune_watchdog = 5*HZ; +#endif void nmi_watchdog_tick (struct pt_regs * regs) { @@ -421,12 +433,24 @@ void nmi_watchdog_tick (struct pt_regs * sum = irq_stat[cpu].apic_timer_irqs; +#ifdef CONFIG_KGDB + if (! in_kgdb(regs) && last_irq_sums[cpu] == sum ) { + +#else if (last_irq_sums[cpu] == sum) { +#endif /* * Ayiee, looks like this CPU is stuck ... * wait a few IRQs (5 seconds) before doing the oops ... */ alert_counter[cpu]++; +#ifdef CONFIG_KGDB + if (alert_counter[cpu] == tune_watchdog) { + kgdb_handle_exception(2, SIGPWR, 0, regs); + last_irq_sums[cpu] = sum; + alert_counter[cpu] = 0; + } +#endif if (alert_counter[cpu] == 5*nmi_hz) { spin_lock(&nmi_print_lock); /* --- diff/arch/i386/kernel/pci-dma.c 2003-01-16 11:30:40.000000000 +0000 +++ source/arch/i386/kernel/pci-dma.c 2004-03-01 15:42:15.000000000 +0000 @@ -20,8 +20,9 @@ void *dma_alloc_coherent(struct device * /* ignore region specifiers */ gfp &= ~(__GFP_DMA | __GFP_HIGHMEM); - if (dev == NULL || (*dev->dma_mask < 0xffffffff)) + if (dev == NULL || (dev->coherent_dma_mask < 0xffffffff)) gfp |= GFP_DMA; + ret = (void *)__get_free_pages(gfp, get_order(size)); if (ret != NULL) { --- diff/arch/i386/kernel/process.c 2004-03-01 14:36:04.000000000 +0000 +++ source/arch/i386/kernel/process.c 2004-03-01 15:42:15.000000000 +0000 @@ -45,6 +45,7 @@ #include #include #include +#include #ifdef CONFIG_MATH_EMULATION #include #endif @@ -302,6 +303,9 @@ void flush_thread(void) struct task_struct *tsk = current; memset(tsk->thread.debugreg, 0, sizeof(unsigned long)*8); +#ifdef CONFIG_X86_HIGH_ENTRY + clear_thread_flag(TIF_DB7); +#endif memset(tsk->thread.tls_array, 0, sizeof(tsk->thread.tls_array)); /* * Forget coprocessor state.. @@ -315,9 +319,8 @@ void release_thread(struct task_struct * if (dead_task->mm) { // temporary debugging check if (dead_task->mm->context.size) { - printk("WARNING: dead process %8s still has LDT? <%p/%d>\n", + printk("WARNING: dead process %8s still has LDT? <%d>\n", dead_task->comm, - dead_task->mm->context.ldt, dead_task->mm->context.size); BUG(); } @@ -352,7 +355,17 @@ int copy_thread(int nr, unsigned long cl p->thread.esp = (unsigned long) childregs; p->thread.esp0 = (unsigned long) (childregs+1); + /* + * get the two stack pages, for the virtual stack. + * + * IMPORTANT: this code relies on the fact that the task + * structure is an 8K aligned piece of physical memory. + */ + p->thread.stack_page0 = virt_to_page((unsigned long)p->thread_info); + p->thread.stack_page1 = virt_to_page((unsigned long)p->thread_info + PAGE_SIZE); + p->thread.eip = (unsigned long) ret_from_fork; + p->thread_info->real_stack = p->thread_info; savesegment(fs,p->thread.fs); savesegment(gs,p->thread.gs); @@ -493,7 +506,7 @@ int dump_task_regs(struct task_struct *t * the task-switch, and shows up in ret_from_fork in entry.S, * for example. */ -struct task_struct * __switch_to(struct task_struct *prev_p, struct task_struct *next_p) +struct task_struct fastcall * __switch_to(struct task_struct *prev_p, struct task_struct *next_p) { struct thread_struct *prev = &prev_p->thread, *next = &next_p->thread; @@ -504,10 +517,41 @@ struct task_struct * __switch_to(struct __unlazy_fpu(prev_p); +#ifdef CONFIG_X86_HIGH_ENTRY + /* + * Set the ptes of the virtual stack. (NOTE: a one-page TLB flush is + * needed because otherwise NMIs could interrupt the + * user-return code with a virtual stack and stale TLBs.) + */ + __kunmap_atomic_type(KM_VSTACK0); + __kunmap_atomic_type(KM_VSTACK1); + __kmap_atomic(next->stack_page0, KM_VSTACK0); + __kmap_atomic(next->stack_page1, KM_VSTACK1); + + /* + * NOTE: here we rely on the task being the stack as well + */ + next_p->thread_info->virtual_stack = + (void *)__kmap_atomic_vaddr(KM_VSTACK0); + +#if defined(CONFIG_PREEMPT) && defined(CONFIG_SMP) + /* + * If next was preempted on entry from userspace to kernel, + * and now it's on a different cpu, we need to adjust %esp. + * This assumes that entry.S does not copy %esp while on the + * virtual stack (with interrupts enabled): which is so, + * except within __SWITCH_KERNELSPACE itself. + */ + if (unlikely(next->esp >= TASK_SIZE)) { + next->esp &= THREAD_SIZE - 1; + next->esp |= (unsigned long) next_p->thread_info->virtual_stack; + } +#endif +#endif /* * Reload esp0, LDT and the page table pointer: */ - load_esp0(tss, next); + load_virtual_esp0(tss, next_p); /* * Load the per-thread Thread-Local Storage descriptor. --- diff/arch/i386/kernel/reboot.c 2004-01-19 10:22:55.000000000 +0000 +++ source/arch/i386/kernel/reboot.c 2004-03-01 15:42:15.000000000 +0000 @@ -155,12 +155,11 @@ void machine_real_restart(unsigned char CMOS_WRITE(0x00, 0x8f); spin_unlock_irqrestore(&rtc_lock, flags); - /* Remap the kernel at virtual address zero, as well as offset zero - from the kernel segment. This assumes the kernel segment starts at - virtual address PAGE_OFFSET. */ - - memcpy (swapper_pg_dir, swapper_pg_dir + USER_PGD_PTRS, - sizeof (swapper_pg_dir [0]) * KERNEL_PGD_PTRS); + /* + * Remap the first 16 MB of RAM (which includes the kernel image) + * at virtual address zero: + */ + setup_identity_mappings(swapper_pg_dir, 0, 16*1024*1024); /* * Use `swapper_pg_dir' as our page directory. --- diff/arch/i386/kernel/setup.c 2004-03-01 14:36:04.000000000 +0000 +++ source/arch/i386/kernel/setup.c 2004-03-01 15:42:15.000000000 +0000 @@ -50,6 +50,11 @@ #include "setup_arch_pre.h" #include "mach_resources.h" +/* This value is set up by the early boot code to point to the value + immediately after the boot time page tables. It contains a *physical* + address, and must not be in the .bss segment! */ +unsigned long init_pg_tables_end __initdata = ~0UL; + int disable_pse __initdata = 0; static inline char * __init machine_specific_memory_setup(void); @@ -115,7 +120,6 @@ extern void early_cpu_init(void); extern void dmi_scan_machine(void); extern void generic_apic_probe(char *); extern int root_mountflags; -extern char _end[]; unsigned long saved_videomode; @@ -569,6 +573,11 @@ static void __init parse_cmdline_early ( acpi_disabled = 0; } + /* acpi=strict disables out-of-spec workarounds */ + else if (!memcmp(from, "acpi=strict", 11)) { + acpi_strict = 1; + } + /* Limit ACPI just to boot-time to enable HT */ else if (!memcmp(from, "acpi=ht", 7)) { acpi_ht = 1; @@ -785,7 +794,7 @@ static unsigned long __init setup_memory * partially used pages are not usable - thus * we are rounding upwards: */ - start_pfn = PFN_UP(__pa(_end)); + start_pfn = PFN_UP(init_pg_tables_end); find_max_pfn(); @@ -1097,7 +1106,7 @@ void __init setup_arch(char **cmdline_p) init_mm.start_code = (unsigned long) _text; init_mm.end_code = (unsigned long) _etext; init_mm.end_data = (unsigned long) _edata; - init_mm.brk = (unsigned long) _end; + init_mm.brk = init_pg_tables_end + PAGE_OFFSET; code_resource.start = virt_to_phys(_text); code_resource.end = virt_to_phys(_etext)-1; --- diff/arch/i386/kernel/signal.c 2003-11-25 15:24:57.000000000 +0000 +++ source/arch/i386/kernel/signal.c 2004-03-01 15:42:15.000000000 +0000 @@ -128,28 +128,29 @@ sys_sigaltstack(const stack_t __user *us */ static int -restore_sigcontext(struct pt_regs *regs, struct sigcontext __user *sc, int *peax) +restore_sigcontext(struct pt_regs *regs, + struct sigcontext __user *__sc, int *peax) { - unsigned int err = 0; + struct sigcontext scratch; /* 88 bytes of scratch area */ /* Always make any pending restarted system calls return -EINTR */ current_thread_info()->restart_block.fn = do_no_restart_syscall; -#define COPY(x) err |= __get_user(regs->x, &sc->x) + if (copy_from_user(&scratch, __sc, sizeof(scratch))) + return -EFAULT; + +#define COPY(x) regs->x = scratch.x #define COPY_SEG(seg) \ - { unsigned short tmp; \ - err |= __get_user(tmp, &sc->seg); \ + { unsigned short tmp = scratch.seg; \ regs->x##seg = tmp; } #define COPY_SEG_STRICT(seg) \ - { unsigned short tmp; \ - err |= __get_user(tmp, &sc->seg); \ + { unsigned short tmp = scratch.seg; \ regs->x##seg = tmp|3; } #define GET_SEG(seg) \ - { unsigned short tmp; \ - err |= __get_user(tmp, &sc->seg); \ + { unsigned short tmp = scratch.seg; \ loadsegment(seg,tmp); } GET_SEG(gs); @@ -168,27 +169,23 @@ restore_sigcontext(struct pt_regs *regs, COPY_SEG_STRICT(ss); { - unsigned int tmpflags; - err |= __get_user(tmpflags, &sc->eflags); + unsigned int tmpflags = scratch.eflags; regs->eflags = (regs->eflags & ~0x40DD5) | (tmpflags & 0x40DD5); regs->orig_eax = -1; /* disable syscall checks */ } { - struct _fpstate __user * buf; - err |= __get_user(buf, &sc->fpstate); + struct _fpstate * buf = scratch.fpstate; if (buf) { if (verify_area(VERIFY_READ, buf, sizeof(*buf))) - goto badframe; - err |= restore_i387(buf); + return -EFAULT; + if (restore_i387(buf)) + return -EFAULT; } } - err |= __get_user(*peax, &sc->eax); - return err; - -badframe: - return 1; + *peax = scratch.eax; + return 0; } asmlinkage int sys_sigreturn(unsigned long __unused) @@ -266,46 +263,47 @@ badframe: */ static int -setup_sigcontext(struct sigcontext __user *sc, struct _fpstate __user *fpstate, +setup_sigcontext(struct sigcontext __user *__sc, struct _fpstate __user *fpstate, struct pt_regs *regs, unsigned long mask) { - int tmp, err = 0; + struct sigcontext sc; /* 88 bytes of scratch area */ + int tmp; tmp = 0; __asm__("movl %%gs,%0" : "=r"(tmp): "0"(tmp)); - err |= __put_user(tmp, (unsigned int *)&sc->gs); + *(unsigned int *)&sc.gs = tmp; __asm__("movl %%fs,%0" : "=r"(tmp): "0"(tmp)); - err |= __put_user(tmp, (unsigned int *)&sc->fs); - - err |= __put_user(regs->xes, (unsigned int *)&sc->es); - err |= __put_user(regs->xds, (unsigned int *)&sc->ds); - err |= __put_user(regs->edi, &sc->edi); - err |= __put_user(regs->esi, &sc->esi); - err |= __put_user(regs->ebp, &sc->ebp); - err |= __put_user(regs->esp, &sc->esp); - err |= __put_user(regs->ebx, &sc->ebx); - err |= __put_user(regs->edx, &sc->edx); - err |= __put_user(regs->ecx, &sc->ecx); - err |= __put_user(regs->eax, &sc->eax); - err |= __put_user(current->thread.trap_no, &sc->trapno); - err |= __put_user(current->thread.error_code, &sc->err); - err |= __put_user(regs->eip, &sc->eip); - err |= __put_user(regs->xcs, (unsigned int *)&sc->cs); - err |= __put_user(regs->eflags, &sc->eflags); - err |= __put_user(regs->esp, &sc->esp_at_signal); - err |= __put_user(regs->xss, (unsigned int *)&sc->ss); + *(unsigned int *)&sc.fs = tmp; + *(unsigned int *)&sc.es = regs->xes; + *(unsigned int *)&sc.ds = regs->xds; + sc.edi = regs->edi; + sc.esi = regs->esi; + sc.ebp = regs->ebp; + sc.esp = regs->esp; + sc.ebx = regs->ebx; + sc.edx = regs->edx; + sc.ecx = regs->ecx; + sc.eax = regs->eax; + sc.trapno = current->thread.trap_no; + sc.err = current->thread.error_code; + sc.eip = regs->eip; + *(unsigned int *)&sc.cs = regs->xcs; + sc.eflags = regs->eflags; + sc.esp_at_signal = regs->esp; + *(unsigned int *)&sc.ss = regs->xss; tmp = save_i387(fpstate); if (tmp < 0) - err = 1; - else - err |= __put_user(tmp ? fpstate : NULL, &sc->fpstate); + return 1; + sc.fpstate = tmp ? fpstate : NULL; /* non-iBCS2 extensions.. */ - err |= __put_user(mask, &sc->oldmask); - err |= __put_user(current->thread.cr2, &sc->cr2); + sc.oldmask = mask; + sc.cr2 = current->thread.cr2; - return err; + if (copy_to_user(__sc, &sc, sizeof(sc))) + return 1; + return 0; } /* @@ -443,7 +441,7 @@ static void setup_rt_frame(int sig, stru /* Create the ucontext. */ err |= __put_user(0, &frame->uc.uc_flags); err |= __put_user(0, &frame->uc.uc_link); - err |= __put_user(current->sas_ss_sp, &frame->uc.uc_stack.ss_sp); + err |= __put_user(current->sas_ss_sp, (unsigned long *)&frame->uc.uc_stack.ss_sp); err |= __put_user(sas_ss_flags(regs->esp), &frame->uc.uc_stack.ss_flags); err |= __put_user(current->sas_ss_size, &frame->uc.uc_stack.ss_size); @@ -551,7 +549,7 @@ handle_signal(unsigned long sig, siginfo * want to handle. Thus you cannot kill init even with a SIGKILL even by * mistake. */ -int do_signal(struct pt_regs *regs, sigset_t *oldset) +int fastcall do_signal(struct pt_regs *regs, sigset_t *oldset) { siginfo_t info; int signr; --- diff/arch/i386/kernel/smp.c 2003-10-09 09:47:16.000000000 +0100 +++ source/arch/i386/kernel/smp.c 2004-03-01 15:42:15.000000000 +0000 @@ -327,10 +327,12 @@ asmlinkage void smp_invalidate_interrupt if (flush_mm == cpu_tlbstate[cpu].active_mm) { if (cpu_tlbstate[cpu].state == TLBSTATE_OK) { +#ifndef CONFIG_X86_SWITCH_PAGETABLES if (flush_va == FLUSH_ALL) local_flush_tlb(); else __flush_tlb_one(flush_va); +#endif } else leave_mm(cpu); } @@ -396,21 +398,6 @@ static void flush_tlb_others(cpumask_t c spin_unlock(&tlbstate_lock); } -void flush_tlb_current_task(void) -{ - struct mm_struct *mm = current->mm; - cpumask_t cpu_mask; - - preempt_disable(); - cpu_mask = mm->cpu_vm_mask; - cpu_clear(smp_processor_id(), cpu_mask); - - local_flush_tlb(); - if (!cpus_empty(cpu_mask)) - flush_tlb_others(cpu_mask, mm, FLUSH_ALL); - preempt_enable(); -} - void flush_tlb_mm (struct mm_struct * mm) { cpumask_t cpu_mask; @@ -442,7 +429,10 @@ void flush_tlb_page(struct vm_area_struc if (current->active_mm == mm) { if(current->mm) - __flush_tlb_one(va); +#ifndef CONFIG_X86_SWITCH_PAGETABLES + __flush_tlb_one(va) +#endif + ; else leave_mm(smp_processor_id()); } @@ -466,7 +456,17 @@ void flush_tlb_all(void) { on_each_cpu(do_flush_tlb_all, 0, 1, 1); } - +#ifdef CONFIG_KGDB +/* + * By using the NMI code instead of a vector we just sneak thru the + * word generator coming out with just what we want. AND it does + * not matter if clustered_apic_mode is set or not. + */ +void smp_send_nmi_allbutself(void) +{ + send_IPI_allbutself(APIC_DM_NMI); +} +#endif /* * this function sends a 'reschedule' IPI to another CPU. * it goes straight through and wastes no time serializing --- diff/arch/i386/kernel/smpboot.c 2004-03-01 14:36:04.000000000 +0000 +++ source/arch/i386/kernel/smpboot.c 2004-03-01 15:42:15.000000000 +0000 @@ -39,6 +39,7 @@ #include #include +#include #include #include #include @@ -934,7 +935,7 @@ static int boot_cpu_logical_apicid; /* Where the IO area was mapped on multiquad, always 0 otherwise */ void *xquad_portio; -int cpu_sibling_map[NR_CPUS] __cacheline_aligned; +cpumask_t cpu_sibling_map[NR_CPUS] __cacheline_aligned; static void __init smp_boot_cpus(unsigned int max_cpus) { @@ -953,6 +954,8 @@ static void __init smp_boot_cpus(unsigne current_thread_info()->cpu = 0; smp_tune_scheduling(); + cpus_clear(cpu_sibling_map[0]); + cpu_set(0, cpu_sibling_map[0]); /* * If we couldn't find an SMP configuration at boot time, @@ -1079,32 +1082,34 @@ static void __init smp_boot_cpus(unsigne Dprintk("Boot done.\n"); /* - * If Hyper-Threading is avaialble, construct cpu_sibling_map[], so - * that we can tell the sibling CPU efficiently. + * construct cpu_sibling_map[], so that we can tell sibling CPUs + * efficiently. */ - if (cpu_has_ht && smp_num_siblings > 1) { - for (cpu = 0; cpu < NR_CPUS; cpu++) - cpu_sibling_map[cpu] = NO_PROC_ID; - - for (cpu = 0; cpu < NR_CPUS; cpu++) { - int i; - if (!cpu_isset(cpu, cpu_callout_map)) - continue; + for (cpu = 0; cpu < NR_CPUS; cpu++) + cpus_clear(cpu_sibling_map[cpu]); + + for (cpu = 0; cpu < NR_CPUS; cpu++) { + int siblings = 0; + int i; + if (!cpu_isset(cpu, cpu_callout_map)) + continue; + if (smp_num_siblings > 1) { for (i = 0; i < NR_CPUS; i++) { - if (i == cpu || !cpu_isset(i, cpu_callout_map)) + if (!cpu_isset(i, cpu_callout_map)) continue; if (phys_proc_id[cpu] == phys_proc_id[i]) { - cpu_sibling_map[cpu] = i; - printk("cpu_sibling_map[%d] = %d\n", cpu, cpu_sibling_map[cpu]); - break; + siblings++; + cpu_set(i, cpu_sibling_map[cpu]); } } - if (cpu_sibling_map[cpu] == NO_PROC_ID) { - smp_num_siblings = 1; - printk(KERN_WARNING "WARNING: No sibling found for CPU %d.\n", cpu); - } + } else { + siblings++; + cpu_set(cpu, cpu_sibling_map[cpu]); } + + if (siblings != smp_num_siblings) + printk(KERN_WARNING "WARNING: %d siblings found for CPU%d, should be %d\n", siblings, cpu, smp_num_siblings); } smpboot_setup_io_apic(); @@ -1118,6 +1123,216 @@ static void __init smp_boot_cpus(unsigne synchronize_tsc_bp(); } +#ifdef CONFIG_SCHED_SMT +#ifdef CONFIG_NUMA +static struct sched_group sched_group_cpus[NR_CPUS]; +static struct sched_group sched_group_phys[NR_CPUS]; +static struct sched_group sched_group_nodes[MAX_NUMNODES]; +static DEFINE_PER_CPU(struct sched_domain, phys_domains); +static DEFINE_PER_CPU(struct sched_domain, node_domains); +__init void arch_init_sched_domains(void) +{ + int i; + struct sched_group *first_cpu = NULL, *last_cpu = NULL; + + /* Set up domains */ + for_each_cpu(i) { + struct sched_domain *cpu_domain = cpu_sched_domain(i); + struct sched_domain *phys_domain = &per_cpu(phys_domains, i); + struct sched_domain *node_domain = &per_cpu(node_domains, i); + int node = cpu_to_node(i); + cpumask_t nodemask = node_to_cpumask(node); + + *cpu_domain = SD_SIBLING_INIT; + cpu_domain->span = cpu_sibling_map[i]; + + *phys_domain = SD_CPU_INIT; + phys_domain->span = nodemask; + + *node_domain = SD_NODE_INIT; + node_domain->span = cpu_possible_map; + } + + /* Set up CPU (sibling) groups */ + for_each_cpu(i) { + struct sched_domain *cpu_domain = cpu_sched_domain(i); + int j; + first_cpu = last_cpu = NULL; + + if (i != first_cpu(cpu_domain->span)) + continue; + + for_each_cpu_mask(j, cpu_domain->span) { + struct sched_group *cpu = &sched_group_cpus[j]; + + cpu->cpumask = CPU_MASK_NONE; + cpu_set(j, cpu->cpumask); + cpu->cpu_power = SCHED_LOAD_SCALE; + + if (!first_cpu) + first_cpu = cpu; + if (last_cpu) + last_cpu->next = cpu; + last_cpu = cpu; + } + last_cpu->next = first_cpu; + } + + for (i = 0; i < MAX_NUMNODES; i++) { + int j; + cpumask_t nodemask; + struct sched_group *node = &sched_group_nodes[i]; + cpus_and(nodemask, node_to_cpumask(i), cpu_possible_map); + + if (cpus_empty(nodemask)) + continue; + + first_cpu = last_cpu = NULL; + /* Set up physical groups */ + for_each_cpu_mask(j, nodemask) { + struct sched_domain *cpu_domain = cpu_sched_domain(j); + struct sched_group *cpu = &sched_group_phys[j]; + + if (j != first_cpu(cpu_domain->span)) + continue; + + cpu->cpumask = cpu_domain->span; + /* + * Make each extra sibling increase power by 10% of + * the basic CPU. This is very arbitrary. + */ + cpu->cpu_power = SCHED_LOAD_SCALE + SCHED_LOAD_SCALE*(cpus_weight(cpu->cpumask)-1) / 10; + node->cpu_power += cpu->cpu_power; + + if (!first_cpu) + first_cpu = cpu; + if (last_cpu) + last_cpu->next = cpu; + last_cpu = cpu; + } + last_cpu->next = first_cpu; + } + + /* Set up nodes */ + first_cpu = last_cpu = NULL; + for (i = 0; i < MAX_NUMNODES; i++) { + struct sched_group *cpu = &sched_group_nodes[i]; + cpumask_t nodemask; + cpus_and(nodemask, node_to_cpumask(i), cpu_possible_map); + + if (cpus_empty(nodemask)) + continue; + + cpu->cpumask = nodemask; + /* ->cpu_power already setup */ + + if (!first_cpu) + first_cpu = cpu; + if (last_cpu) + last_cpu->next = cpu; + last_cpu = cpu; + } + last_cpu->next = first_cpu; + + mb(); + for_each_cpu(i) { + int node = cpu_to_node(i); + struct sched_domain *cpu_domain = cpu_sched_domain(i); + struct sched_domain *phys_domain = &per_cpu(phys_domains, i); + struct sched_domain *node_domain = &per_cpu(node_domains, i); + struct sched_group *cpu_group = &sched_group_cpus[i]; + struct sched_group *phys_group = &sched_group_phys[first_cpu(cpu_domain->span)]; + struct sched_group *node_group = &sched_group_nodes[node]; + + cpu_domain->parent = phys_domain; + phys_domain->parent = node_domain; + + node_domain->groups = node_group; + phys_domain->groups = phys_group; + cpu_domain->groups = cpu_group; + } +} +#else /* CONFIG_NUMA */ +static struct sched_group sched_group_cpus[NR_CPUS]; +static struct sched_group sched_group_phys[NR_CPUS]; +static DEFINE_PER_CPU(struct sched_domain, phys_domains); +__init void arch_init_sched_domains(void) +{ + int i; + struct sched_group *first_cpu = NULL, *last_cpu = NULL; + + /* Set up domains */ + for_each_cpu(i) { + struct sched_domain *cpu_domain = cpu_sched_domain(i); + struct sched_domain *phys_domain = &per_cpu(phys_domains, i); + + *cpu_domain = SD_SIBLING_INIT; + cpu_domain->span = cpu_sibling_map[i]; + + *phys_domain = SD_CPU_INIT; + phys_domain->span = cpu_possible_map; + } + + /* Set up CPU (sibling) groups */ + for_each_cpu(i) { + struct sched_domain *cpu_domain = cpu_sched_domain(i); + int j; + first_cpu = last_cpu = NULL; + + if (i != first_cpu(cpu_domain->span)) + continue; + + for_each_cpu_mask(j, cpu_domain->span) { + struct sched_group *cpu = &sched_group_cpus[j]; + + cpus_clear(cpu->cpumask); + cpu_set(j, cpu->cpumask); + cpu->cpu_power = SCHED_LOAD_SCALE; + + if (!first_cpu) + first_cpu = cpu; + if (last_cpu) + last_cpu->next = cpu; + last_cpu = cpu; + } + last_cpu->next = first_cpu; + } + + first_cpu = last_cpu = NULL; + /* Set up physical groups */ + for_each_cpu(i) { + struct sched_domain *cpu_domain = cpu_sched_domain(i); + struct sched_group *cpu = &sched_group_phys[i]; + + if (i != first_cpu(cpu_domain->span)) + continue; + + cpu->cpumask = cpu_domain->span; + /* See SMT+NUMA setup for comment */ + cpu->cpu_power = SCHED_LOAD_SCALE + SCHED_LOAD_SCALE*(cpus_weight(cpu->cpumask)-1) / 10; + + if (!first_cpu) + first_cpu = cpu; + if (last_cpu) + last_cpu->next = cpu; + last_cpu = cpu; + } + last_cpu->next = first_cpu; + + mb(); + for_each_cpu(i) { + struct sched_domain *cpu_domain = cpu_sched_domain(i); + struct sched_domain *phys_domain = &per_cpu(phys_domains, i); + struct sched_group *cpu_group = &sched_group_cpus[i]; + struct sched_group *phys_group = &sched_group_phys[first_cpu(cpu_domain->span)]; + cpu_domain->parent = phys_domain; + phys_domain->groups = phys_group; + cpu_domain->groups = cpu_group; + } +} +#endif /* CONFIG_NUMA */ +#endif /* CONFIG_SCHED_SMT */ + /* These are wrappers to interface to the new boot process. Someone who understands all this stuff should rewrite it properly. --RR 15/Jul/02 */ void __init smp_prepare_cpus(unsigned int max_cpus) --- diff/arch/i386/kernel/sysenter.c 2003-10-09 09:47:16.000000000 +0100 +++ source/arch/i386/kernel/sysenter.c 2004-03-01 15:42:15.000000000 +0000 @@ -18,13 +18,18 @@ #include #include #include +#include extern asmlinkage void sysenter_entry(void); void enable_sep_cpu(void *info) { int cpu = get_cpu(); +#ifdef CONFIG_X86_HIGH_ENTRY + struct tss_struct *tss = (struct tss_struct *) __fix_to_virt(FIX_TSS_0) + cpu; +#else struct tss_struct *tss = init_tss + cpu; +#endif tss->ss1 = __KERNEL_CS; tss->esp1 = sizeof(struct tss_struct) + (unsigned long) tss; --- diff/arch/i386/kernel/time.c 2004-02-09 10:36:07.000000000 +0000 +++ source/arch/i386/kernel/time.c 2004-03-01 15:42:15.000000000 +0000 @@ -346,7 +346,7 @@ static int time_init_device(void) { int error = sysdev_class_register(&pit_sysclass); if (!error) - error = sys_device_register(&device_i8253); + error = sysdev_register(&device_i8253); return error; } --- diff/arch/i386/kernel/trampoline.S 2003-06-09 14:18:17.000000000 +0100 +++ source/arch/i386/kernel/trampoline.S 2004-03-01 15:42:15.000000000 +0000 @@ -23,9 +23,13 @@ * and IP is zero. Thus, data addresses need to be absolute * (no relocation) and are taken with regard to r_base. * - * If you work on this file, check the object module with objdump - * --full-contents --reloc to make sure there are no relocation - * entries except for the gdt one.. + * If you work on this file, check the object module with + * objdump --reloc to make sure there are no relocation + * entries except for: + * + * TYPE VALUE + * R_386_32 startup_32_smp + * R_386_32 boot_gdt_table */ #include @@ -42,7 +46,6 @@ r_base = . mov %cs, %ax # Code and data in the same place mov %ax, %ds - mov $1, %bx # Flag an SMP trampoline cli # We should be safe anyway movl $0xA5A5A5A5, trampoline_data - r_base @@ -54,22 +57,18 @@ r_base = . xor %ax, %ax inc %ax # protected mode (PE) bit lmsw %ax # into protected mode - jmp flush_instr -flush_instr: - ljmpl $__BOOT_CS, $0x00100000 - # jump to startup_32 in arch/i386/kernel/head.S - -boot_idt: - .word 0 # idt limit = 0 - .word 0, 0 # idt base = 0L + # flush prefetch and jump to startup_32_smp in arch/i386/kernel/head.S + ljmpl $__BOOT_CS, $(startup_32_smp-__PAGE_OFFSET) -# -# NOTE: here we actually use CPU#0's GDT - but that is OK, we reload -# the proper GDT shortly after booting up the secondary CPUs. -# -ENTRY(boot_gdt) + # These need to be in the same 64K segment as the above; + # hence we don't use the boot_gdt_descr defined in head.S +boot_gdt: .word __BOOT_DS + 7 # gdt limit - .long boot_gdt_table-__PAGE_OFFSET # gdt base = gdt (first SMP CPU) + .long boot_gdt_table-__PAGE_OFFSET # gdt base + +boot_idt: + .word 0 # idt limit = 0 + .long 0 # idt base = 0L .globl trampoline_end trampoline_end: --- diff/arch/i386/kernel/traps.c 2004-03-01 14:36:04.000000000 +0000 +++ source/arch/i386/kernel/traps.c 2004-03-01 15:42:15.000000000 +0000 @@ -54,12 +54,8 @@ #include "mach_traps.h" -asmlinkage int system_call(void); -asmlinkage void lcall7(void); -asmlinkage void lcall27(void); - -struct desc_struct default_ldt[] = { { 0, 0 }, { 0, 0 }, { 0, 0 }, - { 0, 0 }, { 0, 0 } }; +struct desc_struct default_ldt[] __attribute__((__section__(".data.default_ldt"))) = { { 0, 0 }, { 0, 0 }, { 0, 0 }, { 0, 0 }, { 0, 0 } }; +struct page *default_ldt_page; /* Do we ignore FPU interrupts ? */ char ignore_fpu_irq = 0; @@ -91,6 +87,41 @@ asmlinkage void alignment_check(void); asmlinkage void spurious_interrupt_bug(void); asmlinkage void machine_check(void); +#ifdef CONFIG_KGDB +extern void sysenter_entry(void); +#include +#include +void set_intr_gate(unsigned int n, void *addr); +static void set_intr_usr_gate(unsigned int n, void *addr); +/* + * Should be able to call this breakpoint() very early in + * bring up. Just hard code the call where needed. + * The breakpoint() code is here because set_?_gate() functions + * are local (static) to trap.c. They need be done only once, + * but it does not hurt to do them over. + */ +void breakpoint(void) +{ + init_entry_mappings(); + set_intr_usr_gate(3,&int3); /* disable ints on trap */ + set_intr_gate(1,&debug); + set_intr_gate(14,&page_fault); + + BREAKPOINT; +} +#define CHK_REMOTE_DEBUG(trapnr,signr,error_code,regs,after) \ + { \ + if (!user_mode(regs) ) \ + { \ + kgdb_handle_exception(trapnr, signr, error_code, regs); \ + after; \ + } else if ((trapnr == 3) && (regs->eflags &0x200)) local_irq_enable(); \ + } +#else +#define CHK_REMOTE_DEBUG(trapnr,signr,error_code,regs,after) +#endif + + static int kstack_depth_to_print = 24; void show_trace(struct task_struct *task, unsigned long * stack) @@ -175,8 +206,9 @@ void show_registers(struct pt_regs *regs ss = regs->xss & 0xffff; } print_modules(); - printk("CPU: %d\nEIP: %04x:[<%08lx>] %s\nEFLAGS: %08lx\n", - smp_processor_id(), 0xffff & regs->xcs, regs->eip, print_tainted(), regs->eflags); + printk("CPU: %d\nEIP: %04x:[<%08lx>] %s VLI\nEFLAGS: %08lx\n", + smp_processor_id(), 0xffff & regs->xcs, + regs->eip, print_tainted(), regs->eflags); print_symbol("EIP is at %s\n", regs->eip); printk("eax: %08lx ebx: %08lx ecx: %08lx edx: %08lx\n", @@ -192,23 +224,27 @@ void show_registers(struct pt_regs *regs * time of the fault.. */ if (in_kernel) { + u8 *eip; printk("\nStack: "); show_stack(NULL, (unsigned long*)esp); printk("Code: "); - if(regs->eip < PAGE_OFFSET) - goto bad; - for(i=0;i<20;i++) - { - unsigned char c; - if(__get_user(c, &((unsigned char*)regs->eip)[i])) { -bad: + eip = (u8 *)regs->eip - 43; + for (i = 0; i < 64; i++, eip++) { + unsigned char c = 0xff; + + if ((user_mode(regs) && get_user(c, eip)) || + (!user_mode(regs) && __direct_get_user(c, eip))) { + printk(" Bad EIP value."); break; } - printk("%02x ", c); + if (eip == (u8 *)regs->eip) + printk("<%02x> ", c); + else + printk("%02x ", c); } } printk("\n"); @@ -276,6 +312,15 @@ void die(const char * str, struct pt_reg #endif if (nl) printk("\n"); +#ifdef CONFIG_KGDB + /* This is about the only place we want to go to kgdb even if in + * user mode. But we must go in via a trap so within kgdb we will + * always be in kernel mode. + */ + if (user_mode(regs)) + BREAKPOINT; +#endif + CHK_REMOTE_DEBUG(0,SIGTRAP,err,regs,) show_registers(regs); bust_spinlocks(0); spin_unlock_irq(&die_lock); @@ -345,6 +390,7 @@ static inline void do_trap(int trapnr, i #define DO_ERROR(trapnr, signr, str, name) \ asmlinkage void do_##name(struct pt_regs * regs, long error_code) \ { \ + CHK_REMOTE_DEBUG(trapnr,signr,error_code,regs,)\ do_trap(trapnr, signr, str, 0, regs, error_code, NULL); \ } @@ -362,7 +408,9 @@ asmlinkage void do_##name(struct pt_regs #define DO_VM86_ERROR(trapnr, signr, str, name) \ asmlinkage void do_##name(struct pt_regs * regs, long error_code) \ { \ + CHK_REMOTE_DEBUG(trapnr, signr, error_code,regs, return)\ do_trap(trapnr, signr, str, 1, regs, error_code, NULL); \ + return; \ } #define DO_VM86_ERROR_INFO(trapnr, signr, str, name, sicode, siaddr) \ @@ -409,8 +457,10 @@ gp_in_vm86: return; gp_in_kernel: - if (!fixup_exception(regs)) + if (!fixup_exception(regs)){ + CHK_REMOTE_DEBUG(13,SIGSEGV,error_code,regs,) die("general protection fault", regs, error_code); + } } static void mem_parity_error(unsigned char reason, struct pt_regs * regs) @@ -549,10 +599,18 @@ asmlinkage void do_debug(struct pt_regs if (regs->eflags & X86_EFLAGS_IF) local_irq_enable(); - /* Mask out spurious debug traps due to lazy DR7 setting */ + /* + * Mask out spurious debug traps due to lazy DR7 setting or + * due to 4G/4G kernel mode: + */ if (condition & (DR_TRAP0|DR_TRAP1|DR_TRAP2|DR_TRAP3)) { if (!tsk->thread.debugreg[7]) goto clear_dr7; + if (!user_mode(regs)) { + // restore upon return-to-userspace: + set_thread_flag(TIF_DB7); + goto clear_dr7; + } } if (regs->eflags & VM_MASK) @@ -572,8 +630,18 @@ asmlinkage void do_debug(struct pt_regs * allowing programs to debug themselves without the ptrace() * interface. */ +#ifdef CONFIG_KGDB + /* + * I think this is the only "real" case of a TF in the kernel + * that really belongs to user space. Others are + * "Ours all ours!" + */ + if (((regs->xcs & 3) == 0) && ((void *)regs->eip == sysenter_entry)) + goto clear_TF_reenable; +#else if ((regs->xcs & 3) == 0) goto clear_TF_reenable; +#endif if ((tsk->ptrace & (PT_DTRACE|PT_PTRACED)) == PT_DTRACE) goto clear_TF; } @@ -585,6 +653,17 @@ asmlinkage void do_debug(struct pt_regs info.si_errno = 0; info.si_code = TRAP_BRKPT; +#ifdef CONFIG_KGDB + /* + * If this is a kernel mode trap, we need to reset db7 to allow us + * to continue sanely ALSO skip the signal delivery + */ + if ((regs->xcs & 3) == 0) + goto clear_dr7; + + /* if not kernel, allow ints but only if they were on */ + if ( regs->eflags & 0x200) local_irq_enable(); +#endif /* If this is a kernel mode trap, save the user PC on entry to * the kernel, that's what the debugger can make sense of. */ @@ -599,6 +678,7 @@ clear_dr7: __asm__("movl %0,%%db7" : /* no output */ : "r" (0)); + CHK_REMOTE_DEBUG(1,SIGTRAP,error_code,regs,) return; debug_vm86: @@ -794,19 +874,53 @@ asmlinkage void math_emulate(long arg) #endif /* CONFIG_MATH_EMULATION */ -#ifdef CONFIG_X86_F00F_BUG -void __init trap_init_f00f_bug(void) +void __init trap_init_virtual_IDT(void) { - __set_fixmap(FIX_F00F_IDT, __pa(&idt_table), PAGE_KERNEL_RO); - /* - * Update the IDT descriptor and reload the IDT so that - * it uses the read-only mapped virtual address. + * "idt" is magic - it overlaps the idt_descr + * variable so that updating idt will automatically + * update the idt descriptor.. */ - idt_descr.address = fix_to_virt(FIX_F00F_IDT); + __set_fixmap(FIX_IDT, __pa(&idt_table), PAGE_KERNEL_RO); + idt_descr.address = __fix_to_virt(FIX_IDT); + __asm__ __volatile__("lidt %0" : : "m" (idt_descr)); } -#endif + +void __init trap_init_virtual_GDT(void) +{ + int cpu = smp_processor_id(); + struct Xgt_desc_struct *gdt_desc = cpu_gdt_descr + cpu; + struct Xgt_desc_struct tmp_desc = {0, 0}; + struct tss_struct * t; + + __asm__ __volatile__("sgdt %0": "=m" (tmp_desc): :"memory"); + +#ifdef CONFIG_X86_HIGH_ENTRY + if (!cpu) { + __set_fixmap(FIX_GDT_0, __pa(cpu_gdt_table), PAGE_KERNEL); + __set_fixmap(FIX_GDT_1, __pa(cpu_gdt_table) + PAGE_SIZE, PAGE_KERNEL); + __set_fixmap(FIX_TSS_0, __pa(init_tss), PAGE_KERNEL); + __set_fixmap(FIX_TSS_1, __pa(init_tss) + 1*PAGE_SIZE, PAGE_KERNEL); + __set_fixmap(FIX_TSS_2, __pa(init_tss) + 2*PAGE_SIZE, PAGE_KERNEL); + __set_fixmap(FIX_TSS_3, __pa(init_tss) + 3*PAGE_SIZE, PAGE_KERNEL); + } + + gdt_desc->address = __fix_to_virt(FIX_GDT_0) + sizeof(cpu_gdt_table[0]) * cpu; +#else + gdt_desc->address = (unsigned long)cpu_gdt_table[cpu]; +#endif + __asm__ __volatile__("lgdt %0": "=m" (*gdt_desc)); + +#ifdef CONFIG_X86_HIGH_ENTRY + t = (struct tss_struct *) __fix_to_virt(FIX_TSS_0) + cpu; +#else + t = init_tss + cpu; +#endif + set_tss_desc(cpu, t); + cpu_gdt_table[cpu][GDT_ENTRY_TSS].b &= 0xfffffdff; + load_TR_desc(); +} #define _set_gate(gate_addr,type,dpl,addr,seg) \ do { \ @@ -833,20 +947,26 @@ void set_intr_gate(unsigned int n, void _set_gate(idt_table+n,14,0,addr,__KERNEL_CS); } -static void __init set_trap_gate(unsigned int n, void *addr) +void __init set_trap_gate(unsigned int n, void *addr) { _set_gate(idt_table+n,15,0,addr,__KERNEL_CS); } -static void __init set_system_gate(unsigned int n, void *addr) +void __init set_system_gate(unsigned int n, void *addr) { _set_gate(idt_table+n,15,3,addr,__KERNEL_CS); } -static void __init set_call_gate(void *a, void *addr) +void __init set_call_gate(void *a, void *addr) { _set_gate(a,12,3,addr,__KERNEL_CS); } +#ifdef CONFIG_KGDB +void set_intr_usr_gate(unsigned int n, void *addr) +{ + _set_gate(idt_table+n,14,3,addr,__KERNEL_CS); +} +#endif static void __init set_task_gate(unsigned int n, unsigned int gdt_entry) { @@ -865,11 +985,16 @@ void __init trap_init(void) #ifdef CONFIG_X86_LOCAL_APIC init_apic_mappings(); #endif + init_entry_mappings(); set_trap_gate(0,÷_error); set_intr_gate(1,&debug); set_intr_gate(2,&nmi); +#ifndef CONFIG_KGDB set_system_gate(3,&int3); /* int3-5 can be called from all */ +#else + set_intr_usr_gate(3,&int3); /* int3-5 can be called from all */ +#endif set_system_gate(4,&overflow); set_system_gate(5,&bounds); set_trap_gate(6,&invalid_op); --- diff/arch/i386/kernel/vm86.c 2004-01-19 10:22:55.000000000 +0000 +++ source/arch/i386/kernel/vm86.c 2004-03-01 15:42:15.000000000 +0000 @@ -95,7 +95,7 @@ #define VM86_REGS_SIZE2 (sizeof(struct kernel_vm86_regs) - VM86_REGS_SIZE1) struct pt_regs * FASTCALL(save_v86_state(struct kernel_vm86_regs * regs)); -struct pt_regs * save_v86_state(struct kernel_vm86_regs * regs) +struct pt_regs * fastcall save_v86_state(struct kernel_vm86_regs * regs) { struct tss_struct *tss; struct pt_regs *ret; @@ -125,7 +125,7 @@ struct pt_regs * save_v86_state(struct k tss = init_tss + get_cpu(); current->thread.esp0 = current->thread.saved_esp0; current->thread.sysenter_cs = __KERNEL_CS; - load_esp0(tss, ¤t->thread); + load_virtual_esp0(tss, current); current->thread.saved_esp0 = 0; put_cpu(); @@ -305,7 +305,7 @@ static void do_sys_vm86(struct kernel_vm tsk->thread.esp0 = (unsigned long) &info->VM86_TSS_ESP0; if (cpu_has_sep) tsk->thread.sysenter_cs = 0; - load_esp0(tss, &tsk->thread); + load_virtual_esp0(tss, tsk); put_cpu(); tsk->thread.screen_bitmap = info->screen_bitmap; --- diff/arch/i386/kernel/vmlinux.lds.S 2003-10-09 09:47:16.000000000 +0100 +++ source/arch/i386/kernel/vmlinux.lds.S 2004-03-01 15:42:15.000000000 +0000 @@ -3,6 +3,9 @@ */ #include +#include +#include +#include OUTPUT_FORMAT("elf32-i386", "elf32-i386", "elf32-i386") OUTPUT_ARCH(i386) @@ -10,7 +13,7 @@ ENTRY(startup_32) jiffies = jiffies_64; SECTIONS { - . = 0xC0000000 + 0x100000; + . = __PAGE_OFFSET + 0x100000; /* read-only */ _text = .; /* Text and read-only data */ .text : { @@ -19,6 +22,19 @@ SECTIONS *(.gnu.warning) } = 0x9090 +#ifdef CONFIG_X86_4G + . = ALIGN(PAGE_SIZE_asm); + __entry_tramp_start = .; + . = FIX_ENTRY_TRAMPOLINE_0_addr; + __start___entry_text = .; + .entry.text : AT (__entry_tramp_start) { *(.entry.text) } + __entry_tramp_end = __entry_tramp_start + SIZEOF(.entry.text); + . = __entry_tramp_end; + . = ALIGN(PAGE_SIZE_asm); +#else + .entry.text : { *(.entry.text) } +#endif + _etext = .; /* End of text section */ . = ALIGN(16); /* Exception table */ @@ -34,15 +50,12 @@ SECTIONS CONSTRUCTORS } - . = ALIGN(4096); + . = ALIGN(PAGE_SIZE_asm); __nosave_begin = .; .data_nosave : { *(.data.nosave) } - . = ALIGN(4096); + . = ALIGN(PAGE_SIZE_asm); __nosave_end = .; - . = ALIGN(4096); - .data.page_aligned : { *(.data.idt) } - . = ALIGN(32); .data.cacheline_aligned : { *(.data.cacheline_aligned) } @@ -52,7 +65,7 @@ SECTIONS .data.init_task : { *(.data.init_task) } /* will be freed after init */ - . = ALIGN(4096); /* Init code and data */ + . = ALIGN(PAGE_SIZE_asm); /* Init code and data */ __init_begin = .; .init.text : { _sinittext = .; @@ -91,7 +104,7 @@ SECTIONS from .altinstructions and .eh_frame */ .exit.text : { *(.exit.text) } .exit.data : { *(.exit.data) } - . = ALIGN(4096); + . = ALIGN(PAGE_SIZE_asm); __initramfs_start = .; .init.ramfs : { *(.init.ramfs) } __initramfs_end = .; @@ -99,16 +112,33 @@ SECTIONS __per_cpu_start = .; .data.percpu : { *(.data.percpu) } __per_cpu_end = .; - . = ALIGN(4096); + . = ALIGN(PAGE_SIZE_asm); __init_end = .; /* freed after init ends here */ - + + . = ALIGN(PAGE_SIZE_asm); + .data.page_aligned_tss : { *(.data.tss) } + + . = ALIGN(PAGE_SIZE_asm); + .data.page_aligned_default_ldt : { *(.data.default_ldt) } + + . = ALIGN(PAGE_SIZE_asm); + .data.page_aligned_idt : { *(.data.idt) } + + . = ALIGN(PAGE_SIZE_asm); + .data.page_aligned_gdt : { *(.data.gdt) } + __bss_start = .; /* BSS */ .bss : { *(.bss) } + . = ALIGN(4); __bss_stop = .; _end = . ; + /* This is where the kernel creates the early boot page tables */ + . = ALIGN(4096); + pg0 = .; + /* Sections to be discarded */ /DISCARD/ : { *(.exitcall.exit) @@ -122,4 +152,6 @@ SECTIONS .stab.index 0 : { *(.stab.index) } .stab.indexstr 0 : { *(.stab.indexstr) } .comment 0 : { *(.comment) } + + } --- diff/arch/i386/kernel/vsyscall-sysenter.S 2003-10-09 09:47:16.000000000 +0100 +++ source/arch/i386/kernel/vsyscall-sysenter.S 2004-03-01 15:42:15.000000000 +0000 @@ -7,6 +7,11 @@ .type __kernel_vsyscall,@function __kernel_vsyscall: .LSTART_vsyscall: + cmpl $192, %eax + jne 1f + int $0x80 + ret +1: push %ecx .Lpush_ecx: push %edx --- diff/arch/i386/kernel/vsyscall.lds 2003-10-09 09:47:16.000000000 +0100 +++ source/arch/i386/kernel/vsyscall.lds 2004-03-01 15:42:15.000000000 +0000 @@ -5,7 +5,7 @@ */ /* This must match . */ -VSYSCALL_BASE = 0xffffe000; +VSYSCALL_BASE = 0xffffd000; SECTIONS { --- diff/arch/i386/lib/Makefile 2003-10-09 09:47:16.000000000 +0100 +++ source/arch/i386/lib/Makefile 2004-03-01 15:42:15.000000000 +0000 @@ -9,4 +9,4 @@ lib-y = checksum.o delay.o \ lib-$(CONFIG_X86_USE_3DNOW) += mmx.o lib-$(CONFIG_HAVE_DEC_LOCK) += dec_and_lock.o -lib-$(CONFIG_DEBUG_IOVIRT) += iodebug.o +lib-$(CONFIG_KGDB) += kgdb_serial.o --- diff/arch/i386/lib/checksum.S 2003-10-09 09:47:16.000000000 +0100 +++ source/arch/i386/lib/checksum.S 2004-03-01 15:42:15.000000000 +0000 @@ -280,14 +280,14 @@ unsigned int csum_partial_copy_generic ( .previous .align 4 -.globl csum_partial_copy_generic +.globl direct_csum_partial_copy_generic #ifndef CONFIG_X86_USE_PPRO_CHECKSUM #define ARGBASE 16 #define FP 12 -csum_partial_copy_generic: +direct_csum_partial_copy_generic: subl $4,%esp pushl %edi pushl %esi @@ -422,7 +422,7 @@ DST( movb %cl, (%edi) ) #define ARGBASE 12 -csum_partial_copy_generic: +direct_csum_partial_copy_generic: pushl %ebx pushl %edi pushl %esi --- diff/arch/i386/lib/dec_and_lock.c 2003-10-09 09:47:16.000000000 +0100 +++ source/arch/i386/lib/dec_and_lock.c 2004-03-01 15:42:15.000000000 +0000 @@ -10,6 +10,7 @@ #include #include +#ifndef ATOMIC_DEC_AND_LOCK int atomic_dec_and_lock(atomic_t *atomic, spinlock_t *lock) { int counter; @@ -38,3 +39,5 @@ slow_path: spin_unlock(lock); return 0; } +#endif + --- diff/arch/i386/lib/getuser.S 2003-10-09 09:47:16.000000000 +0100 +++ source/arch/i386/lib/getuser.S 2004-03-01 15:42:15.000000000 +0000 @@ -9,6 +9,7 @@ * return value. */ #include +#include /* @@ -28,7 +29,7 @@ .globl __get_user_1 __get_user_1: GET_THREAD_INFO(%edx) - cmpl TI_ADDR_LIMIT(%edx),%eax + cmpl TI_addr_limit(%edx),%eax jae bad_get_user 1: movzbl (%eax),%edx xorl %eax,%eax @@ -40,7 +41,7 @@ __get_user_2: addl $1,%eax jc bad_get_user GET_THREAD_INFO(%edx) - cmpl TI_ADDR_LIMIT(%edx),%eax + cmpl TI_addr_limit(%edx),%eax jae bad_get_user 2: movzwl -1(%eax),%edx xorl %eax,%eax @@ -52,7 +53,7 @@ __get_user_4: addl $3,%eax jc bad_get_user GET_THREAD_INFO(%edx) - cmpl TI_ADDR_LIMIT(%edx),%eax + cmpl TI_addr_limit(%edx),%eax jae bad_get_user 3: movl -3(%eax),%edx xorl %eax,%eax --- diff/arch/i386/lib/usercopy.c 2004-01-19 10:22:55.000000000 +0000 +++ source/arch/i386/lib/usercopy.c 2004-03-01 15:42:15.000000000 +0000 @@ -76,7 +76,7 @@ do { \ * and returns @count. */ long -__strncpy_from_user(char *dst, const char __user *src, long count) +__direct_strncpy_from_user(char *dst, const char __user *src, long count) { long res; __do_strncpy_from_user(dst, src, count, res); @@ -102,7 +102,7 @@ __strncpy_from_user(char *dst, const cha * and returns @count. */ long -strncpy_from_user(char *dst, const char __user *src, long count) +direct_strncpy_from_user(char *dst, const char __user *src, long count) { long res = -EFAULT; if (access_ok(VERIFY_READ, src, 1)) @@ -147,7 +147,7 @@ do { \ * On success, this will be zero. */ unsigned long -clear_user(void __user *to, unsigned long n) +direct_clear_user(void __user *to, unsigned long n) { might_sleep(); if (access_ok(VERIFY_WRITE, to, n)) @@ -167,7 +167,7 @@ clear_user(void __user *to, unsigned lon * On success, this will be zero. */ unsigned long -__clear_user(void __user *to, unsigned long n) +__direct_clear_user(void __user *to, unsigned long n) { __do_clear_user(to, n); return n; @@ -184,7 +184,7 @@ __clear_user(void __user *to, unsigned l * On exception, returns 0. * If the string is too long, returns a value greater than @n. */ -long strnlen_user(const char __user *s, long n) +long direct_strnlen_user(const char __user *s, long n) { unsigned long mask = -__addr_ok(s); unsigned long res, tmp; @@ -575,3 +575,4 @@ unsigned long __copy_from_user_ll(void * n = __copy_user_zeroing_intel(to, (const void *) from, n); return n; } + --- diff/arch/i386/mach-voyager/Makefile 2003-02-13 11:46:49.000000000 +0000 +++ source/arch/i386/mach-voyager/Makefile 2004-03-01 15:42:15.000000000 +0000 @@ -1,11 +1,6 @@ # # Makefile for the linux kernel. # -# Note! Dependencies are done automagically by 'make dep', which also -# removes any old dependencies. DON'T put your own dependencies here -# unless it's something special (ie not a .c file). -# -# Note 2! The CFLAGS definitions are now in the main makefile... EXTRA_CFLAGS += -I../kernel obj-y := setup.o voyager_basic.o voyager_thread.o --- diff/arch/i386/math-emu/fpu_system.h 2003-10-09 09:47:16.000000000 +0100 +++ source/arch/i386/math-emu/fpu_system.h 2004-03-01 15:42:15.000000000 +0000 @@ -15,6 +15,7 @@ #include #include #include +#include /* This sets the pointer FPU_info to point to the argument part of the stack frame of math_emulate() */ @@ -22,7 +23,7 @@ /* s is always from a cpu register, and the cpu does bounds checking * during register load --> no further bounds checks needed */ -#define LDT_DESCRIPTOR(s) (((struct desc_struct *)current->mm->context.ldt)[(s) >> 3]) +#define LDT_DESCRIPTOR(s) (((struct desc_struct *)__kmap_atomic_vaddr(KM_LDT_PAGE0))[(s) >> 3]) #define SEG_D_SIZE(x) ((x).b & (3 << 21)) #define SEG_G_BIT(x) ((x).b & (1 << 23)) #define SEG_GRANULARITY(x) (((x).b & (1 << 23)) ? 4096 : 1) --- diff/arch/i386/mm/discontig.c 2003-09-30 15:46:11.000000000 +0100 +++ source/arch/i386/mm/discontig.c 2004-03-01 15:42:15.000000000 +0000 @@ -66,7 +66,7 @@ extern void find_max_pfn(void); extern void one_highpage_init(struct page *, int, int); extern struct e820map e820; -extern char _end; +extern unsigned long init_pg_tables_end; extern unsigned long highend_pfn, highstart_pfn; extern unsigned long max_low_pfn; extern unsigned long totalram_pages; @@ -237,7 +237,7 @@ unsigned long __init setup_memory(void) reserve_pages = calculate_numa_remap_pages(); /* partially used pages are not usable - thus round upwards */ - system_start_pfn = min_low_pfn = PFN_UP(__pa(&_end)); + system_start_pfn = min_low_pfn = PFN_UP(init_pg_tables_end); find_max_pfn(); system_max_low_pfn = max_low_pfn = find_max_low_pfn(); --- diff/arch/i386/mm/fault.c 2003-12-19 09:51:11.000000000 +0000 +++ source/arch/i386/mm/fault.c 2004-03-01 15:42:15.000000000 +0000 @@ -27,6 +27,7 @@ #include #include #include +#include extern void die(const char *,struct pt_regs *,long); @@ -104,8 +105,17 @@ static inline unsigned long get_segment_ if (seg & (1<<2)) { /* Must lock the LDT while reading it. */ down(¤t->mm->context.sem); +#if 1 + /* horrible hack for 4/4 disabled kernels. + I'm not quite sure what the TLB flush is good for, + it's mindlessly copied from the read_ldt code */ + __flush_tlb_global(); + desc = kmap(current->mm->context.ldt_pages[(seg&~7)/PAGE_SIZE]); + desc = (void *)desc + ((seg & ~7) % PAGE_SIZE); +#else desc = current->mm->context.ldt; desc = (void *)desc + (seg & ~7); +#endif } else { /* Must disable preemption while reading the GDT. */ desc = (u32 *)&cpu_gdt_table[get_cpu()]; @@ -118,6 +128,9 @@ static inline unsigned long get_segment_ (desc[1] & 0xff000000); if (seg & (1<<2)) { +#if 1 + kunmap((void *)((unsigned long)desc & PAGE_MASK)); +#endif up(¤t->mm->context.sem); } else put_cpu(); @@ -243,6 +256,19 @@ asmlinkage void do_page_fault(struct pt_ * (error_code & 4) == 0, and that the fault was not a * protection error (error_code & 1) == 0. */ +#ifdef CONFIG_X86_4G + /* + * On 4/4 all kernels faults are either bugs, vmalloc or prefetch + */ + if (unlikely((regs->xcs & 3) == 0)) { + if (error_code & 3) + goto bad_area_nosemaphore; + + /* If it's vm86 fall through */ + if (!(regs->eflags & VM_MASK)) + goto vmalloc_fault; + } +#else if (unlikely(address >= TASK_SIZE)) { if (!(error_code & 5)) goto vmalloc_fault; @@ -252,6 +278,7 @@ asmlinkage void do_page_fault(struct pt_ */ goto bad_area_nosemaphore; } +#endif mm = tsk->mm; @@ -403,6 +430,12 @@ no_context: * Oops. The kernel tried to access some bad page. We'll have to * terminate things with extreme prejudice. */ +#ifdef CONFIG_KGDB + if (!user_mode(regs)){ + kgdb_handle_exception(14,SIGBUS, error_code, regs); + return; + } +#endif bust_spinlocks(1); --- diff/arch/i386/mm/hugetlbpage.c 2004-01-19 10:22:55.000000000 +0000 +++ source/arch/i386/mm/hugetlbpage.c 2004-03-01 15:42:15.000000000 +0000 @@ -61,6 +61,27 @@ static struct page *alloc_fresh_huge_pag static void free_huge_page(struct page *page); +#ifdef CONFIG_NUMA + +static inline void huge_inc_rss(struct mm_struct *mm, struct page *page) +{ + mm->rss += (HPAGE_SIZE / PAGE_SIZE); + mm->pernode_rss[page_nodenum(page)] += (HPAGE_SIZE / PAGE_SIZE); +} + +static inline void huge_dec_rss(struct mm_struct *mm, struct page *page) +{ + mm->rss -= (HPAGE_SIZE / PAGE_SIZE); + mm->pernode_rss[page_nodenum(page)] -= (HPAGE_SIZE / PAGE_SIZE); +} + +#else /* !CONFIG_NUMA */ + +#define huge_inc_rss(mm, page) ((mm)->rss += (HPAGE_SIZE / PAGE_SIZE)) +#define huge_dec_rss(mm, page) ((mm)->rss -= (HPAGE_SIZE / PAGE_SIZE)) + +#endif /* CONFIG_NUMA */ + static struct page *alloc_hugetlb_page(void) { int i; @@ -105,7 +126,7 @@ static void set_huge_pte(struct mm_struc { pte_t entry; - mm->rss += (HPAGE_SIZE / PAGE_SIZE); + huge_inc_rss(mm, page); if (write_access) { entry = pte_mkwrite(pte_mkdirty(mk_pte(page, vma->vm_page_prot))); @@ -145,7 +166,7 @@ int copy_hugetlb_page_range(struct mm_st ptepage = pte_page(entry); get_page(ptepage); set_pte(dst_pte, entry); - dst->rss += (HPAGE_SIZE / PAGE_SIZE); + huge_inc_rss(dst, ptepage); addr += HPAGE_SIZE; } return 0; @@ -314,8 +335,8 @@ void unmap_hugepage_range(struct vm_area page = pte_page(*pte); huge_page_release(page); pte_clear(pte); + huge_dec_rss(mm, page); } - mm->rss -= (end - start) >> PAGE_SHIFT; flush_tlb_range(vma, start, end); } --- diff/arch/i386/mm/init.c 2004-01-19 10:22:55.000000000 +0000 +++ source/arch/i386/mm/init.c 2004-03-01 15:42:15.000000000 +0000 @@ -40,125 +40,13 @@ #include #include #include +#include DEFINE_PER_CPU(struct mmu_gather, mmu_gathers); unsigned long highstart_pfn, highend_pfn; static int do_test_wp_bit(void); -/* - * Creates a middle page table and puts a pointer to it in the - * given global directory entry. This only returns the gd entry - * in non-PAE compilation mode, since the middle layer is folded. - */ -static pmd_t * __init one_md_table_init(pgd_t *pgd) -{ - pmd_t *pmd_table; - -#ifdef CONFIG_X86_PAE - pmd_table = (pmd_t *) alloc_bootmem_low_pages(PAGE_SIZE); - set_pgd(pgd, __pgd(__pa(pmd_table) | _PAGE_PRESENT)); - if (pmd_table != pmd_offset(pgd, 0)) - BUG(); -#else - pmd_table = pmd_offset(pgd, 0); -#endif - - return pmd_table; -} - -/* - * Create a page table and place a pointer to it in a middle page - * directory entry. - */ -static pte_t * __init one_page_table_init(pmd_t *pmd) -{ - if (pmd_none(*pmd)) { - pte_t *page_table = (pte_t *) alloc_bootmem_low_pages(PAGE_SIZE); - set_pmd(pmd, __pmd(__pa(page_table) | _PAGE_TABLE)); - if (page_table != pte_offset_kernel(pmd, 0)) - BUG(); - - return page_table; - } - - return pte_offset_kernel(pmd, 0); -} - -/* - * This function initializes a certain range of kernel virtual memory - * with new bootmem page tables, everywhere page tables are missing in - * the given range. - */ - -/* - * NOTE: The pagetables are allocated contiguous on the physical space - * so we can cache the place of the first one and move around without - * checking the pgd every time. - */ -static void __init page_table_range_init (unsigned long start, unsigned long end, pgd_t *pgd_base) -{ - pgd_t *pgd; - pmd_t *pmd; - int pgd_idx, pmd_idx; - unsigned long vaddr; - - vaddr = start; - pgd_idx = pgd_index(vaddr); - pmd_idx = pmd_index(vaddr); - pgd = pgd_base + pgd_idx; - - for ( ; (pgd_idx < PTRS_PER_PGD) && (vaddr != end); pgd++, pgd_idx++) { - if (pgd_none(*pgd)) - one_md_table_init(pgd); - - pmd = pmd_offset(pgd, vaddr); - for (; (pmd_idx < PTRS_PER_PMD) && (vaddr != end); pmd++, pmd_idx++) { - if (pmd_none(*pmd)) - one_page_table_init(pmd); - - vaddr += PMD_SIZE; - } - pmd_idx = 0; - } -} - -/* - * This maps the physical memory to kernel virtual address space, a total - * of max_low_pfn pages, by creating page tables starting from address - * PAGE_OFFSET. - */ -static void __init kernel_physical_mapping_init(pgd_t *pgd_base) -{ - unsigned long pfn; - pgd_t *pgd; - pmd_t *pmd; - pte_t *pte; - int pgd_idx, pmd_idx, pte_ofs; - - pgd_idx = pgd_index(PAGE_OFFSET); - pgd = pgd_base + pgd_idx; - pfn = 0; - - for (; pgd_idx < PTRS_PER_PGD; pgd++, pgd_idx++) { - pmd = one_md_table_init(pgd); - if (pfn >= max_low_pfn) - continue; - for (pmd_idx = 0; pmd_idx < PTRS_PER_PMD && pfn < max_low_pfn; pmd++, pmd_idx++) { - /* Map with big pages if possible, otherwise create normal page tables. */ - if (cpu_has_pse) { - set_pmd(pmd, pfn_pmd(pfn, PAGE_KERNEL_LARGE)); - pfn += PTRS_PER_PTE; - } else { - pte = one_page_table_init(pmd); - - for (pte_ofs = 0; pte_ofs < PTRS_PER_PTE && pfn < max_low_pfn; pte++, pfn++, pte_ofs++) - set_pte(pte, pfn_pte(pfn, PAGE_KERNEL)); - } - } - } -} - static inline int page_kills_ppro(unsigned long pagenr) { if (pagenr >= 0x70000 && pagenr <= 0x7003F) @@ -206,11 +94,8 @@ static inline int page_is_ram(unsigned l return 0; } -#ifdef CONFIG_HIGHMEM pte_t *kmap_pte; -pgprot_t kmap_prot; -EXPORT_SYMBOL(kmap_prot); EXPORT_SYMBOL(kmap_pte); #define kmap_get_fixmap_pte(vaddr) \ @@ -218,29 +103,7 @@ EXPORT_SYMBOL(kmap_pte); void __init kmap_init(void) { - unsigned long kmap_vstart; - - /* cache the first kmap pte */ - kmap_vstart = __fix_to_virt(FIX_KMAP_BEGIN); - kmap_pte = kmap_get_fixmap_pte(kmap_vstart); - - kmap_prot = PAGE_KERNEL; -} - -void __init permanent_kmaps_init(pgd_t *pgd_base) -{ - pgd_t *pgd; - pmd_t *pmd; - pte_t *pte; - unsigned long vaddr; - - vaddr = PKMAP_BASE; - page_table_range_init(vaddr, vaddr + PAGE_SIZE*LAST_PKMAP, pgd_base); - - pgd = swapper_pg_dir + pgd_index(vaddr); - pmd = pmd_offset(pgd, vaddr); - pte = pte_offset_kernel(pmd, vaddr); - pkmap_page_table = pte; + kmap_pte = kmap_get_fixmap_pte(__fix_to_virt(FIX_KMAP_BEGIN)); } void __init one_highpage_init(struct page *page, int pfn, int bad_ppro) @@ -255,6 +118,8 @@ void __init one_highpage_init(struct pag SetPageReserved(page); } +#ifdef CONFIG_HIGHMEM + #ifndef CONFIG_DISCONTIGMEM void __init set_highmem_pages_init(int bad_ppro) { @@ -266,12 +131,9 @@ void __init set_highmem_pages_init(int b #else extern void set_highmem_pages_init(int); #endif /* !CONFIG_DISCONTIGMEM */ - #else -#define kmap_init() do { } while (0) -#define permanent_kmaps_init(pgd_base) do { } while (0) -#define set_highmem_pages_init(bad_ppro) do { } while (0) -#endif /* CONFIG_HIGHMEM */ +# define set_highmem_pages_init(bad_ppro) do { } while (0) +#endif unsigned long __PAGE_KERNEL = _PAGE_KERNEL; @@ -281,30 +143,125 @@ unsigned long __PAGE_KERNEL = _PAGE_KERN extern void __init remap_numa_kva(void); #endif -static void __init pagetable_init (void) +static __init void prepare_pagetables(pgd_t *pgd_base, unsigned long address) +{ + pgd_t *pgd; + pmd_t *pmd; + pte_t *pte; + + pgd = pgd_base + pgd_index(address); + pmd = pmd_offset(pgd, address); + if (!pmd_present(*pmd)) { + pte = (pte_t *) alloc_bootmem_low_pages(PAGE_SIZE); + set_pmd(pmd, __pmd(_PAGE_TABLE + __pa(pte))); + } +} + +static void __init fixrange_init (unsigned long start, unsigned long end, pgd_t *pgd_base) { unsigned long vaddr; - pgd_t *pgd_base = swapper_pg_dir; + for (vaddr = start; vaddr != end; vaddr += PAGE_SIZE) + prepare_pagetables(pgd_base, vaddr); +} + +void setup_identity_mappings(pgd_t *pgd_base, unsigned long start, unsigned long end) +{ + unsigned long vaddr; + pgd_t *pgd; + int i, j, k; + pmd_t *pmd; + pte_t *pte, *pte_base; + + pgd = pgd_base; + + for (i = 0; i < PTRS_PER_PGD; pgd++, i++) { + vaddr = i*PGDIR_SIZE; + if (end && (vaddr >= end)) + break; + pmd = pmd_offset(pgd, 0); + for (j = 0; j < PTRS_PER_PMD; pmd++, j++) { + vaddr = i*PGDIR_SIZE + j*PMD_SIZE; + if (end && (vaddr >= end)) + break; + if (vaddr < start) + continue; + if (cpu_has_pse) { + unsigned long __pe; + + set_in_cr4(X86_CR4_PSE); + boot_cpu_data.wp_works_ok = 1; + __pe = _KERNPG_TABLE + _PAGE_PSE + vaddr - start; + /* Make it "global" too if supported */ + if (cpu_has_pge) { + set_in_cr4(X86_CR4_PGE); +#if !defined(CONFIG_X86_SWITCH_PAGETABLES) + __pe += _PAGE_GLOBAL; + __PAGE_KERNEL |= _PAGE_GLOBAL; +#endif + } + set_pmd(pmd, __pmd(__pe)); + continue; + } + if (!pmd_present(*pmd)) + pte_base = (pte_t *) alloc_bootmem_low_pages(PAGE_SIZE); + else + pte_base = (pte_t *) page_address(pmd_page(*pmd)); + pte = pte_base; + for (k = 0; k < PTRS_PER_PTE; pte++, k++) { + vaddr = i*PGDIR_SIZE + j*PMD_SIZE + k*PAGE_SIZE; + if (end && (vaddr >= end)) + break; + if (vaddr < start) + continue; + *pte = mk_pte_phys(vaddr-start, PAGE_KERNEL); + } + set_pmd(pmd, __pmd(_KERNPG_TABLE + __pa(pte_base))); + } + } +} + +static void __init pagetable_init (void) +{ + unsigned long vaddr, end; + pgd_t *pgd_base; #ifdef CONFIG_X86_PAE int i; - /* Init entries of the first-level page table to the zero page */ - for (i = 0; i < PTRS_PER_PGD; i++) - set_pgd(pgd_base + i, __pgd(__pa(empty_zero_page) | _PAGE_PRESENT)); #endif - /* Enable PSE if available */ - if (cpu_has_pse) { - set_in_cr4(X86_CR4_PSE); - } + /* + * This can be zero as well - no problem, in that case we exit + * the loops anyway due to the PTRS_PER_* conditions. + */ + end = (unsigned long)__va(max_low_pfn*PAGE_SIZE); - /* Enable PGE if available */ - if (cpu_has_pge) { - set_in_cr4(X86_CR4_PGE); - __PAGE_KERNEL |= _PAGE_GLOBAL; + pgd_base = swapper_pg_dir; +#ifdef CONFIG_X86_PAE + /* + * It causes too many problems if there's no proper pmd set up + * for all 4 entries of the PGD - so we allocate all of them. + * PAE systems will not miss this extra 4-8K anyway ... + */ + for (i = 0; i < PTRS_PER_PGD; i++) { + pmd_t *pmd = (pmd_t *) alloc_bootmem_low_pages(PAGE_SIZE); + set_pgd(pgd_base + i, __pgd(__pa(pmd) + 0x1)); } +#endif + /* + * Set up lowmem-sized identity mappings at PAGE_OFFSET: + */ + setup_identity_mappings(pgd_base, PAGE_OFFSET, end); - kernel_physical_mapping_init(pgd_base); + /* + * Add flat-mode identity-mappings - SMP needs it when + * starting up on an AP from real-mode. (In the non-PAE + * case we already have these mappings through head.S.) + * All user-space mappings are explicitly cleared after + * SMP startup. + */ +#if defined(CONFIG_SMP) && defined(CONFIG_X86_PAE) + setup_identity_mappings(pgd_base, 0, 16*1024*1024); +#endif remap_numa_kva(); /* @@ -312,38 +269,64 @@ static void __init pagetable_init (void) * created - mappings will be set by set_fixmap(): */ vaddr = __fix_to_virt(__end_of_fixed_addresses - 1) & PMD_MASK; - page_table_range_init(vaddr, 0, pgd_base); + fixrange_init(vaddr, 0, pgd_base); - permanent_kmaps_init(pgd_base); +#ifdef CONFIG_HIGHMEM + { + pgd_t *pgd; + pmd_t *pmd; + pte_t *pte; -#ifdef CONFIG_X86_PAE - /* - * Add low memory identity-mappings - SMP needs it when - * starting up on an AP from real-mode. In the non-PAE - * case we already have these mappings through head.S. - * All user-space mappings are explicitly cleared after - * SMP startup. - */ - pgd_base[0] = pgd_base[USER_PTRS_PER_PGD]; + /* + * Permanent kmaps: + */ + vaddr = PKMAP_BASE; + fixrange_init(vaddr, vaddr + PAGE_SIZE*LAST_PKMAP, pgd_base); + + pgd = swapper_pg_dir + pgd_index(vaddr); + pmd = pmd_offset(pgd, vaddr); + pte = pte_offset_kernel(pmd, vaddr); + pkmap_page_table = pte; + } #endif } -void zap_low_mappings (void) +/* + * Clear kernel pagetables in a PMD_SIZE-aligned range. + */ +static void clear_mappings(pgd_t *pgd_base, unsigned long start, unsigned long end) { - int i; + unsigned long vaddr; + pgd_t *pgd; + pmd_t *pmd; + int i, j; + + pgd = pgd_base; + + for (i = 0; i < PTRS_PER_PGD; pgd++, i++) { + vaddr = i*PGDIR_SIZE; + if (end && (vaddr >= end)) + break; + pmd = pmd_offset(pgd, 0); + for (j = 0; j < PTRS_PER_PMD; pmd++, j++) { + vaddr = i*PGDIR_SIZE + j*PMD_SIZE; + if (end && (vaddr >= end)) + break; + if (vaddr < start) + continue; + pmd_clear(pmd); + } + } + flush_tlb_all(); +} + +void zap_low_mappings(void) +{ + printk("zapping low mappings.\n"); /* * Zap initial low-memory mappings. - * - * Note that "pgd_clear()" doesn't do it for - * us, because pgd_clear() is a no-op on i386. */ - for (i = 0; i < USER_PTRS_PER_PGD; i++) -#ifdef CONFIG_X86_PAE - set_pgd(swapper_pg_dir+i, __pgd(1 + __pa(empty_zero_page))); -#else - set_pgd(swapper_pg_dir+i, __pgd(0)); -#endif - flush_tlb_all(); + clear_mappings(swapper_pg_dir, 0, 16*1024*1024); } #ifndef CONFIG_DISCONTIGMEM @@ -393,7 +376,15 @@ void __init paging_init(void) set_in_cr4(X86_CR4_PAE); #endif __flush_tlb_all(); - + /* + * Subtle. SMP is doing it's boot stuff late (because it has to + * fork idle threads) - but it also needs low mappings for the + * protected-mode entry to work. We zap these entries only after + * the WP-bit has been tested. + */ +#ifndef CONFIG_SMP + zap_low_mappings(); +#endif kmap_init(); zone_sizes_init(); } @@ -515,22 +506,18 @@ void __init mem_init(void) if (boot_cpu_data.wp_works_ok < 0) test_wp_bit(); - /* - * Subtle. SMP is doing it's boot stuff late (because it has to - * fork idle threads) - but it also needs low mappings for the - * protected-mode entry to work. We zap these entries only after - * the WP-bit has been tested. - */ -#ifndef CONFIG_SMP - zap_low_mappings(); -#endif + entry_trampoline_setup(); + default_ldt_page = virt_to_page(default_ldt); + load_LDT(&init_mm.context); } -kmem_cache_t *pgd_cache; -kmem_cache_t *pmd_cache; +kmem_cache_t *pgd_cache, *pmd_cache, *kpmd_cache; void __init pgtable_cache_init(void) { + void (*ctor)(void *, kmem_cache_t *, unsigned long); + void (*dtor)(void *, kmem_cache_t *, unsigned long); + if (PTRS_PER_PMD > 1) { pmd_cache = kmem_cache_create("pmd", PTRS_PER_PMD*sizeof(pmd_t), @@ -540,13 +527,36 @@ void __init pgtable_cache_init(void) NULL); if (!pmd_cache) panic("pgtable_cache_init(): cannot create pmd cache"); + + if (TASK_SIZE > PAGE_OFFSET) { + kpmd_cache = kmem_cache_create("kpmd", + PTRS_PER_PMD*sizeof(pmd_t), + 0, + SLAB_HWCACHE_ALIGN | SLAB_MUST_HWCACHE_ALIGN, + kpmd_ctor, + NULL); + if (!kpmd_cache) + panic("pgtable_cache_init(): " + "cannot create kpmd cache"); + } } + + if (PTRS_PER_PMD == 1 || TASK_SIZE <= PAGE_OFFSET) + ctor = pgd_ctor; + else + ctor = NULL; + + if (PTRS_PER_PMD == 1 && TASK_SIZE <= PAGE_OFFSET) + dtor = pgd_dtor; + else + dtor = NULL; + pgd_cache = kmem_cache_create("pgd", PTRS_PER_PGD*sizeof(pgd_t), 0, SLAB_HWCACHE_ALIGN | SLAB_MUST_HWCACHE_ALIGN, - pgd_ctor, - PTRS_PER_PMD == 1 ? pgd_dtor : NULL); + ctor, + dtor); if (!pgd_cache) panic("pgtable_cache_init(): Cannot create pgd cache"); } --- diff/arch/i386/mm/pgtable.c 2003-10-09 09:47:16.000000000 +0100 +++ source/arch/i386/mm/pgtable.c 2004-03-01 15:42:15.000000000 +0000 @@ -21,6 +21,7 @@ #include #include #include +#include void show_mem(void) { @@ -157,11 +158,20 @@ void pmd_ctor(void *pmd, kmem_cache_t *c memset(pmd, 0, PTRS_PER_PMD*sizeof(pmd_t)); } +void kpmd_ctor(void *__pmd, kmem_cache_t *cache, unsigned long flags) +{ + pmd_t *kpmd, *pmd; + kpmd = pmd_offset(&swapper_pg_dir[PTRS_PER_PGD-1], + (PTRS_PER_PMD - NR_SHARED_PMDS)*PMD_SIZE); + pmd = (pmd_t *)__pmd + (PTRS_PER_PMD - NR_SHARED_PMDS); + + memset(__pmd, 0, (PTRS_PER_PMD - NR_SHARED_PMDS)*sizeof(pmd_t)); + memcpy(pmd, kpmd, NR_SHARED_PMDS*sizeof(pmd_t)); +} + /* - * List of all pgd's needed for non-PAE so it can invalidate entries - * in both cached and uncached pgd's; not needed for PAE since the - * kernel pmd is shared. If PAE were not to share the pmd a similar - * tactic would be needed. This is essentially codepath-based locking + * List of all pgd's needed so it can invalidate entries in both cached + * and uncached pgd's. This is essentially codepath-based locking * against pageattr.c; it is the unique case in which a valid change * of kernel pagetables can't be lazily synchronized by vmalloc faults. * vmalloc faults work because attached pagetables are never freed. @@ -170,30 +180,60 @@ void pmd_ctor(void *pmd, kmem_cache_t *c * could be used. The locking scheme was chosen on the basis of * manfred's recommendations and having no core impact whatsoever. * -- wli + * + * The entire issue goes away when XKVA is configured. */ spinlock_t pgd_lock = SPIN_LOCK_UNLOCKED; LIST_HEAD(pgd_list); -void pgd_ctor(void *pgd, kmem_cache_t *cache, unsigned long unused) +/* + * This is not that hard to figure out. + * (a) PTRS_PER_PMD == 1 means non-PAE. + * (b) PTRS_PER_PMD > 1 means PAE. + * (c) TASK_SIZE > PAGE_OFFSET means XKVA. + * (d) TASK_SIZE <= PAGE_OFFSET means non-XKVA. + * + * Do *NOT* back out the preconstruction like the patch I'm cleaning + * up after this very instant did, or at all, for that matter. + * This is never called when PTRS_PER_PMD > 1 && TASK_SIZE > PAGE_OFFSET. + * -- wli + */ +void pgd_ctor(void *__pgd, kmem_cache_t *cache, unsigned long unused) { + pgd_t *pgd = (pgd_t *)__pgd; unsigned long flags; - if (PTRS_PER_PMD == 1) - spin_lock_irqsave(&pgd_lock, flags); + if (PTRS_PER_PMD == 1) { + if (TASK_SIZE <= PAGE_OFFSET) + spin_lock_irqsave(&pgd_lock, flags); + else + memcpy(&pgd[PTRS_PER_PGD - NR_SHARED_PMDS], + &swapper_pg_dir[PTRS_PER_PGD - NR_SHARED_PMDS], + NR_SHARED_PMDS * sizeof(pgd_t)); + } - memcpy((pgd_t *)pgd + USER_PTRS_PER_PGD, - swapper_pg_dir + USER_PTRS_PER_PGD, - (PTRS_PER_PGD - USER_PTRS_PER_PGD) * sizeof(pgd_t)); + if (TASK_SIZE <= PAGE_OFFSET) + memcpy(pgd + USER_PTRS_PER_PGD, + swapper_pg_dir + USER_PTRS_PER_PGD, + (PTRS_PER_PGD - USER_PTRS_PER_PGD) * sizeof(pgd_t)); if (PTRS_PER_PMD > 1) return; - list_add(&virt_to_page(pgd)->lru, &pgd_list); - spin_unlock_irqrestore(&pgd_lock, flags); - memset(pgd, 0, USER_PTRS_PER_PGD*sizeof(pgd_t)); + if (TASK_SIZE > PAGE_OFFSET) + memset(pgd, 0, (PTRS_PER_PGD - NR_SHARED_PMDS)*sizeof(pgd_t)); + else { + list_add(&virt_to_page(pgd)->lru, &pgd_list); + spin_unlock_irqrestore(&pgd_lock, flags); + memset(pgd, 0, USER_PTRS_PER_PGD*sizeof(pgd_t)); + } } -/* never called when PTRS_PER_PMD > 1 */ +/* + * Never called when PTRS_PER_PMD > 1 || TASK_SIZE > PAGE_OFFSET + * for with PAE we would list_del() multiple times, and for non-PAE + * with XKVA all the AGP pgd shootdown code is unnecessary. + */ void pgd_dtor(void *pgd, kmem_cache_t *cache, unsigned long unused) { unsigned long flags; /* can be called from interrupt context */ @@ -203,6 +243,12 @@ void pgd_dtor(void *pgd, kmem_cache_t *c spin_unlock_irqrestore(&pgd_lock, flags); } +/* + * See the comments above pgd_ctor() wrt. preconstruction. + * Do *NOT* memcpy() here. If you do, you back out important + * anti- cache pollution code. + * + */ pgd_t *pgd_alloc(struct mm_struct *mm) { int i; @@ -211,15 +257,33 @@ pgd_t *pgd_alloc(struct mm_struct *mm) if (PTRS_PER_PMD == 1 || !pgd) return pgd; + /* + * In the 4G userspace case alias the top 16 MB virtual + * memory range into the user mappings as well (these + * include the trampoline and CPU data structures). + */ for (i = 0; i < USER_PTRS_PER_PGD; ++i) { - pmd_t *pmd = kmem_cache_alloc(pmd_cache, GFP_KERNEL); + kmem_cache_t *cache; + pmd_t *pmd; + + if (TASK_SIZE > PAGE_OFFSET && i == USER_PTRS_PER_PGD - 1) + cache = kpmd_cache; + else + cache = pmd_cache; + + pmd = kmem_cache_alloc(cache, GFP_KERNEL); if (!pmd) goto out_oom; set_pgd(&pgd[i], __pgd(1 + __pa((u64)((u32)pmd)))); } - return pgd; + return pgd; out_oom: + /* + * we don't have to handle the kpmd_cache here, since it's the + * last allocation, and has either nothing to free or when it + * succeeds the whole operation succeeds. + */ for (i--; i >= 0; i--) kmem_cache_free(pmd_cache, (void *)__va(pgd_val(pgd[i])-1)); kmem_cache_free(pgd_cache, pgd); @@ -230,10 +294,29 @@ void pgd_free(pgd_t *pgd) { int i; - /* in the PAE case user pgd entries are overwritten before usage */ - if (PTRS_PER_PMD > 1) - for (i = 0; i < USER_PTRS_PER_PGD; ++i) - kmem_cache_free(pmd_cache, (void *)__va(pgd_val(pgd[i])-1)); /* in the non-PAE case, clear_page_tables() clears user pgd entries */ + if (PTRS_PER_PMD == 1) + goto out_free; + + /* in the PAE case user pgd entries are overwritten before usage */ + for (i = 0; i < USER_PTRS_PER_PGD; ++i) { + kmem_cache_t *cache; + pmd_t *pmd = __va(pgd_val(pgd[i]) - 1); + + /* + * only userspace pmd's are cleared for us + * by mm/memory.c; it's a slab cache invariant + * that we must separate the kernel pmd slab + * all times, else we'll have bad pmd's. + */ + if (TASK_SIZE > PAGE_OFFSET && i == USER_PTRS_PER_PGD - 1) + cache = kpmd_cache; + else + cache = pmd_cache; + + kmem_cache_free(cache, pmd); + } +out_free: kmem_cache_free(pgd_cache, pgd); } + --- diff/arch/i386/oprofile/nmi_int.c 2004-03-01 14:36:05.000000000 +0000 +++ source/arch/i386/oprofile/nmi_int.c 2004-03-01 15:42:15.000000000 +0000 @@ -65,14 +65,14 @@ static int __init init_driverfs(void) { int error; if (!(error = sysdev_class_register(&oprofile_sysclass))) - error = sys_device_register(&device_oprofile); + error = sysdev_register(&device_oprofile); return error; } static void __exit exit_driverfs(void) { - sys_device_unregister(&device_oprofile); + sysdev_unregister(&device_oprofile); sysdev_class_unregister(&oprofile_sysclass); } --- diff/arch/i386/oprofile/op_model_p4.c 2003-08-26 10:00:51.000000000 +0100 +++ source/arch/i386/oprofile/op_model_p4.c 2004-03-01 15:42:15.000000000 +0000 @@ -382,11 +382,8 @@ static struct p4_event_binding p4_events static unsigned int get_stagger(void) { #ifdef CONFIG_SMP - int cpu; - if (smp_num_siblings > 1) { - cpu = smp_processor_id(); - return (cpu_sibling_map[cpu] > cpu) ? 0 : 1; - } + int cpu = smp_processor_id(); + return (cpu != first_cpu(cpu_sibling_map[cpu])); #endif return 0; } --- diff/arch/i386/pci/Makefile 2003-07-08 09:55:17.000000000 +0100 +++ source/arch/i386/pci/Makefile 2004-03-01 15:42:15.000000000 +0000 @@ -1,6 +1,7 @@ obj-y := i386.o obj-$(CONFIG_PCI_BIOS) += pcbios.o +obj-$(CONFIG_PCI_MMCONFIG) += mmconfig.o obj-$(CONFIG_PCI_DIRECT) += direct.o pci-y := fixup.o --- diff/arch/i386/pci/common.c 2004-02-09 10:36:07.000000000 +0000 +++ source/arch/i386/pci/common.c 2004-03-01 15:42:15.000000000 +0000 @@ -20,7 +20,8 @@ extern void pcibios_sort(void); #endif -unsigned int pci_probe = PCI_PROBE_BIOS | PCI_PROBE_CONF1 | PCI_PROBE_CONF2; +unsigned int pci_probe = PCI_PROBE_BIOS | PCI_PROBE_CONF1 | PCI_PROBE_CONF2 | + PCI_PROBE_MMCONF; int pcibios_last_bus = -1; struct pci_bus *pci_root_bus = NULL; @@ -198,6 +199,12 @@ char * __devinit pcibios_setup(char *st return NULL; } #endif +#ifdef CONFIG_PCI_MMCONFIG + else if (!strcmp(str, "nommconf")) { + pci_probe &= ~PCI_PROBE_MMCONF; + return NULL; + } +#endif else if (!strcmp(str, "noacpi")) { acpi_noirq_set(); return NULL; --- diff/arch/i386/pci/pci.h 2004-02-09 10:36:07.000000000 +0000 +++ source/arch/i386/pci/pci.h 2004-03-01 15:42:15.000000000 +0000 @@ -15,6 +15,9 @@ #define PCI_PROBE_BIOS 0x0001 #define PCI_PROBE_CONF1 0x0002 #define PCI_PROBE_CONF2 0x0004 +#define PCI_PROBE_MMCONF 0x0008 +#define PCI_PROBE_MASK 0x000f + #define PCI_NO_SORT 0x0100 #define PCI_BIOS_SORT 0x0200 #define PCI_NO_CHECKS 0x0400 --- diff/arch/ia64/Kconfig 2004-03-01 14:36:05.000000000 +0000 +++ source/arch/ia64/Kconfig 2004-03-01 15:42:15.000000000 +0000 @@ -288,39 +288,6 @@ config FORCE_MAX_ZONEORDER int default "18" -choice - prompt "Huge TLB page size" - depends on HUGETLB_PAGE - default HUGETLB_PAGE_SIZE_16MB - -config HUGETLB_PAGE_SIZE_4GB - depends on MCKINLEY - bool "4GB" - -config HUGETLB_PAGE_SIZE_1GB - depends on MCKINLEY - bool "1GB" - -config HUGETLB_PAGE_SIZE_256MB - bool "256MB" - -config HUGETLB_PAGE_SIZE_64MB - bool "64MB" - -config HUGETLB_PAGE_SIZE_16MB - bool "16MB" - -config HUGETLB_PAGE_SIZE_4MB - bool "4MB" - -config HUGETLB_PAGE_SIZE_1MB - bool "1MB" - -config HUGETLB_PAGE_SIZE_256KB - bool "256KB" - -endchoice - config IA64_PAL_IDLE bool "Use PAL_HALT_LIGHT in idle loop" help @@ -522,6 +489,13 @@ config BLK_DEV_RAM_SIZE depends on BLK_DEV_RAM default "4096" +config LOCKMETER + bool "Kernel lock metering" + depends on SMP + help + Say Y to enable kernel lock metering, which adds overhead to SMP locks, + but allows you to see various statistics using the lockstat command. + endmenu source "fs/Kconfig" @@ -649,7 +623,11 @@ config DEBUG_INFO debugging info resulting in a larger kernel image. Say Y here only if you plan to use gdb to debug the kernel. If you don't debug the kernel, you can say N. - + +config SYSVIPC_COMPAT + bool + depends on COMPAT && SYSVIPC + default y endmenu source "security/Kconfig" --- diff/arch/ia64/ia32/sys_ia32.c 2004-03-01 14:36:05.000000000 +0000 +++ source/arch/ia64/ia32/sys_ia32.c 2004-03-01 15:42:15.000000000 +0000 @@ -1023,143 +1023,6 @@ sys32_writev (int fd, struct compat_iove return ret; } -/* - * sys32_ipc() is the de-multiplexer for the SysV IPC calls in 32bit emulation.. - * - * This is really horribly ugly. - */ - -struct msgbuf32 { s32 mtype; char mtext[1]; }; - -struct ipc_perm32 { - key_t key; - compat_uid_t uid; - compat_gid_t gid; - compat_uid_t cuid; - compat_gid_t cgid; - compat_mode_t mode; - unsigned short seq; -}; - -struct ipc64_perm32 { - key_t key; - compat_uid32_t uid; - compat_gid32_t gid; - compat_uid32_t cuid; - compat_gid32_t cgid; - compat_mode_t mode; - unsigned short __pad1; - unsigned short seq; - unsigned short __pad2; - unsigned int unused1; - unsigned int unused2; -}; - -struct semid_ds32 { - struct ipc_perm32 sem_perm; /* permissions .. see ipc.h */ - compat_time_t sem_otime; /* last semop time */ - compat_time_t sem_ctime; /* last change time */ - u32 sem_base; /* ptr to first semaphore in array */ - u32 sem_pending; /* pending operations to be processed */ - u32 sem_pending_last; /* last pending operation */ - u32 undo; /* undo requests on this array */ - unsigned short sem_nsems; /* no. of semaphores in array */ -}; - -struct semid64_ds32 { - struct ipc64_perm32 sem_perm; - compat_time_t sem_otime; - unsigned int __unused1; - compat_time_t sem_ctime; - unsigned int __unused2; - unsigned int sem_nsems; - unsigned int __unused3; - unsigned int __unused4; -}; - -struct msqid_ds32 { - struct ipc_perm32 msg_perm; - u32 msg_first; - u32 msg_last; - compat_time_t msg_stime; - compat_time_t msg_rtime; - compat_time_t msg_ctime; - u32 wwait; - u32 rwait; - unsigned short msg_cbytes; - unsigned short msg_qnum; - unsigned short msg_qbytes; - compat_ipc_pid_t msg_lspid; - compat_ipc_pid_t msg_lrpid; -}; - -struct msqid64_ds32 { - struct ipc64_perm32 msg_perm; - compat_time_t msg_stime; - unsigned int __unused1; - compat_time_t msg_rtime; - unsigned int __unused2; - compat_time_t msg_ctime; - unsigned int __unused3; - unsigned int msg_cbytes; - unsigned int msg_qnum; - unsigned int msg_qbytes; - compat_pid_t msg_lspid; - compat_pid_t msg_lrpid; - unsigned int __unused4; - unsigned int __unused5; -}; - -struct shmid_ds32 { - struct ipc_perm32 shm_perm; - int shm_segsz; - compat_time_t shm_atime; - compat_time_t shm_dtime; - compat_time_t shm_ctime; - compat_ipc_pid_t shm_cpid; - compat_ipc_pid_t shm_lpid; - unsigned short shm_nattch; -}; - -struct shmid64_ds32 { - struct ipc64_perm32 shm_perm; - compat_size_t shm_segsz; - compat_time_t shm_atime; - unsigned int __unused1; - compat_time_t shm_dtime; - unsigned int __unused2; - compat_time_t shm_ctime; - unsigned int __unused3; - compat_pid_t shm_cpid; - compat_pid_t shm_lpid; - unsigned int shm_nattch; - unsigned int __unused4; - unsigned int __unused5; -}; - -struct shminfo64_32 { - unsigned int shmmax; - unsigned int shmmin; - unsigned int shmmni; - unsigned int shmseg; - unsigned int shmall; - unsigned int __unused1; - unsigned int __unused2; - unsigned int __unused3; - unsigned int __unused4; -}; - -struct shm_info32 { - int used_ids; - u32 shm_tot, shm_rss, shm_swp; - u32 swap_attempts, swap_successes; -}; - -struct ipc_kludge { - u32 msgp; - s32 msgtyp; -}; - #define SEMOP 1 #define SEMGET 2 #define SEMCTL 3 @@ -1173,454 +1036,6 @@ struct ipc_kludge { #define SHMGET 23 #define SHMCTL 24 -#define IPCOP_MASK(__x) (1UL << (__x)) - -static int -ipc_parse_version32 (int *cmd) -{ - if (*cmd & IPC_64) { - *cmd ^= IPC_64; - return IPC_64; - } else { - return IPC_OLD; - } -} - -static int -semctl32 (int first, int second, int third, void *uptr) -{ - union semun fourth; - u32 pad; - int err = 0, err2; - struct semid64_ds s; - mm_segment_t old_fs; - int version = ipc_parse_version32(&third); - - if (!uptr) - return -EINVAL; - if (get_user(pad, (u32 *)uptr)) - return -EFAULT; - if (third == SETVAL) - fourth.val = (int)pad; - else - fourth.__pad = (void *)A(pad); - switch (third) { - default: - err = -EINVAL; - break; - - case IPC_INFO: - case IPC_RMID: - case IPC_SET: - case SEM_INFO: - case GETVAL: - case GETPID: - case GETNCNT: - case GETZCNT: - case GETALL: - case SETVAL: - case SETALL: - err = sys_semctl(first, second, third, fourth); - break; - - case IPC_STAT: - case SEM_STAT: - fourth.__pad = &s; - old_fs = get_fs(); - set_fs(KERNEL_DS); - err = sys_semctl(first, second, third, fourth); - set_fs(old_fs); - - if (version == IPC_64) { - struct semid64_ds32 *usp64 = (struct semid64_ds32 *) A(pad); - - if (!access_ok(VERIFY_WRITE, usp64, sizeof(*usp64))) { - err = -EFAULT; - break; - } - err2 = __put_user(s.sem_perm.key, &usp64->sem_perm.key); - err2 |= __put_user(s.sem_perm.uid, &usp64->sem_perm.uid); - err2 |= __put_user(s.sem_perm.gid, &usp64->sem_perm.gid); - err2 |= __put_user(s.sem_perm.cuid, &usp64->sem_perm.cuid); - err2 |= __put_user(s.sem_perm.cgid, &usp64->sem_perm.cgid); - err2 |= __put_user(s.sem_perm.mode, &usp64->sem_perm.mode); - err2 |= __put_user(s.sem_perm.seq, &usp64->sem_perm.seq); - err2 |= __put_user(s.sem_otime, &usp64->sem_otime); - err2 |= __put_user(s.sem_ctime, &usp64->sem_ctime); - err2 |= __put_user(s.sem_nsems, &usp64->sem_nsems); - } else { - struct semid_ds32 *usp32 = (struct semid_ds32 *) A(pad); - - if (!access_ok(VERIFY_WRITE, usp32, sizeof(*usp32))) { - err = -EFAULT; - break; - } - err2 = __put_user(s.sem_perm.key, &usp32->sem_perm.key); - err2 |= __put_user(s.sem_perm.uid, &usp32->sem_perm.uid); - err2 |= __put_user(s.sem_perm.gid, &usp32->sem_perm.gid); - err2 |= __put_user(s.sem_perm.cuid, &usp32->sem_perm.cuid); - err2 |= __put_user(s.sem_perm.cgid, &usp32->sem_perm.cgid); - err2 |= __put_user(s.sem_perm.mode, &usp32->sem_perm.mode); - err2 |= __put_user(s.sem_perm.seq, &usp32->sem_perm.seq); - err2 |= __put_user(s.sem_otime, &usp32->sem_otime); - err2 |= __put_user(s.sem_ctime, &usp32->sem_ctime); - err2 |= __put_user(s.sem_nsems, &usp32->sem_nsems); - } - if (err2) - err = -EFAULT; - break; - } - return err; -} - -static int -do_sys32_msgsnd (int first, int second, int third, void *uptr) -{ - struct msgbuf *p = kmalloc(second + sizeof(struct msgbuf), GFP_USER); - struct msgbuf32 *up = (struct msgbuf32 *)uptr; - mm_segment_t old_fs; - int err; - - if (!p) - return -ENOMEM; - err = get_user(p->mtype, &up->mtype); - err |= copy_from_user(p->mtext, &up->mtext, second); - if (err) - goto out; - old_fs = get_fs(); - set_fs(KERNEL_DS); - err = sys_msgsnd(first, p, second, third); - set_fs(old_fs); - out: - kfree(p); - return err; -} - -static int -do_sys32_msgrcv (int first, int second, int msgtyp, int third, int version, void *uptr) -{ - struct msgbuf32 *up; - struct msgbuf *p; - mm_segment_t old_fs; - int err; - - if (!version) { - struct ipc_kludge *uipck = (struct ipc_kludge *)uptr; - struct ipc_kludge ipck; - - err = -EINVAL; - if (!uptr) - goto out; - err = -EFAULT; - if (copy_from_user(&ipck, uipck, sizeof(struct ipc_kludge))) - goto out; - uptr = (void *)A(ipck.msgp); - msgtyp = ipck.msgtyp; - } - err = -ENOMEM; - p = kmalloc(second + sizeof(struct msgbuf), GFP_USER); - if (!p) - goto out; - old_fs = get_fs(); - set_fs(KERNEL_DS); - err = sys_msgrcv(first, p, second, msgtyp, third); - set_fs(old_fs); - if (err < 0) - goto free_then_out; - up = (struct msgbuf32 *)uptr; - if (put_user(p->mtype, &up->mtype) || copy_to_user(&up->mtext, p->mtext, err)) - err = -EFAULT; -free_then_out: - kfree(p); -out: - return err; -} - -static int -msgctl32 (int first, int second, void *uptr) -{ - int err = -EINVAL, err2; - struct msqid64_ds m64; - struct msqid_ds32 *up32 = (struct msqid_ds32 *)uptr; - struct msqid64_ds32 *up64 = (struct msqid64_ds32 *)uptr; - mm_segment_t old_fs; - int version = ipc_parse_version32(&second); - - switch (second) { - case IPC_INFO: - case IPC_RMID: - case MSG_INFO: - err = sys_msgctl(first, second, (struct msqid_ds *)uptr); - break; - - case IPC_SET: - if (version == IPC_64) { - err = get_user(m64.msg_perm.uid, &up64->msg_perm.uid); - err |= get_user(m64.msg_perm.gid, &up64->msg_perm.gid); - err |= get_user(m64.msg_perm.mode, &up64->msg_perm.mode); - err |= get_user(m64.msg_qbytes, &up64->msg_qbytes); - } else { - err = get_user(m64.msg_perm.uid, &up32->msg_perm.uid); - err |= get_user(m64.msg_perm.gid, &up32->msg_perm.gid); - err |= get_user(m64.msg_perm.mode, &up32->msg_perm.mode); - err |= get_user(m64.msg_qbytes, &up32->msg_qbytes); - } - if (err) - break; - old_fs = get_fs(); - set_fs(KERNEL_DS); - err = sys_msgctl(first, second, (struct msqid_ds *)&m64); - set_fs(old_fs); - break; - - case IPC_STAT: - case MSG_STAT: - old_fs = get_fs(); - set_fs(KERNEL_DS); - err = sys_msgctl(first, second, (struct msqid_ds *)&m64); - set_fs(old_fs); - - if (version == IPC_64) { - if (!access_ok(VERIFY_WRITE, up64, sizeof(*up64))) { - err = -EFAULT; - break; - } - err2 = __put_user(m64.msg_perm.key, &up64->msg_perm.key); - err2 |= __put_user(m64.msg_perm.uid, &up64->msg_perm.uid); - err2 |= __put_user(m64.msg_perm.gid, &up64->msg_perm.gid); - err2 |= __put_user(m64.msg_perm.cuid, &up64->msg_perm.cuid); - err2 |= __put_user(m64.msg_perm.cgid, &up64->msg_perm.cgid); - err2 |= __put_user(m64.msg_perm.mode, &up64->msg_perm.mode); - err2 |= __put_user(m64.msg_perm.seq, &up64->msg_perm.seq); - err2 |= __put_user(m64.msg_stime, &up64->msg_stime); - err2 |= __put_user(m64.msg_rtime, &up64->msg_rtime); - err2 |= __put_user(m64.msg_ctime, &up64->msg_ctime); - err2 |= __put_user(m64.msg_cbytes, &up64->msg_cbytes); - err2 |= __put_user(m64.msg_qnum, &up64->msg_qnum); - err2 |= __put_user(m64.msg_qbytes, &up64->msg_qbytes); - err2 |= __put_user(m64.msg_lspid, &up64->msg_lspid); - err2 |= __put_user(m64.msg_lrpid, &up64->msg_lrpid); - if (err2) - err = -EFAULT; - } else { - if (!access_ok(VERIFY_WRITE, up32, sizeof(*up32))) { - err = -EFAULT; - break; - } - err2 = __put_user(m64.msg_perm.key, &up32->msg_perm.key); - err2 |= __put_user(m64.msg_perm.uid, &up32->msg_perm.uid); - err2 |= __put_user(m64.msg_perm.gid, &up32->msg_perm.gid); - err2 |= __put_user(m64.msg_perm.cuid, &up32->msg_perm.cuid); - err2 |= __put_user(m64.msg_perm.cgid, &up32->msg_perm.cgid); - err2 |= __put_user(m64.msg_perm.mode, &up32->msg_perm.mode); - err2 |= __put_user(m64.msg_perm.seq, &up32->msg_perm.seq); - err2 |= __put_user(m64.msg_stime, &up32->msg_stime); - err2 |= __put_user(m64.msg_rtime, &up32->msg_rtime); - err2 |= __put_user(m64.msg_ctime, &up32->msg_ctime); - err2 |= __put_user(m64.msg_cbytes, &up32->msg_cbytes); - err2 |= __put_user(m64.msg_qnum, &up32->msg_qnum); - err2 |= __put_user(m64.msg_qbytes, &up32->msg_qbytes); - err2 |= __put_user(m64.msg_lspid, &up32->msg_lspid); - err2 |= __put_user(m64.msg_lrpid, &up32->msg_lrpid); - if (err2) - err = -EFAULT; - } - break; - } - return err; -} - -static int -shmat32 (int first, int second, int third, int version, void *uptr) -{ - unsigned long raddr; - u32 *uaddr = (u32 *)A((u32)third); - int err; - - if (version == 1) - return -EINVAL; /* iBCS2 emulator entry point: unsupported */ - err = do_shmat(first, uptr, second, &raddr); - if (err) - return err; - return put_user(raddr, uaddr); -} - -static int -shmctl32 (int first, int second, void *uptr) -{ - int err = -EFAULT, err2; - - struct shmid64_ds s64; - struct shmid_ds32 *up32 = (struct shmid_ds32 *)uptr; - struct shmid64_ds32 *up64 = (struct shmid64_ds32 *)uptr; - mm_segment_t old_fs; - struct shm_info32 *uip = (struct shm_info32 *)uptr; - struct shm_info si; - int version = ipc_parse_version32(&second); - struct shminfo64 smi; - struct shminfo *usi32 = (struct shminfo *) uptr; - struct shminfo64_32 *usi64 = (struct shminfo64_32 *) uptr; - - switch (second) { - case IPC_INFO: - old_fs = get_fs(); - set_fs(KERNEL_DS); - err = sys_shmctl(first, second, (struct shmid_ds *)&smi); - set_fs(old_fs); - - if (version == IPC_64) { - if (!access_ok(VERIFY_WRITE, usi64, sizeof(*usi64))) { - err = -EFAULT; - break; - } - err2 = __put_user(smi.shmmax, &usi64->shmmax); - err2 |= __put_user(smi.shmmin, &usi64->shmmin); - err2 |= __put_user(smi.shmmni, &usi64->shmmni); - err2 |= __put_user(smi.shmseg, &usi64->shmseg); - err2 |= __put_user(smi.shmall, &usi64->shmall); - } else { - if (!access_ok(VERIFY_WRITE, usi32, sizeof(*usi32))) { - err = -EFAULT; - break; - } - err2 = __put_user(smi.shmmax, &usi32->shmmax); - err2 |= __put_user(smi.shmmin, &usi32->shmmin); - err2 |= __put_user(smi.shmmni, &usi32->shmmni); - err2 |= __put_user(smi.shmseg, &usi32->shmseg); - err2 |= __put_user(smi.shmall, &usi32->shmall); - } - if (err2) - err = -EFAULT; - break; - - case IPC_RMID: - case SHM_LOCK: - case SHM_UNLOCK: - err = sys_shmctl(first, second, (struct shmid_ds *)uptr); - break; - - case IPC_SET: - if (version == IPC_64) { - err = get_user(s64.shm_perm.uid, &up64->shm_perm.uid); - err |= get_user(s64.shm_perm.gid, &up64->shm_perm.gid); - err |= get_user(s64.shm_perm.mode, &up64->shm_perm.mode); - } else { - err = get_user(s64.shm_perm.uid, &up32->shm_perm.uid); - err |= get_user(s64.shm_perm.gid, &up32->shm_perm.gid); - err |= get_user(s64.shm_perm.mode, &up32->shm_perm.mode); - } - if (err) - break; - old_fs = get_fs(); - set_fs(KERNEL_DS); - err = sys_shmctl(first, second, (struct shmid_ds *)&s64); - set_fs(old_fs); - break; - - case IPC_STAT: - case SHM_STAT: - old_fs = get_fs(); - set_fs(KERNEL_DS); - err = sys_shmctl(first, second, (struct shmid_ds *)&s64); - set_fs(old_fs); - if (err < 0) - break; - if (version == IPC_64) { - if (!access_ok(VERIFY_WRITE, up64, sizeof(*up64))) { - err = -EFAULT; - break; - } - err2 = __put_user(s64.shm_perm.key, &up64->shm_perm.key); - err2 |= __put_user(s64.shm_perm.uid, &up64->shm_perm.uid); - err2 |= __put_user(s64.shm_perm.gid, &up64->shm_perm.gid); - err2 |= __put_user(s64.shm_perm.cuid, &up64->shm_perm.cuid); - err2 |= __put_user(s64.shm_perm.cgid, &up64->shm_perm.cgid); - err2 |= __put_user(s64.shm_perm.mode, &up64->shm_perm.mode); - err2 |= __put_user(s64.shm_perm.seq, &up64->shm_perm.seq); - err2 |= __put_user(s64.shm_atime, &up64->shm_atime); - err2 |= __put_user(s64.shm_dtime, &up64->shm_dtime); - err2 |= __put_user(s64.shm_ctime, &up64->shm_ctime); - err2 |= __put_user(s64.shm_segsz, &up64->shm_segsz); - err2 |= __put_user(s64.shm_nattch, &up64->shm_nattch); - err2 |= __put_user(s64.shm_cpid, &up64->shm_cpid); - err2 |= __put_user(s64.shm_lpid, &up64->shm_lpid); - } else { - if (!access_ok(VERIFY_WRITE, up32, sizeof(*up32))) { - err = -EFAULT; - break; - } - err2 = __put_user(s64.shm_perm.key, &up32->shm_perm.key); - err2 |= __put_user(s64.shm_perm.uid, &up32->shm_perm.uid); - err2 |= __put_user(s64.shm_perm.gid, &up32->shm_perm.gid); - err2 |= __put_user(s64.shm_perm.cuid, &up32->shm_perm.cuid); - err2 |= __put_user(s64.shm_perm.cgid, &up32->shm_perm.cgid); - err2 |= __put_user(s64.shm_perm.mode, &up32->shm_perm.mode); - err2 |= __put_user(s64.shm_perm.seq, &up32->shm_perm.seq); - err2 |= __put_user(s64.shm_atime, &up32->shm_atime); - err2 |= __put_user(s64.shm_dtime, &up32->shm_dtime); - err2 |= __put_user(s64.shm_ctime, &up32->shm_ctime); - err2 |= __put_user(s64.shm_segsz, &up32->shm_segsz); - err2 |= __put_user(s64.shm_nattch, &up32->shm_nattch); - err2 |= __put_user(s64.shm_cpid, &up32->shm_cpid); - err2 |= __put_user(s64.shm_lpid, &up32->shm_lpid); - } - if (err2) - err = -EFAULT; - break; - - case SHM_INFO: - old_fs = get_fs(); - set_fs(KERNEL_DS); - err = sys_shmctl(first, second, (void *)&si); - set_fs(old_fs); - if (err < 0) - break; - - if (!access_ok(VERIFY_WRITE, uip, sizeof(*uip))) { - err = -EFAULT; - break; - } - err2 = __put_user(si.used_ids, &uip->used_ids); - err2 |= __put_user(si.shm_tot, &uip->shm_tot); - err2 |= __put_user(si.shm_rss, &uip->shm_rss); - err2 |= __put_user(si.shm_swp, &uip->shm_swp); - err2 |= __put_user(si.swap_attempts, &uip->swap_attempts); - err2 |= __put_user(si.swap_successes, &uip->swap_successes); - if (err2) - err = -EFAULT; - break; - - } - return err; -} - -extern int sem_ctls[]; -#define sc_semopm (sem_ctls[2]) - -static long -semtimedop32(int semid, struct sembuf *tsops, int nsops, - struct compat_timespec *timeout32) -{ - struct timespec t; - mm_segment_t oldfs; - long ret; - - /* parameter checking precedence should mirror sys_semtimedop() */ - if (nsops < 1 || semid < 0) - return -EINVAL; - if (nsops > sc_semopm) - return -E2BIG; - if (!access_ok(VERIFY_READ, tsops, nsops * sizeof(struct sembuf)) || - get_compat_timespec(&t, timeout32)) - return -EFAULT; - - oldfs = get_fs(); - set_fs(KERNEL_DS); - ret = sys_semtimedop(semid, tsops, nsops, &t); - set_fs(oldfs); - return ret; -} - asmlinkage long sys32_ipc(u32 call, int first, int second, int third, u32 ptr, u32 fifth) { @@ -1632,36 +1047,36 @@ sys32_ipc(u32 call, int first, int secon switch (call) { case SEMTIMEDOP: if (fifth) - return semtimedop32(first, (struct sembuf *)AA(ptr), - second, (struct compat_timespec *)AA(fifth)); + return compat_sys_semtimedop(first, compat_ptr(ptr), + second, compat_ptr(fifth)); /* else fall through for normal semop() */ case SEMOP: /* struct sembuf is the same on 32 and 64bit :)) */ - return sys_semtimedop(first, (struct sembuf *)AA(ptr), second, + return sys_semtimedop(first, compat_ptr(ptr), second, NULL); case SEMGET: return sys_semget(first, second, third); case SEMCTL: - return semctl32(first, second, third, (void *)AA(ptr)); + return compat_sys_semctl(first, second, third, compat_ptr(ptr)); case MSGSND: - return do_sys32_msgsnd(first, second, third, (void *)AA(ptr)); + return compat_sys_msgsnd(first, second, third, compat_ptr(ptr)); case MSGRCV: - return do_sys32_msgrcv(first, second, fifth, third, version, (void *)AA(ptr)); + return compat_sys_msgrcv(first, second, fifth, third, version, compat_ptr(ptr)); case MSGGET: return sys_msgget((key_t) first, second); case MSGCTL: - return msgctl32(first, second, (void *)AA(ptr)); + return compat_sys_msgctl(first, second, compat_ptr(ptr)); case SHMAT: - return shmat32(first, second, third, version, (void *)AA(ptr)); + return compat_sys_shmat(first, second, third, version, compat_ptr(ptr)); break; case SHMDT: - return sys_shmdt((char *)AA(ptr)); + return sys_shmdt(compat_ptr(ptr)); case SHMGET: return sys_shmget(first, second, third); case SHMCTL: - return shmctl32(first, second, (void *)AA(ptr)); + return compat_sys_shmctl(first, second, compat_ptr(ptr)); default: return -ENOSYS; --- diff/arch/ia64/kernel/head.S 2004-01-19 10:22:55.000000000 +0000 +++ source/arch/ia64/kernel/head.S 2004-03-01 15:42:15.000000000 +0000 @@ -816,6 +816,19 @@ GLOBAL_ENTRY(ia64_delay_loop) br.ret.sptk.many rp END(ia64_delay_loop) +GLOBAL_ENTRY(ia64_invoke_kernel_thread_helper) + .prologue + .save rp, r0 // this is the end of the call-chain + .body + alloc r2 = ar.pfs, 0, 0, 2, 0 + mov out0 = r9 + mov out1 = r11;; + br.call.sptk.many rp = kernel_thread_helper;; + mov out0 = r8 + br.call.sptk.many rp = sys_exit;; +1: br.sptk.few 1b // not reached +END(ia64_invoke_kernel_thread_helper) + #ifdef CONFIG_IA64_BRL_EMU /* --- diff/arch/ia64/kernel/iosapic.c 2004-02-18 08:54:07.000000000 +0000 +++ source/arch/ia64/kernel/iosapic.c 2004-03-01 15:42:15.000000000 +0000 @@ -103,6 +103,7 @@ static spinlock_t iosapic_lock = SPIN_LO static struct iosapic_intr_info { char *addr; /* base address of IOSAPIC */ + u32 low32; /* current value of low word of Redirection table entry */ unsigned int gsi_base; /* first GSI assigned to this IOSAPIC */ char rte_index; /* IOSAPIC RTE index (-1 => not an IOSAPIC interrupt) */ unsigned char dmode : 3; /* delivery mode (see iosapic.h) */ @@ -213,6 +214,7 @@ set_rte (unsigned int vector, unsigned i writel(high32, addr + IOSAPIC_WINDOW); writel(IOSAPIC_RTE_LOW(rte_index), addr + IOSAPIC_REG_SELECT); writel(low32, addr + IOSAPIC_WINDOW); + iosapic_intr_info[vector].low32 = low32; } static void @@ -239,9 +241,10 @@ mask_irq (unsigned int irq) spin_lock_irqsave(&iosapic_lock, flags); { writel(IOSAPIC_RTE_LOW(rte_index), addr + IOSAPIC_REG_SELECT); - low32 = readl(addr + IOSAPIC_WINDOW); - low32 |= (1 << IOSAPIC_MASK_SHIFT); /* set only the mask bit */ + /* set only the mask bit */ + low32 = iosapic_intr_info[vec].low32 |= IOSAPIC_MASK; + writel(low32, addr + IOSAPIC_WINDOW); } spin_unlock_irqrestore(&iosapic_lock, flags); @@ -264,9 +267,7 @@ unmask_irq (unsigned int irq) spin_lock_irqsave(&iosapic_lock, flags); { writel(IOSAPIC_RTE_LOW(rte_index), addr + IOSAPIC_REG_SELECT); - low32 = readl(addr + IOSAPIC_WINDOW); - - low32 &= ~(1 << IOSAPIC_MASK_SHIFT); /* clear only the mask bit */ + low32 = iosapic_intr_info[vec].low32 &= ~IOSAPIC_MASK; writel(low32, addr + IOSAPIC_WINDOW); } spin_unlock_irqrestore(&iosapic_lock, flags); @@ -307,9 +308,7 @@ iosapic_set_affinity (unsigned int irq, { /* get current delivery mode by reading the low32 */ writel(IOSAPIC_RTE_LOW(rte_index), addr + IOSAPIC_REG_SELECT); - low32 = readl(addr + IOSAPIC_WINDOW); - - low32 &= ~(7 << IOSAPIC_DELIVERY_SHIFT); + low32 = iosapic_intr_info[vec].low32 & ~(7 << IOSAPIC_DELIVERY_SHIFT); if (redir) /* change delivery mode to lowest priority */ low32 |= (IOSAPIC_LOWEST_PRIORITY << IOSAPIC_DELIVERY_SHIFT); @@ -317,6 +316,7 @@ iosapic_set_affinity (unsigned int irq, /* change delivery mode to fixed */ low32 |= (IOSAPIC_FIXED << IOSAPIC_DELIVERY_SHIFT); + iosapic_intr_info[vec].low32 = low32; writel(IOSAPIC_RTE_HIGH(rte_index), addr + IOSAPIC_REG_SELECT); writel(high32, addr + IOSAPIC_WINDOW); writel(IOSAPIC_RTE_LOW(rte_index), addr + IOSAPIC_REG_SELECT); --- diff/arch/ia64/kernel/irq.c 2004-03-01 14:36:05.000000000 +0000 +++ source/arch/ia64/kernel/irq.c 2004-03-01 15:42:15.000000000 +0000 @@ -455,7 +455,6 @@ unsigned int do_IRQ(unsigned long irq, s unsigned int status; int cpu; - irq_enter(); cpu = smp_processor_id(); /* for CONFIG_PREEMPT, this must come after irq_enter()! */ kstat_cpu(cpu).irqs[irq]++; @@ -525,7 +524,6 @@ unsigned int do_IRQ(unsigned long irq, s desc->handler->end(irq); spin_unlock(&desc->lock); } - irq_exit(); return 1; } --- diff/arch/ia64/kernel/irq_ia64.c 2004-02-09 10:36:07.000000000 +0000 +++ source/arch/ia64/kernel/irq_ia64.c 2004-03-01 15:42:15.000000000 +0000 @@ -120,6 +120,7 @@ ia64_handle_irq (ia64_vector vector, str * 16 (without this, it would be ~240, which could easily lead * to kernel stack overflows). */ + irq_enter(); saved_tpr = ia64_getreg(_IA64_REG_CR_TPR); ia64_srlz_d(); while (vector != IA64_SPURIOUS_INT_VECTOR) { @@ -143,8 +144,7 @@ ia64_handle_irq (ia64_vector vector, str * handler needs to be able to wait for further keyboard interrupts, which can't * come through until ia64_eoi() has been done. */ - if (local_softirq_pending()) - do_softirq(); + irq_exit(); } #ifdef CONFIG_SMP --- diff/arch/ia64/kernel/ivt.S 2004-01-19 10:22:55.000000000 +0000 +++ source/arch/ia64/kernel/ivt.S 2004-03-01 15:42:15.000000000 +0000 @@ -118,10 +118,11 @@ ENTRY(vhpt_miss) #ifdef CONFIG_HUGETLB_PAGE extr.u r26=r25,2,6 ;; - cmp.eq p8,p0=HPAGE_SHIFT,r26 + cmp.ne p8,p0=r18,r26 + sub r27=r26,r18 ;; (p8) dep r25=r18,r25,2,6 -(p8) shr r22=r22,HPAGE_SHIFT-PAGE_SHIFT +(p8) shr r22=r22,r27 #endif ;; cmp.eq p6,p7=5,r17 // is IFA pointing into to region 5? --- diff/arch/ia64/kernel/perfmon.c 2004-02-09 10:36:07.000000000 +0000 +++ source/arch/ia64/kernel/perfmon.c 2004-03-01 15:42:15.000000000 +0000 @@ -82,7 +82,7 @@ #define PFM_REG_IMPL 0x1 /* register implemented */ #define PFM_REG_END 0x2 /* end marker */ #define PFM_REG_MONITOR (0x1<<4|PFM_REG_IMPL) /* a PMC with a pmc.pm field only */ -#define PFM_REG_COUNTING (0x2<<4|PFM_REG_MONITOR|PFM_REG_IMPL) /* a monitor + pmc.oi+ PMD used as a counter */ +#define PFM_REG_COUNTING (0x2<<4|PFM_REG_MONITOR) /* a monitor + pmc.oi+ PMD used as a counter */ #define PFM_REG_CONTROL (0x4<<4|PFM_REG_IMPL) /* PMU control register */ #define PFM_REG_CONFIG (0x8<<4|PFM_REG_IMPL) /* configuration register */ #define PFM_REG_BUFFER (0xc<<4|PFM_REG_IMPL) /* PMD used as buffer */ @@ -109,14 +109,15 @@ #define PMD_PMD_DEP(i) pmu_conf.pmd_desc[i].dep_pmd[0] #define PMC_PMD_DEP(i) pmu_conf.pmc_desc[i].dep_pmd[0] -/* k assumed unsigned (up to 64 registers) */ -#define IBR_IS_IMPL(k) (k< IA64_NUM_DBG_REGS) -#define DBR_IS_IMPL(k) (k< IA64_NUM_DBG_REGS) +#define PFM_NUM_IBRS IA64_NUM_DBG_REGS +#define PFM_NUM_DBRS IA64_NUM_DBG_REGS #define CTX_OVFL_NOBLOCK(c) ((c)->ctx_fl_block == 0) #define CTX_HAS_SMPL(c) ((c)->ctx_fl_is_sampling) #define PFM_CTX_TASK(h) (h)->ctx_task +#define PMU_PMC_OI 5 /* position of pmc.oi bit */ + /* XXX: does not support more than 64 PMDs */ #define CTX_USED_PMD(ctx, mask) (ctx)->ctx_used_pmds[0] |= (mask) #define CTX_IS_USED_PMD(ctx, c) (((ctx)->ctx_used_pmds[0] & (1UL << (c))) != 0UL) @@ -218,6 +219,8 @@ /* * debugging */ +#define PFM_DEBUGGING 1 +#ifdef PFM_DEBUGGING #define DPRINT(a) \ do { \ if (unlikely(pfm_sysctl.debug >0)) { printk("%s.%d: CPU%d [%d] ", __FUNCTION__, __LINE__, smp_processor_id(), current->pid); printk a; } \ @@ -227,18 +230,7 @@ do { \ if (unlikely(pfm_sysctl.debug > 0 && pfm_sysctl.debug_ovfl >0)) { printk("%s.%d: CPU%d [%d] ", __FUNCTION__, __LINE__, smp_processor_id(), current->pid); printk a; } \ } while (0) -/* - * Architected PMC structure - */ -typedef struct { - unsigned long pmc_plm:4; /* privilege level mask */ - unsigned long pmc_ev:1; /* external visibility */ - unsigned long pmc_oi:1; /* overflow interrupt */ - unsigned long pmc_pm:1; /* privileged monitor */ - unsigned long pmc_ig1:1; /* reserved */ - unsigned long pmc_es:8; /* event select */ - unsigned long pmc_ig2:48; /* reserved */ -} pfm_monitor_t; +#endif /* * 64-bit software counter structure @@ -469,20 +461,13 @@ typedef struct { #define PFM_CMD_STOP 0x08 /* command does not work on zombie context */ -#define PFM_CMD_IDX(cmd) (cmd) -#define PFM_CMD_IS_VALID(cmd) ((PFM_CMD_IDX(cmd) >= 0) && (PFM_CMD_IDX(cmd) < PFM_CMD_COUNT) \ - && pfm_cmd_tab[PFM_CMD_IDX(cmd)].cmd_func != NULL) - -#define PFM_CMD_NAME(cmd) pfm_cmd_tab[PFM_CMD_IDX(cmd)].cmd_name -#define PFM_CMD_READ_ARG(cmd) (pfm_cmd_tab[PFM_CMD_IDX(cmd)].cmd_flags & PFM_CMD_ARG_READ) -#define PFM_CMD_RW_ARG(cmd) (pfm_cmd_tab[PFM_CMD_IDX(cmd)].cmd_flags & PFM_CMD_ARG_RW) -#define PFM_CMD_USE_FD(cmd) (pfm_cmd_tab[PFM_CMD_IDX(cmd)].cmd_flags & PFM_CMD_FD) -#define PFM_CMD_STOPPED(cmd) (pfm_cmd_tab[PFM_CMD_IDX(cmd)].cmd_flags & PFM_CMD_STOP) +#define PFM_CMD_NAME(cmd) pfm_cmd_tab[(cmd)].cmd_name +#define PFM_CMD_READ_ARG(cmd) (pfm_cmd_tab[(cmd)].cmd_flags & PFM_CMD_ARG_READ) +#define PFM_CMD_RW_ARG(cmd) (pfm_cmd_tab[(cmd)].cmd_flags & PFM_CMD_ARG_RW) +#define PFM_CMD_USE_FD(cmd) (pfm_cmd_tab[(cmd)].cmd_flags & PFM_CMD_FD) +#define PFM_CMD_STOPPED(cmd) (pfm_cmd_tab[(cmd)].cmd_flags & PFM_CMD_STOP) #define PFM_CMD_ARG_MANY -1 /* cannot be zero */ -#define PFM_CMD_NARG(cmd) (pfm_cmd_tab[PFM_CMD_IDX(cmd)].cmd_narg) -#define PFM_CMD_ARG_SIZE(cmd) (pfm_cmd_tab[PFM_CMD_IDX(cmd)].cmd_argsize) -#define PFM_CMD_GETSIZE(cmd) (pfm_cmd_tab[PFM_CMD_IDX(cmd)].cmd_getsize) typedef struct { int debug; /* turn on/off debugging via syslog */ @@ -2834,10 +2819,11 @@ static int pfm_write_pmcs(pfm_context_t *ctx, void *arg, int count, struct pt_regs *regs) { struct thread_struct *thread = NULL; + struct task_struct *task; pfarg_reg_t *req = (pfarg_reg_t *)arg; - unsigned long value; - unsigned long smpl_pmds, reset_pmds; - unsigned int cnum, reg_flags, flags; + unsigned long value, pmc_pm; + unsigned long smpl_pmds, reset_pmds, impl_pmds; + unsigned int cnum, reg_flags, flags, pmc_type; int i, can_access_pmu = 0, is_loaded, is_system; int is_monitor, is_counting, state; int ret = -EINVAL; @@ -2846,12 +2832,13 @@ pfm_write_pmcs(pfm_context_t *ctx, void state = ctx->ctx_state; is_loaded = state == PFM_CTX_LOADED ? 1 : 0; is_system = ctx->ctx_fl_system; + task = ctx->ctx_task; + impl_pmds = pmu_conf.impl_pmds[0]; if (state == PFM_CTX_TERMINATED || state == PFM_CTX_ZOMBIE) return -EINVAL; - if (is_loaded) { - thread = &ctx->ctx_task->thread; + thread = &task->thread; /* * In system wide and when the context is loaded, access can only happen * when the caller is running on the CPU being monitored by the session. @@ -2861,7 +2848,7 @@ pfm_write_pmcs(pfm_context_t *ctx, void DPRINT(("[%d] should be running on CPU%d\n", current->pid, ctx->ctx_cpu)); return -EBUSY; } - can_access_pmu = GET_PMU_OWNER() == ctx->ctx_task || is_system ? 1 : 0; + can_access_pmu = GET_PMU_OWNER() == task || is_system ? 1 : 0; } for (i = 0; i < count; i++, req++) { @@ -2873,16 +2860,24 @@ pfm_write_pmcs(pfm_context_t *ctx, void reset_pmds = req->reg_reset_pmds[0]; flags = 0; - is_counting = PMC_IS_COUNTING(cnum); - is_monitor = PMC_IS_MONITOR(cnum); + + if (cnum >= PMU_MAX_PMCS) { + DPRINT(("pmc%u is invalid\n", cnum)); + goto error; + } + + pmc_type = pmu_conf.pmc_desc[cnum].type; + pmc_pm = (value >> pmu_conf.pmc_desc[cnum].pm_pos) & 0x1; + is_counting = (pmc_type & PFM_REG_COUNTING) == PFM_REG_COUNTING ? 1 : 0; + is_monitor = (pmc_type & PFM_REG_MONITOR) == PFM_REG_MONITOR ? 1 : 0; /* * we reject all non implemented PMC as well * as attempts to modify PMC[0-3] which are used * as status registers by the PMU */ - if (PMC_IS_IMPL(cnum) == 0 || PMC_IS_CONTROL(cnum)) { - DPRINT(("pmc%u is unimplemented or invalid\n", cnum)); + if ((pmc_type & PFM_REG_IMPL) == 0 || (pmc_type & PFM_REG_CONTROL) == PFM_REG_CONTROL) { + DPRINT(("pmc%u is unimplemented or no-access pmc_type=%x\n", cnum, pmc_type)); goto error; } /* @@ -2890,21 +2885,20 @@ pfm_write_pmcs(pfm_context_t *ctx, void * - system-wide session: PMCx.pm=1 (privileged monitor) * - per-task : PMCx.pm=0 (user monitor) */ - if ((is_monitor || is_counting) && value != PMC_DFL_VAL(cnum) && PFM_CHECK_PMC_PM(ctx, cnum, value)) { - DPRINT(("pmc%u pmc_pm=%ld fl_system=%d\n", + if (is_monitor && value != PMC_DFL_VAL(cnum) && is_system ^ pmc_pm) { + DPRINT(("pmc%u pmc_pm=%lu is_system=%d\n", cnum, - PMC_PM(cnum, value), - ctx->ctx_fl_system)); + pmc_pm, + is_system)); goto error; } if (is_counting) { - pfm_monitor_t *p = (pfm_monitor_t *)&value; /* * enforce generation of overflow interrupt. Necessary on all * CPUs. */ - p->pmc_oi = 1; + value |= 1 << PMU_PMC_OI; if (reg_flags & PFM_REGFL_OVFL_NOTIFY) { flags |= PFM_REGFL_OVFL_NOTIFY; @@ -2913,13 +2907,13 @@ pfm_write_pmcs(pfm_context_t *ctx, void if (reg_flags & PFM_REGFL_RANDOM) flags |= PFM_REGFL_RANDOM; /* verify validity of smpl_pmds */ - if ((smpl_pmds & pmu_conf.impl_pmds[0]) != smpl_pmds) { + if ((smpl_pmds & impl_pmds) != smpl_pmds) { DPRINT(("invalid smpl_pmds 0x%lx for pmc%u\n", smpl_pmds, cnum)); goto error; } /* verify validity of reset_pmds */ - if ((reset_pmds & pmu_conf.impl_pmds[0]) != reset_pmds) { + if ((reset_pmds & impl_pmds) != reset_pmds) { DPRINT(("invalid reset_pmds 0x%lx for pmc%u\n", reset_pmds, cnum)); goto error; } @@ -2935,7 +2929,7 @@ pfm_write_pmcs(pfm_context_t *ctx, void * execute write checker, if any */ if (PMC_WR_FUNC(cnum)) { - ret = PMC_WR_FUNC(cnum)(ctx->ctx_task, ctx, cnum, &value, regs); + ret = PMC_WR_FUNC(cnum)(task, ctx, cnum, &value, regs); if (ret) goto error; ret = -EINVAL; } @@ -2997,7 +2991,7 @@ pfm_write_pmcs(pfm_context_t *ctx, void * * The value in ctx_pmcs[] can only be changed in pfm_write_pmcs(). * - * The value in t->pmc[] may be modified on overflow, i.e., when + * The value in thread->pmcs[] may be modified on overflow, i.e., when * monitoring needs to be stopped. */ if (is_monitor) CTX_USED_MONITOR(ctx, 1UL << cnum); @@ -3056,11 +3050,6 @@ pfm_write_pmcs(pfm_context_t *ctx, void return 0; error: PFM_REG_RETFLAG_SET(req->reg_flags, PFM_REG_RETFL_EINVAL); - - req->reg_flags = PFM_REG_RETFL_EINVAL; - - DPRINT(("pmc[%u]=0x%lx error %d\n", cnum, value, ret)); - return ret; } @@ -3068,6 +3057,7 @@ static int pfm_write_pmds(pfm_context_t *ctx, void *arg, int count, struct pt_regs *regs) { struct thread_struct *thread = NULL; + struct task_struct *task; pfarg_reg_t *req = (pfarg_reg_t *)arg; unsigned long value, hw_value, ovfl_mask; unsigned int cnum; @@ -3080,25 +3070,26 @@ pfm_write_pmds(pfm_context_t *ctx, void is_loaded = state == PFM_CTX_LOADED ? 1 : 0; is_system = ctx->ctx_fl_system; ovfl_mask = pmu_conf.ovfl_val; + task = ctx->ctx_task; - if (state == PFM_CTX_TERMINATED || state == PFM_CTX_ZOMBIE) return -EINVAL; + if (unlikely(state == PFM_CTX_TERMINATED || state == PFM_CTX_ZOMBIE)) return -EINVAL; /* * on both UP and SMP, we can only write to the PMC when the task is * the owner of the local PMU. */ - if (is_loaded) { - thread = &ctx->ctx_task->thread; + if (likely(is_loaded)) { + thread = &task->thread; /* * In system wide and when the context is loaded, access can only happen * when the caller is running on the CPU being monitored by the session. * It does not have to be the owner (ctx_task) of the context per se. */ - if (is_system && ctx->ctx_cpu != smp_processor_id()) { + if (unlikely(is_system && ctx->ctx_cpu != smp_processor_id())) { DPRINT(("[%d] should be running on CPU%d\n", current->pid, ctx->ctx_cpu)); return -EBUSY; } - can_access_pmu = GET_PMU_OWNER() == ctx->ctx_task || is_system ? 1 : 0; + can_access_pmu = GET_PMU_OWNER() == task || is_system ? 1 : 0; } for (i = 0; i < count; i++, req++) { @@ -3118,7 +3109,7 @@ pfm_write_pmds(pfm_context_t *ctx, void if (PMD_WR_FUNC(cnum)) { unsigned long v = value; - ret = PMD_WR_FUNC(cnum)(ctx->ctx_task, ctx, cnum, &v, regs); + ret = PMD_WR_FUNC(cnum)(task, ctx, cnum, &v, regs); if (ret) goto abort_mission; value = v; @@ -3243,16 +3234,6 @@ abort_mission: * for now, we have only one possibility for error */ PFM_REG_RETFLAG_SET(req->reg_flags, PFM_REG_RETFL_EINVAL); - - /* - * we change the return value to EFAULT in case we cannot write register return code. - * The caller first must correct this error, then a resubmission of the request will - * eventually yield the EINVAL. - */ - req->reg_flags = PFM_REG_RETFL_EINVAL; - - DPRINT(("pmd[%u]=0x%lx ret %d\n", cnum, value, ret)); - return ret; } @@ -3269,11 +3250,12 @@ static int pfm_read_pmds(pfm_context_t *ctx, void *arg, int count, struct pt_regs *regs) { struct thread_struct *thread = NULL; - unsigned long val = 0UL, lval, ovfl_mask; + struct task_struct *task; + unsigned long val = 0UL, lval, ovfl_mask, sval; pfarg_reg_t *req = (pfarg_reg_t *)arg; unsigned int cnum, reg_flags = 0; int i, can_access_pmu = 0, state; - int is_loaded, is_system; + int is_loaded, is_system, is_counting; int ret = -EINVAL; /* @@ -3285,32 +3267,33 @@ pfm_read_pmds(pfm_context_t *ctx, void * is_loaded = state == PFM_CTX_LOADED ? 1 : 0; is_system = ctx->ctx_fl_system; ovfl_mask = pmu_conf.ovfl_val; + task = ctx->ctx_task; if (state == PFM_CTX_ZOMBIE) return -EINVAL; - if (is_loaded) { - thread = &ctx->ctx_task->thread; + if (likely(is_loaded)) { + thread = &task->thread; /* * In system wide and when the context is loaded, access can only happen * when the caller is running on the CPU being monitored by the session. * It does not have to be the owner (ctx_task) of the context per se. */ - if (is_system && ctx->ctx_cpu != smp_processor_id()) { + if (unlikely(is_system && ctx->ctx_cpu != smp_processor_id())) { DPRINT(("[%d] should be running on CPU%d\n", current->pid, ctx->ctx_cpu)); return -EBUSY; } /* * this can be true when not self-monitoring only in UP */ - can_access_pmu = GET_PMU_OWNER() == ctx->ctx_task || is_system ? 1 : 0; + can_access_pmu = GET_PMU_OWNER() == task || is_system ? 1 : 0; if (can_access_pmu) ia64_srlz_d(); } - DPRINT(("enter loaded=%d access_pmu=%d ctx_state=%d\n", + DPRINT(("loaded=%d access_pmu=%d ctx_state=%d\n", is_loaded, can_access_pmu, - ctx->ctx_state)); + state)); /* * on both UP and SMP, we can only read the PMD from the hardware register when @@ -3319,11 +3302,10 @@ pfm_read_pmds(pfm_context_t *ctx, void * for (i = 0; i < count; i++, req++) { - lval = 0UL; cnum = req->reg_num; reg_flags = req->reg_flags; - if (!PMD_IS_IMPL(cnum)) goto error; + if (unlikely(!PMD_IS_IMPL(cnum))) goto error; /* * we can only read the register that we use. That includes * the one we explicitely initialize AND the one we want included @@ -3332,7 +3314,11 @@ pfm_read_pmds(pfm_context_t *ctx, void * * Having this restriction allows optimization in the ctxsw routine * without compromising security (leaks) */ - if (!CTX_IS_USED_PMD(ctx, cnum)) goto error; + if (unlikely(!CTX_IS_USED_PMD(ctx, cnum))) goto error; + + sval = ctx->ctx_pmds[cnum].val; + lval = ctx->ctx_pmds[cnum].lval; + is_counting = PMD_IS_COUNTING(cnum); /* * If the task is not the current one, then we check if the @@ -3347,23 +3333,21 @@ pfm_read_pmds(pfm_context_t *ctx, void * * if context is zombie, then task does not exist anymore. * In this case, we use the full value saved in the context (pfm_flush_regs()). */ - val = state == PFM_CTX_LOADED ? thread->pmds[cnum] : 0UL; + val = is_loaded ? thread->pmds[cnum] : 0UL; } - if (PMD_IS_COUNTING(cnum)) { + if (is_counting) { /* * XXX: need to check for overflow when loaded */ val &= ovfl_mask; - val += ctx->ctx_pmds[cnum].val; - - lval = ctx->ctx_pmds[cnum].lval; + val += sval; } /* * execute read checker, if any */ - if (PMD_RD_FUNC(cnum)) { + if (unlikely(PMD_RD_FUNC(cnum))) { unsigned long v = val; ret = PMD_RD_FUNC(cnum)(ctx->ctx_task, ctx, cnum, &v, regs); if (ret) goto error; @@ -3373,12 +3357,7 @@ pfm_read_pmds(pfm_context_t *ctx, void * PFM_REG_RETFLAG_SET(reg_flags, 0); - DPRINT(("pmd[%u]=0x%lx loaded=%d access_pmu=%d ctx_state=%d\n", - cnum, - val, - is_loaded, - can_access_pmu, - ctx->ctx_state)); + DPRINT(("pmd[%u]=0x%lx\n", cnum, val)); /* * update register return value, abort all if problem during copy. @@ -3393,12 +3372,7 @@ pfm_read_pmds(pfm_context_t *ctx, void * return 0; error: - PFM_REG_RETFLAG_SET(reg_flags, PFM_REG_RETFL_EINVAL); - - req->reg_flags = PFM_REG_RETFL_EINVAL; - - DPRINT(("error pmd[%u]=0x%lx\n", cnum, val)); - + PFM_REG_RETFLAG_SET(req->reg_flags, PFM_REG_RETFL_EINVAL); return ret; } @@ -3628,7 +3602,7 @@ pfm_restart(pfm_context_t *ctx, void *ar prefetch(ctx->ctx_smpl_hdr); rst_ctrl.bits.mask_monitoring = 0; - rst_ctrl.bits.reset_ovfl_pmds = 1; + rst_ctrl.bits.reset_ovfl_pmds = 0; if (state == PFM_CTX_LOADED) ret = pfm_buf_fmt_restart_active(fmt, task, &rst_ctrl, ctx->ctx_smpl_hdr, regs); @@ -3748,6 +3722,7 @@ static int pfm_write_ibr_dbr(int mode, pfm_context_t *ctx, void *arg, int count, struct pt_regs *regs) { struct thread_struct *thread = NULL; + struct task_struct *task; pfarg_dbreg_t *req = (pfarg_dbreg_t *)arg; unsigned long flags; dbreg_t dbreg; @@ -3762,6 +3737,7 @@ pfm_write_ibr_dbr(int mode, pfm_context_ state = ctx->ctx_state; is_loaded = state == PFM_CTX_LOADED ? 1 : 0; is_system = ctx->ctx_fl_system; + task = ctx->ctx_task; if (state == PFM_CTX_TERMINATED || state == PFM_CTX_ZOMBIE) return -EINVAL; @@ -3770,17 +3746,17 @@ pfm_write_ibr_dbr(int mode, pfm_context_ * the owner of the local PMU. */ if (is_loaded) { - thread = &ctx->ctx_task->thread; + thread = &task->thread; /* * In system wide and when the context is loaded, access can only happen * when the caller is running on the CPU being monitored by the session. * It does not have to be the owner (ctx_task) of the context per se. */ - if (is_system && ctx->ctx_cpu != smp_processor_id()) { + if (unlikely(is_system && ctx->ctx_cpu != smp_processor_id())) { DPRINT(("[%d] should be running on CPU%d\n", current->pid, ctx->ctx_cpu)); return -EBUSY; } - can_access_pmu = GET_PMU_OWNER() == ctx->ctx_task || is_system ? 1 : 0; + can_access_pmu = GET_PMU_OWNER() == task || is_system ? 1 : 0; } /* @@ -3796,7 +3772,7 @@ pfm_write_ibr_dbr(int mode, pfm_context_ * don't bother if we are loaded and task is being debugged */ if (is_loaded && (thread->flags & IA64_THREAD_DBG_VALID) != 0) { - DPRINT(("debug registers already in use for [%d]\n", ctx->ctx_task->pid)); + DPRINT(("debug registers already in use for [%d]\n", task->pid)); return -EBUSY; } @@ -3837,7 +3813,7 @@ pfm_write_ibr_dbr(int mode, pfm_context_ * is shared by all processes running on it */ if (first_time && can_access_pmu) { - DPRINT(("[%d] clearing ibrs, dbrs\n", ctx->ctx_task->pid)); + DPRINT(("[%d] clearing ibrs, dbrs\n", task->pid)); for (i=0; i < pmu_conf.num_ibrs; i++) { ia64_set_ibr(i, 0UL); ia64_srlz_i(); @@ -3860,7 +3836,7 @@ pfm_write_ibr_dbr(int mode, pfm_context_ ret = -EINVAL; - if ((mode == PFM_CODE_RR && !IBR_IS_IMPL(rnum)) || ((mode == PFM_DATA_RR) && !DBR_IS_IMPL(rnum))) { + if ((mode == PFM_CODE_RR && rnum >= PFM_NUM_IBRS) || ((mode == PFM_DATA_RR) && rnum >= PFM_NUM_DBRS)) { DPRINT(("invalid register %u val=0x%lx mode=%d i=%d count=%d\n", rnum, dbreg.val, mode, i, count)); @@ -4434,6 +4410,7 @@ pfm_context_unload(pfm_context_t *ctx, v struct task_struct *task = PFM_CTX_TASK(ctx); struct pt_regs *tregs; int state, is_system; + int ret; DPRINT(("ctx_state=%d task [%d]\n", ctx->ctx_state, task ? task->pid : -1)); @@ -4451,7 +4428,8 @@ pfm_context_unload(pfm_context_t *ctx, v /* * clear psr and dcr bits */ - pfm_stop(ctx, NULL, 0, regs); + ret = pfm_stop(ctx, NULL, 0, regs); + if (ret) return ret; ctx->ctx_state = state = PFM_CTX_UNLOADED; @@ -4760,37 +4738,45 @@ sys_perfmonctl (int fd, int cmd, void *a void *args_k = NULL; long ret; /* will expand int return types */ size_t base_sz, sz, xtra_sz = 0; - int narg, completed_args = 0, call_made = 0; + int narg, completed_args = 0, call_made = 0, cmd_flags; + int (*func)(pfm_context_t *ctx, void *arg, int count, struct pt_regs *regs); + int (*getsize)(void *arg, size_t *sz); #define PFM_MAX_ARGSIZE 4096 /* - * reject any call if perfmon was disabled at initialization time - mask*/ - if (PFM_IS_DISABLED()) return -ENOSYS; + * reject any call if perfmon was disabled at initialization + */ + if (unlikely(PFM_IS_DISABLED())) return -ENOSYS; - if (unlikely(PFM_CMD_IS_VALID(cmd) == 0)) { + if (unlikely(cmd < 0 || cmd >= PFM_CMD_COUNT)) { DPRINT(("[%d] invalid cmd=%d\n", current->pid, cmd)); return -EINVAL; } - DPRINT(("cmd=%s idx=%d valid=%d narg=0x%x argsz=%lu count=%d\n", + func = pfm_cmd_tab[cmd].cmd_func; + narg = pfm_cmd_tab[cmd].cmd_narg; + base_sz = pfm_cmd_tab[cmd].cmd_argsize; + getsize = pfm_cmd_tab[cmd].cmd_getsize; + cmd_flags = pfm_cmd_tab[cmd].cmd_flags; + + if (unlikely(func == NULL)) { + DPRINT(("[%d] invalid cmd=%d\n", current->pid, cmd)); + return -EINVAL; + } + + DPRINT(("cmd=%s idx=%d narg=0x%x argsz=%lu count=%d\n", PFM_CMD_NAME(cmd), - PFM_CMD_IDX(cmd), - PFM_CMD_IS_VALID(cmd), - PFM_CMD_NARG(cmd), - PFM_CMD_ARG_SIZE(cmd), + cmd, + narg, + base_sz, count)); /* * check if number of arguments matches what the command expects */ - narg = PFM_CMD_NARG(cmd); - if ((narg == PFM_CMD_ARG_MANY && count <= 0) || (narg > 0 && narg != count)) + if (unlikely((narg == PFM_CMD_ARG_MANY && count <= 0) || (narg > 0 && narg != count))) return -EINVAL; - /* get single argument size */ - base_sz = PFM_CMD_ARG_SIZE(cmd); - restart_args: sz = xtra_sz + base_sz*count; /* @@ -4804,7 +4790,7 @@ restart_args: /* * allocate default-sized argument buffer */ - if (count && args_k == NULL) { + if (likely(count && args_k == NULL)) { args_k = kmalloc(PFM_MAX_ARGSIZE, GFP_KERNEL); if (args_k == NULL) return -ENOMEM; } @@ -4824,11 +4810,11 @@ restart_args: /* * check if command supports extra parameters */ - if (completed_args == 0 && PFM_CMD_GETSIZE(cmd)) { + if (completed_args == 0 && getsize) { /* * get extra parameters size (based on main argument) */ - ret = PFM_CMD_GETSIZE(cmd)(args_k, &xtra_sz); + ret = (*getsize)(args_k, &xtra_sz); if (ret) goto error_args; completed_args = 1; @@ -4836,45 +4822,45 @@ restart_args: DPRINT(("[%d] restart_args sz=%lu xtra_sz=%lu\n", current->pid, sz, xtra_sz)); /* retry if necessary */ - if (xtra_sz) goto restart_args; + if (likely(xtra_sz)) goto restart_args; } - if (PFM_CMD_USE_FD(cmd)) { - - ret = -EBADF; + if (unlikely((cmd_flags & PFM_CMD_FD) == 0)) goto skip_fd; - file = fget(fd); - if (file == NULL) { - DPRINT(("[%d] invalid fd %d\n", current->pid, fd)); - goto error_args; - } - if (PFM_IS_FILE(file) == 0) { - DPRINT(("[%d] fd %d not related to perfmon\n", current->pid, fd)); - goto error_args; - } + ret = -EBADF; + file = fget(fd); + if (unlikely(file == NULL)) { + DPRINT(("[%d] invalid fd %d\n", current->pid, fd)); + goto error_args; + } + if (unlikely(PFM_IS_FILE(file) == 0)) { + DPRINT(("[%d] fd %d not related to perfmon\n", current->pid, fd)); + goto error_args; + } - ctx = (pfm_context_t *)file->private_data; - if (ctx == NULL) { - DPRINT(("[%d] no context for fd %d\n", current->pid, fd)); - goto error_args; - } + ctx = (pfm_context_t *)file->private_data; + if (unlikely(ctx == NULL)) { + DPRINT(("[%d] no context for fd %d\n", current->pid, fd)); + goto error_args; + } + prefetch(&ctx->ctx_state); - PROTECT_CTX(ctx, flags); + PROTECT_CTX(ctx, flags); - /* - * check task is stopped - */ - ret = pfm_check_task_state(ctx, cmd, flags); - if (ret) goto abort_locked; - } + /* + * check task is stopped + */ + ret = pfm_check_task_state(ctx, cmd, flags); + if (unlikely(ret)) goto abort_locked; - ret = (*pfm_cmd_tab[PFM_CMD_IDX(cmd)].cmd_func)(ctx, args_k, count, regs); +skip_fd: + ret = (*func)(ctx, args_k, count, regs); call_made = 1; abort_locked: - if (ctx) { + if (likely(ctx)) { DPRINT(("[%d] context unlocked\n", current->pid)); UNPROTECT_CTX(ctx, flags); fput(file); @@ -4907,7 +4893,7 @@ pfm_resume_after_ovfl(pfm_context_t *ctx if (CTX_HAS_SMPL(ctx)) { rst_ctrl.bits.mask_monitoring = 0; - rst_ctrl.bits.reset_ovfl_pmds = 1; + rst_ctrl.bits.reset_ovfl_pmds = 0; if (state == PFM_CTX_LOADED) ret = pfm_buf_fmt_restart_active(fmt, current, &rst_ctrl, ctx->ctx_smpl_hdr, regs); @@ -5096,7 +5082,7 @@ pfm_ovfl_notify_user(pfm_context_t *ctx, msg->pfm_ovfl_msg.msg_ovfl_pmds[1] = 0UL; msg->pfm_ovfl_msg.msg_ovfl_pmds[2] = 0UL; msg->pfm_ovfl_msg.msg_ovfl_pmds[3] = 0UL; - msg->pfm_ovfl_msg.msg_tstamp = ia64_get_itc(); /* relevant on UP only */ + msg->pfm_ovfl_msg.msg_tstamp = 0UL; } DPRINT(("ovfl msg: msg=%p no_msg=%d fd=%d pid=%d ovfl_pmds=0x%lx\n", @@ -5119,10 +5105,12 @@ pfm_end_notify_user(pfm_context_t *ctx) printk(KERN_ERR "perfmon: pfm_end_notify_user no more notification msgs\n"); return -1; } + /* no leak */ + memset(msg, 0, sizeof(*msg)); msg->pfm_end_msg.msg_type = PFM_MSG_END; msg->pfm_end_msg.msg_ctx_fd = ctx->ctx_fd; - msg->pfm_ovfl_msg.msg_tstamp = ia64_get_itc(); /* relevant on UP only */ + msg->pfm_ovfl_msg.msg_tstamp = 0UL; DPRINT(("end msg: msg=%p no_msg=%d ctx_fd=%d pid=%d\n", msg, @@ -5141,8 +5129,8 @@ pfm_overflow_handler(struct task_struct { pfm_ovfl_arg_t ovfl_arg; unsigned long mask; - unsigned long old_val, ovfl_val; - unsigned long ovfl_notify = 0UL, ovfl_pmds = 0UL, smpl_pmds = 0UL; + unsigned long old_val, ovfl_val, new_val; + unsigned long ovfl_notify = 0UL, ovfl_pmds = 0UL, smpl_pmds = 0UL, reset_pmds; unsigned long tstamp; pfm_ovfl_ctrl_t ovfl_ctrl; unsigned int i, has_smpl; @@ -5155,21 +5143,19 @@ pfm_overflow_handler(struct task_struct */ if (unlikely((pmc0 & 0x1) == 0)) goto sanity_check; - tstamp = ia64_get_itc(); - + tstamp = ia64_get_itc(); mask = pmc0 >> PMU_FIRST_COUNTER; ovfl_val = pmu_conf.ovfl_val; + has_smpl = CTX_HAS_SMPL(ctx); DPRINT_ovfl(("pmc0=0x%lx pid=%d iip=0x%lx, %s " - "used_pmds=0x%lx reload_pmcs=0x%lx\n", + "used_pmds=0x%lx\n", pmc0, task ? task->pid: -1, (regs ? regs->cr_iip : 0), CTX_OVFL_NOBLOCK(ctx) ? "nonblocking" : "blocking", - ctx->ctx_used_pmds[0], - ctx->ctx_reload_pmcs[0])); + ctx->ctx_used_pmds[0])); - has_smpl = CTX_HAS_SMPL(ctx); /* * first we update the virtual counters @@ -5180,29 +5166,31 @@ pfm_overflow_handler(struct task_struct /* skip pmd which did not overflow */ if ((mask & 0x1) == 0) continue; - DPRINT_ovfl(("pmd[%d] overflowed hw_pmd=0x%lx ctx_pmd=0x%lx\n", - i, ia64_get_pmd(i), ctx->ctx_pmds[i].val)); - /* * Note that the pmd is not necessarily 0 at this point as qualified events * may have happened before the PMU was frozen. The residual count is not * taken into consideration here but will be with any read of the pmd via * pfm_read_pmds(). */ - old_val = ctx->ctx_pmds[i].val; - ctx->ctx_pmds[i].val += 1 + ovfl_val; + old_val = new_val = ctx->ctx_pmds[i].val; + new_val += 1 + ovfl_val; + ctx->ctx_pmds[i].val = new_val; /* * check for overflow condition */ - if (likely(old_val > ctx->ctx_pmds[i].val)) { + if (likely(old_val > new_val)) { ovfl_pmds |= 1UL << i; if (PMC_OVFL_NOTIFY(ctx, i)) ovfl_notify |= 1UL << i; } - DPRINT_ovfl(("ctx_pmd[%d].val=0x%lx old_val=0x%lx pmd=0x%lx ovfl_pmds=0x%lx ovfl_notify=0x%lx smpl_pmds=0x%lx\n", - i, ctx->ctx_pmds[i].val, old_val, - ia64_get_pmd(i) & ovfl_val, ovfl_pmds, ovfl_notify, smpl_pmds)); + DPRINT_ovfl(("ctx_pmd[%d].val=0x%lx old_val=0x%lx pmd=0x%lx ovfl_pmds=0x%lx ovfl_notify=0x%lx\n", + i, + new_val, + old_val, + ia64_get_pmd(i) & ovfl_val, + ovfl_pmds, + ovfl_notify)); } /* @@ -5214,6 +5202,7 @@ pfm_overflow_handler(struct task_struct * reset all control bits */ ovfl_ctrl.val = 0; + reset_pmds = 0UL; /* * if a sampling format module exists, then we "cache" the overflow by @@ -5225,7 +5214,7 @@ pfm_overflow_handler(struct task_struct int j, k, ret = 0; int this_cpu = smp_processor_id(); - pmd_mask = ovfl_pmds >> PMU_FIRST_COUNTER; + pmd_mask = ovfl_pmds >> PMU_FIRST_COUNTER; prefetch(ctx->ctx_smpl_hdr); @@ -5275,7 +5264,10 @@ pfm_overflow_handler(struct task_struct ovfl_ctrl.bits.notify_user |= ovfl_arg.ovfl_ctrl.bits.notify_user; ovfl_ctrl.bits.block_task |= ovfl_arg.ovfl_ctrl.bits.block_task; ovfl_ctrl.bits.mask_monitoring |= ovfl_arg.ovfl_ctrl.bits.mask_monitoring; - ovfl_ctrl.bits.reset_ovfl_pmds |= ovfl_arg.ovfl_ctrl.bits.reset_ovfl_pmds; /* yes or no */ + /* + * build the bitmask of pmds to reset now + */ + if (ovfl_arg.ovfl_ctrl.bits.reset_ovfl_pmds) reset_pmds |= mask; pfm_stats[this_cpu].pfm_smpl_handler_cycles += end_cycles - start_cycles; } @@ -5287,6 +5279,10 @@ pfm_overflow_handler(struct task_struct current->pid, pmd_mask<pid, + ovfl_pmds, + reset_pmds)); /* - * if we (still) have some overflowed PMD but no notification is requested - * then we use the short reset period. + * reset the requested PMD registers using the short reset values */ - if (ovfl_ctrl.bits.reset_ovfl_pmds) { - unsigned long bm = ovfl_pmds; + if (reset_pmds) { + unsigned long bm = reset_pmds; pfm_reset_regs(ctx, &bm, PFM_PMD_SHORT_RESET); } --- diff/arch/ia64/kernel/perfmon_default_smpl.c 2004-01-19 10:22:55.000000000 +0000 +++ source/arch/ia64/kernel/perfmon_default_smpl.c 2004-03-01 15:42:15.000000000 +0000 @@ -178,6 +178,7 @@ default_handler(struct task_struct *task ent->tstamp = stamp; ent->cpu = smp_processor_id(); ent->set = arg->active_set; + ent->tgid = current->tgid; /* * selectively store PMDs in increasing index number --- diff/arch/ia64/kernel/process.c 2004-02-09 10:36:07.000000000 +0000 +++ source/arch/ia64/kernel/process.c 2004-03-01 15:42:15.000000000 +0000 @@ -259,10 +259,12 @@ ia64_load_extra (struct task_struct *tas * * We get here through the following call chain: * - * - * sys_clone - * do_fork - * copy_thread + * from user-level: from kernel: + * + * + * sys_clone : + * do_fork do_fork + * copy_thread copy_thread * * This means that the stack layout is as follows: * @@ -276,9 +278,6 @@ ia64_load_extra (struct task_struct *tas * | | <-- sp (lowest addr) * +---------------------+ * - * Note: if we get called through kernel_thread() then the memory above "(highest addr)" - * is valid kernel stack memory that needs to be copied as well. - * * Observe that we copy the unat values that are in pt_regs and switch_stack. Spilling an * integer to address X causes bit N in ar.unat to be set to the NaT bit of the register, * with N=(X & 0x1ff)/8. Thus, copying the unat value preserves the NaT bits ONLY if the @@ -291,9 +290,9 @@ copy_thread (int nr, unsigned long clone unsigned long user_stack_base, unsigned long user_stack_size, struct task_struct *p, struct pt_regs *regs) { - unsigned long rbs, child_rbs, rbs_size, stack_offset, stack_top, stack_used; - struct switch_stack *child_stack, *stack; extern char ia64_ret_from_clone, ia32_ret_from_clone; + struct switch_stack *child_stack, *stack; + unsigned long rbs, child_rbs, rbs_size; struct pt_regs *child_ptregs; int retval = 0; @@ -306,16 +305,13 @@ copy_thread (int nr, unsigned long clone return 0; #endif - stack_top = (unsigned long) current + IA64_STK_OFFSET; stack = ((struct switch_stack *) regs) - 1; - stack_used = stack_top - (unsigned long) stack; - stack_offset = IA64_STK_OFFSET - stack_used; - child_stack = (struct switch_stack *) ((unsigned long) p + stack_offset); - child_ptregs = (struct pt_regs *) (child_stack + 1); + child_ptregs = (struct pt_regs *) ((unsigned long) p + IA64_STK_OFFSET) - 1; + child_stack = (struct switch_stack *) child_ptregs - 1; /* copy parent's switch_stack & pt_regs to child: */ - memcpy(child_stack, stack, stack_used); + memcpy(child_stack, stack, sizeof(*child_ptregs) + sizeof(*child_stack)); rbs = (unsigned long) current + IA64_RBS_OFFSET; child_rbs = (unsigned long) p + IA64_RBS_OFFSET; @@ -324,7 +320,7 @@ copy_thread (int nr, unsigned long clone /* copy the parent's register backing store to the child: */ memcpy((void *) child_rbs, (void *) rbs, rbs_size); - if (user_mode(child_ptregs)) { + if (likely(user_mode(child_ptregs))) { if ((clone_flags & CLONE_SETTLS) && !IS_IA32_PROCESS(regs)) child_ptregs->r13 = regs->r16; /* see sys_clone2() in entry.S */ if (user_stack_base) { @@ -341,14 +337,14 @@ copy_thread (int nr, unsigned long clone * been taken care of by the caller of sys_clone() * already. */ - child_ptregs->r12 = (unsigned long) (child_ptregs + 1); /* kernel sp */ + child_ptregs->r12 = (unsigned long) child_ptregs - 16; /* kernel sp */ child_ptregs->r13 = (unsigned long) p; /* set `current' pointer */ } + child_stack->ar_bspstore = child_rbs + rbs_size; if (IS_IA32_PROCESS(regs)) child_stack->b0 = (unsigned long) &ia32_ret_from_clone; else child_stack->b0 = (unsigned long) &ia64_ret_from_clone; - child_stack->ar_bspstore = child_rbs + rbs_size; /* copy parts of thread_struct: */ p->thread.ksp = (unsigned long) child_stack - 16; @@ -358,8 +354,8 @@ copy_thread (int nr, unsigned long clone * therefore we must specify them explicitly here and not include them in * IA64_PSR_BITS_TO_CLEAR. */ - child_ptregs->cr_ipsr = ((child_ptregs->cr_ipsr | IA64_PSR_BITS_TO_SET) - & ~(IA64_PSR_BITS_TO_CLEAR | IA64_PSR_PP | IA64_PSR_UP)); + child_ptregs->cr_ipsr = ((child_ptregs->cr_ipsr | IA64_PSR_BITS_TO_SET) + & ~(IA64_PSR_BITS_TO_CLEAR | IA64_PSR_PP | IA64_PSR_UP)); /* * NOTE: The calling convention considers all floating point @@ -578,27 +574,43 @@ ia64_set_personality (struct elf64_hdr * pid_t kernel_thread (int (*fn)(void *), void *arg, unsigned long flags) { - struct task_struct *parent = current; - int result; - pid_t tid; + extern void ia64_invoke_kernel_thread_helper (void); + unsigned long *helper_fptr = (unsigned long *) &ia64_invoke_kernel_thread_helper; + struct { + struct switch_stack sw; + struct pt_regs pt; + } regs; + + memset(®s, 0, sizeof(regs)); + regs.pt.cr_iip = helper_fptr[0]; /* set entry point (IP) */ + regs.pt.r1 = helper_fptr[1]; /* set GP */ + regs.pt.r9 = (unsigned long) fn; /* 1st argument */ + regs.pt.r11 = (unsigned long) arg; /* 2nd argument */ + /* Preserve PSR bits, except for bits 32-34 and 37-45, which we can't read. */ + regs.pt.cr_ipsr = ia64_getreg(_IA64_REG_PSR) | IA64_PSR_BN; + regs.pt.cr_ifs = 1UL << 63; /* mark as valid, empty frame */ + regs.sw.ar_fpsr = regs.pt.ar_fpsr = ia64_getreg(_IA64_REG_AR_FPSR); + regs.sw.ar_bspstore = (unsigned long) current + IA64_RBS_OFFSET; + + return do_fork(flags | CLONE_VM | CLONE_UNTRACED, 0, ®s.pt, 0, NULL, NULL); +} +EXPORT_SYMBOL(kernel_thread); - tid = clone(flags | CLONE_VM | CLONE_UNTRACED, 0); - if (parent != current) { +/* This gets called from kernel_thread() via ia64_invoke_thread_helper(). */ +int +kernel_thread_helper (int (*fn)(void *), void *arg) +{ #ifdef CONFIG_IA32_SUPPORT - if (IS_IA32_PROCESS(ia64_task_regs(current))) { - /* A kernel thread is always a 64-bit process. */ - current->thread.map_base = DEFAULT_MAP_BASE; - current->thread.task_size = DEFAULT_TASK_SIZE; - ia64_set_kr(IA64_KR_IO_BASE, current->thread.old_iob); - ia64_set_kr(IA64_KR_TSSD, current->thread.old_k1); - } -#endif - result = (*fn)(arg); - _exit(result); + if (IS_IA32_PROCESS(ia64_task_regs(current))) { + /* A kernel thread is always a 64-bit process. */ + current->thread.map_base = DEFAULT_MAP_BASE; + current->thread.task_size = DEFAULT_TASK_SIZE; + ia64_set_kr(IA64_KR_IO_BASE, current->thread.old_iob); + ia64_set_kr(IA64_KR_TSSD, current->thread.old_k1); } - return tid; +#endif + return (*fn)(arg); } -EXPORT_SYMBOL(kernel_thread); /* * Flush thread state. This is called when a thread does an execve(). --- diff/arch/ia64/lib/dec_and_lock.c 2004-01-19 10:22:55.000000000 +0000 +++ source/arch/ia64/lib/dec_and_lock.c 2004-03-01 15:42:15.000000000 +0000 @@ -13,6 +13,7 @@ #include #include +#ifndef CONFIG_LOCKMETER /* * Decrement REFCOUNT and if the count reaches zero, acquire the spinlock. Both of these * operations have to be done atomically, so that the count doesn't drop to zero without @@ -40,3 +41,4 @@ atomic_dec_and_lock (atomic_t *refcount, } EXPORT_SYMBOL(atomic_dec_and_lock); +#endif --- diff/arch/ia64/lib/swiotlb.c 2003-06-09 14:18:17.000000000 +0100 +++ source/arch/ia64/lib/swiotlb.c 2004-03-01 15:42:15.000000000 +0000 @@ -47,7 +47,7 @@ #define IO_TLB_SHIFT 11 /* - * Used to do a quick range check in swiotlb_unmap_single and swiotlb_sync_single, to see + * Used to do a quick range check in swiotlb_unmap_single and swiotlb_sync_single_*, to see * if the memory was in fact allocated by this API. */ static char *io_tlb_start, *io_tlb_end; @@ -381,11 +381,24 @@ swiotlb_unmap_single (struct device *hwd * * If you perform a swiotlb_map_single() but wish to interrogate the buffer using the cpu, * yet do not wish to teardown the PCI dma mapping, you must call this function before - * doing so. At the next point you give the PCI dma address back to the card, the device - * again owns the buffer. + * doing so. At the next point you give the PCI dma address back to the card, you must + * first perform a swiotlb_dma_sync_for_device, and then the device again owns the buffer */ void -swiotlb_sync_single (struct device *hwdev, dma_addr_t dev_addr, size_t size, int dir) +swiotlb_sync_single_for_cpu (struct device *hwdev, dma_addr_t dev_addr, size_t size, int dir) +{ + char *dma_addr = phys_to_virt(dev_addr); + + if (dir == DMA_NONE) + BUG(); + if (dma_addr >= io_tlb_start && dma_addr < io_tlb_end) + sync_single(hwdev, dma_addr, size, dir); + else if (dir == DMA_FROM_DEVICE) + mark_clean(dma_addr, size); +} + +void +swiotlb_sync_single_for_device (struct device *hwdev, dma_addr_t dev_addr, size_t size, int dir) { char *dma_addr = phys_to_virt(dev_addr); @@ -456,11 +469,24 @@ swiotlb_unmap_sg (struct device *hwdev, * Make physical memory consistent for a set of streaming mode DMA translations after a * transfer. * - * The same as swiotlb_dma_sync_single but for a scatter-gather list, same rules and + * The same as swiotlb_sync_single_* but for a scatter-gather list, same rules and * usage. */ void -swiotlb_sync_sg (struct device *hwdev, struct scatterlist *sg, int nelems, int dir) +swiotlb_sync_sg_for_cpu (struct device *hwdev, struct scatterlist *sg, int nelems, int dir) +{ + int i; + + if (dir == DMA_NONE) + BUG(); + + for (i = 0; i < nelems; i++, sg++) + if (sg->dma_address != SG_ENT_PHYS_ADDRESS(sg)) + sync_single(hwdev, (void *) sg->dma_address, sg->dma_length, dir); +} + +void +swiotlb_sync_sg_for_device (struct device *hwdev, struct scatterlist *sg, int nelems, int dir) { int i; @@ -488,8 +514,10 @@ EXPORT_SYMBOL(swiotlb_map_single); EXPORT_SYMBOL(swiotlb_unmap_single); EXPORT_SYMBOL(swiotlb_map_sg); EXPORT_SYMBOL(swiotlb_unmap_sg); -EXPORT_SYMBOL(swiotlb_sync_single); -EXPORT_SYMBOL(swiotlb_sync_sg); +EXPORT_SYMBOL(swiotlb_sync_single_for_cpu); +EXPORT_SYMBOL(swiotlb_sync_single_for_device); +EXPORT_SYMBOL(swiotlb_sync_sg_for_cpu); +EXPORT_SYMBOL(swiotlb_sync_sg_for_device); EXPORT_SYMBOL(swiotlb_alloc_coherent); EXPORT_SYMBOL(swiotlb_free_coherent); EXPORT_SYMBOL(swiotlb_dma_supported); --- diff/arch/ia64/mm/hugetlbpage.c 2004-01-19 10:22:55.000000000 +0000 +++ source/arch/ia64/mm/hugetlbpage.c 2004-03-01 15:42:15.000000000 +0000 @@ -1,7 +1,11 @@ /* * IA-64 Huge TLB Page Support for Kernel. * - * Copyright (C) 2002, Rohit Seth + * Copyright (C) 2002-2004 Rohit Seth + * Copyright (C) 2003-2004 Ken Chen + * + * Sep, 2003: add numa support + * Feb, 2004: dynamic hugetlb page size via boot parameter */ #include @@ -18,11 +22,10 @@ #include #include -#define TASK_HPAGE_BASE (REGION_HPAGE << REGION_SHIFT) - static long htlbpagemem; int htlbpage_max; static long htlbzone_pages; +unsigned int hpage_shift=HPAGE_SHIFT_DEFAULT; static struct list_head hugepage_freelists[MAX_NUMNODES]; static spinlock_t htlbpage_lock = SPIN_LOCK_UNLOCKED; @@ -407,7 +410,7 @@ unsigned long hugetlb_get_unmapped_area( return -EINVAL; /* This code assumes that REGION_HPAGE != 0. */ if ((REGION_NUMBER(addr) != REGION_HPAGE) || (addr & (HPAGE_SIZE - 1))) - addr = TASK_HPAGE_BASE; + addr = HPAGE_REGION_BASE; else addr = ALIGN(addr, HPAGE_SIZE); for (vmm = find_vma(current->mm, addr); ; vmm = vmm->vm_next) { @@ -520,6 +523,35 @@ static int __init hugetlb_setup(char *s) } __setup("hugepages=", hugetlb_setup); +static int __init hugetlb_setup_sz(char *str) +{ + u64 tr_pages; + unsigned long long size; + + if (ia64_pal_vm_page_size(&tr_pages, NULL) != 0) + /* + * shouldn't happen, but just in case. + */ + tr_pages = 0x15557000UL; + + size = memparse(str, &str); + if (*str || (size & (size-1)) || !(tr_pages & size) || + size <= PAGE_SIZE || + size >= (1UL << PAGE_SHIFT << MAX_ORDER)) { + printk(KERN_WARNING "Invalid huge page size specified\n"); + return 1; + } + + hpage_shift = __ffs(size); + /* + * boot cpu already executed ia64_mmu_init, and has HPAGE_SHIFT_DEFAULT + * override here with new page shift. + */ + ia64_set_rr(HPAGE_REGION_BASE, hpage_shift << 2); + return 1; +} +__setup("hugepagesz=", hugetlb_setup_sz); + static int __init hugetlb_init(void) { int i; @@ -540,7 +572,7 @@ static int __init hugetlb_init(void) printk("Total HugeTLB memory allocated, %ld\n", htlbpagemem); return 0; } -module_init(hugetlb_init); +__initcall(hugetlb_init); int hugetlb_report_meminfo(char *buf) { --- diff/arch/ia64/mm/init.c 2004-02-09 10:36:07.000000000 +0000 +++ source/arch/ia64/mm/init.c 2004-03-01 15:42:15.000000000 +0000 @@ -342,6 +342,10 @@ ia64_mmu_init (void *my_cpu_data) ia64_tlb_init(); +#ifdef CONFIG_HUGETLB_PAGE + ia64_set_rr(HPAGE_REGION_BASE, HPAGE_SHIFT << 2); +#endif + #ifdef CONFIG_IA64_MCA cpu = smp_processor_id(); --- diff/arch/ia64/pci/pci.c 2004-03-01 14:36:05.000000000 +0000 +++ source/arch/ia64/pci/pci.c 2004-03-01 15:42:15.000000000 +0000 @@ -57,17 +57,16 @@ struct pci_fixup pcibios_fixups[1]; ((u64)(seg << 24) | (u64)(bus << 16) | \ (u64)(devfn << 8) | (u64)(reg)) - static int pci_sal_read (int seg, int bus, int devfn, int reg, int len, u32 *value) { int result = 0; u64 data = 0; - if (!value || (seg > 255) || (bus > 255) || (devfn > 255) || (reg > 255)) + if ((seg > 255) || (bus > 255) || (devfn > 255) || (reg > 255)) return -EINVAL; - result = ia64_sal_pci_config_read(PCI_SAL_ADDRESS(seg, bus, devfn, reg), len, &data); + result = ia64_sal_pci_config_read(PCI_SAL_ADDRESS(seg, bus, devfn, reg), 0, len, &data); *value = (u32) data; @@ -80,15 +79,61 @@ pci_sal_write (int seg, int bus, int dev if ((seg > 255) || (bus > 255) || (devfn > 255) || (reg > 255)) return -EINVAL; - return ia64_sal_pci_config_write(PCI_SAL_ADDRESS(seg, bus, devfn, reg), len, value); + return ia64_sal_pci_config_write(PCI_SAL_ADDRESS(seg, bus, devfn, reg), 0, len, value); } -struct pci_raw_ops pci_sal_ops = { +static struct pci_raw_ops pci_sal_ops = { .read = pci_sal_read, .write = pci_sal_write }; -struct pci_raw_ops *raw_pci_ops = &pci_sal_ops; /* default to SAL */ +/* SAL 3.2 adds support for extended config space. */ + +#define PCI_SAL_EXT_ADDRESS(seg, bus, devfn, reg) \ + ((u64)(seg << 28) | (u64)(bus << 20) | \ + (u64)(devfn << 12) | (u64)(reg)) + +static int +pci_sal_ext_read (int seg, int bus, int devfn, int reg, int len, u32 *value) +{ + int result = 0; + u64 data = 0; + + if ((seg > 65535) || (bus > 255) || (devfn > 255) || (reg > 4095)) + return -EINVAL; + + result = ia64_sal_pci_config_read(PCI_SAL_EXT_ADDRESS(seg, bus, devfn, reg), 1, len, &data); + + *value = (u32) data; + + return result; +} + +static int +pci_sal_ext_write (int seg, int bus, int devfn, int reg, int len, u32 value) +{ + if ((seg > 65535) || (bus > 255) || (devfn > 255) || (reg > 4095)) + return -EINVAL; + + return ia64_sal_pci_config_write(PCI_SAL_EXT_ADDRESS(seg, bus, devfn, reg), 1, len, value); +} + +static struct pci_raw_ops pci_sal_ext_ops = { + .read = pci_sal_ext_read, + .write = pci_sal_ext_write +}; + +struct pci_raw_ops *raw_pci_ops = &pci_sal_ops; /* default to SAL < 3.2 */ + +static int __init pci_set_sal_ops(void) +{ + if (sal_check_revision(3, 2)) { + raw_pci_ops = &pci_sal_ext_ops; + } + return 0; +} + +arch_initcall(pci_set_sal_ops); static int --- diff/arch/ia64/sn/io/machvec/pci_dma.c 2004-03-01 14:36:05.000000000 +0000 +++ source/arch/ia64/sn/io/machvec/pci_dma.c 2004-03-01 15:42:15.000000000 +0000 @@ -437,7 +437,8 @@ sn_pci_unmap_single(struct pci_dev *hwde } /** - * sn_pci_dma_sync_single - make sure all DMAs have completed + * sn_pci_dma_sync_single_* - make sure all DMAs or CPU accesses + * have completed * @hwdev: device to sync * @dma_handle: DMA address to sync * @size: size of region @@ -448,14 +449,19 @@ sn_pci_unmap_single(struct pci_dev *hwde * anything on our platform. */ void -sn_pci_dma_sync_single(struct pci_dev *hwdev, dma_addr_t dma_handle, size_t size, int direction) +sn_pci_dma_sync_single_for_cpu(struct pci_dev *hwdev, dma_addr_t dma_handle, size_t size, int direction) { return; +} +void +sn_pci_dma_sync_single_for_device(struct pci_dev *hwdev, dma_addr_t dma_handle, size_t size, int direction) +{ + return; } /** - * sn_pci_dma_sync_sg - make sure all DMAs have completed + * sn_pci_dma_sync_sg_* - make sure all DMAs or CPU accesses have completed * @hwdev: device to sync * @sg: scatterlist to sync * @nents: number of entries in the scatterlist @@ -466,10 +472,15 @@ sn_pci_dma_sync_single(struct pci_dev *h * on our platform. */ void -sn_pci_dma_sync_sg(struct pci_dev *hwdev, struct scatterlist *sg, int nents, int direction) +sn_pci_dma_sync_sg_for_cpu(struct pci_dev *hwdev, struct scatterlist *sg, int nents, int direction) { return; +} +void +sn_pci_dma_sync_sg_for_device(struct pci_dev *hwdev, struct scatterlist *sg, int nents, int direction) +{ + return; } /** @@ -602,28 +613,51 @@ sn_dma_unmap_sg(struct device *dev, stru EXPORT_SYMBOL(sn_dma_unmap_sg); void -sn_dma_sync_single(struct device *dev, dma_addr_t dma_handle, size_t size, +sn_dma_sync_single_for_cpu(struct device *dev, dma_addr_t dma_handle, size_t size, + int direction) +{ + BUG_ON(dev->bus != &pci_bus_type); + + sn_pci_dma_sync_single_for_cpu(to_pci_dev(dev), dma_handle, size, (int)direction); +} +EXPORT_SYMBOL(sn_dma_sync_single_for_cpu); + +void +sn_dma_sync_single_for_device(struct device *dev, dma_addr_t dma_handle, size_t size, int direction) { BUG_ON(dev->bus != &pci_bus_type); - sn_pci_dma_sync_single(to_pci_dev(dev), dma_handle, size, (int)direction); + sn_pci_dma_sync_single_for_device(to_pci_dev(dev), dma_handle, size, (int)direction); +} +EXPORT_SYMBOL(sn_dma_sync_single_for_device); + +void +sn_dma_sync_sg_for_cpu(struct device *dev, struct scatterlist *sg, int nelems, + int direction) +{ + BUG_ON(dev->bus != &pci_bus_type); + + sn_pci_dma_sync_sg_for_cpu(to_pci_dev(dev), sg, nelems, (int)direction); } -EXPORT_SYMBOL(sn_dma_sync_single); +EXPORT_SYMBOL(sn_dma_sync_sg_for_cpu); void -sn_dma_sync_sg(struct device *dev, struct scatterlist *sg, int nelems, +sn_dma_sync_sg_for_device(struct device *dev, struct scatterlist *sg, int nelems, int direction) { BUG_ON(dev->bus != &pci_bus_type); - sn_pci_dma_sync_sg(to_pci_dev(dev), sg, nelems, (int)direction); + sn_pci_dma_sync_sg_for_device(to_pci_dev(dev), sg, nelems, (int)direction); } -EXPORT_SYMBOL(sn_dma_sync_sg); +EXPORT_SYMBOL(sn_dma_sync_sg_for_device); EXPORT_SYMBOL(sn_pci_unmap_single); EXPORT_SYMBOL(sn_pci_map_single); -EXPORT_SYMBOL(sn_pci_dma_sync_single); +EXPORT_SYMBOL(sn_pci_dma_sync_single_for_cpu); +EXPORT_SYMBOL(sn_pci_dma_sync_single_for_device); +EXPORT_SYMBOL(sn_pci_dma_sync_sg_for_cpu); +EXPORT_SYMBOL(sn_pci_dma_sync_sg_for_device); EXPORT_SYMBOL(sn_pci_map_sg); EXPORT_SYMBOL(sn_pci_unmap_sg); EXPORT_SYMBOL(sn_pci_alloc_consistent); --- diff/arch/m68k/kernel/entry.S 2004-03-01 14:36:05.000000000 +0000 +++ source/arch/m68k/kernel/entry.S 2004-03-01 15:42:15.000000000 +0000 @@ -528,7 +528,7 @@ sys_call_table: .long sys_ni_syscall /* old profil syscall holder */ .long sys_statfs .long sys_fstatfs /* 100 */ - .long sys_ioperm + .long sys_ni_syscall /* ioperm for i386 */ .long sys_socketcall .long sys_syslog .long sys_setitimer --- diff/arch/m68k/kernel/sys_m68k.c 2004-03-01 14:36:05.000000000 +0000 +++ source/arch/m68k/kernel/sys_m68k.c 2004-03-01 15:42:15.000000000 +0000 @@ -261,12 +261,6 @@ asmlinkage int sys_ipc (uint call, int f return -EINVAL; } -asmlinkage long sys_ioperm(unsigned long from, unsigned long num, int on) -{ - return -ENOSYS; -} - - /* Convert virtual (user) address VADDR to physical address PADDR */ #define virt_to_phys_040(vaddr) \ ({ \ --- diff/arch/m68knommu/kernel/sys_m68k.c 2004-03-01 14:36:05.000000000 +0000 +++ source/arch/m68knommu/kernel/sys_m68k.c 2004-03-01 15:42:15.000000000 +0000 @@ -193,12 +193,6 @@ asmlinkage int sys_ipc (uint call, int f return -EINVAL; } -asmlinkage long sys_ioperm(unsigned long from, unsigned long num, int on) -{ - return -ENOSYS; -} - - /* sys_cacheflush -- flush (part of) the processor cache. */ asmlinkage int sys_cacheflush (unsigned long addr, int scope, int cache, unsigned long len) --- diff/arch/m68knommu/kernel/syscalltable.S 2003-05-21 11:49:45.000000000 +0100 +++ source/arch/m68knommu/kernel/syscalltable.S 2004-03-01 15:42:15.000000000 +0000 @@ -120,7 +120,7 @@ ENTRY(sys_call_table) .long sys_ni_syscall /* old profil syscall holder */ .long sys_statfs .long sys_fstatfs /* 100 */ - .long sys_ioperm + .long sys_ni_syscall /* ioperm for i386 */ .long sys_socketcall .long sys_syslog .long sys_setitimer --- diff/arch/mips/au1000/csb250/Makefile 2004-03-01 14:36:05.000000000 +0000 +++ source/arch/mips/au1000/csb250/Makefile 2004-03-01 15:42:15.000000000 +0000 @@ -4,10 +4,6 @@ # # Makefile for the Cogent CSB250 Au1500 board. Copied from Pb1500. # -# Note! Dependencies are done automagically by 'make dep', which also -# removes any old dependencies. DON'T put your own dependencies here -# unless it's something special (ie not a .c file). -# USE_STANDARD_AS_RULE := true --- diff/arch/mips/au1000/hydrogen3/Makefile 2004-03-01 14:36:05.000000000 +0000 +++ source/arch/mips/au1000/hydrogen3/Makefile 2004-03-01 15:42:15.000000000 +0000 @@ -5,10 +5,6 @@ # # Makefile for the Alchemy Semiconductor PB1000 board. # -# Note! Dependencies are done automagically by 'make dep', which also -# removes any old dependencies. DON'T put your own dependencies here -# unless it's something special (ie not a .c file). -# .S.s: $(CPP) $(CFLAGS) $< -o $*.s --- diff/arch/mips/au1000/mtx-1/Makefile 2004-03-01 14:36:05.000000000 +0000 +++ source/arch/mips/au1000/mtx-1/Makefile 2004-03-01 15:42:15.000000000 +0000 @@ -6,9 +6,5 @@ # # Makefile for 4G Systems MTX-1 board. # -# Note! Dependencies are done automagically by 'make dep', which also -# removes any old dependencies. DON'T put your own dependencies here -# unless it's something special (ie not a .c file). -# lib-y := init.o board_setup.o irqmap.o --- diff/arch/mips/au1000/pb1550/Makefile 2004-03-01 14:36:05.000000000 +0000 +++ source/arch/mips/au1000/pb1550/Makefile 2004-03-01 15:42:15.000000000 +0000 @@ -5,10 +5,6 @@ # # Makefile for the Alchemy Semiconductor PB1000 board. # -# Note! Dependencies are done automagically by 'make dep', which also -# removes any old dependencies. DON'T put your own dependencies here -# unless it's something special (ie not a .c file). -# .S.s: $(CPP) $(CFLAGS) $< -o $*.s --- diff/arch/mips/au1000/xxs1500/Makefile 2004-03-01 14:36:05.000000000 +0000 +++ source/arch/mips/au1000/xxs1500/Makefile 2004-03-01 15:42:15.000000000 +0000 @@ -5,9 +5,5 @@ # # Makefile for MyCable XXS1500 board. # -# Note! Dependencies are done automagically by 'make dep', which also -# removes any old dependencies. DON'T put your own dependencies here -# unless it's something special (ie not a .c file). -# lib-y := init.o board_setup.o irqmap.o --- diff/arch/mips/kernel/gdb-stub.c 2004-03-01 14:36:05.000000000 +0000 +++ source/arch/mips/kernel/gdb-stub.c 2004-03-01 15:42:15.000000000 +0000 @@ -95,7 +95,7 @@ * Example: * $ cd ~/linux * $ make menuconfig - * $ make dep; make vmlinux + * $ make * * Step 3: * Download the kernel to the remote target and start --- diff/arch/mips/kernel/i8259.c 2004-03-01 14:36:05.000000000 +0000 +++ source/arch/mips/kernel/i8259.c 2004-03-01 15:42:15.000000000 +0000 @@ -242,7 +242,7 @@ static int __init i8259A_init_sysfs(void { int error = sysdev_class_register(&i8259_sysdev_class); if (!error) - error = sys_device_register(&device_i8259A); + error = sysdev_register(&device_i8259A); return error; } --- diff/arch/mips/mm/dma-coherent.c 2004-03-01 14:36:05.000000000 +0000 +++ source/arch/mips/mm/dma-coherent.c 2004-03-01 15:42:15.000000000 +0000 @@ -119,30 +119,55 @@ void dma_unmap_sg(struct device *dev, st EXPORT_SYMBOL(dma_unmap_sg); -void dma_sync_single(struct device *dev, dma_addr_t dma_handle, size_t size, +void dma_sync_single_for_cpu(struct device *dev, dma_addr_t dma_handle, size_t size, enum dma_data_direction direction) { BUG_ON(direction == DMA_NONE); } -EXPORT_SYMBOL(dma_sync_single); +EXPORT_SYMBOL(dma_sync_single_for_cpu); -void dma_sync_single_range(struct device *dev, dma_addr_t dma_handle, +void dma_sync_single_for_device(struct device *dev, dma_addr_t dma_handle, size_t size, + enum dma_data_direction direction) +{ + BUG_ON(direction == DMA_NONE); +} + +EXPORT_SYMBOL(dma_sync_single_for_device); + +void dma_sync_single_range_for_cpu(struct device *dev, dma_addr_t dma_handle, + unsigned long offset, size_t size, + enum dma_data_direction direction) +{ + BUG_ON(direction == DMA_NONE); +} + +EXPORT_SYMBOL(dma_sync_single_range_for_cpu); + +void dma_sync_single_range_for_device(struct device *dev, dma_addr_t dma_handle, unsigned long offset, size_t size, enum dma_data_direction direction) { BUG_ON(direction == DMA_NONE); } -EXPORT_SYMBOL(dma_sync_single_range); +EXPORT_SYMBOL(dma_sync_single_range_for_device); -void dma_sync_sg(struct device *dev, struct scatterlist *sg, int nelems, +void dma_sync_sg_for_cpu(struct device *dev, struct scatterlist *sg, int nelems, enum dma_data_direction direction) { BUG_ON(direction == DMA_NONE); } -EXPORT_SYMBOL(dma_sync_sg); +EXPORT_SYMBOL(dma_sync_sg_for_cpu); + +void dma_sync_sg_for_device(struct device *dev, struct scatterlist *sg, int nelems, + enum dma_data_direction direction) +{ + BUG_ON(direction == DMA_NONE); +} + +EXPORT_SYMBOL(dma_sync_sg_for_device); int dma_supported(struct device *dev, u64 mask) { @@ -204,12 +229,20 @@ unsigned long pci_dac_dma_to_offset(stru EXPORT_SYMBOL(pci_dac_dma_to_offset); -void pci_dac_dma_sync_single(struct pci_dev *pdev, +void pci_dac_dma_sync_single_for_cpu(struct pci_dev *pdev, + dma64_addr_t dma_addr, size_t len, int direction) +{ + BUG_ON(direction == PCI_DMA_NONE); +} + +EXPORT_SYMBOL(pci_dac_dma_sync_single_for_cpu); + +void pci_dac_dma_sync_single_for_device(struct pci_dev *pdev, dma64_addr_t dma_addr, size_t len, int direction) { BUG_ON(direction == PCI_DMA_NONE); } -EXPORT_SYMBOL(pci_dac_dma_sync_single); +EXPORT_SYMBOL(pci_dac_dma_sync_single_for_device); #endif /* CONFIG_PCI */ --- diff/arch/mips/mm/dma-ip27.c 2004-03-01 14:36:05.000000000 +0000 +++ source/arch/mips/mm/dma-ip27.c 2004-03-01 15:42:15.000000000 +0000 @@ -125,30 +125,55 @@ void dma_unmap_sg(struct device *dev, st EXPORT_SYMBOL(dma_unmap_sg); -void dma_sync_single(struct device *dev, dma_addr_t dma_handle, size_t size, +void dma_sync_single_for_cpu(struct device *dev, dma_addr_t dma_handle, size_t size, enum dma_data_direction direction) { BUG_ON(direction == DMA_NONE); } -EXPORT_SYMBOL(dma_sync_single); +EXPORT_SYMBOL(dma_sync_single_for_cpu); -void dma_sync_single_range(struct device *dev, dma_addr_t dma_handle, +void dma_sync_single_for_device(struct device *dev, dma_addr_t dma_handle, size_t size, + enum dma_data_direction direction) +{ + BUG_ON(direction == DMA_NONE); +} + +EXPORT_SYMBOL(dma_sync_single_for_device); + +void dma_sync_single_range_for_cpu(struct device *dev, dma_addr_t dma_handle, + unsigned long offset, size_t size, + enum dma_data_direction direction) +{ + BUG_ON(direction == DMA_NONE); +} + +EXPORT_SYMBOL(dma_sync_single_range_for_cpu); + +void dma_sync_single_range_for_device(struct device *dev, dma_addr_t dma_handle, unsigned long offset, size_t size, enum dma_data_direction direction) { BUG_ON(direction == DMA_NONE); } -EXPORT_SYMBOL(dma_sync_single_range); +EXPORT_SYMBOL(dma_sync_single_range_for_device); -void dma_sync_sg(struct device *dev, struct scatterlist *sg, int nelems, +void dma_sync_sg_for_cpu(struct device *dev, struct scatterlist *sg, int nelems, enum dma_data_direction direction) { BUG_ON(direction == DMA_NONE); } -EXPORT_SYMBOL(dma_sync_sg); +EXPORT_SYMBOL(dma_sync_sg_for_cpu); + +void dma_sync_sg_for_device(struct device *dev, struct scatterlist *sg, int nelems, + enum dma_data_direction direction) +{ + BUG_ON(direction == DMA_NONE); +} + +EXPORT_SYMBOL(dma_sync_sg_for_device); int dma_supported(struct device *dev, u64 mask) { @@ -208,10 +233,18 @@ unsigned long pci_dac_dma_to_offset(stru EXPORT_SYMBOL(pci_dac_dma_to_offset); -void pci_dac_dma_sync_single(struct pci_dev *pdev, +void pci_dac_dma_sync_single_for_cpu(struct pci_dev *pdev, + dma64_addr_t dma_addr, size_t len, int direction) +{ + BUG_ON(direction == PCI_DMA_NONE); +} + +EXPORT_SYMBOL(pci_dac_dma_sync_single_for_cpu); + +void pci_dac_dma_sync_single_for_device(struct pci_dev *pdev, dma64_addr_t dma_addr, size_t len, int direction) { BUG_ON(direction == PCI_DMA_NONE); } -EXPORT_SYMBOL(pci_dac_dma_sync_single); +EXPORT_SYMBOL(pci_dac_dma_sync_single_for_device); --- diff/arch/mips/mm/dma-noncoherent.c 2004-03-01 14:36:05.000000000 +0000 +++ source/arch/mips/mm/dma-noncoherent.c 2004-03-01 15:42:15.000000000 +0000 @@ -226,7 +226,7 @@ void dma_unmap_sg(struct device *dev, st EXPORT_SYMBOL(dma_unmap_sg); -void dma_sync_single(struct device *dev, dma_addr_t dma_handle, size_t size, +void dma_sync_single_for_cpu(struct device *dev, dma_addr_t dma_handle, size_t size, enum dma_data_direction direction) { unsigned long addr; @@ -237,9 +237,35 @@ void dma_sync_single(struct device *dev, __dma_sync(addr, size, direction); } -EXPORT_SYMBOL(dma_sync_single); +EXPORT_SYMBOL(dma_sync_single_for_cpu); -void dma_sync_single_range(struct device *dev, dma_addr_t dma_handle, +void dma_sync_single_for_device(struct device *dev, dma_addr_t dma_handle, size_t size, + enum dma_data_direction direction) +{ + unsigned long addr; + + BUG_ON(direction == DMA_NONE); + + addr = dma_handle + PAGE_OFFSET; + __dma_sync(addr, size, direction); +} + +EXPORT_SYMBOL(dma_sync_single_for_device); + +void dma_sync_single_range_for_cpu(struct device *dev, dma_addr_t dma_handle, + unsigned long offset, size_t size, enum dma_data_direction direction) +{ + unsigned long addr; + + BUG_ON(direction == DMA_NONE); + + addr = dma_handle + offset + PAGE_OFFSET; + __dma_sync(addr, size, direction); +} + +EXPORT_SYMBOL(dma_sync_single_range_for_cpu); + +void dma_sync_single_range_for_device(struct device *dev, dma_addr_t dma_handle, unsigned long offset, size_t size, enum dma_data_direction direction) { unsigned long addr; @@ -250,9 +276,9 @@ void dma_sync_single_range(struct device __dma_sync(addr, size, direction); } -EXPORT_SYMBOL(dma_sync_single_range); +EXPORT_SYMBOL(dma_sync_single_range_for_device); -void dma_sync_sg(struct device *dev, struct scatterlist *sg, int nelems, +void dma_sync_sg_for_cpu(struct device *dev, struct scatterlist *sg, int nelems, enum dma_data_direction direction) { int i; @@ -265,7 +291,22 @@ void dma_sync_sg(struct device *dev, str sg->length, direction); } -EXPORT_SYMBOL(dma_sync_sg); +EXPORT_SYMBOL(dma_sync_sg_for_cpu); + +void dma_sync_sg_for_device(struct device *dev, struct scatterlist *sg, int nelems, + enum dma_data_direction direction) +{ + int i; + + BUG_ON(direction == DMA_NONE); + + /* Make sure that gcc doesn't leave the empty loop body. */ + for (i = 0; i < nelems; i++, sg++) + __dma_sync((unsigned long)page_address(sg->page), + sg->length, direction); +} + +EXPORT_SYMBOL(dma_sync_sg_for_device); int dma_supported(struct device *dev, u64 mask) { @@ -329,7 +370,17 @@ unsigned long pci_dac_dma_to_offset(stru EXPORT_SYMBOL(pci_dac_dma_to_offset); -void pci_dac_dma_sync_single(struct pci_dev *pdev, +void pci_dac_dma_sync_single_for_cpu(struct pci_dev *pdev, + dma64_addr_t dma_addr, size_t len, int direction) +{ + BUG_ON(direction == PCI_DMA_NONE); + + dma_cache_wback_inv(dma_addr + PAGE_OFFSET, len); +} + +EXPORT_SYMBOL(pci_dac_dma_sync_single_for_cpu); + +void pci_dac_dma_sync_single_for_device(struct pci_dev *pdev, dma64_addr_t dma_addr, size_t len, int direction) { BUG_ON(direction == PCI_DMA_NONE); @@ -337,6 +388,6 @@ void pci_dac_dma_sync_single(struct pci_ dma_cache_wback_inv(dma_addr + PAGE_OFFSET, len); } -EXPORT_SYMBOL(pci_dac_dma_sync_single); +EXPORT_SYMBOL(pci_dac_dma_sync_single_for_device); #endif /* CONFIG_PCI */ --- diff/arch/mips/momentum/jaguar_atx/Makefile 2004-03-01 14:36:05.000000000 +0000 +++ source/arch/mips/momentum/jaguar_atx/Makefile 2004-03-01 15:42:15.000000000 +0000 @@ -1,10 +1,6 @@ # # Makefile for Momentum Computer's Jaguar-ATX board. # -# Note! Dependencies are done automagically by 'make dep', which also -# removes any old dependencies. DON'T put your own dependencies here -# unless it's something special (ie not a .c file). -# obj-y += mv-irq.o int-handler.o irq.o pci-irq.o prom.o reset.o setup.o obj-$(CONFIG_PCI) += pci.o --- diff/arch/mips/tx4927/common/Makefile 2003-07-08 09:55:17.000000000 +0100 +++ source/arch/mips/tx4927/common/Makefile 2004-03-01 15:42:15.000000000 +0000 @@ -1,10 +1,6 @@ # # Makefile for common code for Toshiba TX4927 based systems # -# Note! Dependencies are done automagically by 'make dep', which also -# removes any old dependencies. DON'T put your own dependencies here -# unless it's something special (ie not a .c file). -# obj-y := tx4927_prom.o obj-y += tx4927_setup.o --- diff/arch/parisc/kernel/drivers.c 2004-02-18 08:54:07.000000000 +0000 +++ source/arch/parisc/kernel/drivers.c 2004-03-01 15:42:15.000000000 +0000 @@ -618,6 +618,7 @@ static void parisc_generic_device_regist tmp1); /* make the generic dma mask a pointer to the parisc one */ dev->dev.dma_mask = &dev->dma_mask; + dev->dev.coherent_dma_mask = dev->dma_mask; pr_debug("device_register(%s)\n", dev->dev.bus_id); device_register(&dev->dev); } --- diff/arch/parisc/kernel/pci-dma.c 2004-02-18 08:54:07.000000000 +0000 +++ source/arch/parisc/kernel/pci-dma.c 2004-03-01 15:42:15.000000000 +0000 @@ -372,7 +372,7 @@ static void * pa11_dma_alloc_consistent ** ISA cards will certainly only support 24-bit DMA addressing. ** Not clear if we can, want, or need to support ISA. */ - if (!dev || *dev->dma_mask != 0xffffffff) + if (!dev || *dev->coherent_dma_mask < 0xffffffff) gfp |= GFP_DMA; #endif return (void *)vaddr; @@ -413,7 +413,7 @@ static void pa11_dma_unmap_single(struct /* * For PCI_DMA_FROMDEVICE this flush is not necessary for the * simple map/unmap case. However, it IS necessary if if - * pci_dma_sync_single has been called and the buffer reused. + * pci_dma_sync_single_* has been called and the buffer reused. */ flush_kernel_dcache_range((unsigned long) phys_to_virt(dma_handle), size); @@ -453,7 +453,7 @@ static void pa11_dma_unmap_sg(struct dev return; } -static void pa11_dma_sync_single(struct device *dev, dma_addr_t dma_handle, unsigned long offset, size_t size, enum dma_data_direction direction) +static void pa11_dma_sync_single_for_cpu(struct device *dev, dma_addr_t dma_handle, unsigned long offset, size_t size, enum dma_data_direction direction) { if (direction == DMA_NONE) BUG(); @@ -461,7 +461,25 @@ static void pa11_dma_sync_single(struct flush_kernel_dcache_range((unsigned long) phys_to_virt(dma_handle) + offset, size); } -static void pa11_dma_sync_sg(struct device *dev, struct scatterlist *sglist, int nents, enum dma_data_direction direction) +static void pa11_dma_sync_single_for_device(struct device *dev, dma_addr_t dma_handle, unsigned long offset, size_t size, enum dma_data_direction direction) +{ + if (direction == DMA_NONE) + BUG(); + + flush_kernel_dcache_range((unsigned long) phys_to_virt(dma_handle) + offset, size); +} + +static void pa11_dma_sync_sg_for_cpu(struct device *dev, struct scatterlist *sglist, int nents, enum dma_data_direction direction) +{ + int i; + + /* once we do combining we'll need to use phys_to_virt(sg_dma_address(sglist)) */ + + for (i = 0; i < nents; i++, sglist++ ) + flush_kernel_dcache_range(sg_virt_addr(sglist), sglist->length); +} + +static void pa11_dma_sync_sg_for_device(struct device *dev, struct scatterlist *sglist, int nents, enum dma_data_direction direction) { int i; @@ -480,8 +498,10 @@ struct hppa_dma_ops pcxl_dma_ops = { .unmap_single = pa11_dma_unmap_single, .map_sg = pa11_dma_map_sg, .unmap_sg = pa11_dma_unmap_sg, - .dma_sync_single = pa11_dma_sync_single, - .dma_sync_sg = pa11_dma_sync_sg, + .dma_sync_single_for_cpu = pa11_dma_sync_single_for_cpu, + .dma_sync_single_for_device = pa11_dma_sync_single_for_device, + .dma_sync_sg_for_cpu = pa11_dma_sync_sg_for_cpu, + .dma_sync_sg_for_device = pa11_dma_sync_sg_for_device, }; static void *fail_alloc_consistent(struct device *dev, size_t size, @@ -519,8 +539,10 @@ struct hppa_dma_ops pcx_dma_ops = { .unmap_single = pa11_dma_unmap_single, .map_sg = pa11_dma_map_sg, .unmap_sg = pa11_dma_unmap_sg, - .dma_sync_single = pa11_dma_sync_single, - .dma_sync_sg = pa11_dma_sync_sg, + .dma_sync_single_cpu = pa11_dma_sync_single_cpu, + .dma_sync_single_device = pa11_dma_sync_single_device, + .dma_sync_sg_cpu = pa11_dma_sync_sg_cpu, + .dma_sync_sg_device = pa11_dma_sync_sg_device, }; --- diff/arch/parisc/kernel/process.c 2004-02-18 08:54:07.000000000 +0000 +++ source/arch/parisc/kernel/process.c 2004-03-01 15:42:15.000000000 +0000 @@ -1,5 +1,3 @@ -/* - * PARISC Architecture-dependent parts of process handling * based on the work for i386 * * Copyright (C) 1999-2003 Matthew Wilcox @@ -32,7 +30,6 @@ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA */ -#define __KERNEL_SYSCALLS__ #include #include --- diff/arch/parisc/kernel/smp.c 2004-01-19 10:22:55.000000000 +0000 +++ source/arch/parisc/kernel/smp.c 2004-03-01 15:42:15.000000000 +0000 @@ -16,7 +16,6 @@ ** the Free Software Foundation; either version 2 of the License, or ** (at your option) any later version. */ -#define __KERNEL_SYSCALLS__ #undef ENTRY_SYS_CPUS /* syscall support for iCOD-like functionality */ #include --- diff/arch/parisc/kernel/sys_parisc.c 2004-03-01 14:36:05.000000000 +0000 +++ source/arch/parisc/kernel/sys_parisc.c 2004-03-01 15:42:15.000000000 +0000 @@ -242,14 +242,6 @@ asmlinkage ssize_t parisc_readahead(int return sys_readahead(fd, (loff_t)high << 32 | low, count); } -/* - * This changes the io permissions bitmap in the current task. - */ -asmlinkage long sys_ioperm(unsigned long from, unsigned long num, int turn_on) -{ - return -ENOSYS; -} - asmlinkage unsigned long sys_alloc_hugepages(int key, unsigned long addr, unsigned long len, int prot, int flag) { return -ENOMEM; --- diff/arch/ppc/boot/ld.script 2003-10-09 09:47:16.000000000 +0100 +++ source/arch/ppc/boot/ld.script 2004-03-01 15:42:15.000000000 +0000 @@ -82,6 +82,7 @@ SECTIONS *(__ksymtab) *(__ksymtab_strings) *(__bug_table) + *(__kcrctab) } } --- diff/arch/ppc/kernel/misc.S 2004-02-18 08:54:07.000000000 +0000 +++ source/arch/ppc/kernel/misc.S 2004-03-01 15:42:15.000000000 +0000 @@ -1108,17 +1108,7 @@ _GLOBAL(name) \ li r3,-1; \ blr -#define __NR__exit __NR_exit - -SYSCALL(setsid) -SYSCALL(open) -SYSCALL(read) -SYSCALL(write) -SYSCALL(lseek) -SYSCALL(close) -SYSCALL(dup) SYSCALL(execve) -SYSCALL(waitpid) /* Why isn't this a) automatic, b) written in 'C'? */ .data --- diff/arch/ppc/kernel/ppc_ksyms.c 2004-02-18 08:54:07.000000000 +0000 +++ source/arch/ppc/kernel/ppc_ksyms.c 2004-03-01 15:42:15.000000000 +0000 @@ -32,8 +32,6 @@ #include #include #include -#define __KERNEL_SYSCALLS__ -#include #include #include #include @@ -189,10 +187,6 @@ EXPORT_SYMBOL(consistent_sync); EXPORT_SYMBOL(flush_dcache_all); #endif -EXPORT_SYMBOL(open); -EXPORT_SYMBOL(read); -EXPORT_SYMBOL(lseek); -EXPORT_SYMBOL(close); EXPORT_SYMBOL(start_thread); EXPORT_SYMBOL(kernel_thread); --- diff/arch/ppc/kernel/smp.c 2004-03-01 14:36:05.000000000 +0000 +++ source/arch/ppc/kernel/smp.c 2004-03-01 15:42:15.000000000 +0000 @@ -17,8 +17,6 @@ #include #include #include -#define __KERNEL_SYSCALLS__ -#include #include #include #include --- diff/arch/ppc/platforms/chrp_smp.c 2003-05-21 11:49:59.000000000 +0100 +++ source/arch/ppc/platforms/chrp_smp.c 2004-03-01 15:42:15.000000000 +0000 @@ -16,8 +16,6 @@ #include #include #include -#define __KERNEL_SYSCALLS__ -#include #include #include --- diff/arch/ppc/platforms/pmac_pic.c 2004-02-18 08:54:08.000000000 +0000 +++ source/arch/ppc/platforms/pmac_pic.c 2004-03-01 15:42:15.000000000 +0000 @@ -646,7 +646,7 @@ static int __init init_pmacpic_sysfs(voi printk(KERN_DEBUG "Registering pmac pic with sysfs...\n"); sysdev_class_register(&pmacpic_sysclass); - sys_device_register(&device_pmacpic); + sysdev_register(&device_pmacpic); sysdev_driver_register(&pmacpic_sysclass, &driver_pmacpic); return 0; } --- diff/arch/ppc/platforms/pmac_smp.c 2004-02-18 08:54:08.000000000 +0000 +++ source/arch/ppc/platforms/pmac_smp.c 2004-03-01 15:42:15.000000000 +0000 @@ -29,8 +29,6 @@ #include #include #include -#define __KERNEL_SYSCALLS__ -#include #include #include #include --- diff/arch/ppc/syslib/open_pic.c 2004-02-18 08:54:08.000000000 +0000 +++ source/arch/ppc/syslib/open_pic.c 2004-03-01 15:42:15.000000000 +0000 @@ -1032,7 +1032,7 @@ static int __init init_openpic_sysfs(voi printk(KERN_ERR "Failed registering openpic sys class\n"); return -ENODEV; } - rc = sys_device_register(&device_openpic); + rc = sysdev_register(&device_openpic); if (rc) { printk(KERN_ERR "Failed registering openpic sys device\n"); return -ENODEV; --- diff/arch/ppc/syslib/open_pic2.c 2004-02-18 08:54:08.000000000 +0000 +++ source/arch/ppc/syslib/open_pic2.c 2004-03-01 15:42:15.000000000 +0000 @@ -699,7 +699,7 @@ static int __init init_openpic2_sysfs(vo printk(KERN_ERR "Failed registering openpic sys class\n"); return -ENODEV; } - rc = sys_device_register(&device_openpic2); + rc = sysdev_register(&device_openpic2); if (rc) { printk(KERN_ERR "Failed registering openpic sys device\n"); return -ENODEV; --- diff/arch/ppc64/kernel/misc.S 2004-03-01 14:36:05.000000000 +0000 +++ source/arch/ppc64/kernel/misc.S 2004-03-01 15:42:15.000000000 +0000 @@ -582,17 +582,7 @@ _GLOBAL(name) \ li r3,-1; \ blr -#define __NR__exit __NR_exit - -SYSCALL(setsid) -SYSCALL(open) -SYSCALL(read) -SYSCALL(write) -SYSCALL(lseek) -SYSCALL(close) -SYSCALL(dup) SYSCALL(execve) -SYSCALL(waitpid) #ifdef CONFIG_PPC_ISERIES /* hack hack hack */ #define ppc_rtas sys_ni_syscall --- diff/arch/ppc64/kernel/pSeries_htab.c 2004-03-01 14:36:05.000000000 +0000 +++ source/arch/ppc64/kernel/pSeries_htab.c 2004-03-01 15:42:15.000000000 +0000 @@ -103,7 +103,7 @@ long pSeries_hpte_insert(unsigned long h __asm__ __volatile__ ("ptesync" : : : "memory"); - return i; + return i | (secondary << 3); } static long pSeries_hpte_remove(unsigned long hpte_group) --- diff/arch/ppc64/kernel/pSeries_lpar.c 2004-03-01 14:36:05.000000000 +0000 +++ source/arch/ppc64/kernel/pSeries_lpar.c 2004-03-01 15:42:15.000000000 +0000 @@ -379,7 +379,10 @@ long pSeries_lpar_hpte_insert(unsigned l if (lpar_rc != H_Success) return -2; - return slot; + /* Because of iSeries, we have to pass down the secondary + * bucket bit here as well + */ + return (slot & 7) | (secondary << 3); } static spinlock_t pSeries_lpar_tlbie_lock = SPIN_LOCK_UNLOCKED; --- diff/arch/ppc64/kernel/pmac_smp.c 2004-03-01 14:36:06.000000000 +0000 +++ source/arch/ppc64/kernel/pmac_smp.c 2004-03-01 15:42:15.000000000 +0000 @@ -29,8 +29,6 @@ #include #include #include -#define __KERNEL_SYSCALLS__ -#include #include #include #include --- diff/arch/ppc64/kernel/prom.c 2004-03-01 14:36:06.000000000 +0000 +++ source/arch/ppc64/kernel/prom.c 2004-03-01 15:42:15.000000000 +0000 @@ -516,6 +516,9 @@ prom_initialize_naca(unsigned long mem) return mem; } +#ifdef CONFIG_PMAC_DART +static int dart_force_on; +#endif static unsigned long __init prom_initialize_lmb(unsigned long mem) @@ -539,10 +542,12 @@ prom_initialize_lmb(unsigned long mem) prom_print(opt); prom_print(RELOC("\n")); opt += 6; - while(*opt && *opt == ' ') + while (*opt && *opt == ' ') opt++; if (!strncmp(opt, RELOC("off"), 3)) nodart = 1; + else if (!strncmp(opt, RELOC("force"), 5)) + RELOC(dart_force_on) = 1; } #else nodart = 1; @@ -763,8 +768,10 @@ void prom_initialize_dart_table(void) extern unsigned long dart_tablebase; extern unsigned long dart_tablesize; - /* Only reserve DART space if machine has more than 2Gb of RAM */ - if (lmb_end_of_DRAM() <= 0x80000000ull) + /* Only reserve DART space if machine has more than 2GB of RAM + * or if requested with iommu=on on cmdline. + */ + if (lmb_end_of_DRAM() <= 0x80000000ull && !RELOC(dart_force_on)) return; /* 512 pages is max DART tablesize. */ --- diff/arch/ppc64/mm/hash_low.S 2004-02-18 08:54:08.000000000 +0000 +++ source/arch/ppc64/mm/hash_low.S 2004-03-01 15:42:15.000000000 +0000 @@ -176,7 +176,6 @@ _GLOBAL(htab_call_hpte_insert1) beq- htab_pte_insert_failure /* Now try secondary slot */ - ori r30,r30,_PAGE_SECONDARY /* page number in r5 */ rldicl r5,r31,64-PTE_SHIFT,PTE_SHIFT @@ -215,8 +214,8 @@ _GLOBAL(htab_call_hpte_remove) b htab_insert_pte htab_pte_insert_ok: - /* Insert slot number in PTE */ - rldimi r30,r3,12,63-14 + /* Insert slot number & secondary bit in PTE */ + rldimi r30,r3,12,63-15 /* Write out the PTE with a normal write * (maybe add eieio may be good still ?) --- diff/arch/ppc64/mm/numa.c 2004-03-01 14:36:06.000000000 +0000 +++ source/arch/ppc64/mm/numa.c 2004-03-01 15:42:15.000000000 +0000 @@ -33,7 +33,10 @@ bootmem_data_t plat_node_bdata[MAX_NUMNO static unsigned long node0_io_hole_size; EXPORT_SYMBOL(node_data); +EXPORT_SYMBOL(numa_cpu_lookup_table); EXPORT_SYMBOL(numa_memory_lookup_table); +EXPORT_SYMBOL(numa_cpumask_lookup_table); +EXPORT_SYMBOL(nr_cpus_in_node); static inline void map_cpu_to_node(int cpu, int node) { --- diff/arch/s390/Kconfig 2004-03-01 14:36:06.000000000 +0000 +++ source/arch/s390/Kconfig 2004-03-01 15:42:15.000000000 +0000 @@ -143,6 +143,11 @@ config COMPAT depends on S390_SUPPORT default y +config SYSVIPC_COMPAT + bool + depends on COMPAT && SYSVIPC + default y + config BINFMT_ELF32 tristate "Kernel support for 31 bit ELF binaries" depends on S390_SUPPORT --- diff/arch/s390/kernel/compat_linux.c 2004-03-01 14:36:06.000000000 +0000 +++ source/arch/s390/kernel/compat_linux.c 2004-03-01 15:42:15.000000000 +0000 @@ -293,541 +293,6 @@ static inline long put_tv32(struct compa __put_user(i->tv_usec, &o->tv_usec))); } -struct msgbuf32 { s32 mtype; char mtext[1]; }; - -struct ipc64_perm_ds32 -{ - __kernel_key_t key; - __kernel_uid32_t uid; - __kernel_gid32_t gid; - __kernel_uid32_t cuid; - __kernel_gid32_t cgid; - compat_mode_t mode; - unsigned short __pad1; - unsigned short seq; - unsigned short __pad2; - unsigned int __unused1; - unsigned int __unused2; -}; - -struct ipc_perm32 -{ - key_t key; - compat_uid_t uid; - compat_gid_t gid; - compat_uid_t cuid; - compat_gid_t cgid; - compat_mode_t mode; - unsigned short seq; -}; - -struct semid_ds32 { - struct ipc_perm32 sem_perm; /* permissions .. see ipc.h */ - compat_time_t sem_otime; /* last semop time */ - compat_time_t sem_ctime; /* last change time */ - u32 sem_base; /* ptr to first semaphore in array */ - u32 sem_pending; /* pending operations to be processed */ - u32 sem_pending_last; /* last pending operation */ - u32 undo; /* undo requests on this array */ - unsigned short sem_nsems; /* no. of semaphores in array */ -}; - -struct semid64_ds32 { - struct ipc64_perm_ds32 sem_perm; - unsigned int __pad1; - compat_time_t sem_otime; - unsigned int __pad2; - compat_time_t sem_ctime; - u32 sem_nsems; - u32 __unused1; - u32 __unused2; -}; - -struct msqid_ds32 -{ - struct ipc_perm32 msg_perm; - u32 msg_first; - u32 msg_last; - compat_time_t msg_stime; - compat_time_t msg_rtime; - compat_time_t msg_ctime; - u32 wwait; - u32 rwait; - unsigned short msg_cbytes; - unsigned short msg_qnum; - unsigned short msg_qbytes; - compat_ipc_pid_t msg_lspid; - compat_ipc_pid_t msg_lrpid; -}; - -struct msqid64_ds32 { - struct ipc64_perm_ds32 msg_perm; - unsigned int __pad1; - compat_time_t msg_stime; - unsigned int __pad2; - compat_time_t msg_rtime; - unsigned int __pad3; - compat_time_t msg_ctime; - unsigned int msg_cbytes; - unsigned int msg_qnum; - unsigned int msg_qbytes; - compat_pid_t msg_lspid; - compat_pid_t msg_lrpid; - unsigned int __unused1; - unsigned int __unused2; -}; - - -struct shmid_ds32 { - struct ipc_perm32 shm_perm; - int shm_segsz; - compat_time_t shm_atime; - compat_time_t shm_dtime; - compat_time_t shm_ctime; - compat_ipc_pid_t shm_cpid; - compat_ipc_pid_t shm_lpid; - unsigned short shm_nattch; -}; - -struct shmid64_ds32 { - struct ipc64_perm_ds32 shm_perm; - compat_size_t shm_segsz; - compat_time_t shm_atime; - unsigned int __unused1; - compat_time_t shm_dtime; - unsigned int __unused2; - compat_time_t shm_ctime; - unsigned int __unused3; - compat_pid_t shm_cpid; - compat_pid_t shm_lpid; - unsigned int shm_nattch; - unsigned int __unused4; - unsigned int __unused5; -}; - -extern int sem_ctls[]; -#define sc_semopm (sem_ctls[2]) -#define SEMOPM_FAST 64 /* ~ 372 bytes on stack */ - -static long -do_sys32_semtimedop (int semid, struct sembuf *tsops, int nsops, - struct compat_timespec *timeout32) -{ - struct sembuf *sops, fast_sops[SEMOPM_FAST]; - struct timespec t; - mm_segment_t oldfs; - long ret; - - /* parameter checking precedence should mirror sys_semtimedop() */ - if (nsops < 1 || semid < 0) - return -EINVAL; - if (nsops > sc_semopm) - return -E2BIG; - if (nsops <= SEMOPM_FAST) - sops = fast_sops; - else { - sops = kmalloc(nsops * sizeof(*sops), GFP_KERNEL); - if (sops == NULL) - return -ENOMEM; - } - if (copy_from_user(sops, tsops, nsops * sizeof(*tsops)) || - get_compat_timespec(&t, timeout32)) - ret = -EFAULT; - else { - oldfs = get_fs(); - set_fs(KERNEL_DS); - ret = sys_semtimedop(semid, sops, nsops, &t); - set_fs(oldfs); - } - if (sops != fast_sops) - kfree(sops); - return ret; -} - -#define IPCOP_MASK(__x) (1UL << (__x)) -static int do_sys32_semctl(int first, int second, int third, void *uptr) -{ - union semun fourth; - u32 pad; - int err = -EINVAL; - - if (!uptr) - goto out; - err = -EFAULT; - if (get_user (pad, (u32 *)uptr)) - goto out; - if(third == SETVAL) - fourth.val = (int)pad; - else - fourth.__pad = (void *)A(pad); - if (IPCOP_MASK (third) & - (IPCOP_MASK (IPC_INFO) | IPCOP_MASK (SEM_INFO) | IPCOP_MASK (GETVAL) | - IPCOP_MASK (GETPID) | IPCOP_MASK (GETNCNT) | IPCOP_MASK (GETZCNT) | - IPCOP_MASK (GETALL) | IPCOP_MASK (SETALL) | IPCOP_MASK (IPC_RMID))) { - err = sys_semctl (first, second, third, fourth); - } else if (third & IPC_64) { - struct semid64_ds s; - struct semid64_ds32 *usp = (struct semid64_ds32 *)A(pad); - mm_segment_t old_fs; - int need_back_translation; - - if (third == (IPC_SET|IPC_64)) { - err = get_user (s.sem_perm.uid, &usp->sem_perm.uid); - err |= __get_user (s.sem_perm.gid, &usp->sem_perm.gid); - err |= __get_user (s.sem_perm.mode, &usp->sem_perm.mode); - if (err) - goto out; - fourth.__pad = &s; - } - need_back_translation = - (IPCOP_MASK (third) & - (IPCOP_MASK (SEM_STAT) | IPCOP_MASK (IPC_STAT))) != 0; - if (need_back_translation) - fourth.__pad = &s; - old_fs = get_fs (); - set_fs (KERNEL_DS); - err = sys_semctl (first, second, third, fourth); - set_fs (old_fs); - if (need_back_translation) { - int err2 = put_user (s.sem_perm.key, &usp->sem_perm.key); - err2 |= __put_user (high2lowuid(s.sem_perm.uid), &usp->sem_perm.uid); - err2 |= __put_user (high2lowgid(s.sem_perm.gid), &usp->sem_perm.gid); - err2 |= __put_user (high2lowuid(s.sem_perm.cuid), &usp->sem_perm.cuid); - err2 |= __put_user (high2lowgid(s.sem_perm.cgid), &usp->sem_perm.cgid); - err2 |= __put_user (s.sem_perm.mode, &usp->sem_perm.mode); - err2 |= __put_user (s.sem_perm.seq, &usp->sem_perm.seq); - err2 |= __put_user (s.sem_otime, &usp->sem_otime); - err2 |= __put_user (s.sem_ctime, &usp->sem_ctime); - err2 |= __put_user (s.sem_nsems, &usp->sem_nsems); - if (err2) err = -EFAULT; - } - } else { - struct semid_ds s; - struct semid_ds32 *usp = (struct semid_ds32 *)A(pad); - mm_segment_t old_fs; - int need_back_translation; - - if (third == IPC_SET) { - err = get_user (s.sem_perm.uid, &usp->sem_perm.uid); - err |= __get_user (s.sem_perm.gid, &usp->sem_perm.gid); - err |= __get_user (s.sem_perm.mode, &usp->sem_perm.mode); - if (err) - goto out; - fourth.__pad = &s; - } - need_back_translation = - (IPCOP_MASK (third) & - (IPCOP_MASK (SEM_STAT) | IPCOP_MASK (IPC_STAT))) != 0; - if (need_back_translation) - fourth.__pad = &s; - old_fs = get_fs (); - set_fs (KERNEL_DS); - err = sys_semctl (first, second, third, fourth); - set_fs (old_fs); - if (need_back_translation) { - int err2 = put_user (s.sem_perm.key, &usp->sem_perm.key); - err2 |= __put_user (high2lowuid(s.sem_perm.uid), &usp->sem_perm.uid); - err2 |= __put_user (high2lowgid(s.sem_perm.gid), &usp->sem_perm.gid); - err2 |= __put_user (high2lowuid(s.sem_perm.cuid), &usp->sem_perm.cuid); - err2 |= __put_user (high2lowgid(s.sem_perm.cgid), &usp->sem_perm.cgid); - err2 |= __put_user (s.sem_perm.mode, &usp->sem_perm.mode); - err2 |= __put_user (s.sem_perm.seq, &usp->sem_perm.seq); - err2 |= __put_user (s.sem_otime, &usp->sem_otime); - err2 |= __put_user (s.sem_ctime, &usp->sem_ctime); - err2 |= __put_user (s.sem_nsems, &usp->sem_nsems); - if (err2) err = -EFAULT; - } - } -out: - return err; -} - -static int do_sys32_msgsnd (int first, int second, int third, void *uptr) -{ - struct msgbuf *p = kmalloc (second + sizeof (struct msgbuf), GFP_USER); - struct msgbuf32 *up = (struct msgbuf32 *)uptr; - mm_segment_t old_fs; - int err; - - if (!p) - return -ENOMEM; - - err = -EINVAL; - if (second > MSGMAX || first < 0 || second < 0) - goto out; - - err = -EFAULT; - if (!uptr) - goto out; - if (get_user (p->mtype, &up->mtype) || - __copy_from_user (p->mtext, &up->mtext, second)) - goto out; - old_fs = get_fs (); - set_fs (KERNEL_DS); - err = sys_msgsnd (first, p, second, third); - set_fs (old_fs); -out: - kfree (p); - return err; -} - -static int do_sys32_msgrcv (int first, int second, int msgtyp, int third, - int version, void *uptr) -{ - struct msgbuf32 *up; - struct msgbuf *p; - mm_segment_t old_fs; - int err; - - if (first < 0 || second < 0) - return -EINVAL; - - if (!version) { - struct ipc_kludge_32 *uipck = (struct ipc_kludge_32 *)uptr; - struct ipc_kludge_32 ipck; - - err = -EINVAL; - if (!uptr) - goto out; - err = -EFAULT; - if (copy_from_user (&ipck, uipck, sizeof (struct ipc_kludge_32))) - goto out; - uptr = (void *)A(ipck.msgp); - msgtyp = ipck.msgtyp; - } - err = -ENOMEM; - p = kmalloc (second + sizeof (struct msgbuf), GFP_USER); - if (!p) - goto out; - old_fs = get_fs (); - set_fs (KERNEL_DS); - err = sys_msgrcv (first, p, second, msgtyp, third); - set_fs (old_fs); - if (err < 0) - goto free_then_out; - up = (struct msgbuf32 *)uptr; - if (put_user (p->mtype, &up->mtype) || - __copy_to_user (&up->mtext, p->mtext, err)) - err = -EFAULT; -free_then_out: - kfree (p); -out: - return err; -} - -static int do_sys32_msgctl (int first, int second, void *uptr) -{ - int err; - - if (IPCOP_MASK (second) & - (IPCOP_MASK (IPC_INFO) | IPCOP_MASK (MSG_INFO) | - IPCOP_MASK (IPC_RMID))) { - err = sys_msgctl (first, second, (struct msqid_ds *)uptr); - } else if (second & IPC_64) { - struct msqid64_ds m; - struct msqid64_ds32 *up = (struct msqid64_ds32 *)uptr; - mm_segment_t old_fs; - - if (second == (IPC_SET|IPC_64)) { - err = get_user (m.msg_perm.uid, &up->msg_perm.uid); - err |= __get_user (m.msg_perm.gid, &up->msg_perm.gid); - err |= __get_user (m.msg_perm.mode, &up->msg_perm.mode); - err |= __get_user (m.msg_qbytes, &up->msg_qbytes); - if (err) - goto out; - } - old_fs = get_fs (); - set_fs (KERNEL_DS); - err = sys_msgctl (first, second, (struct msqid_ds *)&m); - set_fs (old_fs); - if (IPCOP_MASK (second) & - (IPCOP_MASK (MSG_STAT) | IPCOP_MASK (IPC_STAT))) { - int err2 = put_user (m.msg_perm.key, &up->msg_perm.key); - err2 |= __put_user (high2lowuid(m.msg_perm.uid), &up->msg_perm.uid); - err2 |= __put_user (high2lowgid(m.msg_perm.gid), &up->msg_perm.gid); - err2 |= __put_user (high2lowuid(m.msg_perm.cuid), &up->msg_perm.cuid); - err2 |= __put_user (high2lowgid(m.msg_perm.cgid), &up->msg_perm.cgid); - err2 |= __put_user (m.msg_perm.mode, &up->msg_perm.mode); - err2 |= __put_user (m.msg_perm.seq, &up->msg_perm.seq); - err2 |= __put_user (m.msg_stime, &up->msg_stime); - err2 |= __put_user (m.msg_rtime, &up->msg_rtime); - err2 |= __put_user (m.msg_ctime, &up->msg_ctime); - err2 |= __put_user (m.msg_cbytes, &up->msg_cbytes); - err2 |= __put_user (m.msg_qnum, &up->msg_qnum); - err2 |= __put_user (m.msg_qbytes, &up->msg_qbytes); - err2 |= __put_user (m.msg_lspid, &up->msg_lspid); - err2 |= __put_user (m.msg_lrpid, &up->msg_lrpid); - if (err2) - err = -EFAULT; - } - } else { - struct msqid_ds m; - struct msqid_ds32 *up = (struct msqid_ds32 *)uptr; - mm_segment_t old_fs; - - if (second == IPC_SET) { - err = get_user (m.msg_perm.uid, &up->msg_perm.uid); - err |= __get_user (m.msg_perm.gid, &up->msg_perm.gid); - err |= __get_user (m.msg_perm.mode, &up->msg_perm.mode); - err |= __get_user (m.msg_qbytes, &up->msg_qbytes); - if (err) - goto out; - } - old_fs = get_fs (); - set_fs (KERNEL_DS); - err = sys_msgctl (first, second, &m); - set_fs (old_fs); - if (IPCOP_MASK (second) & - (IPCOP_MASK (MSG_STAT) | IPCOP_MASK (IPC_STAT))) { - int err2 = put_user (m.msg_perm.key, &up->msg_perm.key); - err2 |= __put_user (high2lowuid(m.msg_perm.uid), &up->msg_perm.uid); - err2 |= __put_user (high2lowgid(m.msg_perm.gid), &up->msg_perm.gid); - err2 |= __put_user (high2lowuid(m.msg_perm.cuid), &up->msg_perm.cuid); - err2 |= __put_user (high2lowgid(m.msg_perm.cgid), &up->msg_perm.cgid); - err2 |= __put_user (m.msg_perm.mode, &up->msg_perm.mode); - err2 |= __put_user (m.msg_perm.seq, &up->msg_perm.seq); - err2 |= __put_user (m.msg_stime, &up->msg_stime); - err2 |= __put_user (m.msg_rtime, &up->msg_rtime); - err2 |= __put_user (m.msg_ctime, &up->msg_ctime); - err2 |= __put_user (m.msg_cbytes, &up->msg_cbytes); - err2 |= __put_user (m.msg_qnum, &up->msg_qnum); - err2 |= __put_user (m.msg_qbytes, &up->msg_qbytes); - err2 |= __put_user (m.msg_lspid, &up->msg_lspid); - err2 |= __put_user (m.msg_lrpid, &up->msg_lrpid); - if (err2) - err = -EFAULT; - } - } - -out: - return err; -} - -static int do_sys32_shmat (int first, int second, int third, int version, void *uptr) -{ - unsigned long raddr; - u32 *uaddr = (u32 *)A((u32)third); - int err = -EINVAL; - - if (version == 1) - goto out; - err = do_shmat (first, uptr, second, &raddr); - if (err) - goto out; - err = put_user (raddr, uaddr); -out: - return err; -} - -static int do_sys32_shmctl (int first, int second, void *uptr) -{ - int err; - - if (IPCOP_MASK (second) & - (IPCOP_MASK (IPC_INFO) | IPCOP_MASK (SHM_LOCK) | IPCOP_MASK (SHM_UNLOCK) | - IPCOP_MASK (IPC_RMID))) { - if (second == (IPC_INFO|IPC_64)) - second = IPC_INFO; /* So that we don't have to translate it */ - err = sys_shmctl (first, second, (struct shmid_ds *)uptr); - } else if ((second & IPC_64) && second != (SHM_INFO|IPC_64)) { - struct shmid64_ds s; - struct shmid64_ds32 *up = (struct shmid64_ds32 *)uptr; - mm_segment_t old_fs; - - if (second == (IPC_SET|IPC_64)) { - err = get_user (s.shm_perm.uid, &up->shm_perm.uid); - err |= __get_user (s.shm_perm.gid, &up->shm_perm.gid); - err |= __get_user (s.shm_perm.mode, &up->shm_perm.mode); - if (err) - goto out; - } - old_fs = get_fs (); - set_fs (KERNEL_DS); - err = sys_shmctl (first, second, (struct shmid_ds *)&s); - set_fs (old_fs); - if (err < 0) - goto out; - - /* Mask it even in this case so it becomes a CSE. */ - if (IPCOP_MASK (second) & - (IPCOP_MASK (SHM_STAT) | IPCOP_MASK (IPC_STAT))) { - int err2 = put_user (s.shm_perm.key, &up->shm_perm.key); - err2 |= __put_user (high2lowuid(s.shm_perm.uid), &up->shm_perm.uid); - err2 |= __put_user (high2lowgid(s.shm_perm.gid), &up->shm_perm.gid); - err2 |= __put_user (high2lowuid(s.shm_perm.cuid), &up->shm_perm.cuid); - err2 |= __put_user (high2lowgid(s.shm_perm.cgid), &up->shm_perm.cgid); - err2 |= __put_user (s.shm_perm.mode, &up->shm_perm.mode); - err2 |= __put_user (s.shm_perm.seq, &up->shm_perm.seq); - err2 |= __put_user (s.shm_atime, &up->shm_atime); - err2 |= __put_user (s.shm_dtime, &up->shm_dtime); - err2 |= __put_user (s.shm_ctime, &up->shm_ctime); - err2 |= __put_user (s.shm_segsz, &up->shm_segsz); - err2 |= __put_user (s.shm_nattch, &up->shm_nattch); - err2 |= __put_user (s.shm_cpid, &up->shm_cpid); - err2 |= __put_user (s.shm_lpid, &up->shm_lpid); - if (err2) - err = -EFAULT; - } - } else { - struct shmid_ds s; - struct shmid_ds32 *up = (struct shmid_ds32 *)uptr; - mm_segment_t old_fs; - - second &= ~IPC_64; - if (second == IPC_SET) { - err = get_user (s.shm_perm.uid, &up->shm_perm.uid); - err |= __get_user (s.shm_perm.gid, &up->shm_perm.gid); - err |= __get_user (s.shm_perm.mode, &up->shm_perm.mode); - if (err) - goto out; - } - old_fs = get_fs (); - set_fs (KERNEL_DS); - err = sys_shmctl (first, second, &s); - set_fs (old_fs); - if (err < 0) - goto out; - - /* Mask it even in this case so it becomes a CSE. */ - if (second == SHM_INFO) { - struct shm_info32 { - int used_ids; - u32 shm_tot, shm_rss, shm_swp; - u32 swap_attempts, swap_successes; - } *uip = (struct shm_info32 *)uptr; - struct shm_info *kp = (struct shm_info *)&s; - int err2 = put_user (kp->used_ids, &uip->used_ids); - err2 |= __put_user (kp->shm_tot, &uip->shm_tot); - err2 |= __put_user (kp->shm_rss, &uip->shm_rss); - err2 |= __put_user (kp->shm_swp, &uip->shm_swp); - err2 |= __put_user (kp->swap_attempts, &uip->swap_attempts); - err2 |= __put_user (kp->swap_successes, &uip->swap_successes); - if (err2) - err = -EFAULT; - } else if (IPCOP_MASK (second) & - (IPCOP_MASK (SHM_STAT) | IPCOP_MASK (IPC_STAT))) { - int err2 = put_user (s.shm_perm.key, &up->shm_perm.key); - err2 |= __put_user (high2lowuid(s.shm_perm.uid), &up->shm_perm.uid); - err2 |= __put_user (high2lowgid(s.shm_perm.gid), &up->shm_perm.gid); - err2 |= __put_user (high2lowuid(s.shm_perm.cuid), &up->shm_perm.cuid); - err2 |= __put_user (high2lowgid(s.shm_perm.cgid), &up->shm_perm.cgid); - err2 |= __put_user (s.shm_perm.mode, &up->shm_perm.mode); - err2 |= __put_user (s.shm_perm.seq, &up->shm_perm.seq); - err2 |= __put_user (s.shm_atime, &up->shm_atime); - err2 |= __put_user (s.shm_dtime, &up->shm_dtime); - err2 |= __put_user (s.shm_ctime, &up->shm_ctime); - err2 |= __put_user (s.shm_segsz, &up->shm_segsz); - err2 |= __put_user (s.shm_nattch, &up->shm_nattch); - err2 |= __put_user (s.shm_cpid, &up->shm_cpid); - err2 |= __put_user (s.shm_lpid, &up->shm_lpid); - if (err2) - err = -EFAULT; - } - } -out: - return err; -} - /* * sys32_ipc() is the de-multiplexer for the SysV IPC calls in 32bit emulation. * @@ -835,84 +300,64 @@ out: */ asmlinkage int sys32_ipc (u32 call, int first, int second, int third, u32 ptr) { - int version, err; + if(call >> 16) /* hack for backward compatibility */ + return -EINVAL; - version = call >> 16; /* hack for backward compatibility */ call &= 0xffff; - if(version) - return -EINVAL; - if (call <= SEMTIMEDOP) switch (call) { case SEMTIMEDOP: - if (third) { - err = do_sys32_semtimedop(first, - (struct sembuf *)AA(ptr), - second, - (struct compat_timespec *) - AA((u32)third)); - goto out; - } + if (third) + return compat_sys_semtimedop(first, + compat_ptr(ptr), second, + compat_ptr(third)); /* else fall through for normal semop() */ case SEMOP: /* struct sembuf is the same on 32 and 64bit :)) */ - err = sys_semtimedop (first, (struct sembuf *)AA(ptr), + return sys_semtimedop (first, compat_ptr(ptr), second, NULL); - goto out; case SEMGET: - err = sys_semget (first, second, third); - goto out; + return sys_semget (first, second, third); case SEMCTL: - err = do_sys32_semctl (first, second, third, (void *)AA(ptr)); - goto out; + return compat_sys_semctl (first, second, third, + compat_ptr(ptr)); default: - err = -EINVAL; - goto out; + return -EINVAL; }; if (call <= MSGCTL) switch (call) { case MSGSND: - err = do_sys32_msgsnd (first, second, third, (void *)AA(ptr)); - goto out; + return compat_sys_msgsnd (first, second, third, + compat_ptr(ptr)); case MSGRCV: - err = do_sys32_msgrcv (first, second, 0, third, - version, (void *)AA(ptr)); - goto out; + return compat_sys_msgrcv (first, second, 0, third, + 0, compat_ptr(ptr)); case MSGGET: - err = sys_msgget ((key_t) first, second); - goto out; + return sys_msgget ((key_t) first, second); case MSGCTL: - err = do_sys32_msgctl (first, second, (void *)AA(ptr)); - goto out; + return compat_sys_msgctl (first, second, + compat_ptr(ptr)); default: - err = -EINVAL; - goto out; + return -EINVAL; } if (call <= SHMCTL) switch (call) { case SHMAT: - err = do_sys32_shmat (first, second, third, - version, (void *)AA(ptr)); - goto out; + return compat_sys_shmat (first, second, third, + 0, compat_ptr(ptr)); case SHMDT: - err = sys_shmdt ((char *)AA(ptr)); - goto out; + return sys_shmdt(compat_ptr(ptr)); case SHMGET: - err = sys_shmget (first, second, third); - goto out; + return sys_shmget(first, second, third); case SHMCTL: - err = do_sys32_shmctl (first, second, (void *)AA(ptr)); - goto out; + return compat_sys_shmctl(first, second, + compat_ptr(ptr)); default: - err = -EINVAL; - goto out; + return -EINVAL; } - err = -EINVAL; - -out: - return err; + return -EINVAL; } asmlinkage int sys32_truncate64(const char * path, unsigned long high, unsigned long low) --- diff/arch/s390/kernel/sys_s390.c 2004-03-01 14:36:06.000000000 +0000 +++ source/arch/s390/kernel/sys_s390.c 2004-03-01 15:42:15.000000000 +0000 @@ -289,11 +289,6 @@ asmlinkage int sys_olduname(struct oldol return error; } -asmlinkage long sys_ioperm(unsigned long from, unsigned long num, int on) -{ - return -ENOSYS; -} - #else /* CONFIG_ARCH_S390X */ asmlinkage int s390x_newuname(struct new_utsname * name) --- diff/arch/s390/kernel/syscalls.S 2004-02-09 10:36:09.000000000 +0000 +++ source/arch/s390/kernel/syscalls.S 2004-03-01 15:42:15.000000000 +0000 @@ -109,7 +109,7 @@ SYSCALL(sys_setpriority,sys_setpriority, NI_SYSCALL /* old profil syscall */ SYSCALL(sys_statfs,sys_statfs,compat_sys_statfs_wrapper) SYSCALL(sys_fstatfs,sys_fstatfs,compat_sys_fstatfs_wrapper) /* 100 */ -SYSCALL(sys_ioperm,sys_ni_syscall,sys_ni_syscall) +NI_SYSCALL /* ioperm for i386 */ SYSCALL(sys_socketcall,sys_socketcall,compat_sys_socketcall_wrapper) SYSCALL(sys_syslog,sys_syslog,sys32_syslog_wrapper) SYSCALL(sys_setitimer,sys_setitimer,compat_sys_setitimer_wrapper) --- diff/arch/sh/boards/adx/Makefile 2004-02-09 10:36:09.000000000 +0000 +++ source/arch/sh/boards/adx/Makefile 2004-03-01 15:42:15.000000000 +0000 @@ -1,10 +1,6 @@ # # Makefile for ADX boards # -# Note! Dependencies are done automagically by 'make dep', which also -# removes any old dependencies. DON'T put your own dependencies here -# unless it's something special (ie not a .c file). -# obj-y := setup.o irq.o irq_maskreq.o --- diff/arch/sh/boards/bigsur/Makefile 2004-02-09 10:36:09.000000000 +0000 +++ source/arch/sh/boards/bigsur/Makefile 2004-03-01 15:42:15.000000000 +0000 @@ -1,10 +1,6 @@ # # Makefile for the BigSur specific parts of the kernel # -# Note! Dependencies are done automagically by 'make dep', which also -# removes any old dependencies. DON'T put your own dependencies here -# unless it's something special (ie not a .c file). -# obj-y := setup.o io.o irq.o led.o --- diff/arch/sh/boards/cat68701/Makefile 2004-02-09 10:36:09.000000000 +0000 +++ source/arch/sh/boards/cat68701/Makefile 2004-03-01 15:42:15.000000000 +0000 @@ -1,10 +1,6 @@ # # Makefile for the CAT-68701 specific parts of the kernel # -# Note! Dependencies are done automagically by 'make dep', which also -# removes any old dependencies. DON'T put your own dependencies here -# unless it's something special (ie not a .c file). -# obj-y := setup.o irq.o --- diff/arch/sh/boards/cqreek/Makefile 2004-02-09 10:36:09.000000000 +0000 +++ source/arch/sh/boards/cqreek/Makefile 2004-03-01 15:42:15.000000000 +0000 @@ -1,10 +1,6 @@ # # Makefile for the CqREEK specific parts of the kernel # -# Note! Dependencies are done automagically by 'make dep', which also -# removes any old dependencies. DON'T put your own dependencies here -# unless it's something special (ie not a .c file). -# obj-y := setup.o irq.o --- diff/arch/sh/boards/dmida/Makefile 2003-06-30 10:07:19.000000000 +0100 +++ source/arch/sh/boards/dmida/Makefile 2004-03-01 15:42:15.000000000 +0000 @@ -2,10 +2,6 @@ # Makefile for the DataMyte Industrial Digital Assistant(tm) specific parts # of the kernel # -# Note! Dependencies are done automagically by 'make dep', which also -# removes any old dependencies. DON'T put your own dependencies here -# unless it's something special (ie not a .c file). -# obj-y := mach.o --- diff/arch/sh/boards/dreamcast/Makefile 2004-02-09 10:36:09.000000000 +0000 +++ source/arch/sh/boards/dreamcast/Makefile 2004-03-01 15:42:15.000000000 +0000 @@ -1,10 +1,6 @@ # # Makefile for the Sega Dreamcast specific parts of the kernel # -# Note! Dependencies are done automagically by 'make dep', which also -# removes any old dependencies. DON'T put your own dependencies here -# unless it's something special (ie not a .c file). -# obj-y := setup.o irq.o rtc.o --- diff/arch/sh/boards/ec3104/Makefile 2004-02-09 10:36:09.000000000 +0000 +++ source/arch/sh/boards/ec3104/Makefile 2004-03-01 15:42:15.000000000 +0000 @@ -1,10 +1,6 @@ # # Makefile for the EC3104 specific parts of the kernel # -# Note! Dependencies are done automagically by 'make dep', which also -# removes any old dependencies. DON'T put your own dependencies here -# unless it's something special (ie not a .c file). -# obj-y := setup.o io.o irq.o --- diff/arch/sh/boards/harp/Makefile 2003-06-30 10:07:19.000000000 +0100 +++ source/arch/sh/boards/harp/Makefile 2004-03-01 15:42:15.000000000 +0000 @@ -1,10 +1,6 @@ # # Makefile for STMicroelectronics board specific parts of the kernel # -# Note! Dependencies are done automagically by 'make dep', which also -# removes any old dependencies. DON'T put your own dependencies here -# unless it's something special (ie not a .c file). -# obj-y := irq.o setup.o mach.o led.o --- diff/arch/sh/boards/hp6xx/hp620/Makefile 2003-06-30 10:07:19.000000000 +0100 +++ source/arch/sh/boards/hp6xx/hp620/Makefile 2004-03-01 15:42:15.000000000 +0000 @@ -1,10 +1,6 @@ # # Makefile for the HP620 specific parts of the kernel # -# Note! Dependencies are done automagically by 'make dep', which also -# removes any old dependencies. DON'T put your own dependencies here -# unless it's something special (ie not a .c file). -# obj-y := mach.o --- diff/arch/sh/boards/hp6xx/hp680/Makefile 2004-02-18 08:54:08.000000000 +0000 +++ source/arch/sh/boards/hp6xx/hp680/Makefile 2004-03-01 15:42:15.000000000 +0000 @@ -1,10 +1,6 @@ # # Makefile for the HP680 specific parts of the kernel # -# Note! Dependencies are done automagically by 'make dep', which also -# removes any old dependencies. DON'T put your own dependencies here -# unless it's something special (ie not a .c file). -# obj-y := mach.o setup.o --- diff/arch/sh/boards/hp6xx/hp690/Makefile 2003-06-30 10:07:19.000000000 +0100 +++ source/arch/sh/boards/hp6xx/hp690/Makefile 2004-03-01 15:42:15.000000000 +0000 @@ -1,10 +1,6 @@ # # Makefile for the HP690 specific parts of the kernel # -# Note! Dependencies are done automagically by 'make dep', which also -# removes any old dependencies. DON'T put your own dependencies here -# unless it's something special (ie not a .c file). -# obj-y := mach.o --- diff/arch/sh/boards/mpc1211/Makefile 2004-02-09 10:36:09.000000000 +0000 +++ source/arch/sh/boards/mpc1211/Makefile 2004-03-01 15:42:15.000000000 +0000 @@ -1,10 +1,6 @@ # # Makefile for the Interface (CTP/PCI/MPC-SH02) specific parts of the kernel # -# Note! Dependencies are done automagically by 'make dep', which also -# removes any old dependencies. DON'T put your own dependencies here -# unless it's something special (ie not a .c file). -# obj-y := setup.o rtc.o led.o --- diff/arch/sh/boards/overdrive/Makefile 2003-06-30 10:07:19.000000000 +0100 +++ source/arch/sh/boards/overdrive/Makefile 2004-03-01 15:42:15.000000000 +0000 @@ -1,10 +1,6 @@ # # Makefile for the STMicroelectronics Overdrive specific parts of the kernel # -# Note! Dependencies are done automagically by 'make dep', which also -# removes any old dependencies. DON'T put your own dependencies here -# unless it's something special (ie not a .c file). -# obj-y := mach.o setup.o io.o irq.o led.o time.o --- diff/arch/sh/boards/saturn/Makefile 2004-02-09 10:36:09.000000000 +0000 +++ source/arch/sh/boards/saturn/Makefile 2004-03-01 15:42:15.000000000 +0000 @@ -1,10 +1,6 @@ # # Makefile for the Sega Saturn specific parts of the kernel # -# Note! Dependencies are done automagically by 'make dep', which also -# removes any old dependencies. DON'T put your own dependencies here -# unless it's something special (ie not a .c file). -# obj-y := setup.o io.o irq.o --- diff/arch/sh/boards/se/770x/Makefile 2003-06-30 10:07:19.000000000 +0100 +++ source/arch/sh/boards/se/770x/Makefile 2004-03-01 15:42:15.000000000 +0000 @@ -1,10 +1,6 @@ # # Makefile for the 770x SolutionEngine specific parts of the kernel # -# Note! Dependencies are done automagically by 'make dep', which also -# removes any old dependencies. DON'T put your own dependencies here -# unless it's something special (ie not a .c file). -# obj-y := mach.o setup.o io.o irq.o led.o --- diff/arch/sh/boards/se/7751/Makefile 2003-06-30 10:07:19.000000000 +0100 +++ source/arch/sh/boards/se/7751/Makefile 2004-03-01 15:42:15.000000000 +0000 @@ -1,10 +1,6 @@ # # Makefile for the 7751 SolutionEngine specific parts of the kernel # -# Note! Dependencies are done automagically by 'make dep', which also -# removes any old dependencies. DON'T put your own dependencies here -# unless it's something special (ie not a .c file). -# obj-y := mach.o setup.o io.o irq.o led.o --- diff/arch/sh/boards/sh2000/Makefile 2004-02-09 10:36:09.000000000 +0000 +++ source/arch/sh/boards/sh2000/Makefile 2004-03-01 15:42:15.000000000 +0000 @@ -1,10 +1,6 @@ # # Makefile for the SH2000 specific parts of the kernel # -# Note! Dependencies are done automagically by 'make dep', which also -# removes any old dependencies. DON'T put your own dependencies here -# unless it's something special (ie not a .c file). -# obj-y := setup.o --- diff/arch/sh/boards/snapgear/Makefile 2004-02-09 10:36:09.000000000 +0000 +++ source/arch/sh/boards/snapgear/Makefile 2004-03-01 15:42:15.000000000 +0000 @@ -1,10 +1,6 @@ # # Makefile for the SnapGear specific parts of the kernel # -# Note! Dependencies are done automagically by 'make dep', which also -# removes any old dependencies. DON'T put your own dependencies here -# unless it's something special (ie not a .c file). -# obj-y := setup.o io.o rtc.o --- diff/arch/sh/boards/systemh/Makefile 2004-02-09 10:36:09.000000000 +0000 +++ source/arch/sh/boards/systemh/Makefile 2004-03-01 15:42:15.000000000 +0000 @@ -1,10 +1,6 @@ # # Makefile for the SystemH specific parts of the kernel # -# Note! Dependencies are done automagically by 'make dep', which also -# removes any old dependencies. DON'T put your own dependencies here -# unless it's something special (ie not a .c file). -# obj-y := setup.o irq.o io.o --- diff/arch/sh/boards/unknown/Makefile 2003-06-30 10:07:19.000000000 +0100 +++ source/arch/sh/boards/unknown/Makefile 2004-03-01 15:42:15.000000000 +0000 @@ -1,10 +1,6 @@ # # Makefile for unknown SH boards # -# Note! Dependencies are done automagically by 'make dep', which also -# removes any old dependencies. DON'T put your own dependencies here -# unless it's something special (ie not a .c file). -# obj-y := mach.o io.o setup.o --- diff/arch/sh/cchips/hd6446x/hd64461/Makefile 2003-06-30 10:07:19.000000000 +0100 +++ source/arch/sh/cchips/hd6446x/hd64461/Makefile 2004-03-01 15:42:15.000000000 +0000 @@ -1,10 +1,6 @@ # # Makefile for the HD64461 # -# Note! Dependencies are done automagically by 'make dep', which also -# removes any old dependencies. DON'T put your own dependencies here -# unless it's something special (ie not a .c file). -# obj-y := setup.o io.o --- diff/arch/sh/cchips/hd6446x/hd64465/Makefile 2003-06-30 10:07:19.000000000 +0100 +++ source/arch/sh/cchips/hd6446x/hd64465/Makefile 2004-03-01 15:42:15.000000000 +0000 @@ -1,10 +1,6 @@ # # Makefile for the HD64465 # -# Note! Dependencies are done automagically by 'make dep', which also -# removes any old dependencies. DON'T put your own dependencies here -# unless it's something special (ie not a .c file). -# obj-y := setup.o io.o gpio.o --- diff/arch/sparc/kernel/ioport.c 2003-09-17 12:28:03.000000000 +0100 +++ source/arch/sparc/kernel/ioport.c 2004-03-01 15:42:15.000000000 +0000 @@ -360,7 +360,7 @@ void sbus_unmap_sg(struct sbus_dev *sdev /* */ -void sbus_dma_sync_single(struct sbus_dev *sdev, dma_addr_t ba, size_t size, int direction) +void sbus_dma_sync_single_for_cpu(struct sbus_dev *sdev, dma_addr_t ba, size_t size, int direction) { #if 0 unsigned long va; @@ -380,9 +380,34 @@ void sbus_dma_sync_single(struct sbus_de #endif } -void sbus_dma_sync_sg(struct sbus_dev *sdev, struct scatterlist *sg, int n, int direction) +void sbus_dma_sync_single_for_device(struct sbus_dev *sdev, dma_addr_t ba, size_t size, int direction) { - printk("sbus_dma_sync_sg: not implemented yet\n"); +#if 0 + unsigned long va; + struct resource *res; + + /* We do not need the resource, just print a message if invalid. */ + res = _sparc_find_resource(&_sparc_dvma, ba); + if (res == NULL) + panic("sbus_dma_sync_single: 0x%x\n", ba); + + va = page_address(mmu_translate_dvma(ba)); /* XXX higmem */ + /* + * XXX This bogosity will be fixed with the iommu rewrite coming soon + * to a kernel near you. - Anton + */ + /* mmu_inval_dma_area(va, (size + PAGE_SIZE-1) & PAGE_MASK); */ +#endif +} + +void sbus_dma_sync_sg_for_cpu(struct sbus_dev *sdev, struct scatterlist *sg, int n, int direction) +{ + printk("sbus_dma_sync_sg_for_cpu: not implemented yet\n"); +} + +void sbus_dma_sync_sg_for_device(struct sbus_dev *sdev, struct scatterlist *sg, int n, int direction) +{ + printk("sbus_dma_sync_sg_for_device: not implemented yet\n"); } #endif /* CONFIG_SBUS */ @@ -482,7 +507,7 @@ void pci_free_consistent(struct pci_dev * The 32-bit bus address to use is returned. * * Once the device is given the dma address, the device owns this memory - * until either pci_unmap_single or pci_dma_sync_single is performed. + * until either pci_unmap_single or pci_dma_sync_single_* is performed. */ dma_addr_t pci_map_single(struct pci_dev *hwdev, void *ptr, size_t size, int direction) @@ -591,10 +616,21 @@ void pci_unmap_sg(struct pci_dev *hwdev, * If you perform a pci_map_single() but wish to interrogate the * buffer using the cpu, yet do not wish to teardown the PCI dma * mapping, you must call this function before doing so. At the - * next point you give the PCI dma address back to the card, the + * next point you give the PCI dma address back to the card, you + * must first perform a pci_dma_sync_for_device, and then the * device again owns the buffer. */ -void pci_dma_sync_single(struct pci_dev *hwdev, dma_addr_t ba, size_t size, int direction) +void pci_dma_sync_single_for_cpu(struct pci_dev *hwdev, dma_addr_t ba, size_t size, int direction) +{ + if (direction == PCI_DMA_NONE) + BUG(); + if (direction != PCI_DMA_TODEVICE) { + mmu_inval_dma_area((unsigned long)phys_to_virt(ba), + (size + PAGE_SIZE-1) & PAGE_MASK); + } +} + +void pci_dma_sync_single_for_device(struct pci_dev *hwdev, dma_addr_t ba, size_t size, int direction) { if (direction == PCI_DMA_NONE) BUG(); @@ -607,10 +643,27 @@ void pci_dma_sync_single(struct pci_dev /* Make physical memory consistent for a set of streaming * mode DMA translations after a transfer. * - * The same as pci_dma_sync_single but for a scatter-gather list, + * The same as pci_dma_sync_single_* but for a scatter-gather list, * same rules and usage. */ -void pci_dma_sync_sg(struct pci_dev *hwdev, struct scatterlist *sg, int nents, int direction) +void pci_dma_sync_sg_for_cpu(struct pci_dev *hwdev, struct scatterlist *sg, int nents, int direction) +{ + int n; + + if (direction == PCI_DMA_NONE) + BUG(); + if (direction != PCI_DMA_TODEVICE) { + for (n = 0; n < nents; n++) { + if (page_address(sg->page) == NULL) BUG(); + mmu_inval_dma_area( + (unsigned long) page_address(sg->page), + (sg->length + PAGE_SIZE-1) & PAGE_MASK); + sg++; + } + } +} + +void pci_dma_sync_sg_for_device(struct pci_dev *hwdev, struct scatterlist *sg, int nents, int direction) { int n; --- diff/arch/sparc/kernel/process.c 2004-03-01 14:36:06.000000000 +0000 +++ source/arch/sparc/kernel/process.c 2004-03-01 15:42:15.000000000 +0000 @@ -9,7 +9,6 @@ * This file handles the architecture-dependent parts of process handling.. */ -#define __KERNEL_SYSCALLS__ #include #include @@ -19,7 +18,6 @@ #include #include #include -#include #include #include #include --- diff/arch/sparc/kernel/setup.c 2004-03-01 14:36:06.000000000 +0000 +++ source/arch/sparc/kernel/setup.c 2004-03-01 15:42:15.000000000 +0000 @@ -390,11 +390,6 @@ static int __init set_preferred_console( } console_initcall(set_preferred_console); -asmlinkage long sys_ioperm(unsigned long from, unsigned long num, int on) -{ - return -EIO; -} - extern char *sparc_cpu_type[]; extern char *sparc_fpu_type[]; --- diff/arch/sparc/kernel/smp.c 2004-03-01 14:36:06.000000000 +0000 +++ source/arch/sparc/kernel/smp.c 2004-03-01 15:42:15.000000000 +0000 @@ -33,9 +33,6 @@ #include #include -#define __KERNEL_SYSCALLS__ -#include - #define IRQ_RESCHEDULE 13 #define IRQ_STOP_CPU 14 #define IRQ_CROSS_CALL 15 --- diff/arch/sparc/kernel/sparc_ksyms.c 2004-03-01 14:36:06.000000000 +0000 +++ source/arch/sparc/kernel/sparc_ksyms.c 2004-03-01 15:42:15.000000000 +0000 @@ -206,8 +206,10 @@ EXPORT_SYMBOL(sbus_map_single); EXPORT_SYMBOL(sbus_unmap_single); EXPORT_SYMBOL(sbus_map_sg); EXPORT_SYMBOL(sbus_unmap_sg); -EXPORT_SYMBOL(sbus_dma_sync_single); -EXPORT_SYMBOL(sbus_dma_sync_sg); +EXPORT_SYMBOL(sbus_dma_sync_single_for_cpu); +EXPORT_SYMBOL(sbus_dma_sync_single_for_device); +EXPORT_SYMBOL(sbus_dma_sync_sg_for_cpu); +EXPORT_SYMBOL(sbus_dma_sync_sg_for_device); EXPORT_SYMBOL(sbus_iounmap); EXPORT_SYMBOL(sbus_ioremap); #endif @@ -219,7 +221,10 @@ EXPORT_SYMBOL(pci_alloc_consistent); EXPORT_SYMBOL(pci_free_consistent); EXPORT_SYMBOL(pci_map_single); EXPORT_SYMBOL(pci_unmap_single); -EXPORT_SYMBOL(pci_dma_sync_single); +EXPORT_SYMBOL(pci_dma_sync_single_for_cpu); +EXPORT_SYMBOL(pci_dma_sync_single_for_device); +EXPORT_SYMBOL(pci_dma_sync_sg_for_cpu); +EXPORT_SYMBOL(pci_dma_sync_sg_for_device); /* Actually, ioremap/iounmap are not PCI specific. But it is ok for drivers. */ EXPORT_SYMBOL(ioremap); EXPORT_SYMBOL(iounmap); --- diff/arch/sparc/kernel/sun4d_smp.c 2003-09-30 15:46:12.000000000 +0100 +++ source/arch/sparc/kernel/sun4d_smp.c 2004-03-01 15:42:15.000000000 +0000 @@ -32,9 +32,6 @@ #include #include -#define __KERNEL_SYSCALLS__ -#include - #define IRQ_CROSS_CALL 15 extern ctxd_t *srmmu_ctx_table_phys; --- diff/arch/sparc/kernel/sun4m_smp.c 2003-05-21 11:50:00.000000000 +0100 +++ source/arch/sparc/kernel/sun4m_smp.c 2004-03-01 15:42:15.000000000 +0000 @@ -27,9 +27,6 @@ #include #include -#define __KERNEL_SYSCALLS__ -#include - #define IRQ_RESCHEDULE 13 #define IRQ_STOP_CPU 14 #define IRQ_CROSS_CALL 15 --- diff/arch/sparc64/Kconfig 2004-03-01 14:36:06.000000000 +0000 +++ source/arch/sparc64/Kconfig 2004-03-01 15:42:15.000000000 +0000 @@ -700,12 +700,19 @@ config DEBUG_BOOTMEM depends on DEBUG_KERNEL bool "Debug BOOTMEM initialization" +config LOCKMETER + bool "Kernel lock metering" + depends on SMP && !PREEMPT + help + Say Y to enable kernel lock metering, which adds overhead to SMP locks, + but allows you to see various statistics using the lockstat command. + # We have a custom atomic_dec_and_lock() implementation but it's not # compatible with spinlock debugging so we need to fall back on # the generic version in that case. config HAVE_DEC_LOCK bool - depends on SMP && !DEBUG_SPINLOCK + depends on SMP && !DEBUG_SPINLOCK && !LOCKMETER default y config MCOUNT --- diff/arch/sparc64/kernel/pci_iommu.c 2003-08-20 14:16:26.000000000 +0100 +++ source/arch/sparc64/kernel/pci_iommu.c 2004-03-01 15:42:15.000000000 +0000 @@ -661,7 +661,7 @@ void pci_unmap_sg(struct pci_dev *pdev, /* Make physical memory consistent for a single * streaming mode DMA translation after a transfer. */ -void pci_dma_sync_single(struct pci_dev *pdev, dma_addr_t bus_addr, size_t sz, int direction) +void pci_dma_sync_single_for_cpu(struct pci_dev *pdev, dma_addr_t bus_addr, size_t sz, int direction) { struct pcidev_cookie *pcp; struct pci_iommu *iommu; @@ -722,7 +722,7 @@ void pci_dma_sync_single(struct pci_dev /* Make physical memory consistent for a set of streaming * mode DMA translations after a transfer. */ -void pci_dma_sync_sg(struct pci_dev *pdev, struct scatterlist *sglist, int nelems, int direction) +void pci_dma_sync_sg_for_cpu(struct pci_dev *pdev, struct scatterlist *sglist, int nelems, int direction) { struct pcidev_cookie *pcp; struct pci_iommu *iommu; --- diff/arch/sparc64/kernel/process.c 2003-10-09 09:47:33.000000000 +0100 +++ source/arch/sparc64/kernel/process.c 2004-03-01 15:42:15.000000000 +0000 @@ -10,7 +10,6 @@ * This file handles the architecture-dependent parts of process handling.. */ -#define __KERNEL_SYSCALLS__ #include #include @@ -22,7 +21,6 @@ #include #include #include -#include #include #include #include --- diff/arch/sparc64/kernel/sbus.c 2003-05-21 11:50:14.000000000 +0100 +++ source/arch/sparc64/kernel/sbus.c 2004-03-01 15:42:15.000000000 +0000 @@ -540,7 +540,7 @@ void sbus_unmap_sg(struct sbus_dev *sdev spin_unlock_irqrestore(&iommu->lock, flags); } -void sbus_dma_sync_single(struct sbus_dev *sdev, dma_addr_t base, size_t size, int direction) +void sbus_dma_sync_single_for_cpu(struct sbus_dev *sdev, dma_addr_t base, size_t size, int direction) { struct sbus_iommu *iommu = sdev->bus->iommu; unsigned long flags; @@ -552,7 +552,11 @@ void sbus_dma_sync_single(struct sbus_de spin_unlock_irqrestore(&iommu->lock, flags); } -void sbus_dma_sync_sg(struct sbus_dev *sdev, struct scatterlist *sg, int nents, int direction) +void sbus_dma_sync_single_for_device(struct sbus_dev *sdev, dma_addr_t base, size_t size, int direction) +{ +} + +void sbus_dma_sync_sg_for_cpu(struct sbus_dev *sdev, struct scatterlist *sg, int nents, int direction) { struct sbus_iommu *iommu = sdev->bus->iommu; unsigned long flags, size; @@ -572,6 +576,10 @@ void sbus_dma_sync_sg(struct sbus_dev *s spin_unlock_irqrestore(&iommu->lock, flags); } +void sbus_dma_sync_sg_for_device(struct sbus_dev *sdev, struct scatterlist *sg, int nents, int direction) +{ +} + /* Enable 64-bit DVMA mode for the given device. */ void sbus_set_sbus64(struct sbus_dev *sdev, int bursts) { --- diff/arch/sparc64/kernel/setup.c 2004-03-01 14:36:06.000000000 +0000 +++ source/arch/sparc64/kernel/setup.c 2004-03-01 15:42:15.000000000 +0000 @@ -603,11 +603,6 @@ static int __init set_preferred_console( } console_initcall(set_preferred_console); -asmlinkage long sys_ioperm(unsigned long from, unsigned long num, int on) -{ - return -EIO; -} - /* BUFFER is PAGE_SIZE bytes long. */ extern char *sparc_cpu_type; --- diff/arch/sparc64/kernel/smp.c 2004-02-09 10:36:09.000000000 +0000 +++ source/arch/sparc64/kernel/smp.c 2004-03-01 15:42:15.000000000 +0000 @@ -36,9 +36,6 @@ #include #include -#define __KERNEL_SYSCALLS__ -#include - extern int linux_num_cpus; extern void calibrate_delay(void); @@ -46,7 +43,6 @@ extern void calibrate_delay(void); static unsigned char boot_cpu_id; cpumask_t cpu_online_map = CPU_MASK_NONE; -atomic_t sparc64_num_cpus_possible = ATOMIC_INIT(0); cpumask_t phys_cpu_present_map = CPU_MASK_NONE; static cpumask_t smp_commenced_mask; static cpumask_t cpu_callout_map; @@ -1236,20 +1232,17 @@ void __init smp_prepare_cpus(unsigned in instance = 0; while (!cpu_find_by_instance(instance, NULL, &mid)) { - if (mid < max_cpus) { + if (mid < max_cpus) cpu_set(mid, phys_cpu_present_map); - atomic_inc(&sparc64_num_cpus_possible); - } instance++; } - if (atomic_read(&sparc64_num_cpus_possible) > max_cpus) { + if (num_possible_cpus() > max_cpus) { instance = 0; while (!cpu_find_by_instance(instance, NULL, &mid)) { if (mid != boot_cpu_id) { cpu_clear(mid, phys_cpu_present_map); - atomic_dec(&sparc64_num_cpus_possible); - if (atomic_read(&sparc64_num_cpus_possible) <= max_cpus) + if (num_possible_cpus() <= max_cpus) break; } instance++; --- diff/arch/sparc64/kernel/sparc64_ksyms.c 2004-03-01 14:36:06.000000000 +0000 +++ source/arch/sparc64/kernel/sparc64_ksyms.c 2004-03-01 15:42:15.000000000 +0000 @@ -145,7 +145,6 @@ EXPORT_SYMBOL_NOVERS(mcount); /* CPU online map and active count. */ EXPORT_SYMBOL(cpu_online_map); EXPORT_SYMBOL(phys_cpu_present_map); -EXPORT_SYMBOL(sparc64_num_cpus_possible); /* Spinlock debugging library, optional. */ #ifdef CONFIG_DEBUG_SPINLOCK @@ -214,8 +213,8 @@ EXPORT_SYMBOL(sbus_map_single); EXPORT_SYMBOL(sbus_unmap_single); EXPORT_SYMBOL(sbus_map_sg); EXPORT_SYMBOL(sbus_unmap_sg); -EXPORT_SYMBOL(sbus_dma_sync_single); -EXPORT_SYMBOL(sbus_dma_sync_sg); +EXPORT_SYMBOL(sbus_dma_sync_single_for_cpu); +EXPORT_SYMBOL(sbus_dma_sync_sg_for_cpu); #endif EXPORT_SYMBOL(outsb); EXPORT_SYMBOL(outsw); @@ -233,8 +232,8 @@ EXPORT_SYMBOL(pci_map_single); EXPORT_SYMBOL(pci_unmap_single); EXPORT_SYMBOL(pci_map_sg); EXPORT_SYMBOL(pci_unmap_sg); -EXPORT_SYMBOL(pci_dma_sync_single); -EXPORT_SYMBOL(pci_dma_sync_sg); +EXPORT_SYMBOL(pci_dma_sync_single_for_cpu); +EXPORT_SYMBOL(pci_dma_sync_sg_for_cpu); EXPORT_SYMBOL(pci_dma_supported); #endif --- diff/arch/sparc64/kernel/sys_sparc32.c 2004-03-01 14:36:06.000000000 +0000 +++ source/arch/sparc64/kernel/sys_sparc32.c 2004-03-01 15:42:15.000000000 +0000 @@ -282,11 +282,6 @@ static inline long put_tv32(struct compa __put_user(i->tv_usec, &o->tv_usec))); } -asmlinkage long sys32_ioperm(u32 from, u32 num, int on) -{ - return sys_ioperm((unsigned long)from, (unsigned long)num, on); -} - struct msgbuf32 { s32 mtype; char mtext[1]; }; struct ipc_perm32 --- diff/arch/sparc64/lib/rwlock.S 2003-11-25 15:24:57.000000000 +0000 +++ source/arch/sparc64/lib/rwlock.S 2004-03-01 15:42:15.000000000 +0000 @@ -85,5 +85,20 @@ __write_trylock_succeed: __write_trylock_fail: retl mov 0, %o0 + + .globl __read_trylock +__read_trylock: /* %o0 = lock_ptr */ + ldsw [%o0], %g5 + brlz,pn %g5, 100f + add %g5, 1, %g7 + cas [%o0], %g5, %g7 + cmp %g5, %g7 + bne,pn %icc, __read_trylock + membar #StoreLoad | #StoreStore + retl + mov 1, %o0 +100: retl + mov 0, %o0 + rwlock_impl_end: --- diff/arch/v850/kernel/rte_mb_a_pci.c 2003-08-20 14:16:26.000000000 +0100 +++ source/arch/v850/kernel/rte_mb_a_pci.c 2004-03-01 15:42:15.000000000 +0000 @@ -687,10 +687,11 @@ void pci_unmap_single (struct pci_dev *p If you perform a pci_map_single() but wish to interrogate the buffer using the cpu, yet do not wish to teardown the PCI dma mapping, you must call this function before doing so. At the next - point you give the PCI dma address back to the card, the device - again owns the buffer. */ + point you give the PCI dma address back to the card, you must first + perform a pci_dma_sync_for_device, and then the device again owns + the buffer. */ void -pci_dma_sync_single (struct pci_dev *pdev, dma_addr_t dma_addr, size_t size, +pci_dma_sync_single_for_cpu (struct pci_dev *pdev, dma_addr_t dma_addr, size_t size, int dir) { void *mb_sram_addr = PCI_TO_MB_SRAM (dma_addr); @@ -700,6 +701,22 @@ pci_dma_sync_single (struct pci_dev *pde if (dir == PCI_DMA_FROMDEVICE) memcpy (mapping->cpu_addr, mb_sram_addr, size); else if (dir == PCI_DMA_TODEVICE) + ; /* nothing to do */ + else + panic("pci_dma_sync_single: unsupported sync dir: %d", dir); +} + +void +pci_dma_sync_single_for_device (struct pci_dev *pdev, dma_addr_t dma_addr, size_t size, + int dir) +{ + void *mb_sram_addr = PCI_TO_MB_SRAM (dma_addr); + struct dma_mapping *mapping = find_dma_mapping (mb_sram_addr); + + /* Synchronize the DMA buffer with the CPU buffer if necessary. */ + if (dir == PCI_DMA_FROMDEVICE) + ; /* nothing to do */ + else if (dir == PCI_DMA_TODEVICE) memcpy (mb_sram_addr, mapping->cpu_addr, size); else panic("pci_dma_sync_single: unsupported sync dir: %d", dir); @@ -724,11 +741,18 @@ pci_unmap_sg (struct pci_dev *pdev, stru } /* Make physical memory consistent for a set of streaming mode DMA - translations after a transfer. The same as pci_dma_sync_single but + translations after a transfer. The same as pci_dma_sync_single_* but for a scatter-gather list, same rules and usage. */ void -pci_dma_sync_sg (struct pci_dev *dev, struct scatterlist *sg, int sg_len, +pci_dma_sync_sg_for_cpu (struct pci_dev *dev, struct scatterlist *sg, int sg_len, + int dir) +{ + BUG (); +} + +void +pci_dma_sync_sg_for_device (struct pci_dev *dev, struct scatterlist *sg, int sg_len, int dir) { BUG (); @@ -770,4 +794,5 @@ EXPORT_SYMBOL (pci_map_single); EXPORT_SYMBOL (pci_unmap_single); EXPORT_SYMBOL (pci_alloc_consistent); EXPORT_SYMBOL (pci_free_consistent); -EXPORT_SYMBOL (pci_dma_sync_single); +EXPORT_SYMBOL (pci_dma_sync_single_for_cpu); +EXPORT_SYMBOL (pci_dma_sync_single_for_device); --- diff/arch/x86_64/Kconfig 2004-03-01 14:36:06.000000000 +0000 +++ source/arch/x86_64/Kconfig 2004-03-01 15:42:15.000000000 +0000 @@ -381,6 +381,10 @@ config COMPAT depends on IA32_EMULATION default y +config SYSVIPC_COMPAT + bool + depends on COMPAT && SYSVIPC + default y config UID16 bool @@ -452,6 +456,7 @@ config INIT_DEBUG config DEBUG_INFO bool "Compile the kernel with debug info" depends on DEBUG_KERNEL + default n help If you say Y here the resulting kernel image will include debugging info resulting in a larger kernel image. @@ -483,9 +488,8 @@ config IOMMU_LEAK help Add a simple leak tracer to the IOMMU code. This is useful when you are debugging a buggy device driver that leaks IOMMU mappings. - -#config X86_REMOTE_DEBUG -# bool "kgdb debugging stub" + +source "arch/x86_64/Kconfig.kgdb" endmenu --- diff/arch/x86_64/Makefile 2004-03-01 14:36:06.000000000 +0000 +++ source/arch/x86_64/Makefile 2004-03-01 15:42:15.000000000 +0000 @@ -38,7 +38,7 @@ OBJCOPYFLAGS := -O binary -R .note -R .c LDFLAGS_vmlinux := -e stext cflags-$(CONFIG_MK8) += $(call check_gcc,-march=k8,) -cflags-$(CONFIG_MPSC) += $(call check_gcc,-march=pentium4,) +cflags-$(CONFIG_MPSC) += $(call check_gcc,-march=prescott,) CFLAGS += $(cflags-y) CFLAGS += -mno-red-zone --- diff/arch/x86_64/defconfig 2004-03-01 14:36:06.000000000 +0000 +++ source/arch/x86_64/defconfig 2004-03-01 15:42:15.000000000 +0000 @@ -27,6 +27,7 @@ CONFIG_SYSVIPC=y # CONFIG_BSD_PROCESS_ACCT is not set CONFIG_SYSCTL=y CONFIG_LOG_BUF_SHIFT=18 +# CONFIG_HOTPLUG is not set CONFIG_IKCONFIG=y CONFIG_IKCONFIG_PROC=y # CONFIG_EMBEDDED is not set @@ -106,6 +107,7 @@ CONFIG_ACPI_POWER=y CONFIG_ACPI_PCI=y CONFIG_ACPI_SYSTEM=y # CONFIG_ACPI_RELAXED_AML is not set +# CONFIG_X86_PM_TIMER is not set # # CPU Frequency scaling @@ -119,7 +121,6 @@ CONFIG_PCI=y CONFIG_PCI_DIRECT=y # CONFIG_PCI_LEGACY_PROC is not set # CONFIG_PCI_NAMES is not set -# CONFIG_HOTPLUG is not set # # Executable file formats / Emulations @@ -168,6 +169,7 @@ CONFIG_BLK_DEV_RAM=y CONFIG_BLK_DEV_RAM_SIZE=4096 CONFIG_BLK_DEV_INITRD=y CONFIG_LBD=y +# CONFIG_DCSSBLK is not set # # ATA/ATAPI/MFM/RLL support @@ -306,7 +308,7 @@ CONFIG_FUSION_MAX_SGE=40 # CONFIG_FUSION_CTL is not set # -# IEEE 1394 (FireWire) support (EXPERIMENTAL) +# IEEE 1394 (FireWire) support # # CONFIG_IEEE1394 is not set @@ -489,7 +491,7 @@ CONFIG_TIGON3=y # # ISDN subsystem # -# CONFIG_ISDN_BOOL is not set +# CONFIG_ISDN is not set # # Telephony Support @@ -562,7 +564,8 @@ CONFIG_SERIAL_8250_NR_UARTS=4 CONFIG_SERIAL_CORE=y CONFIG_SERIAL_CORE_CONSOLE=y CONFIG_UNIX98_PTYS=y -CONFIG_UNIX98_PTY_COUNT=256 +CONFIG_LEGACY_PTYS=y +CONFIG_LEGACY_PTY_COUNT=256 # # Mice @@ -718,7 +721,6 @@ CONFIG_ISO9660_FS=y CONFIG_PROC_FS=y CONFIG_PROC_KCORE=y # CONFIG_DEVFS_FS is not set -CONFIG_DEVPTS_FS=y # CONFIG_DEVPTS_FS_XATTR is not set CONFIG_TMPFS=y CONFIG_HUGETLBFS=y @@ -731,6 +733,7 @@ CONFIG_RAMFS=y # CONFIG_ADFS_FS is not set # CONFIG_AFFS_FS is not set # CONFIG_HFS_FS is not set +# CONFIG_HFSPLUS_FS is not set # CONFIG_BEFS_FS is not set # CONFIG_BFS_FS is not set # CONFIG_EFS_FS is not set @@ -761,7 +764,6 @@ CONFIG_SUNRPC=y # CONFIG_CIFS is not set # CONFIG_NCP_FS is not set # CONFIG_CODA_FS is not set -# CONFIG_INTERMEZZO_FS is not set # CONFIG_AFS_FS is not set # --- diff/arch/x86_64/ia32/ipc32.c 2004-03-01 14:36:06.000000000 +0000 +++ source/arch/x86_64/ia32/ipc32.c 2004-03-01 15:42:15.000000000 +0000 @@ -1,656 +1,19 @@ #include -#include -#include -#include +#include +#include #include +#include #include #include -#include #include -#include #include #include -#include -#include -#include -#include -#include - -#include - -/* - * sys32_ipc() is the de-multiplexer for the SysV IPC calls in 32bit emulation.. - * - * This is really horribly ugly. - */ - -struct msgbuf32 { - s32 mtype; - char mtext[1]; -}; - -struct ipc_perm32 { - int key; - compat_uid_t uid; - compat_gid_t gid; - compat_uid_t cuid; - compat_gid_t cgid; - unsigned short mode; - unsigned short seq; -}; - -struct ipc64_perm32 { - unsigned key; - compat_uid32_t uid; - compat_gid32_t gid; - compat_uid32_t cuid; - compat_gid32_t cgid; - unsigned short mode; - unsigned short __pad1; - unsigned short seq; - unsigned short __pad2; - unsigned int unused1; - unsigned int unused2; -}; - -struct semid_ds32 { - struct ipc_perm32 sem_perm; /* permissions .. see ipc.h */ - compat_time_t sem_otime; /* last semop time */ - compat_time_t sem_ctime; /* last change time */ - u32 sem_base; /* ptr to first semaphore in array */ - u32 sem_pending; /* pending operations to be processed */ - u32 sem_pending_last; /* last pending operation */ - u32 undo; /* undo requests on this array */ - unsigned short sem_nsems; /* no. of semaphores in array */ -}; - -struct semid64_ds32 { - struct ipc64_perm32 sem_perm; - compat_time_t sem_otime; - unsigned int __unused1; - compat_time_t sem_ctime; - unsigned int __unused2; - unsigned int sem_nsems; - unsigned int __unused3; - unsigned int __unused4; -}; - -struct msqid_ds32 { - struct ipc_perm32 msg_perm; - u32 msg_first; - u32 msg_last; - compat_time_t msg_stime; - compat_time_t msg_rtime; - compat_time_t msg_ctime; - u32 wwait; - u32 rwait; - unsigned short msg_cbytes; - unsigned short msg_qnum; - unsigned short msg_qbytes; - compat_ipc_pid_t msg_lspid; - compat_ipc_pid_t msg_lrpid; -}; - -struct msqid64_ds32 { - struct ipc64_perm32 msg_perm; - compat_time_t msg_stime; - unsigned int __unused1; - compat_time_t msg_rtime; - unsigned int __unused2; - compat_time_t msg_ctime; - unsigned int __unused3; - unsigned int msg_cbytes; - unsigned int msg_qnum; - unsigned int msg_qbytes; - compat_pid_t msg_lspid; - compat_pid_t msg_lrpid; - unsigned int __unused4; - unsigned int __unused5; -}; - -struct shmid_ds32 { - struct ipc_perm32 shm_perm; - int shm_segsz; - compat_time_t shm_atime; - compat_time_t shm_dtime; - compat_time_t shm_ctime; - compat_ipc_pid_t shm_cpid; - compat_ipc_pid_t shm_lpid; - unsigned short shm_nattch; -}; - -struct shmid64_ds32 { - struct ipc64_perm32 shm_perm; - compat_size_t shm_segsz; - compat_time_t shm_atime; - unsigned int __unused1; - compat_time_t shm_dtime; - unsigned int __unused2; - compat_time_t shm_ctime; - unsigned int __unused3; - compat_pid_t shm_cpid; - compat_pid_t shm_lpid; - unsigned int shm_nattch; - unsigned int __unused4; - unsigned int __unused5; -}; - -struct shminfo64_32 { - unsigned int shmmax; - unsigned int shmmin; - unsigned int shmmni; - unsigned int shmseg; - unsigned int shmall; - unsigned int __unused1; - unsigned int __unused2; - unsigned int __unused3; - unsigned int __unused4; -}; - -struct shm_info32 { - int used_ids; - u32 shm_tot, shm_rss, shm_swp; - u32 swap_attempts, swap_successes; -}; - -struct ipc_kludge { - u32 msgp; - s32 msgtyp; -}; - - -#define A(__x) ((unsigned long)(__x)) -#define AA(__x) ((unsigned long)(__x)) - -#define SEMOP 1 -#define SEMGET 2 -#define SEMCTL 3 -#define TIMEDSEMOP 4 -#define MSGSND 11 -#define MSGRCV 12 -#define MSGGET 13 -#define MSGCTL 14 -#define SHMAT 21 -#define SHMDT 22 -#define SHMGET 23 -#define SHMCTL 24 - -#define IPCOP_MASK(__x) (1UL << (__x)) - -static int -ipc_parse_version32 (int *cmd) -{ - if (*cmd & IPC_64) { - *cmd ^= IPC_64; - return IPC_64; - } else { - return IPC_OLD; - } -} - -static int put_semid(void *user_semid, struct semid64_ds *s, int version) -{ - int err2; - switch (version) { - case IPC_64: { - struct semid64_ds32 *usp64 = (struct semid64_ds32 *) user_semid; - - if (!access_ok(VERIFY_WRITE, usp64, sizeof(*usp64))) { - err2 = -EFAULT; - break; - } - err2 = __put_user(s->sem_perm.key, &usp64->sem_perm.key); - err2 |= __put_user(s->sem_perm.uid, &usp64->sem_perm.uid); - err2 |= __put_user(s->sem_perm.gid, &usp64->sem_perm.gid); - err2 |= __put_user(s->sem_perm.cuid, &usp64->sem_perm.cuid); - err2 |= __put_user(s->sem_perm.cgid, &usp64->sem_perm.cgid); - err2 |= __put_user(s->sem_perm.mode, &usp64->sem_perm.mode); - err2 |= __put_user(s->sem_perm.seq, &usp64->sem_perm.seq); - err2 |= __put_user(s->sem_otime, &usp64->sem_otime); - err2 |= __put_user(s->sem_ctime, &usp64->sem_ctime); - err2 |= __put_user(s->sem_nsems, &usp64->sem_nsems); - break; - } - default: { - struct semid_ds32 *usp32 = (struct semid_ds32 *) user_semid; - - if (!access_ok(VERIFY_WRITE, usp32, sizeof(*usp32))) { - err2 = -EFAULT; - break; - } - err2 = __put_user(s->sem_perm.key, &usp32->sem_perm.key); - err2 |= __put_user(s->sem_perm.uid, &usp32->sem_perm.uid); - err2 |= __put_user(s->sem_perm.gid, &usp32->sem_perm.gid); - err2 |= __put_user(s->sem_perm.cuid, &usp32->sem_perm.cuid); - err2 |= __put_user(s->sem_perm.cgid, &usp32->sem_perm.cgid); - err2 |= __put_user(s->sem_perm.mode, &usp32->sem_perm.mode); - err2 |= __put_user(s->sem_perm.seq, &usp32->sem_perm.seq); - err2 |= __put_user(s->sem_otime, &usp32->sem_otime); - err2 |= __put_user(s->sem_ctime, &usp32->sem_ctime); - err2 |= __put_user(s->sem_nsems, &usp32->sem_nsems); - break; - } - } - return err2; -} - -static int -semctl32 (int first, int second, int third, void *uptr) -{ - union semun fourth; - u32 pad; - int err; - struct semid64_ds s; - mm_segment_t old_fs; - int version = ipc_parse_version32(&third); - - if (!uptr) - return -EINVAL; - if (get_user(pad, (u32 *)uptr)) - return -EFAULT; - if (third == SETVAL) - fourth.val = (int)pad; - else - fourth.__pad = (void *)A(pad); - switch (third) { - case IPC_INFO: - case IPC_RMID: - case IPC_SET: - case SEM_INFO: - case GETVAL: - case GETPID: - case GETNCNT: - case GETZCNT: - case GETALL: - case SETVAL: - case SETALL: - err = sys_semctl(first, second, third, fourth); - break; - - case IPC_STAT: - case SEM_STAT: - fourth.__pad = &s; - old_fs = get_fs(); - set_fs(KERNEL_DS); - err = sys_semctl(first, second, third, fourth); - set_fs(old_fs); - if (!err) - err = put_semid((void *)A(pad), &s, version); - break; - default: - err = -EINVAL; - break; - } - return err; -} - -#define MAXBUF (64*1024) - -static int -do_sys32_msgsnd (int first, int second, int third, void *uptr) -{ - struct msgbuf *p; - struct msgbuf32 *up = (struct msgbuf32 *)uptr; - mm_segment_t old_fs; - int err; - - if (second >= MAXBUF-sizeof(struct msgbuf)) - return -EINVAL; - p = kmalloc(second + sizeof(struct msgbuf), GFP_USER); - if (!p) - return -ENOMEM; - err = get_user(p->mtype, &up->mtype); - err |= (copy_from_user(p->mtext, &up->mtext, second) ? -EFAULT : 0); - if (err) - goto out; - old_fs = get_fs(); - set_fs(KERNEL_DS); - err = sys_msgsnd(first, p, second, third); - set_fs(old_fs); - out: - kfree(p); - return err; -} - -static int -do_sys32_msgrcv (int first, int second, int msgtyp, int third, int version, void *uptr) -{ - struct msgbuf32 *up; - struct msgbuf *p; - mm_segment_t old_fs; - int err; - - if (!version) { - struct ipc_kludge *uipck = (struct ipc_kludge *)uptr; - struct ipc_kludge ipck; - - err = -EINVAL; - if (!uptr) - goto out; - err = -EFAULT; - if (copy_from_user(&ipck, uipck, sizeof(struct ipc_kludge))) - goto out; - uptr = (void *)A(ipck.msgp); - msgtyp = ipck.msgtyp; - } - if (second >= MAXBUF-sizeof(struct msgbuf)) - return -EINVAL; - err = -ENOMEM; - p = kmalloc(second + sizeof(struct msgbuf), GFP_USER); - if (!p) - goto out; - old_fs = get_fs(); - set_fs(KERNEL_DS); - err = sys_msgrcv(first, p, second, msgtyp, third); - set_fs(old_fs); - if (err < 0) - goto free_then_out; - up = (struct msgbuf32 *)uptr; - if (put_user(p->mtype, &up->mtype) || copy_to_user(&up->mtext, p->mtext, err)) - err = -EFAULT; -free_then_out: - kfree(p); -out: - return err; -} - - -static int -msgctl32 (int first, int second, void *uptr) -{ - int err = -EINVAL, err2; - struct msqid_ds m; - struct msqid64_ds m64; - struct msqid_ds32 *up32 = (struct msqid_ds32 *)uptr; - struct msqid64_ds32 *up64 = (struct msqid64_ds32 *)uptr; - mm_segment_t old_fs; - int version = ipc_parse_version32(&second); - - switch (second) { - case IPC_INFO: - case IPC_RMID: - case MSG_INFO: - err = sys_msgctl(first, second, (struct msqid_ds *)uptr); - break; - - case IPC_SET: - if (version == IPC_64) { - err = get_user(m.msg_perm.uid, &up64->msg_perm.uid); - err |= get_user(m.msg_perm.gid, &up64->msg_perm.gid); - err |= get_user(m.msg_perm.mode, &up64->msg_perm.mode); - err |= get_user(m.msg_qbytes, &up64->msg_qbytes); - } else { - err = get_user(m.msg_perm.uid, &up32->msg_perm.uid); - err |= get_user(m.msg_perm.gid, &up32->msg_perm.gid); - err |= get_user(m.msg_perm.mode, &up32->msg_perm.mode); - err |= get_user(m.msg_qbytes, &up32->msg_qbytes); - } - if (err) - break; - old_fs = get_fs(); - set_fs(KERNEL_DS); - err = sys_msgctl(first, second, &m); - set_fs(old_fs); - break; - - case IPC_STAT: - case MSG_STAT: - old_fs = get_fs(); - set_fs(KERNEL_DS); - err = sys_msgctl(first, second, (void *) &m64); - set_fs(old_fs); - if (version == IPC_64) { - if (!access_ok(VERIFY_WRITE, up64, sizeof(*up64))) { - err = -EFAULT; - break; - } - err2 = __put_user(m64.msg_perm.key, &up64->msg_perm.key); - err2 |= __put_user(m64.msg_perm.uid, &up64->msg_perm.uid); - err2 |= __put_user(m64.msg_perm.gid, &up64->msg_perm.gid); - err2 |= __put_user(m64.msg_perm.cuid, &up64->msg_perm.cuid); - err2 |= __put_user(m64.msg_perm.cgid, &up64->msg_perm.cgid); - err2 |= __put_user(m64.msg_perm.mode, &up64->msg_perm.mode); - err2 |= __put_user(m64.msg_perm.seq, &up64->msg_perm.seq); - err2 |= __put_user(m64.msg_stime, &up64->msg_stime); - err2 |= __put_user(m64.msg_rtime, &up64->msg_rtime); - err2 |= __put_user(m64.msg_ctime, &up64->msg_ctime); - err2 |= __put_user(m64.msg_cbytes, &up64->msg_cbytes); - err2 |= __put_user(m64.msg_qnum, &up64->msg_qnum); - err2 |= __put_user(m64.msg_qbytes, &up64->msg_qbytes); - err2 |= __put_user(m64.msg_lspid, &up64->msg_lspid); - err2 |= __put_user(m64.msg_lrpid, &up64->msg_lrpid); - if (err2) - err = -EFAULT; - } else { - if (!access_ok(VERIFY_WRITE, up32, sizeof(*up32))) { - err = -EFAULT; - break; - } - err2 = __put_user(m64.msg_perm.key, &up32->msg_perm.key); - err2 |= __put_user(m64.msg_perm.uid, &up32->msg_perm.uid); - err2 |= __put_user(m64.msg_perm.gid, &up32->msg_perm.gid); - err2 |= __put_user(m64.msg_perm.cuid, &up32->msg_perm.cuid); - err2 |= __put_user(m64.msg_perm.cgid, &up32->msg_perm.cgid); - err2 |= __put_user(m64.msg_perm.mode, &up32->msg_perm.mode); - err2 |= __put_user(m64.msg_perm.seq, &up32->msg_perm.seq); - err2 |= __put_user(m64.msg_stime, &up32->msg_stime); - err2 |= __put_user(m64.msg_rtime, &up32->msg_rtime); - err2 |= __put_user(m64.msg_ctime, &up32->msg_ctime); - err2 |= __put_user(m64.msg_cbytes, &up32->msg_cbytes); - err2 |= __put_user(m64.msg_qnum, &up32->msg_qnum); - err2 |= __put_user(m64.msg_qbytes, &up32->msg_qbytes); - err2 |= __put_user(m64.msg_lspid, &up32->msg_lspid); - err2 |= __put_user(m64.msg_lrpid, &up32->msg_lrpid); - if (err2) - err = -EFAULT; - } - break; - } - return err; -} - -static int -shmat32 (int first, int second, int third, int version, void *uptr) -{ - unsigned long raddr; - u32 *uaddr = (u32 *)A((u32)third); - int err; - - if (version == 1) - return -EINVAL; /* iBCS2 emulator entry point: unsupported */ - err = do_shmat(first, uptr, second, &raddr); - if (err) - return err; - return put_user(raddr, uaddr); -} - -static int put_shmid64(struct shmid64_ds *s64p, void *uptr, int version) -{ - int err2; -#define s64 (*s64p) - if (version == IPC_64) { - struct shmid64_ds32 *up64 = (struct shmid64_ds32 *)uptr; - - if (!access_ok(VERIFY_WRITE, up64, sizeof(*up64))) - return -EFAULT; - - err2 = __put_user(s64.shm_perm.key, &up64->shm_perm.key); - err2 |= __put_user(s64.shm_perm.uid, &up64->shm_perm.uid); - err2 |= __put_user(s64.shm_perm.gid, &up64->shm_perm.gid); - err2 |= __put_user(s64.shm_perm.cuid, &up64->shm_perm.cuid); - err2 |= __put_user(s64.shm_perm.cgid, &up64->shm_perm.cgid); - err2 |= __put_user(s64.shm_perm.mode, &up64->shm_perm.mode); - err2 |= __put_user(s64.shm_perm.seq, &up64->shm_perm.seq); - err2 |= __put_user(s64.shm_atime, &up64->shm_atime); - err2 |= __put_user(s64.shm_dtime, &up64->shm_dtime); - err2 |= __put_user(s64.shm_ctime, &up64->shm_ctime); - err2 |= __put_user(s64.shm_segsz, &up64->shm_segsz); - err2 |= __put_user(s64.shm_nattch, &up64->shm_nattch); - err2 |= __put_user(s64.shm_cpid, &up64->shm_cpid); - err2 |= __put_user(s64.shm_lpid, &up64->shm_lpid); - } else { - struct shmid_ds32 *up32 = (struct shmid_ds32 *)uptr; - - if (!access_ok(VERIFY_WRITE, up32, sizeof(*up32))) - return -EFAULT; - - err2 = __put_user(s64.shm_perm.key, &up32->shm_perm.key); - err2 |= __put_user(s64.shm_perm.uid, &up32->shm_perm.uid); - err2 |= __put_user(s64.shm_perm.gid, &up32->shm_perm.gid); - err2 |= __put_user(s64.shm_perm.cuid, &up32->shm_perm.cuid); - err2 |= __put_user(s64.shm_perm.cgid, &up32->shm_perm.cgid); - err2 |= __put_user(s64.shm_perm.mode, &up32->shm_perm.mode); - err2 |= __put_user(s64.shm_perm.seq, &up32->shm_perm.seq); - err2 |= __put_user(s64.shm_atime, &up32->shm_atime); - err2 |= __put_user(s64.shm_dtime, &up32->shm_dtime); - err2 |= __put_user(s64.shm_ctime, &up32->shm_ctime); - err2 |= __put_user(s64.shm_segsz, &up32->shm_segsz); - err2 |= __put_user(s64.shm_nattch, &up32->shm_nattch); - err2 |= __put_user(s64.shm_cpid, &up32->shm_cpid); - err2 |= __put_user(s64.shm_lpid, &up32->shm_lpid); - } -#undef s64 - return err2 ? -EFAULT : 0; -} -static int -shmctl32 (int first, int second, void *uptr) -{ - int err = -EFAULT, err2; - struct shmid_ds s; - struct shmid64_ds s64; - mm_segment_t old_fs; - struct shm_info32 *uip = (struct shm_info32 *)uptr; - struct shm_info si; - int version = ipc_parse_version32(&second); - struct shminfo64 smi; - struct shminfo *usi32 = (struct shminfo *) uptr; - struct shminfo64_32 *usi64 = (struct shminfo64_32 *) uptr; - - switch (second) { - case IPC_INFO: - old_fs = get_fs(); - set_fs(KERNEL_DS); - err = sys_shmctl(first, second, (struct shmid_ds *)&smi); - set_fs(old_fs); - - if (version == IPC_64) { - if (!access_ok(VERIFY_WRITE, usi64, sizeof(*usi64))) { - err = -EFAULT; - break; - } - err2 = __put_user(smi.shmmax, &usi64->shmmax); - err2 |= __put_user(smi.shmmin, &usi64->shmmin); - err2 |= __put_user(smi.shmmni, &usi64->shmmni); - err2 |= __put_user(smi.shmseg, &usi64->shmseg); - err2 |= __put_user(smi.shmall, &usi64->shmall); - } else { - if (!access_ok(VERIFY_WRITE, usi32, sizeof(*usi32))) { - err = -EFAULT; - break; - } - err2 = __put_user(smi.shmmax, &usi32->shmmax); - err2 |= __put_user(smi.shmmin, &usi32->shmmin); - err2 |= __put_user(smi.shmmni, &usi32->shmmni); - err2 |= __put_user(smi.shmseg, &usi32->shmseg); - err2 |= __put_user(smi.shmall, &usi32->shmall); - } - if (err2) - err = -EFAULT; - break; - - case IPC_RMID: - case SHM_LOCK: - case SHM_UNLOCK: - err = sys_shmctl(first, second, (struct shmid_ds *)uptr); - break; - - case IPC_SET: - if (version == IPC_64) { - struct shmid64_ds32 *up64 = (struct shmid64_ds32 *)uptr; - err = get_user(s.shm_perm.uid, &up64->shm_perm.uid); - err |= get_user(s.shm_perm.gid, &up64->shm_perm.gid); - err |= get_user(s.shm_perm.mode, &up64->shm_perm.mode); - } else { - struct shmid_ds32 *up32 = (struct shmid_ds32 *)uptr; - err = get_user(s.shm_perm.uid, &up32->shm_perm.uid); - err |= get_user(s.shm_perm.gid, &up32->shm_perm.gid); - err |= get_user(s.shm_perm.mode, &up32->shm_perm.mode); - } - if (err) - break; - old_fs = get_fs(); - set_fs(KERNEL_DS); - err = sys_shmctl(first, second, &s); - set_fs(old_fs); - break; - - case IPC_STAT: - case SHM_STAT: - old_fs = get_fs(); - set_fs(KERNEL_DS); - err = sys_shmctl(first, second, (void *) &s64); - set_fs(old_fs); - - if (err < 0) - break; - err2 = put_shmid64(&s64, uptr, version); - if (err2) - err = err2; - break; - - case SHM_INFO: - old_fs = get_fs(); - set_fs(KERNEL_DS); - err = sys_shmctl(first, second, (void *)&si); - set_fs(old_fs); - if (err < 0) - break; - - if (!access_ok(VERIFY_WRITE, uip, sizeof(*uip))) { - err = -EFAULT; - break; - } - err2 = __put_user(si.used_ids, &uip->used_ids); - err2 |= __put_user(si.shm_tot, &uip->shm_tot); - err2 |= __put_user(si.shm_rss, &uip->shm_rss); - err2 |= __put_user(si.shm_swp, &uip->shm_swp); - err2 |= __put_user(si.swap_attempts, &uip->swap_attempts); - err2 |= __put_user(si.swap_successes, &uip->swap_successes); - if (err2) - err = -EFAULT; - break; - default: - err = -EINVAL; - break; - } - return err; -} - -extern int sem_ctls[]; - -static long semtimedop32(int semid, struct sembuf *sb, - unsigned nsops, struct compat_timespec *ts32) -{ - struct timespec ts; - mm_segment_t oldfs = get_fs(); - long ret; - - if (nsops > sem_ctls[2]) - return -E2BIG; - if (!access_ok(VERIFY_READ, sb, nsops * sizeof(struct sembuf))) - return -EFAULT; - if (ts32 && get_compat_timespec(&ts, ts32)) - return -EFAULT; - - set_fs(KERNEL_DS); - ret = sys_semtimedop(semid, sb, nsops, ts32 ? &ts : NULL); - set_fs(oldfs); - return ret; -} +#include asmlinkage long -sys32_ipc (u32 call, int first, int second, int third, u32 ptr, u32 fifth) +sys32_ipc(u32 call, int first, int second, int third, + compat_uptr_t ptr, u32 fifth) { int version; @@ -660,35 +23,35 @@ sys32_ipc (u32 call, int first, int seco switch (call) { case SEMOP: /* struct sembuf is the same on 32 and 64bit :)) */ - return sys_semtimedop(first, (struct sembuf *)AA(ptr), second, - NULL); - case TIMEDSEMOP: - return semtimedop32(first, (struct sembuf *)AA(ptr), second, - (struct compat_timespec *)AA(fifth)); + return sys_semtimedop(first, compat_ptr(ptr), second, NULL); + case SEMTIMEDOP: + return compat_sys_semtimedop(first, compat_ptr(ptr), second, + compat_ptr(fifth)); case SEMGET: return sys_semget(first, second, third); case SEMCTL: - return semctl32(first, second, third, (void *)AA(ptr)); + return compat_sys_semctl(first, second, third, compat_ptr(ptr)); case MSGSND: - return do_sys32_msgsnd(first, second, third, (void *)AA(ptr)); + return compat_sys_msgsnd(first, second, third, compat_ptr(ptr)); case MSGRCV: - return do_sys32_msgrcv(first, second, fifth, third, version, (void *)AA(ptr)); + return compat_sys_msgrcv(first, second, fifth, third, + version, compat_ptr(ptr)); case MSGGET: return sys_msgget((key_t) first, second); case MSGCTL: - return msgctl32(first, second, (void *)AA(ptr)); + return compat_sys_msgctl(first, second, compat_ptr(ptr)); case SHMAT: - return shmat32(first, second, third, version, (void *)AA(ptr)); + return compat_sys_shmat(first, second, third, version, + compat_ptr(ptr)); break; case SHMDT: - return sys_shmdt((char *)AA(ptr)); + return sys_shmdt(compat_ptr(ptr)); case SHMGET: return sys_shmget(first, second, third); case SHMCTL: - return shmctl32(first, second, (void *)AA(ptr)); + return compat_sys_shmctl(first, second, compat_ptr(ptr)); } return -ENOSYS; } - --- diff/arch/x86_64/kernel/Makefile 2004-03-01 14:36:06.000000000 +0000 +++ source/arch/x86_64/kernel/Makefile 2004-03-01 15:42:15.000000000 +0000 @@ -27,6 +27,7 @@ obj-$(CONFIG_DUMMY_IOMMU) += pci-nommu.o obj-$(CONFIG_SWIOTLB) += swiotlb.o obj-$(CONFIG_MODULES) += module.o +obj-$(CONFIG_KGDB) += kgdb_stub.o obj-y += topology.o --- diff/arch/x86_64/kernel/acpi/boot.c 2004-02-18 08:54:08.000000000 +0000 +++ source/arch/x86_64/kernel/acpi/boot.c 2004-03-01 15:42:15.000000000 +0000 @@ -48,11 +48,12 @@ #define PREFIX "ACPI: " -int acpi_noirq __initdata = 0; /* skip ACPI IRQ initialization */ +int acpi_noirq __initdata; /* skip ACPI IRQ initialization */ int acpi_ht __initdata = 1; /* enable HT */ int acpi_lapic; int acpi_ioapic; +int acpi_strict; /* -------------------------------------------------------------------------- Boot-time Configuration @@ -264,7 +265,7 @@ acpi_parse_hpet ( * programs the PIC-mode SCI to Level Trigger. * (NO-OP if the BIOS set Level Trigger already) * - * If a PIC-mode SCI is not recogznied or gives spurious IRQ7's + * If a PIC-mode SCI is not recognized or gives spurious IRQ7's * it may require Edge Trigger -- use "acpi_pic_sci=edge" * (NO-OP if the BIOS set Edge Trigger already) * --- diff/arch/x86_64/kernel/apic.c 2004-03-01 14:36:06.000000000 +0000 +++ source/arch/x86_64/kernel/apic.c 2004-03-01 15:42:15.000000000 +0000 @@ -553,7 +553,7 @@ static int __init init_lapic_sysfs(void) /* XXX: remove suspend/resume procs if !apic_pm_state.active? */ error = sysdev_class_register(&lapic_sysclass); if (!error) - error = sys_device_register(&device_lapic); + error = sysdev_register(&device_lapic); return error; } device_initcall(init_lapic_sysfs); --- diff/arch/x86_64/kernel/i8259.c 2004-01-19 10:22:55.000000000 +0000 +++ source/arch/x86_64/kernel/i8259.c 2004-03-01 15:42:15.000000000 +0000 @@ -423,14 +423,14 @@ static struct sysdev_class timer_sysclas static struct sys_device device_timer = { .id = 0, - .cls &timer_sysclass, + .cls = &timer_sysclass, }; static int __init init_timer_sysfs(void) { int error = sysdev_class_register(&timer_sysclass); if (!error) - error = sys_device_register(&device_timer); + error = sysdev_register(&device_timer); return error; } --- diff/arch/x86_64/kernel/irq.c 2004-03-01 14:36:06.000000000 +0000 +++ source/arch/x86_64/kernel/irq.c 2004-03-01 15:42:15.000000000 +0000 @@ -405,6 +405,9 @@ out: spin_unlock(&desc->lock); irq_exit(); + + kgdb_process_breakpoint(); + return 1; } --- diff/arch/x86_64/kernel/mce.c 2004-03-01 14:36:06.000000000 +0000 +++ source/arch/x86_64/kernel/mce.c 2004-03-01 15:42:15.000000000 +0000 @@ -73,7 +73,9 @@ static void print_mce(struct mce *m) printk("CPU %d: Machine Check Exception: %16Lx Bank %d: %016Lx\n", m->cpu, m->mcgstatus, m->bank, m->status); if (m->rip) { - printk("RIP %02x:<%016Lx> ", m->cs, m->rip); + printk("RIP%s %02x:<%016Lx> ", + !(m->mcgstatus & MCG_STATUS_EIPV) ? " !INEXACT!" : "", + m->cs, m->rip); if (m->cs == __KERNEL_CS) print_symbol("{%s}", m->rip); printk("\n"); @@ -133,7 +135,7 @@ void do_machine_check(struct pt_regs * r return; if (!(m.mcgstatus & MCG_STATUS_RIPV)) kill_it = 1; - if (regs && (m.mcgstatus & MCG_STATUS_EIPV)) { + if (regs) { m.rip = regs->rip; m.cs = regs->cs; } @@ -448,7 +450,7 @@ static __init int mce_init_device(void) return -EIO; err = sysdev_class_register(&mce_sysclass); if (!err) - err = sys_device_register(&device_mce); + err = sysdev_register(&device_mce); if (!err) { /* could create per CPU objects, but is not worth it. */ sysdev_create_file(&device_mce, &attr_disabled_banks); --- diff/arch/x86_64/kernel/mpparse.c 2004-03-01 14:36:06.000000000 +0000 +++ source/arch/x86_64/kernel/mpparse.c 2004-03-01 15:42:15.000000000 +0000 @@ -996,7 +996,7 @@ void __init mp_parse_prt (void) continue; } if ((1<consistent_dma_mask; + dma_mask = hwdev->dev.coherent_dma_mask; } if (dma_mask == 0) --- diff/arch/x86_64/kernel/process.c 2004-03-01 14:36:06.000000000 +0000 +++ source/arch/x86_64/kernel/process.c 2004-03-01 15:42:15.000000000 +0000 @@ -16,7 +16,6 @@ * This file handles the architecture-dependent parts of process handling.. */ -#define __KERNEL_SYSCALLS__ #include #include @@ -25,7 +24,6 @@ #include #include #include -#include #include #include #include --- diff/arch/x86_64/kernel/setup.c 2004-03-01 14:36:06.000000000 +0000 +++ source/arch/x86_64/kernel/setup.c 2004-03-01 15:42:15.000000000 +0000 @@ -610,8 +610,7 @@ static void __init detect_ht(void) * At this point we only support two siblings per * processor package. */ -#define NR_SIBLINGS 2 - if (smp_num_siblings != NR_SIBLINGS) { + if (smp_num_siblings > NR_CPUS) { printk(KERN_WARNING "CPU: Unsupported number of the siblings %d", smp_num_siblings); smp_num_siblings = 1; return; --- diff/arch/x86_64/kernel/smp.c 2003-11-25 15:24:57.000000000 +0000 +++ source/arch/x86_64/kernel/smp.c 2004-03-01 15:42:15.000000000 +0000 @@ -362,6 +362,18 @@ void smp_send_reschedule(int cpu) send_IPI_mask(cpumask_of_cpu(cpu), RESCHEDULE_VECTOR); } +#ifdef CONFIG_KGDB +/* + * By using the NMI code instead of a vector we just sneak thru the + * word generator coming out with just what we want. AND it does + * not matter if clustered_apic_mode is set or not. + */ +void smp_send_nmi_allbutself(void) +{ + send_IPI_allbutself(APIC_DM_NMI); +} +#endif + /* * Structure and data for smp_call_function(). This is designed to minimise * static memory requirements. It also looks cleaner. --- diff/arch/x86_64/kernel/time.c 2004-03-01 14:36:06.000000000 +0000 +++ source/arch/x86_64/kernel/time.c 2004-03-01 15:42:15.000000000 +0000 @@ -26,6 +26,7 @@ #include #include #include +#include #include #include #include @@ -353,11 +354,11 @@ static irqreturn_t timer_interrupt(int i } if (lost) { - if (report_lost_ticks) + if (report_lost_ticks) { printk(KERN_WARNING "time.c: Lost %ld timer " - "tick(s)! (rip %016lx)\n", - (offset - vxtime.last) / hpet_tick - 1, - regs->rip); + "tick(s)! ", lost); + print_symbol("rip %s)\n", regs->rip); + } jiffies += lost; } @@ -399,8 +400,19 @@ static irqreturn_t timer_interrupt(int i return IRQ_HANDLED; } -/* RED-PEN: calculation is done in 32bits with multiply for performance - and could overflow, it may be better (but slower)to use an 64bit division. */ +static unsigned int cyc2ns_scale; +#define CYC2NS_SCALE_FACTOR 10 /* 2^10, carefully chosen */ + +static inline void set_cyc2ns_scale(unsigned long cpu_mhz) +{ + cyc2ns_scale = (1000 << CYC2NS_SCALE_FACTOR)/cpu_mhz; +} + +static inline unsigned long long cycles_2_ns(unsigned long long cyc) +{ + return (cyc * cyc2ns_scale) >> CYC2NS_SCALE_FACTOR; +} + unsigned long long sched_clock(void) { unsigned long a = 0; @@ -420,7 +432,7 @@ unsigned long long sched_clock(void) purposes. */ rdtscll(a); - return (a * vxtime.tsc_quot) >> 32; + return cycles_2_ns(a); } unsigned long get_cmos_time(void) @@ -527,6 +539,8 @@ static int time_cpufreq_notifier(struct vxtime.tsc_quot = (1000L << 32) / cpu_khz; } + set_cyc2ns_scale(cpu_khz_ref / 1000); + return 0; } @@ -725,6 +739,8 @@ void __init time_init(void) rdtscll_sync(&vxtime.last_tsc); setup_irq(0, &irq0); + set_cyc2ns_scale(cpu_khz / 1000); + #ifdef CONFIG_CPU_FREQ cpufreq_register_notifier(&time_cpufreq_notifier_block, CPUFREQ_TRANSITION_NOTIFIER); @@ -788,7 +804,7 @@ static int time_init_device(void) { int error = sysdev_class_register(&pit_sysclass); if (!error) - error = sys_device_register(&device_i8253); + error = sysdev_register(&device_i8253); return error; } --- diff/arch/x86_64/kernel/traps.c 2004-03-01 14:36:06.000000000 +0000 +++ source/arch/x86_64/kernel/traps.c 2004-03-01 15:42:15.000000000 +0000 @@ -45,6 +45,9 @@ #include #include +#ifdef CONFIG_KGDB +#include +#endif extern struct gate_struct idt_table[256]; @@ -351,9 +354,24 @@ void oops_end(void) void __die(const char * str, struct pt_regs * regs, long err) { + int nl = 0; static int die_counter; printk(KERN_EMERG "%s: %04lx [%u]\n", str, err & 0xffff,++die_counter); notify_die(DIE_OOPS, (char *)str, regs, err, 255, SIGSEGV); +#ifdef CONFIG_PREEMPT + printk("PREEMPT "); + nl = 1; +#endif +#ifdef CONFIG_SMP + printk("SMP "); + nl = 1; +#endif +#ifdef CONFIG_DEBUG_PAGEALLOC + printk("DEBUG_PAGEALLOC"); + nl = 1; +#endif + if (nl) + printk("\n"); show_registers(regs); /* Executive summary in case the oops scrolled away */ printk("RIP "); --- diff/arch/x86_64/kernel/x8664_ksyms.c 2004-03-01 14:36:06.000000000 +0000 +++ source/arch/x86_64/kernel/x8664_ksyms.c 2004-03-01 15:42:15.000000000 +0000 @@ -63,10 +63,6 @@ EXPORT_SYMBOL(pm_idle); EXPORT_SYMBOL(pm_power_off); EXPORT_SYMBOL(get_cmos_time); -#ifdef CONFIG_IO_DEBUG -EXPORT_SYMBOL(__io_virt_debug); -#endif - EXPORT_SYMBOL_NOVERS(__down_failed); EXPORT_SYMBOL_NOVERS(__down_failed_interruptible); EXPORT_SYMBOL_NOVERS(__down_failed_trylock); @@ -196,6 +192,7 @@ EXPORT_SYMBOL(die_chain); #ifdef CONFIG_SMP EXPORT_SYMBOL(cpu_sibling_map); +EXPORT_SYMBOL(smp_num_siblings); #endif extern void do_softirq_thunk(void); --- diff/arch/x86_64/lib/Makefile 2003-06-30 10:07:20.000000000 +0100 +++ source/arch/x86_64/lib/Makefile 2004-03-01 15:42:15.000000000 +0000 @@ -9,5 +9,5 @@ lib-y := csum-partial.o csum-copy.o csum thunk.o io.o clear_page.o copy_page.o bitstr.o lib-y += memcpy.o memmove.o memset.o copy_user.o -lib-$(CONFIG_IO_DEBUG) += iodebug.o lib-$(CONFIG_HAVE_DEC_LOCK) += dec_and_lock.o +lib-$(CONFIG_KGDB) += kgdb_serial.o --- diff/arch/x86_64/lib/io.c 2002-12-30 10:17:12.000000000 +0000 +++ source/arch/x86_64/lib/io.c 2004-03-01 15:42:15.000000000 +0000 @@ -4,12 +4,12 @@ void *memcpy_toio(void *dst,const void*src,unsigned len) { - return __inline_memcpy(__io_virt(dst),src,len); + return __inline_memcpy(dst,src,len); } void *memcpy_fromio(void *dst,const void*src,unsigned len) { - return __inline_memcpy(dst,__io_virt(src),len); + return __inline_memcpy(dst,src,len); } EXPORT_SYMBOL(memcpy_toio); --- diff/crypto/Makefile 2003-09-17 12:28:03.000000000 +0100 +++ source/crypto/Makefile 2004-03-01 15:42:15.000000000 +0000 @@ -4,7 +4,7 @@ proc-crypto-$(CONFIG_PROC_FS) = proc.o -obj-$(CONFIG_CRYPTO) += api.o cipher.o digest.o compress.o \ +obj-$(CONFIG_CRYPTO) += api.o scatterwalk.o cipher.o digest.o compress.o \ $(proc-crypto-y) obj-$(CONFIG_CRYPTO_HMAC) += hmac.o --- diff/crypto/cipher.c 2003-08-26 10:00:52.000000000 +0100 +++ source/crypto/cipher.c 2004-03-01 15:42:15.000000000 +0000 @@ -4,7 +4,6 @@ * Cipher operations. * * Copyright (c) 2002 James Morris - * Generic scatterwalk code by Adam J. Richter . * * This program is free software; you can redistribute it and/or modify it * under the terms of the GNU General Public License as published by the Free @@ -17,30 +16,13 @@ #include #include #include -#include -#include #include #include "internal.h" +#include "scatterwalk.h" typedef void (cryptfn_t)(void *, u8 *, const u8 *); typedef void (procfn_t)(struct crypto_tfm *, u8 *, - u8*, cryptfn_t, int enc, void *); - -struct scatter_walk { - struct scatterlist *sg; - struct page *page; - void *data; - unsigned int len_this_page; - unsigned int len_this_segment; - unsigned int offset; -}; - -enum km_type crypto_km_types[] = { - KM_USER0, - KM_USER1, - KM_SOFTIRQ0, - KM_SOFTIRQ1, -}; + u8*, cryptfn_t, int enc, void *, int); static inline void xor_64(u8 *a, const u8 *b) { @@ -57,108 +39,6 @@ static inline void xor_128(u8 *a, const } -/* Define sg_next is an inline routine now in case we want to change - scatterlist to a linked list later. */ -static inline struct scatterlist *sg_next(struct scatterlist *sg) -{ - return sg + 1; -} - -void *which_buf(struct scatter_walk *walk, unsigned int nbytes, void *scratch) -{ - if (nbytes <= walk->len_this_page && - (((unsigned long)walk->data) & (PAGE_CACHE_SIZE - 1)) + nbytes <= - PAGE_CACHE_SIZE) - return walk->data; - else - return scratch; -} - -static void memcpy_dir(void *buf, void *sgdata, size_t nbytes, int out) -{ - if (out) - memcpy(sgdata, buf, nbytes); - else - memcpy(buf, sgdata, nbytes); -} - -static void scatterwalk_start(struct scatter_walk *walk, struct scatterlist *sg) -{ - unsigned int rest_of_page; - - walk->sg = sg; - - walk->page = sg->page; - walk->len_this_segment = sg->length; - - rest_of_page = PAGE_CACHE_SIZE - (sg->offset & (PAGE_CACHE_SIZE - 1)); - walk->len_this_page = min(sg->length, rest_of_page); - walk->offset = sg->offset; -} - -static void scatterwalk_map(struct scatter_walk *walk, int out) -{ - walk->data = crypto_kmap(walk->page, out) + walk->offset; -} - -static void scatter_page_done(struct scatter_walk *walk, int out, - unsigned int more) -{ - /* walk->data may be pointing the first byte of the next page; - however, we know we transfered at least one byte. So, - walk->data - 1 will be a virutual address in the mapped page. */ - - if (out) - flush_dcache_page(walk->page); - - if (more) { - walk->len_this_segment -= walk->len_this_page; - - if (walk->len_this_segment) { - walk->page++; - walk->len_this_page = min(walk->len_this_segment, - (unsigned)PAGE_CACHE_SIZE); - walk->offset = 0; - } - else - scatterwalk_start(walk, sg_next(walk->sg)); - } -} - -static void scatter_done(struct scatter_walk *walk, int out, int more) -{ - crypto_kunmap(walk->data, out); - if (walk->len_this_page == 0 || !more) - scatter_page_done(walk, out, more); -} - -/* - * Do not call this unless the total length of all of the fragments - * has been verified as multiple of the block size. - */ -static int copy_chunks(void *buf, struct scatter_walk *walk, - size_t nbytes, int out) -{ - if (buf != walk->data) { - while (nbytes > walk->len_this_page) { - memcpy_dir(buf, walk->data, walk->len_this_page, out); - buf += walk->len_this_page; - nbytes -= walk->len_this_page; - - crypto_kunmap(walk->data, out); - scatter_page_done(walk, out, 1); - scatterwalk_map(walk, out); - } - - memcpy_dir(buf, walk->data, nbytes, out); - } - - walk->offset += nbytes; - walk->len_this_page -= nbytes; - walk->len_this_segment -= nbytes; - return 0; -} - /* * Generic encrypt/decrypt wrapper for ciphers, handles operations across * multiple page boundaries by using temporary blocks. In user context, @@ -191,19 +71,21 @@ static int crypt(struct crypto_tfm *tfm, scatterwalk_map(&walk_in, 0); scatterwalk_map(&walk_out, 1); - src_p = which_buf(&walk_in, bsize, tmp_src); - dst_p = which_buf(&walk_out, bsize, tmp_dst); + src_p = scatterwalk_whichbuf(&walk_in, bsize, tmp_src); + dst_p = scatterwalk_whichbuf(&walk_out, bsize, tmp_dst); nbytes -= bsize; - copy_chunks(src_p, &walk_in, bsize, 0); + scatterwalk_copychunks(src_p, &walk_in, bsize, 0); - prfn(tfm, dst_p, src_p, crfn, enc, info); + prfn(tfm, dst_p, src_p, crfn, enc, info, + scatterwalk_samebuf(&walk_in, &walk_out, + src_p, dst_p)); - scatter_done(&walk_in, 0, nbytes); + scatterwalk_done(&walk_in, 0, nbytes); - copy_chunks(dst_p, &walk_out, bsize, 1); - scatter_done(&walk_out, 1, nbytes); + scatterwalk_copychunks(dst_p, &walk_out, bsize, 1); + scatterwalk_done(&walk_out, 1, nbytes); if (!nbytes) return 0; @@ -212,8 +94,8 @@ static int crypt(struct crypto_tfm *tfm, } } -static void cbc_process(struct crypto_tfm *tfm, - u8 *dst, u8 *src, cryptfn_t fn, int enc, void *info) +static void cbc_process(struct crypto_tfm *tfm, u8 *dst, u8 *src, + cryptfn_t fn, int enc, void *info, int in_place) { u8 *iv = info; @@ -226,10 +108,9 @@ static void cbc_process(struct crypto_tf fn(crypto_tfm_ctx(tfm), dst, iv); memcpy(iv, dst, crypto_tfm_alg_blocksize(tfm)); } else { - const int need_stack = (src == dst); - u8 stack[need_stack ? crypto_tfm_alg_blocksize(tfm) : 0]; - u8 *buf = need_stack ? stack : dst; - + u8 stack[in_place ? crypto_tfm_alg_blocksize(tfm) : 0]; + u8 *buf = in_place ? stack : dst; + fn(crypto_tfm_ctx(tfm), buf, src); tfm->crt_u.cipher.cit_xor_block(buf, iv); memcpy(iv, src, crypto_tfm_alg_blocksize(tfm)); @@ -239,7 +120,7 @@ static void cbc_process(struct crypto_tf } static void ecb_process(struct crypto_tfm *tfm, u8 *dst, u8 *src, - cryptfn_t fn, int enc, void *info) + cryptfn_t fn, int enc, void *info, int in_place) { fn(crypto_tfm_ctx(tfm), dst, src); } --- diff/crypto/internal.h 2003-09-17 12:28:03.000000000 +0100 +++ source/crypto/internal.h 2004-03-01 15:42:15.000000000 +0000 @@ -11,6 +11,7 @@ */ #ifndef _CRYPTO_INTERNAL_H #define _CRYPTO_INTERNAL_H +#include #include #include #include --- diff/drivers/Kconfig 2004-02-18 08:54:08.000000000 +0000 +++ source/drivers/Kconfig 2004-03-01 15:42:15.000000000 +0000 @@ -42,7 +42,7 @@ source "drivers/char/Kconfig" source "drivers/i2c/Kconfig" -# source "drivers/misc/Kconfig" +source "drivers/misc/Kconfig" source "drivers/media/Kconfig" --- diff/drivers/acpi/Kconfig 2004-03-01 14:36:06.000000000 +0000 +++ source/drivers/acpi/Kconfig 2004-03-01 15:42:15.000000000 +0000 @@ -251,18 +251,6 @@ config ACPI_SYSTEM This driver will enable your system to shut down using ACPI, and dump your ACPI DSDT table using /proc/acpi/dsdt. -config ACPI_RELAXED_AML - bool "Relaxed AML" - depends on ACPI_INTERPRETER - depends on !IA64_SGI_SN - default n - help - If you say `Y' here, the ACPI interpreter will relax its checking - for valid AML and will ignore some AML mistakes, such as off-by-one - errors in region sizes. Some laptops may require this option. In - particular, many Toshiba laptops require this for correct operation - of the AC module. - config X86_PM_TIMER bool "Power Management Timer Support" depends on X86 && ACPI --- diff/drivers/acpi/executer/exfldio.c 2004-03-01 14:36:06.000000000 +0000 +++ source/drivers/acpi/executer/exfldio.c 2004-03-01 15:42:15.000000000 +0000 @@ -154,8 +154,7 @@ acpi_ex_setup_region ( field_datum_byte_offset, obj_desc->common_field.access_byte_width, acpi_ut_get_node_name (rgn_desc->region.node), rgn_desc->region.length)); - #ifdef CONFIG_ACPI_RELAXED_AML - { + if (!acpi_strict) { /* * Allow access to the field if it is within the region size * rounded up to a multiple of the access byte width. This @@ -186,9 +185,9 @@ acpi_ex_setup_region ( return_ACPI_STATUS (AE_OK); } } - #else + else { return_ACPI_STATUS (AE_AML_REGION_LIMIT); - #endif + } } return_ACPI_STATUS (AE_OK); --- diff/drivers/acpi/hardware/hwgpe.c 2004-02-09 10:36:10.000000000 +0000 +++ source/drivers/acpi/hardware/hwgpe.c 2004-03-01 15:42:15.000000000 +0000 @@ -528,6 +528,14 @@ acpi_hw_enable_non_wakeup_gpe_block ( /* Examine each GPE register within the block */ for (i = 0; i < gpe_block->register_count; i++) { + /* Clear the entire status register */ + + status = acpi_hw_low_level_write (8, 0xFF, + &gpe_block->register_info[i].status_address); + if (ACPI_FAILURE (status)) { + return (status); + } + /* * We previously stored the enabled status of all GPEs. * Blast them back in. --- diff/drivers/acpi/hardware/hwregs.c 2004-02-09 10:36:10.000000000 +0000 +++ source/drivers/acpi/hardware/hwregs.c 2004-03-01 15:42:15.000000000 +0000 @@ -152,11 +152,11 @@ acpi_get_sleep_type_data ( /* * Evaluate the namespace object containing the values for this state */ - status = acpi_ns_evaluate_by_name ((char *) acpi_gbl_db_sleep_states[sleep_state], + status = acpi_ns_evaluate_by_name ((char *) acpi_gbl_sleep_state_names[sleep_state], NULL, &obj_desc); if (ACPI_FAILURE (status)) { ACPI_DEBUG_PRINT ((ACPI_DB_EXEC, "%s while evaluating sleep_state [%s]\n", - acpi_format_exception (status), acpi_gbl_db_sleep_states[sleep_state])); + acpi_format_exception (status), acpi_gbl_sleep_state_names[sleep_state])); return_ACPI_STATUS (status); } @@ -201,7 +201,7 @@ acpi_get_sleep_type_data ( if (ACPI_FAILURE (status)) { ACPI_DEBUG_PRINT ((ACPI_DB_ERROR, "While evaluating sleep_state [%s], bad Sleep object %p type %s\n", - acpi_gbl_db_sleep_states[sleep_state], obj_desc, acpi_ut_get_object_type_name (obj_desc))); + acpi_gbl_sleep_state_names[sleep_state], obj_desc, acpi_ut_get_object_type_name (obj_desc))); } acpi_ut_remove_reference (obj_desc); --- diff/drivers/acpi/hardware/hwsleep.c 2004-03-01 14:36:06.000000000 +0000 +++ source/drivers/acpi/hardware/hwsleep.c 2004-03-01 15:42:15.000000000 +0000 @@ -48,6 +48,19 @@ ACPI_MODULE_NAME ("hwsleep") +#define METHOD_NAME__BFS "\\_BFS" +#define METHOD_NAME__GTS "\\_GTS" +#define METHOD_NAME__PTS "\\_PTS" +#define METHOD_NAME__SST "\\_SI._SST" +#define METHOD_NAME__WAK "\\_WAK" + +#define ACPI_SST_INDICATOR_OFF 0 +#define ACPI_SST_WORKING 1 +#define ACPI_SST_WAKING 2 +#define ACPI_SST_SLEEPING 3 +#define ACPI_SST_SLEEP_CONTEXT 4 + + /****************************************************************************** * * FUNCTION: acpi_set_firmware_waking_vector @@ -171,19 +184,41 @@ acpi_enter_sleep_state_prep ( /* Run the _PTS and _GTS methods */ - status = acpi_evaluate_object (NULL, "\\_PTS", &arg_list, NULL); + status = acpi_evaluate_object (NULL, METHOD_NAME__PTS, &arg_list, NULL); if (ACPI_FAILURE (status) && status != AE_NOT_FOUND) { return_ACPI_STATUS (status); } - status = acpi_evaluate_object (NULL, "\\_GTS", &arg_list, NULL); + status = acpi_evaluate_object (NULL, METHOD_NAME__GTS, &arg_list, NULL); if (ACPI_FAILURE (status) && status != AE_NOT_FOUND) { return_ACPI_STATUS (status); } + /* Setup the argument to _SST */ + + switch (sleep_state) { + case ACPI_STATE_S0: + arg.integer.value = ACPI_SST_WORKING; + break; + + case ACPI_STATE_S1: + case ACPI_STATE_S2: + case ACPI_STATE_S3: + arg.integer.value = ACPI_SST_SLEEPING; + break; + + case ACPI_STATE_S4: + arg.integer.value = ACPI_SST_SLEEP_CONTEXT; + break; + + default: + arg.integer.value = ACPI_SST_INDICATOR_OFF; /* Default is indicator off */ + break; + } + /* Set the system indicators to show the desired sleep state. */ - status = acpi_evaluate_object (NULL, "\\_SI._SST", &arg_list, NULL); + status = acpi_evaluate_object (NULL, METHOD_NAME__SST, &arg_list, NULL); if (ACPI_FAILURE (status) && status != AE_NOT_FOUND) { ACPI_REPORT_ERROR (("Method _SST failed, %s\n", acpi_format_exception (status))); } @@ -477,19 +512,19 @@ acpi_leave_sleep_state ( /* Ignore any errors from these methods */ - arg.integer.value = 0; - status = acpi_evaluate_object (NULL, "\\_SI._SST", &arg_list, NULL); + arg.integer.value = ACPI_SST_WAKING; + status = acpi_evaluate_object (NULL, METHOD_NAME__SST, &arg_list, NULL); if (ACPI_FAILURE (status) && status != AE_NOT_FOUND) { ACPI_REPORT_ERROR (("Method _SST failed, %s\n", acpi_format_exception (status))); } arg.integer.value = sleep_state; - status = acpi_evaluate_object (NULL, "\\_BFS", &arg_list, NULL); + status = acpi_evaluate_object (NULL, METHOD_NAME__BFS, &arg_list, NULL); if (ACPI_FAILURE (status) && status != AE_NOT_FOUND) { ACPI_REPORT_ERROR (("Method _BFS failed, %s\n", acpi_format_exception (status))); } - status = acpi_evaluate_object (NULL, "\\_WAK", &arg_list, NULL); + status = acpi_evaluate_object (NULL, METHOD_NAME__WAK, &arg_list, NULL); if (ACPI_FAILURE (status) && status != AE_NOT_FOUND) { ACPI_REPORT_ERROR (("Method _WAK failed, %s\n", acpi_format_exception (status))); } @@ -501,8 +536,25 @@ acpi_leave_sleep_state ( return_ACPI_STATUS (status); } + /* Enable power button */ + + acpi_set_register(acpi_gbl_fixed_event_info[ACPI_EVENT_POWER_BUTTON].enable_register_id, + 1, ACPI_MTX_DO_NOT_LOCK); + acpi_set_register(acpi_gbl_fixed_event_info[ACPI_EVENT_POWER_BUTTON].status_register_id, + 1, ACPI_MTX_DO_NOT_LOCK); + /* Enable BM arbitration */ status = acpi_set_register (ACPI_BITREG_ARB_DISABLE, 0, ACPI_MTX_LOCK); + if (ACPI_FAILURE (status)) { + return_ACPI_STATUS (status); + } + + arg.integer.value = ACPI_SST_WORKING; + status = acpi_evaluate_object (NULL, METHOD_NAME__SST, &arg_list, NULL); + if (ACPI_FAILURE (status) && status != AE_NOT_FOUND) { + ACPI_REPORT_ERROR (("Method _SST failed, %s\n", acpi_format_exception (status))); + } + return_ACPI_STATUS (status); } --- diff/drivers/acpi/namespace/nseval.c 2004-02-09 10:36:10.000000000 +0000 +++ source/drivers/acpi/namespace/nseval.c 2004-03-01 15:42:15.000000000 +0000 @@ -110,7 +110,7 @@ acpi_ns_evaluate_relative ( status = acpi_ut_acquire_mutex (ACPI_MTX_NAMESPACE); if (ACPI_FAILURE (status)) { - return_ACPI_STATUS (status); + goto cleanup; } prefix_node = acpi_ns_map_handle_to_node (handle); @@ -197,7 +197,7 @@ acpi_ns_evaluate_by_name ( status = acpi_ut_acquire_mutex (ACPI_MTX_NAMESPACE); if (ACPI_FAILURE (status)) { - return_ACPI_STATUS (status); + goto cleanup; } /* Lookup the name in the namespace */ --- diff/drivers/acpi/namespace/nsutils.c 2004-02-09 10:36:10.000000000 +0000 +++ source/drivers/acpi/namespace/nsutils.c 2004-03-01 15:42:15.000000000 +0000 @@ -918,7 +918,7 @@ acpi_ns_get_node_by_path ( status = acpi_ut_acquire_mutex (ACPI_MTX_NAMESPACE); if (ACPI_FAILURE (status)) { - return_ACPI_STATUS (status); + goto cleanup; } /* Setup lookup scope (search starting point) */ @@ -936,10 +936,10 @@ acpi_ns_get_node_by_path ( internal_path, acpi_format_exception (status))); } - /* Cleanup */ - (void) acpi_ut_release_mutex (ACPI_MTX_NAMESPACE); +cleanup: + /* Cleanup */ if (internal_path) { ACPI_MEM_FREE (internal_path); } --- diff/drivers/acpi/namespace/nsxfname.c 2004-02-09 10:36:10.000000000 +0000 +++ source/drivers/acpi/namespace/nsxfname.c 2004-03-01 15:42:15.000000000 +0000 @@ -326,6 +326,13 @@ acpi_get_object_info ( info.valid |= ACPI_VALID_ADR; } + /* Execute the Device._sx_d methods */ + + status = acpi_ut_execute_sxds (node, info.highest_dstates); + if (ACPI_SUCCESS (status)) { + info.valid |= ACPI_VALID_STA; + } + status = AE_OK; } --- diff/drivers/acpi/pci_link.c 2004-02-18 08:54:08.000000000 +0000 +++ source/drivers/acpi/pci_link.c 2004-03-01 15:42:16.000000000 +0000 @@ -768,7 +768,6 @@ __setup("acpi_irq_pci=", acpi_irq_pci); static int __init acpi_irq_nobalance_set(char *str) { -printk("ACPI STATIC SET\n"); acpi_irq_balance = 0; return(1); } @@ -776,7 +775,6 @@ __setup("acpi_irq_nobalance", acpi_irq_n int __init acpi_irq_balance_set(char *str) { -printk("ACPI BALANCE SET\n"); acpi_irq_balance = 1; return(1); } --- diff/drivers/acpi/resources/rsxface.c 2004-02-09 10:36:10.000000000 +0000 +++ source/drivers/acpi/resources/rsxface.c 2004-03-01 15:42:16.000000000 +0000 @@ -239,6 +239,7 @@ acpi_walk_resources ( acpi_status status; struct acpi_buffer buffer = {ACPI_ALLOCATE_BUFFER, NULL}; struct acpi_resource *resource; + struct acpi_resource *buffer_end; ACPI_FUNCTION_TRACE ("acpi_walk_resources"); @@ -255,7 +256,13 @@ acpi_walk_resources ( return_ACPI_STATUS (status); } - resource = (struct acpi_resource *) buffer.pointer; + /* Setup pointers */ + + resource = (struct acpi_resource *) buffer.pointer; + buffer_end = (struct acpi_resource *) ((u8 *) buffer.pointer + buffer.length); + + /* Walk the resource list */ + for (;;) { if (!resource || resource->id == ACPI_RSTYPE_END_TAG) { break; @@ -268,6 +275,7 @@ acpi_walk_resources ( case AE_CTRL_DEPTH: /* Just keep going */ + status = AE_OK; break; @@ -285,7 +293,15 @@ acpi_walk_resources ( goto cleanup; } + /* Get the next resource descriptor */ + resource = ACPI_NEXT_RESOURCE (resource); + + /* Check for end-of-buffer */ + + if (resource >= buffer_end) { + goto cleanup; + } } cleanup: --- diff/drivers/acpi/tables.c 2004-02-18 08:54:08.000000000 +0000 +++ source/drivers/acpi/tables.c 2004-03-01 15:42:16.000000000 +0000 @@ -58,6 +58,7 @@ static char *acpi_table_signatures[ACPI_ [ACPI_SSDT] = "SSDT", [ACPI_SPMI] = "SPMI", [ACPI_HPET] = "HPET", + [ACPI_MCFG] = "MCFG", }; static char *mps_inti_flags_polarity[] = { "dfl", "high", "res", "low" }; --- diff/drivers/acpi/utilities/uteval.c 2004-02-09 10:36:10.000000000 +0000 +++ source/drivers/acpi/utilities/uteval.c 2004-03-01 15:42:16.000000000 +0000 @@ -562,3 +562,63 @@ acpi_ut_execute_STA ( acpi_ut_remove_reference (obj_desc); return_ACPI_STATUS (status); } + + +/******************************************************************************* + * + * FUNCTION: acpi_ut_execute_Sxds + * + * PARAMETERS: device_node - Node for the device + * *Flags - Where the status flags are returned + * + * RETURN: Status + * + * DESCRIPTION: Executes _STA for selected device and stores results in + * *Flags. + * + * NOTE: Internal function, no parameter validation + * + ******************************************************************************/ + +acpi_status +acpi_ut_execute_sxds ( + struct acpi_namespace_node *device_node, + u8 *highest) +{ + union acpi_operand_object *obj_desc; + acpi_status status; + u32 i; + + + ACPI_FUNCTION_TRACE ("ut_execute_Sxds"); + + + for (i = 0; i < 4; i++) { + highest[i] = 0xFF; + status = acpi_ut_evaluate_object (device_node, + (char *) acpi_gbl_highest_dstate_names[i], + ACPI_BTYPE_INTEGER, &obj_desc); + if (ACPI_FAILURE (status)) { + if (status != AE_NOT_FOUND) { + ACPI_DEBUG_PRINT ((ACPI_DB_EXEC, + "%s on Device %4.4s, %s\n", + (char *) acpi_gbl_highest_dstate_names[i], + acpi_ut_get_node_name (device_node), + acpi_format_exception (status))); + + return_ACPI_STATUS (status); + } + } + else { + /* Extract the Dstate value */ + + highest[i] = (u8) obj_desc->integer.value; + + /* Delete the return object */ + + acpi_ut_remove_reference (obj_desc); + } + } + + return_ACPI_STATUS (AE_OK); +} --- diff/drivers/acpi/utilities/utglobal.c 2004-02-09 10:36:10.000000000 +0000 +++ source/drivers/acpi/utilities/utglobal.c 2004-03-01 15:42:16.000000000 +0000 @@ -171,7 +171,7 @@ u8 acpi const u8 acpi_gbl_decode_to8bit [8] = {1,2,4,8,16,32,64,128}; -const char *acpi_gbl_db_sleep_states[ACPI_S_STATE_COUNT] = { +const char *acpi_gbl_sleep_state_names[ACPI_S_STATE_COUNT] = { "\\_S0_", "\\_S1_", "\\_S2_", @@ -179,6 +179,11 @@ const char *acp "\\_S4_", "\\_S5_"}; +const char *acpi_gbl_highest_dstate_names[4] = { + "_S1D", + "_S2D", + "_S3D", + "_S4D"}; /****************************************************************************** * --- diff/drivers/atm/fore200e.c 2004-02-18 08:54:08.000000000 +0000 +++ source/drivers/atm/fore200e.c 2004-03-01 15:42:16.000000000 +0000 @@ -482,11 +482,19 @@ fore200e_pca_dma_unmap(struct fore200e* static void -fore200e_pca_dma_sync(struct fore200e* fore200e, u32 dma_addr, int size, int direction) +fore200e_pca_dma_sync_for_cpu(struct fore200e* fore200e, u32 dma_addr, int size, int direction) { DPRINTK(3, "PCI DVMA sync: dma_addr = 0x%08x, size = %d, direction = %d\n", dma_addr, size, direction); - pci_dma_sync_single((struct pci_dev*)fore200e->bus_dev, dma_addr, size, direction); + pci_dma_sync_single_for_cpu((struct pci_dev*)fore200e->bus_dev, dma_addr, size, direction); +} + +static void +fore200e_pca_dma_sync_for_device(struct fore200e* fore200e, u32 dma_addr, int size, int direction) +{ + DPRINTK(3, "PCI DVMA sync: dma_addr = 0x%08x, size = %d, direction = %d\n", dma_addr, size, direction); + + pci_dma_sync_single_for_device((struct pci_dev*)fore200e->bus_dev, dma_addr, size, direction); } @@ -761,11 +769,19 @@ fore200e_sba_dma_unmap(struct fore200e* static void -fore200e_sba_dma_sync(struct fore200e* fore200e, u32 dma_addr, int size, int direction) +fore200e_sba_dma_sync_for_cpu(struct fore200e* fore200e, u32 dma_addr, int size, int direction) { DPRINTK(3, "SBUS DVMA sync: dma_addr = 0x%08x, size = %d, direction = %d\n", dma_addr, size, direction); - sbus_dma_sync_single((struct sbus_dev*)fore200e->bus_dev, dma_addr, size, direction); + sbus_dma_sync_single_for_cpu((struct sbus_dev*)fore200e->bus_dev, dma_addr, size, direction); +} + +static void +fore200e_sba_dma_sync_for_device(struct fore200e* fore200e, u32 dma_addr, int size, int direction) +{ + DPRINTK(3, "SBUS DVMA sync: dma_addr = 0x%08x, size = %d, direction = %d\n", dma_addr, size, direction); + + sbus_dma_sync_single_for_device((struct sbus_dev*)fore200e->bus_dev, dma_addr, size, direction); } @@ -1149,10 +1165,13 @@ fore200e_push_rpd(struct fore200e* fore2 /* rebuild rx buffer address from rsd handle */ buffer = FORE200E_HDL2BUF(rpd->rsd[ i ].handle); - /* ensure DMA synchronisation */ - fore200e->bus->dma_sync(fore200e, buffer->data.dma_addr, rpd->rsd[ i ].length, FORE200E_DMA_FROMDEVICE); + /* Make device DMA transfer visible to CPU. */ + fore200e->bus->dma_sync_for_cpu(fore200e, buffer->data.dma_addr, rpd->rsd[ i ].length, FORE200E_DMA_FROMDEVICE); memcpy(skb_put(skb, rpd->rsd[ i ].length), buffer->data.align_addr, rpd->rsd[ i ].length); + + /* Now let the device get at it again. */ + fore200e->bus->dma_sync_for_device(fore200e, buffer->data.dma_addr, rpd->rsd[ i ].length, FORE200E_DMA_FROMDEVICE); } DPRINTK(3, "rx skb: len = %d, truesize = %d\n", skb->len, skb->truesize); @@ -1584,8 +1603,9 @@ fore200e_send(struct atm_vcc *vcc, struc tasklet_enable(&fore200e->tasklet); - /* ensure DMA synchronisation */ - fore200e->bus->dma_sync(fore200e, tpd->tsd[ 0 ].buffer, tpd->tsd[ 0 ].length, FORE200E_DMA_TODEVICE); + /* The dma_map call above implies a dma_sync so the device can use it, + * thus no explicit dma_sync call is necessary here. + */ DPRINTK(3, "tx on %d.%d.%d:%d, len = %u (%u)\n", vcc->itf, vcc->vpi, vcc->vci, fore200e_atm2fore_aal(vcc->qos.aal), @@ -2918,7 +2938,8 @@ static const struct fore200e_bus fore200 fore200e_pca_write, fore200e_pca_dma_map, fore200e_pca_dma_unmap, - fore200e_pca_dma_sync, + fore200e_pca_dma_sync_for_cpu, + fore200e_pca_dma_sync_for_device, fore200e_pca_dma_chunk_alloc, fore200e_pca_dma_chunk_free, fore200e_pca_detect, @@ -2940,7 +2961,8 @@ static const struct fore200e_bus fore200 fore200e_sba_write, fore200e_sba_dma_map, fore200e_sba_dma_unmap, - fore200e_sba_dma_sync, + fore200e_sba_dma_sync_for_cpu, + fore200e_sba_dma_sync_for_device, fore200e_sba_dma_chunk_alloc, fore200e_sba_dma_chunk_free, fore200e_sba_detect, --- diff/drivers/atm/fore200e.h 2002-10-16 04:28:23.000000000 +0100 +++ source/drivers/atm/fore200e.h 2004-03-01 15:42:16.000000000 +0000 @@ -801,7 +801,8 @@ typedef struct fore200e_bus { void (*write)(u32, volatile u32*); u32 (*dma_map)(struct fore200e*, void*, int, int); void (*dma_unmap)(struct fore200e*, u32, int, int); - void (*dma_sync)(struct fore200e*, u32, int, int); + void (*dma_sync_for_cpu)(struct fore200e*, u32, int, int); + void (*dma_sync_for_device)(struct fore200e*, u32, int, int); int (*dma_chunk_alloc)(struct fore200e*, struct chunk*, int, int, int); void (*dma_chunk_free)(struct fore200e*, struct chunk*); struct fore200e* (*detect)(const struct fore200e_bus*, int); --- diff/drivers/atm/idt77252.c 2004-02-18 08:54:08.000000000 +0000 +++ source/drivers/atm/idt77252.c 2004-03-01 15:42:16.000000000 +0000 @@ -1064,8 +1064,8 @@ dequeue_rx(struct idt77252_dev *card, st vcc = vc->rx_vcc; - pci_dma_sync_single(card->pcidev, IDT77252_PRV_PADDR(skb), - skb->end - skb->data, PCI_DMA_FROMDEVICE); + pci_dma_sync_single_for_cpu(card->pcidev, IDT77252_PRV_PADDR(skb), + skb->end - skb->data, PCI_DMA_FROMDEVICE); if ((vcc->qos.aal == ATM_AAL0) || (vcc->qos.aal == ATM_AAL34)) { @@ -1903,6 +1903,9 @@ recycle_rx_skb(struct idt77252_dev *card u32 handle = IDT77252_PRV_POOL(skb); int err; + pci_dma_sync_single_for_device(card->pcidev, IDT77252_PRV_PADDR(skb), + skb->end - skb->data, PCI_DMA_FROMDEVICE); + err = push_rx_skb(card, skb, POOL_QUEUE(handle)); if (err) { pci_unmap_single(card->pcidev, IDT77252_PRV_PADDR(skb), --- diff/drivers/base/cpu.c 2003-10-27 09:20:43.000000000 +0000 +++ source/drivers/base/cpu.c 2004-03-01 15:42:16.000000000 +0000 @@ -7,6 +7,7 @@ #include #include #include +#include struct sysdev_class cpu_sysdev_class = { @@ -14,6 +15,46 @@ struct sysdev_class cpu_sysdev_class = { }; EXPORT_SYMBOL(cpu_sysdev_class); +#ifdef CONFIG_HOTPLUG_CPU +static ssize_t show_online(struct sys_device *dev, char *buf) +{ + struct cpu *cpu = container_of(dev, struct cpu, sysdev); + + return sprintf(buf, "%u\n", !!cpu_online(cpu->sysdev.id)); +} + +static ssize_t store_online(struct sys_device *dev, const char *buf, + size_t count) +{ + struct cpu *cpu = container_of(dev, struct cpu, sysdev); + ssize_t ret; + + switch (buf[0]) { + case '0': + ret = cpu_down(cpu->sysdev.id); + break; + case '1': + ret = cpu_up(cpu->sysdev.id); + break; + default: + ret = -EINVAL; + } + + if (ret >= 0) + ret = count; + return ret; +} +static SYSDEV_ATTR(online, 0600, show_online, store_online); + +static void __init register_cpu_control(struct cpu *cpu) +{ + sysdev_create_file(&cpu->sysdev, &attr_online); +} +#else /* ... !CONFIG_HOTPLUG_CPU */ +static void __init register_cpu_control(struct cpu *cpu) +{ +} +#endif /* CONFIG_HOTPLUG_CPU */ /* * register_cpu - Setup a driverfs device for a CPU. @@ -29,11 +70,13 @@ int __init register_cpu(struct cpu *cpu, cpu->sysdev.id = num; cpu->sysdev.cls = &cpu_sysdev_class; - error = sys_device_register(&cpu->sysdev); + error = sysdev_register(&cpu->sysdev); if (!error && root) error = sysfs_create_link(&root->sysdev.kobj, &cpu->sysdev.kobj, kobject_name(&cpu->sysdev.kobj)); + if (!error) + register_cpu_control(cpu); return error; } --- diff/drivers/base/firmware_class.c 2004-03-01 14:36:06.000000000 +0000 +++ source/drivers/base/firmware_class.c 2004-03-01 15:42:16.000000000 +0000 @@ -27,6 +27,7 @@ enum { FW_STATUS_LOADING, FW_STATUS_DONE, FW_STATUS_ABORT, + FW_STATUS_READY, }; static int loading_timeout = 10; /* In seconds */ @@ -96,6 +97,9 @@ firmware_class_hotplug(struct class_devi int i = 0; char *scratch = buffer; + if (!test_bit(FW_STATUS_READY, &fw_priv->status)) + return -ENODEV; + if (buffer_size < (FIRMWARE_NAME_MAX + 10)) return -ENOMEM; if (num_envp < 1) @@ -263,6 +267,8 @@ fw_class_dev_release(struct class_device kfree(fw_priv); kfree(class_dev); + + module_put(THIS_MODULE); } static void @@ -325,6 +331,7 @@ error_kfree: kfree(class_dev); return retval; } + static int fw_setup_class_device(struct firmware *fw, struct class_device **class_dev_p, const char *fw_name, struct device *device) @@ -338,6 +345,9 @@ fw_setup_class_device(struct firmware *f if (retval) goto out; + /* Need to pin this module until class device is destroyed */ + __module_get(THIS_MODULE); + fw_priv = class_get_devdata(class_dev); fw_priv->fw = fw; @@ -356,6 +366,7 @@ fw_setup_class_device(struct firmware *f goto error_unreg; } + set_bit(FW_STATUS_READY, &fw_priv->status); *class_dev_p = class_dev; goto out; @@ -409,6 +420,7 @@ request_firmware(const struct firmware * add_timer(&fw_priv->timeout); } + kobject_hotplug("add", &class_dev->kobj); wait_for_completion(&fw_priv->completion); set_bit(FW_STATUS_DONE, &fw_priv->status); --- diff/drivers/base/node.c 2004-03-01 14:36:06.000000000 +0000 +++ source/drivers/base/node.c 2004-03-01 15:42:16.000000000 +0000 @@ -69,7 +69,7 @@ int __init register_node(struct node *no node->cpumap = node_to_cpumask(num); node->sysdev.id = num; node->sysdev.cls = &node_class; - error = sys_device_register(&node->sysdev); + error = sysdev_register(&node->sysdev); if (!error){ sysdev_create_file(&node->sysdev, &attr_cpumap); --- diff/drivers/base/sys.c 2003-09-17 12:28:03.000000000 +0100 +++ source/drivers/base/sys.c 2004-03-01 15:42:16.000000000 +0000 @@ -8,7 +8,7 @@ * * This exports a 'system' bus type. * By default, a 'sys' bus gets added to the root of the system. There will - * always be core system devices. Devices can use sys_device_register() to + * always be core system devices. Devices can use sysdev_register() to * add themselves as children of the system bus. */ @@ -164,11 +164,11 @@ EXPORT_SYMBOL(sysdev_driver_unregister); /** - * sys_device_register - add a system device to the tree + * sysdev_register - add a system device to the tree * @sysdev: device in question * */ -int sys_device_register(struct sys_device * sysdev) +int sysdev_register(struct sys_device * sysdev) { int error; struct sysdev_class * cls = sysdev->cls; @@ -212,7 +212,7 @@ int sys_device_register(struct sys_devic return error; } -void sys_device_unregister(struct sys_device * sysdev) +void sysdev_unregister(struct sys_device * sysdev) { struct sysdev_driver * drv; @@ -390,5 +390,5 @@ int __init sys_bus_init(void) return subsystem_register(&system_subsys); } -EXPORT_SYMBOL(sys_device_register); -EXPORT_SYMBOL(sys_device_unregister); +EXPORT_SYMBOL(sysdev_register); +EXPORT_SYMBOL(sysdev_unregister); --- diff/drivers/block/Kconfig 2004-03-01 14:36:06.000000000 +0000 +++ source/drivers/block/Kconfig 2004-03-01 15:42:16.000000000 +0000 @@ -321,6 +321,7 @@ config BLK_DEV_RAM_SIZE config BLK_DEV_INITRD bool "Initial RAM disk (initrd) support" + depends on BLK_DEV_RAM && BLK_DEV_RAM!=m help The initial RAM disk is a RAM disk that is loaded by the boot loader (loadlin or lilo) and that is mounted as root before the normal boot --- diff/drivers/block/Kconfig.iosched 2003-10-09 09:47:16.000000000 +0100 +++ source/drivers/block/Kconfig.iosched 2004-03-01 15:42:16.000000000 +0000 @@ -27,3 +27,10 @@ config IOSCHED_DEADLINE a disk at any one time, its behaviour is almost identical to the anticipatory I/O scheduler and so is a good choice. +config IOSCHED_CFQ + bool "CFQ I/O scheduler" if EMBEDDED + default y + ---help--- + The CFQ I/O scheduler tries to distribute bandwidth equally + among all processes in the system. It should provide a fair + working environment, suitable for desktop systems. --- diff/drivers/block/Makefile 2004-03-01 14:36:06.000000000 +0000 +++ source/drivers/block/Makefile 2004-03-01 15:42:16.000000000 +0000 @@ -18,6 +18,7 @@ obj-y := elevator.o ll_rw_blk.o ioctl.o obj-$(CONFIG_IOSCHED_NOOP) += noop-iosched.o obj-$(CONFIG_IOSCHED_AS) += as-iosched.o obj-$(CONFIG_IOSCHED_DEADLINE) += deadline-iosched.o +obj-$(CONFIG_IOSCHED_CFQ) += cfq-iosched.o obj-$(CONFIG_MAC_FLOPPY) += swim3.o obj-$(CONFIG_BLK_DEV_FD) += floppy.o obj-$(CONFIG_BLK_DEV_FD98) += floppy98.o --- diff/drivers/block/floppy.c 2004-02-18 08:54:08.000000000 +0000 +++ source/drivers/block/floppy.c 2004-03-01 15:42:16.000000000 +0000 @@ -4242,6 +4242,15 @@ int __init floppy_init(void) disks[i] = alloc_disk(1); if (!disks[i]) goto Enomem; + + disks[i]->major = FLOPPY_MAJOR; + disks[i]->first_minor = TOMINOR(i); + disks[i]->fops = &floppy_fops; + sprintf(disks[i]->disk_name, "fd%d", i); + + init_timer(&motor_off_timer[i]); + motor_off_timer[i].data = i; + motor_off_timer[i].function = motor_off_callback; } devfs_mk_dir ("floppy"); @@ -4255,13 +4264,6 @@ int __init floppy_init(void) goto fail_queue; } - for (i=0; imajor = FLOPPY_MAJOR; - disks[i]->first_minor = TOMINOR(i); - disks[i]->fops = &floppy_fops; - sprintf(disks[i]->disk_name, "fd%d", i); - } - blk_register_region(MKDEV(FLOPPY_MAJOR, 0), 256, THIS_MODULE, floppy_find, NULL, NULL); @@ -4366,9 +4368,6 @@ int __init floppy_init(void) } for (drive = 0; drive < N_DRIVE; drive++) { - init_timer(&motor_off_timer[drive]); - motor_off_timer[drive].data = drive; - motor_off_timer[drive].function = motor_off_callback; if (!(allowed_drive_mask & (1 << drive))) continue; if (fdc_state[FDC(drive)].version == FDC_NONE) --- diff/drivers/block/ll_rw_blk.c 2004-02-09 10:36:10.000000000 +0000 +++ source/drivers/block/ll_rw_blk.c 2004-03-01 15:42:16.000000000 +0000 @@ -27,6 +27,7 @@ #include #include #include +#include static void blk_unplug_work(void *data); static void blk_unplug_timeout(unsigned long data); @@ -521,10 +522,10 @@ init_tag_map(request_queue_t *q, struct { int bits, i; - if (depth > q->nr_requests * 2) { - depth = q->nr_requests * 2; - printk(KERN_ERR "%s: adjusted depth to %d\n", - __FUNCTION__, depth); + if (depth > q->nr_requests / 2) { + q->nr_requests = depth * 2; + printk(KERN_INFO "%s: large TCQ depth: adjusted nr_requests " + "to %lu\n", __FUNCTION__, q->nr_requests); } tags->tag_index = kmalloc(depth * sizeof(struct request *), GFP_ATOMIC); @@ -1334,6 +1335,8 @@ static elevator_t *chosen_elevator = &iosched_as; #elif defined(CONFIG_IOSCHED_DEADLINE) &iosched_deadline; +#elif defined(CONFIG_IOSCHED_CFQ) + &iosched_cfq; #elif defined(CONFIG_IOSCHED_NOOP) &elevator_noop; #else @@ -1352,6 +1355,10 @@ static int __init elevator_setup(char *s if (!strcmp(str, "as")) chosen_elevator = &iosched_as; #endif +#ifdef CONFIG_IOSCHED_CFQ + if (!strcmp(str, "cfq")) + chosen_elevator = &iosched_cfq; +#endif #ifdef CONFIG_IOSCHED_NOOP if (!strcmp(str, "noop")) chosen_elevator = &elevator_noop; @@ -1884,18 +1891,22 @@ EXPORT_SYMBOL(blk_put_request); * Waits for up to @timeout jiffies for a queue (any queue) to exit congestion. * If no queues are congested then just wait for the next request to be * returned. + * + * Returns the number of jiffies remaining, this is zero, unless we returned + * before @timeout expired. */ -void blk_congestion_wait(int rw, long timeout) +long blk_congestion_wait(int rw, long timeout) { + long ret; DEFINE_WAIT(wait); wait_queue_head_t *wqh = &congestion_wqh[rw]; blk_run_queues(); prepare_to_wait(wqh, &wait, TASK_UNINTERRUPTIBLE); - io_schedule_timeout(timeout); + ret = io_schedule_timeout(timeout); finish_wait(wqh, &wait); + return ret; } - EXPORT_SYMBOL(blk_congestion_wait); /* @@ -2305,6 +2316,15 @@ int submit_bio(int rw, struct bio *bio) mod_page_state(pgpgout, count); else mod_page_state(pgpgin, count); + + if (unlikely(block_dump)) { + char b[BDEVNAME_SIZE]; + printk("%s(%d): %s block %Lu on %s\n", + current->comm, current->pid, + (rw & WRITE) ? "WRITE" : "READ", + (unsigned long long)bio->bi_sector, bdevname(bio->bi_bdev,b)); + } + generic_make_request(bio); return 1; } @@ -2592,10 +2612,22 @@ void end_that_request_last(struct reques unsigned long duration = jiffies - req->start_time; switch (rq_data_dir(req)) { case WRITE: + /* + * schedule the writeout of pending dirty data when the disk is idle. + * (Writeback is not postponed by writes, only by reads.) + */ + if (unlikely(laptop_mode)) + disk_is_spun_up(0); disk_stat_inc(disk, writes); disk_stat_add(disk, write_ticks, duration); break; case READ: + /* + * schedule the writeout of pending dirty data when the disk is idle. + * (postpone writeback until system is quiescent again.) + */ + if (unlikely(laptop_mode)) + disk_is_spun_up(1); disk_stat_inc(disk, reads); disk_stat_add(disk, read_ticks, duration); break; --- diff/drivers/block/loop.c 2004-03-01 14:36:06.000000000 +0000 +++ source/drivers/block/loop.c 2004-03-01 15:42:16.000000000 +0000 @@ -66,6 +66,7 @@ #include #include #include /* for invalidate_bdev() */ +#include #include @@ -148,14 +149,12 @@ static struct loop_func_table *xfer_func &xor_funcs }; -static int -figure_loop_size(struct loop_device *lo) +static loff_t get_loop_size(struct loop_device *lo, struct file *file) { loff_t size, offset, loopsize; - sector_t x; /* Compute loopsize in bytes */ - size = i_size_read(lo->lo_backing_file->f_mapping->host); + size = i_size_read(file->f_mapping->host); offset = lo->lo_offset; loopsize = size - offset; if (lo->lo_sizelimit > 0 && lo->lo_sizelimit < loopsize) @@ -165,8 +164,14 @@ figure_loop_size(struct loop_device *lo) * Unfortunately, if we want to do I/O on the device, * the number of 512-byte sectors has to fit into a sector_t. */ - size = loopsize >> 9; - x = (sector_t)size; + return loopsize >> 9; +} + +static int +figure_loop_size(struct loop_device *lo) +{ + loff_t size = get_loop_size(lo, lo->lo_backing_file); + sector_t x = (sector_t)size; if ((loff_t)x != size) return -EFBIG; @@ -429,12 +434,24 @@ inactive: goto out; } +struct switch_request { + struct file *file; + struct completion wait; +}; + +static void do_loop_switch(struct loop_device *, struct switch_request *); + static inline void loop_handle_bio(struct loop_device *lo, struct bio *bio) { int ret; - ret = do_bio_filebacked(lo, bio); - bio_endio(bio, bio->bi_size, ret); + if (unlikely(!bio->bi_bdev)) { + do_loop_switch(lo, bio->bi_private); + bio_put(bio); + } else { + ret = do_bio_filebacked(lo, bio); + bio_endio(bio, bio->bi_size, ret); + } } /* @@ -495,6 +512,103 @@ static int loop_thread(void *data) return 0; } +/* + * loop_switch performs the hard work of switching a backing store. + * First it needs to flush existing IO, it does this by sending a magic + * BIO down the pipe. The completion of this BIO does the actual switch. + */ +static int loop_switch(struct loop_device *lo, struct file *file) +{ + struct switch_request w; + struct bio *bio = bio_alloc(GFP_KERNEL, 1); + if (!bio) + return -ENOMEM; + init_completion(&w.wait); + w.file = file; + bio->bi_private = &w; + bio->bi_bdev = NULL; + loop_make_request(lo->lo_queue, bio); + wait_for_completion(&w.wait); + return 0; +} + +/* + * Do the actual switch; called from the BIO completion routine + */ +static void do_loop_switch(struct loop_device *lo, struct switch_request *p) +{ + struct file *file = p->file; + struct file *old_file = lo->lo_backing_file; + struct address_space *mapping = file->f_mapping; + + mapping_set_gfp_mask(old_file->f_mapping, lo->old_gfp_mask); + lo->lo_backing_file = file; + lo->lo_blocksize = mapping->host->i_blksize; + lo->old_gfp_mask = mapping_gfp_mask(mapping); + mapping_set_gfp_mask(mapping, lo->old_gfp_mask & ~(__GFP_IO|__GFP_FS)); + complete(&p->wait); +} + + +/* + * loop_change_fd switched the backing store of a loopback device to + * a new file. This is useful for operating system installers to free up + * the original file and in High Availability environments to switch to + * an alternative location for the content in case of server meltdown. + * This can only work if the loop device is used read-only, and if the + * new backing store is the same size and type as the old backing store. + */ +static int loop_change_fd(struct loop_device *lo, struct file *lo_file, + struct block_device *bdev, unsigned int arg) +{ + struct file *file, *old_file; + struct inode *inode; + int error; + + error = -ENXIO; + if (lo->lo_state != Lo_bound) + goto out; + + /* the loop device has to be read-only */ + error = -EINVAL; + if (lo->lo_flags != LO_FLAGS_READ_ONLY) + goto out; + + error = -EBADF; + file = fget(arg); + if (!file) + goto out; + + inode = file->f_mapping->host; + old_file = lo->lo_backing_file; + + error = -EINVAL; + + if (!S_ISREG(inode->i_mode) && !S_ISBLK(inode->i_mode)) + goto out_putf; + + /* new backing store needs to support loop (eg sendfile) */ + if (!inode->i_fop->sendfile) + goto out_putf; + + /* size of the new backing store needs to be the same */ + if (get_loop_size(lo, file) != get_loop_size(lo, old_file)) + goto out_putf; + + /* and ... switch */ + error = loop_switch(lo, file); + if (error) + goto out_putf; + + fput(old_file); + return 0; + + out_putf: + fput(file); + out: + return error; +} + static int loop_set_fd(struct loop_device *lo, struct file *lo_file, struct block_device *bdev, unsigned int arg) { @@ -505,6 +619,7 @@ static int loop_set_fd(struct loop_devic unsigned lo_blocksize; int lo_flags = 0; int error; + loff_t size; /* This is safe, since we have a reference from open(). */ __module_get(THIS_MODULE); @@ -543,6 +658,13 @@ static int loop_set_fd(struct loop_devic goto out_putf; } + size = get_loop_size(lo, file); + + if ((loff_t)(sector_t)size != size) { + error = -EFBIG; + goto out_putf; + } + if (!(lo_file->f_mode & FMODE_WRITE)) lo_flags |= LO_FLAGS_READ_ONLY; @@ -555,10 +677,6 @@ static int loop_set_fd(struct loop_devic lo->transfer = NULL; lo->ioctl = NULL; lo->lo_sizelimit = 0; - if (figure_loop_size(lo)) { - error = -EFBIG; - goto out_putf; - } lo->old_gfp_mask = mapping_gfp_mask(mapping); mapping_set_gfp_mask(mapping, lo->old_gfp_mask & ~(__GFP_IO|__GFP_FS)); @@ -571,6 +689,8 @@ static int loop_set_fd(struct loop_devic blk_queue_make_request(lo->lo_queue, loop_make_request); lo->lo_queue->queuedata = lo; + set_capacity(disks[lo->lo_number], size); + set_blocksize(bdev, lo_blocksize); kernel_thread(loop_thread, lo, CLONE_KERNEL); @@ -881,6 +1001,9 @@ static int lo_ioctl(struct inode * inode case LOOP_SET_FD: err = loop_set_fd(lo, file, inode->i_bdev, arg); break; + case LOOP_CHANGE_FD: + err = loop_change_fd(lo, file, inode->i_bdev, arg); + break; case LOOP_CLR_FD: err = loop_clr_fd(lo, inode->i_bdev); break; --- diff/drivers/block/scsi_ioctl.c 2004-02-18 08:54:08.000000000 +0000 +++ source/drivers/block/scsi_ioctl.c 2004-03-01 15:42:16.000000000 +0000 @@ -312,7 +312,7 @@ static int sg_scsi_ioctl(request_queue_t return -EFAULT; if (in_len > PAGE_SIZE || out_len > PAGE_SIZE) return -EINVAL; - if (get_user(opcode, sic->data)) + if (get_user(opcode, (int *)sic->data)) return -EFAULT; bytes = max(in_len, out_len); --- diff/drivers/char/cyclades.c 2004-03-01 14:36:06.000000000 +0000 +++ source/drivers/char/cyclades.c 2004-03-01 15:42:16.000000000 +0000 @@ -5411,6 +5411,7 @@ cy_init(void) cy_serial_driver->owner = THIS_MODULE; cy_serial_driver->driver_name = "cyclades"; cy_serial_driver->name = "ttyC"; + cy_serial_driver->devfs_name = "tts/C"; cy_serial_driver->major = CYCLADES_MAJOR; cy_serial_driver->minor_start = 0; cy_serial_driver->type = TTY_DRIVER_TYPE_SERIAL; --- diff/drivers/char/drm/Kconfig 2004-02-09 10:36:10.000000000 +0000 +++ source/drivers/char/drm/Kconfig 2004-03-01 15:42:16.000000000 +0000 @@ -76,7 +76,7 @@ config DRM_SIS tristate "SiS video cards" depends on DRM && AGP help - Choose this option if you have a SiS 630 or compatibel video + Choose this option if you have a SiS 630 or compatible video chipset. If M is selected the module will be called sis. AGP support is required for this driver to work. --- diff/drivers/char/drm/drm.h 2003-07-22 18:54:27.000000000 +0100 +++ source/drivers/char/drm/drm.h 2004-03-01 15:42:16.000000000 +0000 @@ -580,6 +580,16 @@ typedef struct drm_scatter_gather { unsigned long handle; /**< Used for mapping / unmapping */ } drm_scatter_gather_t; +/** + * DRM_IOCTL_SET_VERSION ioctl argument type. + */ +typedef struct drm_set_version { + int drm_di_major; + int drm_di_minor; + int drm_dd_major; + int drm_dd_minor; +} drm_set_version_t; + #define DRM_IOCTL_BASE 'd' #define DRM_IO(nr) _IO(DRM_IOCTL_BASE,nr) @@ -594,6 +604,7 @@ typedef struct drm_scatter_gather { #define DRM_IOCTL_GET_MAP DRM_IOWR(0x04, drm_map_t) #define DRM_IOCTL_GET_CLIENT DRM_IOWR(0x05, drm_client_t) #define DRM_IOCTL_GET_STATS DRM_IOR( 0x06, drm_stats_t) +#define DRM_IOCTL_SET_VERSION DRM_IOWR(0x07, drm_set_version_t) #define DRM_IOCTL_SET_UNIQUE DRM_IOW( 0x10, drm_unique_t) #define DRM_IOCTL_AUTH_MAGIC DRM_IOW( 0x11, drm_auth_t) --- diff/drivers/char/drm/drmP.h 2004-01-19 10:22:56.000000000 +0000 +++ source/drivers/char/drm/drmP.h 2004-03-01 15:42:16.000000000 +0000 @@ -92,8 +92,8 @@ #ifndef __HAVE_DMA #define __HAVE_DMA 0 #endif -#ifndef __HAVE_DMA_IRQ -#define __HAVE_DMA_IRQ 0 +#ifndef __HAVE_IRQ +#define __HAVE_IRQ 0 #endif #ifndef __HAVE_DMA_WAITLIST #define __HAVE_DMA_WAITLIST 0 @@ -324,6 +324,7 @@ do { \ #define DRM_BUFCOUNT(x) ((x)->count - DRM_LEFTCOUNT(x)) #define DRM_WAITCOUNT(dev,idx) DRM_BUFCOUNT(&dev->queuelist[idx]->waitlist) +#define DRM_IF_VERSION(maj, min) ((maj) << 16 | (min)) /** * Get the private SAREA mapping. * @@ -362,10 +363,13 @@ do { \ typedef int drm_ioctl_t( struct inode *inode, struct file *filp, unsigned int cmd, unsigned long arg ); -typedef struct drm_pci_list { - u16 vendor; - u16 device; -} drm_pci_list_t; +typedef struct drm_pci_id_list +{ + int vendor; + int device; + long driver_private; + char *name; +} drm_pci_id_list_t; typedef struct drm_ioctl_desc { drm_ioctl_t *func; @@ -488,6 +492,9 @@ typedef struct drm_file { struct drm_device *dev; int remove_auth_on_close; unsigned long lock_count; +#ifdef DRIVER_FILE_FIELDS + DRIVER_FILE_FIELDS; +#endif } drm_file_t; /** Wait queue */ @@ -622,6 +629,8 @@ typedef struct drm_device { int unique_len; /**< Length of unique field */ dev_t device; /**< Device number for mknod */ char *devname; /**< For /proc/interrupts */ + int minor; /**< Minor device number */ + int if_version; /**< Highest interface version set */ int blocked; /**< Blocked due to VC switch? */ struct proc_dir_entry *root; /**< Root for this device's entries */ @@ -679,6 +688,7 @@ typedef struct drm_device { /** \name Context support */ /*@{*/ int irq; /**< Interrupt used by board */ + int irq_enabled; /**< True if irq handler is enabled */ __volatile__ long context_flag; /**< Context swapping flag */ __volatile__ long interrupt_flag; /**< Interruption handler flag */ __volatile__ long dma_flag; /**< DMA dispatch flag */ @@ -714,7 +724,12 @@ typedef struct drm_device { #if __REALLY_HAVE_AGP drm_agp_head_t *agp; /**< AGP data */ #endif - struct pci_dev *pdev; /**< PCI device structure */ + + struct pci_dev *pdev; /**< PCI device structure */ + int pci_domain; /**< PCI bus domain number */ + int pci_bus; /**< PCI bus number */ + int pci_slot; /**< PCI slot number */ + int pci_func; /**< PCI function number */ #ifdef __alpha__ #if LINUX_VERSION_CODE < KERNEL_VERSION(2,4,3) struct pci_controler *hose; @@ -804,8 +819,8 @@ extern int DRM(unbind_agp)(DRM #endif /* Misc. IOCTL support (drm_ioctl.h) */ -extern int DRM(irq_busid)(struct inode *inode, struct file *filp, - unsigned int cmd, unsigned long arg); +extern int DRM(irq_by_busid)(struct inode *inode, struct file *filp, + unsigned int cmd, unsigned long arg); extern int DRM(getunique)(struct inode *inode, struct file *filp, unsigned int cmd, unsigned long arg); extern int DRM(setunique)(struct inode *inode, struct file *filp, @@ -816,6 +831,8 @@ extern int DRM(getclient)(struct in unsigned int cmd, unsigned long arg); extern int DRM(getstats)(struct inode *inode, struct file *filp, unsigned int cmd, unsigned long arg); +extern int DRM(setversion)(struct inode *inode, struct file *filp, + unsigned int cmd, unsigned long arg); /* Context IOCTL support (drm_context.h) */ extern int DRM(resctx)( struct inode *inode, struct file *filp, @@ -900,12 +917,17 @@ extern int DRM(dma_setup)(drm_devic extern void DRM(dma_takedown)(drm_device_t *dev); extern void DRM(free_buffer)(drm_device_t *dev, drm_buf_t *buf); extern void DRM(reclaim_buffers)( struct file *filp ); -#if __HAVE_DMA_IRQ +#endif /* __HAVE_DMA */ + + /* IRQ support (drm_irq.h) */ +#if __HAVE_IRQ || __HAVE_DMA extern int DRM(control)( struct inode *inode, struct file *filp, unsigned int cmd, unsigned long arg ); -extern int DRM(irq_install)( drm_device_t *dev, int irq ); +#endif +#if __HAVE_IRQ +extern int DRM(irq_install)( drm_device_t *dev ); extern int DRM(irq_uninstall)( drm_device_t *dev ); -extern irqreturn_t DRM(dma_service)( DRM_IRQ_ARGS ); +extern irqreturn_t DRM(irq_handler)( DRM_IRQ_ARGS ); extern void DRM(driver_irq_preinstall)( drm_device_t *dev ); extern void DRM(driver_irq_postinstall)( drm_device_t *dev ); extern void DRM(driver_irq_uninstall)( drm_device_t *dev ); @@ -915,12 +937,11 @@ extern int DRM(wait_vblank)(st extern int DRM(vblank_wait)(drm_device_t *dev, unsigned int *vbl_seq); extern void DRM(vbl_send_signals)( drm_device_t *dev ); #endif -#if __HAVE_DMA_IRQ_BH -extern void DRM(dma_immediate_bh)( void *dev ); +#if __HAVE_IRQ_BH +extern void DRM(irq_immediate_bh)( void *dev ); #endif #endif -#endif /* __HAVE_DMA */ #if __REALLY_HAVE_AGP /* AGP/GART support (drm_agpsupport.h) */ --- diff/drivers/char/drm/drm_bufs.h 2003-07-22 18:54:27.000000000 +0100 +++ source/drivers/char/drm/drm_bufs.h 2004-03-01 15:42:16.000000000 +0000 @@ -147,7 +147,9 @@ int DRM(addmap)( struct inode *inode, st MTRR_TYPE_WRCOMB, 1 ); } #endif - map->handle = DRM(ioremap)( map->offset, map->size, dev ); + if (map->type == _DRM_REGISTERS) + map->handle = DRM(ioremap)( map->offset, map->size, + dev ); break; case _DRM_SHM: @@ -160,6 +162,12 @@ int DRM(addmap)( struct inode *inode, st } map->offset = (unsigned long)map->handle; if ( map->flags & _DRM_CONTAINS_LOCK ) { + /* Prevent a 2nd X Server from creating a 2nd lock */ + if (dev->lock.hw_lock != NULL) { + vfree( map->handle ); + DRM(free)( map, sizeof(*map), DRM_MEM_MAPS ); + return -EBUSY; + } dev->sigdata.lock = dev->lock.hw_lock = map->handle; /* Pointer to lock */ } --- diff/drivers/char/drm/drm_dma.h 2003-07-22 18:54:27.000000000 +0100 +++ source/drivers/char/drm/drm_dma.h 2004-03-01 15:42:16.000000000 +0000 @@ -43,15 +43,6 @@ #ifndef __HAVE_DMA_RECLAIM #define __HAVE_DMA_RECLAIM 0 #endif -#ifndef __HAVE_SHARED_IRQ -#define __HAVE_SHARED_IRQ 0 -#endif - -#if __HAVE_SHARED_IRQ -#define DRM_IRQ_TYPE SA_SHIRQ -#else -#define DRM_IRQ_TYPE 0 -#endif #if __HAVE_DMA @@ -214,293 +205,11 @@ void DRM(reclaim_buffers)( struct file * } #endif - - - -#if __HAVE_DMA_IRQ - -/** - * Install IRQ handler. - * - * \param dev DRM device. - * \param irq IRQ number. - * - * Initializes the IRQ related data, and setups drm_device::vbl_queue. Installs the handler, calling the driver - * \c DRM(driver_irq_preinstall)() and \c DRM(driver_irq_postinstall)() functions - * before and after the installation. - */ -int DRM(irq_install)( drm_device_t *dev, int irq ) -{ - int ret; - - if ( !irq ) - return -EINVAL; - - down( &dev->struct_sem ); - - /* Driver must have been initialized */ - if ( !dev->dev_private ) { - up( &dev->struct_sem ); - return -EINVAL; - } - - if ( dev->irq ) { - up( &dev->struct_sem ); - return -EBUSY; - } - dev->irq = irq; - up( &dev->struct_sem ); - - DRM_DEBUG( "%s: irq=%d\n", __FUNCTION__, irq ); - - dev->context_flag = 0; - dev->interrupt_flag = 0; - dev->dma_flag = 0; - - dev->dma->next_buffer = NULL; - dev->dma->next_queue = NULL; - dev->dma->this_buffer = NULL; - -#if __HAVE_DMA_IRQ_BH - INIT_WORK(&dev->work, DRM(dma_immediate_bh), dev); -#endif - -#if __HAVE_VBL_IRQ - init_waitqueue_head(&dev->vbl_queue); - - spin_lock_init( &dev->vbl_lock ); - - INIT_LIST_HEAD( &dev->vbl_sigs.head ); - - dev->vbl_pending = 0; -#endif - - /* Before installing handler */ - DRM(driver_irq_preinstall)(dev); - - /* Install handler */ - ret = request_irq( dev->irq, DRM(dma_service), - DRM_IRQ_TYPE, dev->devname, dev ); - if ( ret < 0 ) { - down( &dev->struct_sem ); - dev->irq = 0; - up( &dev->struct_sem ); - return ret; - } - - /* After installing handler */ - DRM(driver_irq_postinstall)(dev); - - return 0; -} - -/** - * Uninstall the IRQ handler. - * - * \param dev DRM device. - * - * Calls the driver's \c DRM(driver_irq_uninstall)() function, and stops the irq. - */ -int DRM(irq_uninstall)( drm_device_t *dev ) -{ - int irq; - - down( &dev->struct_sem ); - irq = dev->irq; - dev->irq = 0; - up( &dev->struct_sem ); - - if ( !irq ) - return -EINVAL; - - DRM_DEBUG( "%s: irq=%d\n", __FUNCTION__, irq ); - - DRM(driver_irq_uninstall)( dev ); - - free_irq( irq, dev ); - - return 0; -} - -/** - * IRQ control ioctl. - * - * \param inode device inode. - * \param filp file pointer. - * \param cmd command. - * \param arg user argument, pointing to a drm_control structure. - * \return zero on success or a negative number on failure. - * - * Calls irq_install() or irq_uninstall() according to \p arg. - */ -int DRM(control)( struct inode *inode, struct file *filp, - unsigned int cmd, unsigned long arg ) -{ - drm_file_t *priv = filp->private_data; - drm_device_t *dev = priv->dev; - drm_control_t ctl; - - if ( copy_from_user( &ctl, (drm_control_t *)arg, sizeof(ctl) ) ) - return -EFAULT; - - switch ( ctl.func ) { - case DRM_INST_HANDLER: - return DRM(irq_install)( dev, ctl.irq ); - case DRM_UNINST_HANDLER: - return DRM(irq_uninstall)( dev ); - default: - return -EINVAL; - } -} - -#if __HAVE_VBL_IRQ - -/** - * Wait for VBLANK. - * - * \param inode device inode. - * \param filp file pointer. - * \param cmd command. - * \param data user argument, pointing to a drm_wait_vblank structure. - * \return zero on success or a negative number on failure. - * - * Verifies the IRQ is installed. - * - * If a signal is requested checks if this task has already scheduled the same signal - * for the same vblank sequence number - nothing to be done in - * that case. If the number of tasks waiting for the interrupt exceeds 100 the - * function fails. Otherwise adds a new entry to drm_device::vbl_sigs for this - * task. - * - * If a signal is not requested, then calls vblank_wait(). - */ -int DRM(wait_vblank)( DRM_IOCTL_ARGS ) -{ - drm_file_t *priv = filp->private_data; - drm_device_t *dev = priv->dev; - drm_wait_vblank_t vblwait; - struct timeval now; - int ret = 0; - unsigned int flags; - - if (!dev->irq) - return -EINVAL; - - DRM_COPY_FROM_USER_IOCTL( vblwait, (drm_wait_vblank_t *)data, - sizeof(vblwait) ); - - switch ( vblwait.request.type & ~_DRM_VBLANK_FLAGS_MASK ) { - case _DRM_VBLANK_RELATIVE: - vblwait.request.sequence += atomic_read( &dev->vbl_received ); - vblwait.request.type &= ~_DRM_VBLANK_RELATIVE; - case _DRM_VBLANK_ABSOLUTE: - break; - default: - return -EINVAL; - } - - flags = vblwait.request.type & _DRM_VBLANK_FLAGS_MASK; - - if ( flags & _DRM_VBLANK_SIGNAL ) { - unsigned long irqflags; - drm_vbl_sig_t *vbl_sig; - - vblwait.reply.sequence = atomic_read( &dev->vbl_received ); - - spin_lock_irqsave( &dev->vbl_lock, irqflags ); - - /* Check if this task has already scheduled the same signal - * for the same vblank sequence number; nothing to be done in - * that case - */ - list_for_each_entry( vbl_sig, &dev->vbl_sigs.head, head ) { - if (vbl_sig->sequence == vblwait.request.sequence - && vbl_sig->info.si_signo == vblwait.request.signal - && vbl_sig->task == current) - { - spin_unlock_irqrestore( &dev->vbl_lock, irqflags ); - goto done; - } - } - - if ( dev->vbl_pending >= 100 ) { - spin_unlock_irqrestore( &dev->vbl_lock, irqflags ); - return -EBUSY; - } - - dev->vbl_pending++; - - spin_unlock_irqrestore( &dev->vbl_lock, irqflags ); - - if ( !( vbl_sig = DRM_MALLOC( sizeof( drm_vbl_sig_t ) ) ) ) { - return -ENOMEM; - } - - memset( (void *)vbl_sig, 0, sizeof(*vbl_sig) ); - - vbl_sig->sequence = vblwait.request.sequence; - vbl_sig->info.si_signo = vblwait.request.signal; - vbl_sig->task = current; - - spin_lock_irqsave( &dev->vbl_lock, irqflags ); - - list_add_tail( (struct list_head *) vbl_sig, &dev->vbl_sigs.head ); - - spin_unlock_irqrestore( &dev->vbl_lock, irqflags ); - } else { - ret = DRM(vblank_wait)( dev, &vblwait.request.sequence ); - - do_gettimeofday( &now ); - vblwait.reply.tval_sec = now.tv_sec; - vblwait.reply.tval_usec = now.tv_usec; - } - -done: - DRM_COPY_TO_USER_IOCTL( (drm_wait_vblank_t *)data, vblwait, - sizeof(vblwait) ); - - return ret; -} - -/** - * Send the VBLANK signals. - * - * \param dev DRM device. - * - * Sends a signal for each task in drm_device::vbl_sigs and empties the list. - * - * If a signal is not requested, then calls vblank_wait(). +#if !__HAVE_IRQ +/* This stub DRM_IOCTL_CONTROL handler is for the drivers that used to require + * IRQs for DMA but no longer do. It maintains compatibility with the X Servers + * that try to use the control ioctl by simply returning success. */ -void DRM(vbl_send_signals)( drm_device_t *dev ) -{ - struct list_head *list, *tmp; - drm_vbl_sig_t *vbl_sig; - unsigned int vbl_seq = atomic_read( &dev->vbl_received ); - unsigned long flags; - - spin_lock_irqsave( &dev->vbl_lock, flags ); - - list_for_each_safe( list, tmp, &dev->vbl_sigs.head ) { - vbl_sig = list_entry( list, drm_vbl_sig_t, head ); - if ( ( vbl_seq - vbl_sig->sequence ) <= (1<<23) ) { - vbl_sig->info.si_code = vbl_seq; - send_sig_info( vbl_sig->info.si_signo, &vbl_sig->info, vbl_sig->task ); - - list_del( list ); - - DRM_FREE( vbl_sig, sizeof(*vbl_sig) ); - - dev->vbl_pending--; - } - } - - spin_unlock_irqrestore( &dev->vbl_lock, flags ); -} - -#endif /* __HAVE_VBL_IRQ */ - -#else - int DRM(control)( struct inode *inode, struct file *filp, unsigned int cmd, unsigned long arg ) { @@ -517,7 +226,6 @@ int DRM(control)( struct inode *inode, s return -EINVAL; } } - -#endif /* __HAVE_DMA_IRQ */ +#endif #endif /* __HAVE_DMA */ --- diff/drivers/char/drm/drm_drv.h 2003-10-27 09:20:43.000000000 +0000 +++ source/drivers/char/drm/drm_drv.h 2004-03-01 15:42:16.000000000 +0000 @@ -58,8 +58,8 @@ #ifndef __HAVE_CTX_BITMAP #define __HAVE_CTX_BITMAP 0 #endif -#ifndef __HAVE_DMA_IRQ -#define __HAVE_DMA_IRQ 0 +#ifndef __HAVE_IRQ +#define __HAVE_IRQ 0 #endif #ifndef __HAVE_DMA_QUEUE #define __HAVE_DMA_QUEUE 0 @@ -126,6 +126,9 @@ #ifndef DRIVER_IOCTLS #define DRIVER_IOCTLS #endif +#ifndef DRIVER_OPEN_HELPER +#define DRIVER_OPEN_HELPER( priv, dev ) +#endif #ifndef DRIVER_FOPS #define DRIVER_FOPS \ static struct file_operations DRM(fops) = { \ @@ -159,15 +162,8 @@ __setup( DRIVER_NAME "=", DRM_OPTIONS_FU #undef DRM_OPTIONS_FUNC #endif -/** - * The default number of instances (minor numbers) to initialize. - */ -#ifndef DRIVER_NUM_CARDS -#define DRIVER_NUM_CARDS 1 -#endif - -static drm_device_t *DRM(device); -static int *DRM(minor); +#define MAX_DEVICES 4 +static drm_device_t DRM(device)[MAX_DEVICES]; static int DRM(numdevs) = 0; DRIVER_FOPS; @@ -177,10 +173,13 @@ static drm_ioctl_desc_t DRM(ioctls)[] [DRM_IOCTL_NR(DRM_IOCTL_VERSION)] = { DRM(version), 0, 0 }, [DRM_IOCTL_NR(DRM_IOCTL_GET_UNIQUE)] = { DRM(getunique), 0, 0 }, [DRM_IOCTL_NR(DRM_IOCTL_GET_MAGIC)] = { DRM(getmagic), 0, 0 }, - [DRM_IOCTL_NR(DRM_IOCTL_IRQ_BUSID)] = { DRM(irq_busid), 0, 1 }, +#if __HAVE_IRQ + [DRM_IOCTL_NR(DRM_IOCTL_IRQ_BUSID)] = { DRM(irq_by_busid), 0, 1 }, +#endif [DRM_IOCTL_NR(DRM_IOCTL_GET_MAP)] = { DRM(getmap), 0, 0 }, [DRM_IOCTL_NR(DRM_IOCTL_GET_CLIENT)] = { DRM(getclient), 0, 0 }, [DRM_IOCTL_NR(DRM_IOCTL_GET_STATS)] = { DRM(getstats), 0, 0 }, + [DRM_IOCTL_NR(DRM_IOCTL_SET_VERSION)] = { DRM(setversion), 0, 1 }, [DRM_IOCTL_NR(DRM_IOCTL_SET_UNIQUE)] = { DRM(setunique), 1, 1 }, [DRM_IOCTL_NR(DRM_IOCTL_BLOCK)] = { DRM(noop), 1, 1 }, @@ -222,9 +221,9 @@ static drm_ioctl_desc_t DRM(ioctls)[] [DRM_IOCTL_NR(DRM_IOCTL_INFO_BUFS)] = { DRM(infobufs), 1, 0 }, [DRM_IOCTL_NR(DRM_IOCTL_MAP_BUFS)] = { DRM(mapbufs), 1, 0 }, [DRM_IOCTL_NR(DRM_IOCTL_FREE_BUFS)] = { DRM(freebufs), 1, 0 }, - - /* The DRM_IOCTL_DMA ioctl should be defined by the driver. - */ + /* The DRM_IOCTL_DMA ioctl should be defined by the driver. */ +#endif +#if __HAVE_IRQ || __HAVE_DMA [DRM_IOCTL_NR(DRM_IOCTL_CONTROL)] = { DRM(control), 1, 1 }, #endif @@ -337,7 +336,7 @@ static int DRM(setup)( drm_device_t *dev dev->queue_reserved = 0; dev->queue_slots = 0; dev->queuelist = NULL; - dev->irq = 0; + dev->irq_enabled = 0; dev->context_flag = 0; dev->interrupt_flag = 0; dev->dma_flag = 0; @@ -345,6 +344,7 @@ static int DRM(setup)( drm_device_t *dev dev->last_switch = 0; dev->last_checked = 0; init_waitqueue_head( &dev->context_wait ); + dev->if_version = 0; dev->ctx_start = 0; dev->lck_start = 0; @@ -391,8 +391,8 @@ static int DRM(takedown)( drm_device_t * DRM_DEBUG( "\n" ); DRIVER_PRETAKEDOWN(); -#if __HAVE_DMA_IRQ - if ( dev->irq ) DRM(irq_uninstall)( dev ); +#if __HAVE_IRQ + if ( dev->irq_enabled ) DRM(irq_uninstall)( dev ); #endif down( &dev->struct_sem ); @@ -534,52 +534,111 @@ static int DRM(takedown)( drm_device_t * return 0; } -/** - * Figure out how many instances to initialize. - * - * \return number of cards found. - * - * Searches for every PCI card in \c DRIVER_CARD_LIST with matching vendor and device ids. - */ -static int drm_count_cards(void) +static drm_pci_id_list_t DRM(pciidlist)[] = { + DRIVER_PCI_IDS +}; + +static int DRM(probe)(struct pci_dev *pdev) { - int num = 0; -#if defined(DRIVER_CARD_LIST) - int i; - drm_pci_list_t *l; - u16 device, vendor; - struct pci_dev *pdev = NULL; + drm_device_t *dev; +#if __HAVE_CTX_BITMAP + int retcode; #endif + int i; + char *desc = NULL; DRM_DEBUG( "\n" ); -#if defined(DRIVER_COUNT_CARDS) - num = DRIVER_COUNT_CARDS(); -#elif defined(DRIVER_CARD_LIST) - for (i = 0, l = DRIVER_CARD_LIST; l[i].vendor != 0; i++) { - pdev = NULL; - vendor = l[i].vendor; - device = l[i].device; - if(device == 0xffff) device = PCI_ANY_ID; - if(vendor == 0xffff) vendor = PCI_ANY_ID; - while ((pdev = pci_find_device(vendor, device, pdev))) { - num++; + for (i = 0; DRM(pciidlist)[i].vendor != 0; i++) { + if ((DRM(pciidlist)[i].vendor == pdev->vendor) && + (DRM(pciidlist)[i].device == pdev->device)) { + desc = DRM(pciidlist)[i].name; } } + if (desc == NULL) + return -ENODEV; + + if (DRM(numdevs) >= MAX_DEVICES) + return -ENODEV; + + dev = &(DRM(device)[DRM(numdevs)]); + + memset( (void *)dev, 0, sizeof(*dev) ); + dev->count_lock = SPIN_LOCK_UNLOCKED; + init_timer( &dev->timer ); + sema_init( &dev->struct_sem, 1 ); + + if ((dev->minor = DRM(stub_register)(DRIVER_NAME, &DRM(fops),dev)) < 0) + return -EPERM; + dev->device = MKDEV(DRM_MAJOR, dev->minor ); + dev->name = DRIVER_NAME; + + dev->pdev = pdev; +#ifdef __alpha__ + dev->hose = pdev->sysdata; + dev->pci_domain = dev->hose->bus->number; #else - num = DRIVER_NUM_CARDS; + dev->pci_domain = 0; +#endif + dev->pci_bus = pdev->bus->number; + dev->pci_slot = PCI_SLOT(pdev->devfn); + dev->pci_func = PCI_FUNC(pdev->devfn); + dev->irq = pdev->irq; + + DRIVER_PREINIT(); + +#if __REALLY_HAVE_AGP + dev->agp = DRM(agp_init)(); +#if __MUST_HAVE_AGP + if ( dev->agp == NULL ) { + DRM_ERROR( "Cannot initialize the agpgart module.\n" ); + DRM(stub_unregister)(dev->minor); + DRM(takedown)( dev ); + return -ENOMEM; + } +#endif +#if __REALLY_HAVE_MTRR + if (dev->agp) + dev->agp->agp_mtrr = mtrr_add( dev->agp->agp_info.aper_base, + dev->agp->agp_info.aper_size*1024*1024, + MTRR_TYPE_WRCOMB, + 1 ); +#endif +#endif + +#if __HAVE_CTX_BITMAP + retcode = DRM(ctxbitmap_init)( dev ); + if( retcode ) { + DRM_ERROR( "Cannot allocate memory for context bitmap.\n" ); + DRM(stub_unregister)(dev->minor); + DRM(takedown)( dev ); + return retcode; + } #endif - DRM_DEBUG("numdevs = %d\n", num); - return num; + DRM(numdevs)++; /* no errors, mark it reserved */ + + DRM_INFO( "Initialized %s %d.%d.%d %s on minor %d: %s\n", + DRIVER_NAME, + DRIVER_MAJOR, + DRIVER_MINOR, + DRIVER_PATCHLEVEL, + DRIVER_DATE, + dev->minor, + desc ); + + DRIVER_POSTINIT(); + + return 0; } + /** * Module initialization. Called via init_module at module load time, or via * linux/init/main.c (this is not currently supported). * * \return zero on success or a negative number on failure. * - * Allocates and initialize an array of drm_device structures, and attempts to + * Initializes an array of drm_device structures, and attempts to * initialize all available devices, using consecutive minors, registering the * stubs and initializing the AGP device. * @@ -588,88 +647,19 @@ static int drm_count_cards(void) */ static int __init drm_init( void ) { + struct pci_dev *pdev = NULL; - drm_device_t *dev; - int i; -#if __HAVE_CTX_BITMAP - int retcode; -#endif DRM_DEBUG( "\n" ); #ifdef MODULE DRM(parse_options)( drm_opts ); #endif - DRM(numdevs) = drm_count_cards(); - /* Force at least one instance. */ - if (DRM(numdevs) <= 0) - DRM(numdevs) = 1; - - DRM(device) = kmalloc(sizeof(*DRM(device)) * DRM(numdevs), GFP_KERNEL); - if (!DRM(device)) { - return -ENOMEM; - } - DRM(minor) = kmalloc(sizeof(*DRM(minor)) * DRM(numdevs), GFP_KERNEL); - if (!DRM(minor)) { - kfree(DRM(device)); - return -ENOMEM; - } - - DRIVER_PREINIT(); - DRM(mem_init)(); - for (i = 0; i < DRM(numdevs); i++) { - dev = &(DRM(device)[i]); - memset( (void *)dev, 0, sizeof(*dev) ); - dev->count_lock = SPIN_LOCK_UNLOCKED; - init_timer( &dev->timer ); - sema_init( &dev->struct_sem, 1 ); - - if ((DRM(minor)[i] = DRM(stub_register)(DRIVER_NAME, &DRM(fops),dev)) < 0) - return -EPERM; - dev->device = MKDEV(DRM_MAJOR, DRM(minor)[i] ); - dev->name = DRIVER_NAME; - -#if __REALLY_HAVE_AGP - dev->agp = DRM(agp_init)(); -#if __MUST_HAVE_AGP - if ( dev->agp == NULL ) { - DRM_ERROR( "Cannot initialize the agpgart module.\n" ); - DRM(stub_unregister)(DRM(minor)[i]); - DRM(takedown)( dev ); - return -EINVAL; - } -#endif -#if __REALLY_HAVE_MTRR - if (dev->agp) - dev->agp->agp_mtrr = mtrr_add( dev->agp->agp_info.aper_base, - dev->agp->agp_info.aper_size*1024*1024, - MTRR_TYPE_WRCOMB, - 1 ); -#endif -#endif - -#if __HAVE_CTX_BITMAP - retcode = DRM(ctxbitmap_init)( dev ); - if( retcode ) { - DRM_ERROR( "Cannot allocate memory for context bitmap.\n" ); - DRM(stub_unregister)(DRM(minor)[i]); - DRM(takedown)( dev ); - return retcode; - } -#endif - DRM_INFO( "Initialized %s %d.%d.%d %s on minor %d\n", - DRIVER_NAME, - DRIVER_MAJOR, - DRIVER_MINOR, - DRIVER_PATCHLEVEL, - DRIVER_DATE, - DRM(minor)[i] ); + while ((pdev = pci_find_device(PCI_ANY_ID, PCI_ANY_ID, pdev)) != NULL) { + DRM(probe)(pdev); } - - DRIVER_POSTINIT(); - return 0; } @@ -689,10 +679,10 @@ static void __exit drm_cleanup( void ) for (i = DRM(numdevs) - 1; i >= 0; i--) { dev = &(DRM(device)[i]); - if ( DRM(stub_unregister)(DRM(minor)[i]) ) { + if ( DRM(stub_unregister)(dev->minor) ) { DRM_ERROR( "Cannot unload module\n" ); } else { - DRM_DEBUG("minor %d unregistered\n", DRM(minor)[i]); + DRM_DEBUG("minor %d unregistered\n", dev->minor); if (i == 0) { DRM_INFO( "Module unloaded\n" ); } @@ -722,8 +712,6 @@ static void __exit drm_cleanup( void ) #endif } DRIVER_POSTCLEANUP(); - kfree(DRM(minor)); - kfree(DRM(device)); DRM(numdevs) = 0; } @@ -795,7 +783,7 @@ int DRM(open)( struct inode *inode, stru int i; for (i = 0; i < DRM(numdevs); i++) { - if (iminor(inode) == DRM(minor)[i]) { + if (iminor(inode) == DRM(device)[i].minor) { dev = &(DRM(device)[i]); break; } --- diff/drivers/char/drm/drm_fops.h 2003-10-09 09:47:16.000000000 +0100 +++ source/drivers/char/drm/drm_fops.h 2004-03-01 15:42:16.000000000 +0000 @@ -72,6 +72,8 @@ int DRM(open_helper)(struct inode *inode priv->authenticated = capable(CAP_SYS_ADMIN); priv->lock_count = 0; + DRIVER_OPEN_HELPER( priv, dev ); + down(&dev->struct_sem); if (!dev->file_last) { priv->next = NULL; --- diff/drivers/char/drm/drm_ioctl.h 2003-07-22 18:54:27.000000000 +0100 +++ source/drivers/char/drm/drm_ioctl.h 2004-03-01 15:42:16.000000000 +0000 @@ -35,69 +35,7 @@ #include "drmP.h" - -/** - * Get interrupt from bus id. - * - * \param inode device inode. - * \param filp file pointer. - * \param cmd command. - * \param arg user argument, pointing to a drm_irq_busid structure. - * \return zero on success or a negative number on failure. - * - * Finds the PCI device with the specified bus id and gets its IRQ number. - */ -int DRM(irq_busid)(struct inode *inode, struct file *filp, - unsigned int cmd, unsigned long arg) -{ - drm_irq_busid_t p; - struct pci_dev *dev; - - if (copy_from_user(&p, (drm_irq_busid_t *)arg, sizeof(p))) - return -EFAULT; -#ifdef __alpha__ - { - int domain = p.busnum >> 8; - p.busnum &= 0xff; - - /* - * Find the hose the device is on (the domain number is the - * hose index) and offset the bus by the root bus of that - * hose. - */ - for(dev = pci_find_device(PCI_ANY_ID,PCI_ANY_ID,NULL); - dev; - dev = pci_find_device(PCI_ANY_ID,PCI_ANY_ID,dev)) { - struct pci_controller *hose = dev->sysdata; - - if (hose->index == domain) { - p.busnum += hose->bus->number; - break; - } - } - } -#endif - dev = pci_find_slot(p.busnum, PCI_DEVFN(p.devnum, p.funcnum)); - if (!dev) { - DRM_ERROR("pci_find_slot failed for %d:%d:%d\n", - p.busnum, p.devnum, p.funcnum); - p.irq = 0; - goto out; - } - if (pci_enable_device(dev) != 0) { - DRM_ERROR("pci_enable_device failed for %d:%d:%d\n", - p.busnum, p.devnum, p.funcnum); - p.irq = 0; - goto out; - } - p.irq = dev->irq; - out: - DRM_DEBUG("%d:%d:%d => IRQ %d\n", - p.busnum, p.devnum, p.funcnum, p.irq); - if (copy_to_user((drm_irq_busid_t *)arg, &p, sizeof(p))) - return -EFAULT; - return 0; -} +#include /** * Get the bus id. @@ -138,8 +76,10 @@ int DRM(getunique)(struct inode *inode, * \param arg user argument, pointing to a drm_unique structure. * \return zero on success or a negative number on failure. * - * Copies the bus id from userspace into drm_device::unique, and searches for - * the respective PCI device, updating drm_device::pdev. + * Copies the bus id from userspace into drm_device::unique, and verifies that + * it matches the device this DRM is attached to (EINVAL otherwise). Deprecated + * in interface version 1.1 and will return EBUSY when setversion has requested + * version 1.1 or greater. */ int DRM(setunique)(struct inode *inode, struct file *filp, unsigned int cmd, unsigned long arg) @@ -147,6 +87,7 @@ int DRM(setunique)(struct inode *inode, drm_file_t *priv = filp->private_data; drm_device_t *dev = priv->dev; drm_unique_t u; + int domain, bus, slot, func, ret; if (dev->unique_len || dev->unique) return -EBUSY; @@ -164,55 +105,42 @@ int DRM(setunique)(struct inode *inode, dev->devname = DRM(alloc)(strlen(dev->name) + strlen(dev->unique) + 2, DRM_MEM_DRIVER); - if(!dev->devname) { - DRM(free)(dev->devname, sizeof(*dev->devname), DRM_MEM_DRIVER); + if (!dev->devname) return -ENOMEM; - } + sprintf(dev->devname, "%s@%s", dev->name, dev->unique); - do { - struct pci_dev *pci_dev; - int domain, b, d, f; - char *p; - - for(p = dev->unique; p && *p && *p != ':'; p++); - if (!p || !*p) break; - b = (int)simple_strtoul(p+1, &p, 10); - if (*p != ':') break; - d = (int)simple_strtoul(p+1, &p, 10); - if (*p != ':') break; - f = (int)simple_strtoul(p+1, &p, 10); - if (*p) break; - - domain = b >> 8; - b &= 0xff; - -#ifdef __alpha__ - /* - * Find the hose the device is on (the domain number is the - * hose index) and offset the bus by the root bus of that - * hose. - */ - for(pci_dev = pci_find_device(PCI_ANY_ID,PCI_ANY_ID,NULL); - pci_dev; - pci_dev = pci_find_device(PCI_ANY_ID,PCI_ANY_ID,pci_dev)) { - struct pci_controller *hose = pci_dev->sysdata; - - if (hose->index == domain) { - b += hose->bus->number; - break; - } - } -#endif + /* Return error if the busid submitted doesn't match the device's actual + * busid. + */ + ret = sscanf(dev->unique, "PCI:%d:%d:%d", &bus, &slot, &func); + if (ret != 3) + return DRM_ERR(EINVAL); + domain = bus >> 8; + bus &= 0xff; + + if ((domain != dev->pci_domain) || + (bus != dev->pci_bus) || + (slot != dev->pci_slot) || + (func != dev->pci_func)) + return -EINVAL; - pci_dev = pci_find_slot(b, PCI_DEVFN(d,f)); - if (pci_dev) { - dev->pdev = pci_dev; -#ifdef __alpha__ - dev->hose = pci_dev->sysdata; -#endif - } - } while(0); + return 0; +} + +static int +DRM(set_busid)(drm_device_t *dev) +{ + if (dev->unique != NULL) + return EBUSY; + + dev->unique_len = 20; + dev->unique = DRM(alloc)(dev->unique_len + 1, DRM_MEM_DRIVER); + if (dev->unique == NULL) + return ENOMEM; + + snprintf(dev->unique, dev->unique_len, "pci:%04x:%02x:%02x.%d", + dev->pci_domain, dev->pci_bus, dev->pci_slot, dev->pci_func); return 0; } @@ -363,3 +291,48 @@ int DRM(getstats)( struct inode *inode, return -EFAULT; return 0; } + +#define DRM_IF_MAJOR 1 +#define DRM_IF_MINOR 2 + +int DRM(setversion)(DRM_IOCTL_ARGS) +{ + DRM_DEVICE; + drm_set_version_t sv; + drm_set_version_t retv; + int if_version; + + DRM_COPY_FROM_USER_IOCTL(sv, (drm_set_version_t *)data, sizeof(sv)); + + retv.drm_di_major = DRM_IF_MAJOR; + retv.drm_di_minor = DRM_IF_MINOR; + retv.drm_dd_major = DRIVER_MAJOR; + retv.drm_dd_minor = DRIVER_MINOR; + + DRM_COPY_TO_USER_IOCTL((drm_set_version_t *)data, retv, sizeof(sv)); + + if (sv.drm_di_major != -1) { + if (sv.drm_di_major != DRM_IF_MAJOR || + sv.drm_di_minor < 0 || sv.drm_di_minor > DRM_IF_MINOR) + return EINVAL; + if_version = DRM_IF_VERSION(sv.drm_di_major, sv.drm_dd_minor); + dev->if_version = DRM_MAX(if_version, dev->if_version); + if (sv.drm_di_minor >= 1) { + /* + * Version 1.1 includes tying of DRM to specific device + */ + DRM(set_busid)(dev); + } + } + + if (sv.drm_dd_major != -1) { + if (sv.drm_dd_major != DRIVER_MAJOR || + sv.drm_dd_minor < 0 || sv.drm_dd_minor > DRIVER_MINOR) + return EINVAL; +#ifdef DRIVER_SETVERSION + DRIVER_SETVERSION(dev, &sv); +#endif + } + return 0; +} + --- diff/drivers/char/drm/drm_os_linux.h 2003-09-30 15:46:12.000000000 +0100 +++ source/drivers/char/drm/drm_os_linux.h 2004-03-01 15:42:16.000000000 +0000 @@ -62,8 +62,12 @@ verify_area( VERIFY_READ, uaddr, size ) #define DRM_COPY_FROM_USER_UNCHECKED(arg1, arg2, arg3) \ __copy_from_user(arg1, arg2, arg3) +#define DRM_COPY_TO_USER_UNCHECKED(arg1, arg2, arg3) \ + __copy_to_user(arg1, arg2, arg3) #define DRM_GET_USER_UNCHECKED(val, uaddr) \ __get_user(val, uaddr) +#define DRM_PUT_USER_UNCHECKED(uaddr, val) \ + __put_user(val, uaddr) /** 'malloc' without the overhead of DRM(alloc)() */ @@ -71,6 +75,8 @@ /** 'free' without the overhead of DRM(free)() */ #define DRM_FREE(x,size) kfree(x) +#define DRM_GET_PRIV_WITH_RETURN(_priv, _filp) _priv = _filp->private_data + /** * Get the pointer to the SAREA. * --- diff/drivers/char/drm/gamma.h 2003-05-21 11:50:14.000000000 +0100 +++ source/drivers/char/drm/gamma.h 2004-03-01 15:42:16.000000000 +0000 @@ -53,6 +53,10 @@ [DRM_IOCTL_NR(DRM_IOCTL_GAMMA_INIT)] = { gamma_dma_init, 1, 1 }, \ [DRM_IOCTL_NR(DRM_IOCTL_GAMMA_COPY)] = { gamma_dma_copy, 1, 1 } +#define DRIVER_PCI_IDS \ + {0x3d3d, 0x0008, 0, "3DLabs GLINT Gamma G1"}, \ + {0, 0, 0, NULL} + #define IOCTL_TABLE_NAME DRM(ioctls) #define IOCTL_FUNC_NAME DRM(ioctl) @@ -104,8 +108,8 @@ return 0; \ } while (0) -#define __HAVE_DMA_IRQ 1 -#define __HAVE_DMA_IRQ_BH 1 +#define __HAVE_IRQ 1 +#define __HAVE_IRQ_BH 1 #define DRIVER_AGP_BUFFERS_MAP( dev ) \ ((drm_gamma_private_t *)((dev)->dev_private))->buffers --- diff/drivers/char/drm/gamma_dma.c 2003-09-30 15:46:12.000000000 +0100 +++ source/drivers/char/drm/gamma_dma.c 2004-03-01 15:42:16.000000000 +0000 @@ -116,7 +116,7 @@ static inline int gamma_dma_is_ready(drm return (!GAMMA_READ(GAMMA_DMACOUNT)); } -irqreturn_t gamma_dma_service( DRM_IRQ_ARGS ) +irqreturn_t gamma_irq_handler( DRM_IRQ_ARGS ) { drm_device_t *dev = (drm_device_t *)arg; drm_device_dma_t *dma = dev->dma; @@ -262,7 +262,7 @@ static void gamma_dma_timer_bh(unsigned gamma_dma_schedule((drm_device_t *)dev, 0); } -void gamma_dma_immediate_bh(void *dev) +void gamma_irq_immediate_bh(void *dev) { gamma_dma_schedule(dev, 0); } @@ -656,12 +656,12 @@ int gamma_do_cleanup_dma( drm_device_t * { DRM_DEBUG( "%s\n", __FUNCTION__ ); -#if _HAVE_DMA_IRQ +#if __HAVE_IRQ /* Make sure interrupts are disabled here because the uninstall ioctl * may not have been called from userspace and after dev_private * is freed, it's too late. */ - if ( dev->irq ) DRM(irq_uninstall)(dev); + if ( dev->irq_enabled ) DRM(irq_uninstall)(dev); #endif if ( dev->dev_private ) { --- diff/drivers/char/drm/gamma_drv.c 2003-05-21 11:50:14.000000000 +0100 +++ source/drivers/char/drm/gamma_drv.c 2004-03-01 15:42:16.000000000 +0000 @@ -48,6 +48,7 @@ #include "drm_fops.h" #include "drm_init.h" #include "drm_ioctl.h" +#include "drm_irq.h" #include "gamma_lists.h" /* NOTE */ #include "drm_lock.h" #include "gamma_lock.h" /* NOTE */ --- diff/drivers/char/drm/i810.h 2003-08-26 10:00:52.000000000 +0100 +++ source/drivers/char/drm/i810.h 2004-03-01 15:42:16.000000000 +0000 @@ -77,7 +77,14 @@ [DRM_IOCTL_NR(DRM_IOCTL_I810_MC)] = { i810_dma_mc, 1, 1 }, \ [DRM_IOCTL_NR(DRM_IOCTL_I810_RSTATUS)] = { i810_rstatus, 1, 0 }, \ [DRM_IOCTL_NR(DRM_IOCTL_I810_FLIP)] = { i810_flip_bufs, 1, 0 } - + +#define DRIVER_PCI_IDS \ + {0x8086, 0x7121, 0, "Intel i810 GMCH"}, \ + {0x8086, 0x7123, 0, "Intel i810-DC100 GMCH"}, \ + {0x8086, 0x7125, 0, "Intel i810E GMCH"}, \ + {0x8086, 0x1132, 0, "Intel i815 GMCH"}, \ + {0, 0, 0, NULL} + #define __HAVE_COUNTERS 4 #define __HAVE_COUNTER6 _DRM_STAT_IRQ @@ -112,7 +119,7 @@ * a noop stub is generated for compatibility. */ /* XXX: Add vblank support? */ -#define __HAVE_DMA_IRQ 0 +#define __HAVE_IRQ 0 /* Buffer customization: */ --- diff/drivers/char/drm/i810_dma.c 2003-09-30 15:46:12.000000000 +0100 +++ source/drivers/char/drm/i810_dma.c 2004-03-01 15:42:16.000000000 +0000 @@ -53,41 +53,41 @@ static inline void i810_print_status_page(drm_device_t *dev) { - drm_device_dma_t *dma = dev->dma; - drm_i810_private_t *dev_priv = dev->dev_private; + drm_device_dma_t *dma = dev->dma; + drm_i810_private_t *dev_priv = dev->dev_private; u32 *temp = dev_priv->hw_status_page; - int i; + int i; - DRM_DEBUG( "hw_status: Interrupt Status : %x\n", temp[0]); - DRM_DEBUG( "hw_status: LpRing Head ptr : %x\n", temp[1]); - DRM_DEBUG( "hw_status: IRing Head ptr : %x\n", temp[2]); - DRM_DEBUG( "hw_status: Reserved : %x\n", temp[3]); + DRM_DEBUG( "hw_status: Interrupt Status : %x\n", temp[0]); + DRM_DEBUG( "hw_status: LpRing Head ptr : %x\n", temp[1]); + DRM_DEBUG( "hw_status: IRing Head ptr : %x\n", temp[2]); + DRM_DEBUG( "hw_status: Reserved : %x\n", temp[3]); DRM_DEBUG( "hw_status: Last Render: %x\n", temp[4]); - DRM_DEBUG( "hw_status: Driver Counter : %d\n", temp[5]); - for(i = 6; i < dma->buf_count + 6; i++) { - DRM_DEBUG( "buffer status idx : %d used: %d\n", i - 6, temp[i]); + DRM_DEBUG( "hw_status: Driver Counter : %d\n", temp[5]); + for(i = 6; i < dma->buf_count + 6; i++) { + DRM_DEBUG( "buffer status idx : %d used: %d\n", i - 6, temp[i]); } } static drm_buf_t *i810_freelist_get(drm_device_t *dev) { - drm_device_dma_t *dma = dev->dma; + drm_device_dma_t *dma = dev->dma; int i; - int used; + int used; /* Linear search might not be the best solution */ - for (i = 0; i < dma->buf_count; i++) { - drm_buf_t *buf = dma->buflist[ i ]; - drm_i810_buf_priv_t *buf_priv = buf->dev_private; + for (i = 0; i < dma->buf_count; i++) { + drm_buf_t *buf = dma->buflist[ i ]; + drm_i810_buf_priv_t *buf_priv = buf->dev_private; /* In use is already a pointer */ - used = cmpxchg(buf_priv->in_use, I810_BUF_FREE, + used = cmpxchg(buf_priv->in_use, I810_BUF_FREE, I810_BUF_CLIENT); if (used == I810_BUF_FREE) { return buf; } } - return NULL; + return NULL; } /* This should only be called if the buffer is not sent to the hardware @@ -96,17 +96,17 @@ static drm_buf_t *i810_freelist_get(drm_ static int i810_freelist_put(drm_device_t *dev, drm_buf_t *buf) { - drm_i810_buf_priv_t *buf_priv = buf->dev_private; - int used; + drm_i810_buf_priv_t *buf_priv = buf->dev_private; + int used; - /* In use is already a pointer */ - used = cmpxchg(buf_priv->in_use, I810_BUF_CLIENT, I810_BUF_FREE); + /* In use is already a pointer */ + used = cmpxchg(buf_priv->in_use, I810_BUF_CLIENT, I810_BUF_FREE); if (used != I810_BUF_CLIENT) { - DRM_ERROR("Freeing buffer thats not in use : %d\n", buf->idx); - return -EINVAL; + DRM_ERROR("Freeing buffer thats not in use : %d\n", buf->idx); + return -EINVAL; } - return 0; + return 0; } static struct file_operations i810_buffer_fops = { @@ -135,7 +135,7 @@ int i810_mmap_buffers(struct file *filp, vma->vm_flags |= (VM_IO | VM_DONTCOPY); vma->vm_file = filp; - buf_priv->currently_mapped = I810_BUF_MAPPED; + buf_priv->currently_mapped = I810_BUF_MAPPED; unlock_kernel(); if (remap_page_range(DRM_RPR_ARG(vma) vma->vm_start, @@ -150,8 +150,8 @@ static int i810_map_buffer(drm_buf_t *bu drm_file_t *priv = filp->private_data; drm_device_t *dev = priv->dev; drm_i810_buf_priv_t *buf_priv = buf->dev_private; - drm_i810_private_t *dev_priv = dev->dev_private; - struct file_operations *old_fops; + drm_i810_private_t *dev_priv = dev->dev_private; + struct file_operations *old_fops; int retcode = 0; if (buf_priv->currently_mapped == I810_BUF_MAPPED) @@ -192,8 +192,8 @@ static int i810_unmap_buffer(drm_buf_t * (size_t) buf->total); up_write(¤t->mm->mmap_sem); - buf_priv->currently_mapped = I810_BUF_UNMAPPED; - buf_priv->virtual = 0; + buf_priv->currently_mapped = I810_BUF_UNMAPPED; + buf_priv->virtual = 0; return retcode; } @@ -208,22 +208,22 @@ static int i810_dma_get_buffer(drm_devic buf = i810_freelist_get(dev); if (!buf) { retcode = -ENOMEM; - DRM_DEBUG("retcode=%d\n", retcode); + DRM_DEBUG("retcode=%d\n", retcode); return retcode; } retcode = i810_map_buffer(buf, filp); if (retcode) { i810_freelist_put(dev, buf); - DRM_ERROR("mapbuf failed, retcode %d\n", retcode); + DRM_ERROR("mapbuf failed, retcode %d\n", retcode); return retcode; } buf->filp = filp; buf_priv = buf->dev_private; d->granted = 1; - d->request_idx = buf->idx; - d->request_size = buf->total; - d->virtual = buf_priv->virtual; + d->request_idx = buf->idx; + d->request_size = buf->total; + d->virtual = buf_priv->virtual; return retcode; } @@ -232,33 +232,33 @@ int i810_dma_cleanup(drm_device_t *dev) { drm_device_dma_t *dma = dev->dma; -#if _HAVE_DMA_IRQ +#if __HAVE_IRQ /* Make sure interrupts are disabled here because the uninstall ioctl * may not have been called from userspace and after dev_private * is freed, it's too late. */ - if (dev->irq) DRM(irq_uninstall)(dev); + if (dev->irq_enabled) DRM(irq_uninstall)(dev); #endif if (dev->dev_private) { int i; - drm_i810_private_t *dev_priv = - (drm_i810_private_t *) dev->dev_private; + drm_i810_private_t *dev_priv = + (drm_i810_private_t *) dev->dev_private; if (dev_priv->ring.virtual_start) { - DRM(ioremapfree)((void *) dev_priv->ring.virtual_start, + DRM(ioremapfree)((void *) dev_priv->ring.virtual_start, dev_priv->ring.Size, dev); } - if (dev_priv->hw_status_page) { - pci_free_consistent(dev->pdev, PAGE_SIZE, + if (dev_priv->hw_status_page) { + pci_free_consistent(dev->pdev, PAGE_SIZE, dev_priv->hw_status_page, dev_priv->dma_status_page); - /* Need to rewrite hardware status page */ - I810_WRITE(0x02080, 0x1ffff000); + /* Need to rewrite hardware status page */ + I810_WRITE(0x02080, 0x1ffff000); } - DRM(free)(dev->dev_private, sizeof(drm_i810_private_t), + DRM(free)(dev->dev_private, sizeof(drm_i810_private_t), DRM_MEM_DRIVER); - dev->dev_private = NULL; + dev->dev_private = NULL; for (i = 0; i < dma->buf_count; i++) { drm_buf_t *buf = dma->buflist[ i ]; @@ -267,73 +267,73 @@ int i810_dma_cleanup(drm_device_t *dev) DRM(ioremapfree)(buf_priv->kernel_virtual, buf->total, dev); } } - return 0; + return 0; } static int i810_wait_ring(drm_device_t *dev, int n) { - drm_i810_private_t *dev_priv = dev->dev_private; - drm_i810_ring_buffer_t *ring = &(dev_priv->ring); - int iters = 0; - unsigned long end; + drm_i810_private_t *dev_priv = dev->dev_private; + drm_i810_ring_buffer_t *ring = &(dev_priv->ring); + int iters = 0; + unsigned long end; unsigned int last_head = I810_READ(LP_RING + RING_HEAD) & HEAD_ADDR; end = jiffies + (HZ*3); - while (ring->space < n) { - ring->head = I810_READ(LP_RING + RING_HEAD) & HEAD_ADDR; - ring->space = ring->head - (ring->tail+8); + while (ring->space < n) { + ring->head = I810_READ(LP_RING + RING_HEAD) & HEAD_ADDR; + ring->space = ring->head - (ring->tail+8); if (ring->space < 0) ring->space += ring->Size; - + if (ring->head != last_head) { end = jiffies + (HZ*3); last_head = ring->head; } - iters++; + iters++; if (time_before(end, jiffies)) { - DRM_ERROR("space: %d wanted %d\n", ring->space, n); - DRM_ERROR("lockup\n"); - goto out_wait_ring; + DRM_ERROR("space: %d wanted %d\n", ring->space, n); + DRM_ERROR("lockup\n"); + goto out_wait_ring; } udelay(1); } out_wait_ring: - return iters; + return iters; } static void i810_kernel_lost_context(drm_device_t *dev) { - drm_i810_private_t *dev_priv = dev->dev_private; - drm_i810_ring_buffer_t *ring = &(dev_priv->ring); + drm_i810_private_t *dev_priv = dev->dev_private; + drm_i810_ring_buffer_t *ring = &(dev_priv->ring); - ring->head = I810_READ(LP_RING + RING_HEAD) & HEAD_ADDR; - ring->tail = I810_READ(LP_RING + RING_TAIL); - ring->space = ring->head - (ring->tail+8); - if (ring->space < 0) ring->space += ring->Size; + ring->head = I810_READ(LP_RING + RING_HEAD) & HEAD_ADDR; + ring->tail = I810_READ(LP_RING + RING_TAIL); + ring->space = ring->head - (ring->tail+8); + if (ring->space < 0) ring->space += ring->Size; } static int i810_freelist_init(drm_device_t *dev, drm_i810_private_t *dev_priv) { - drm_device_dma_t *dma = dev->dma; - int my_idx = 24; - u32 *hw_status = (u32 *)(dev_priv->hw_status_page + my_idx); - int i; + drm_device_dma_t *dma = dev->dma; + int my_idx = 24; + u32 *hw_status = (u32 *)(dev_priv->hw_status_page + my_idx); + int i; if (dma->buf_count > 1019) { - /* Not enough space in the status page for the freelist */ - return -EINVAL; + /* Not enough space in the status page for the freelist */ + return -EINVAL; } - for (i = 0; i < dma->buf_count; i++) { - drm_buf_t *buf = dma->buflist[ i ]; - drm_i810_buf_priv_t *buf_priv = buf->dev_private; + for (i = 0; i < dma->buf_count; i++) { + drm_buf_t *buf = dma->buflist[ i ]; + drm_i810_buf_priv_t *buf_priv = buf->dev_private; - buf_priv->in_use = hw_status++; - buf_priv->my_use_idx = my_idx; - my_idx += 4; + buf_priv->in_use = hw_status++; + buf_priv->my_use_idx = my_idx; + my_idx += 4; - *buf_priv->in_use = I810_BUF_FREE; + *buf_priv->in_use = I810_BUF_FREE; buf_priv->kernel_virtual = DRM(ioremap)(buf->bus_address, buf->total, dev); @@ -347,7 +347,7 @@ static int i810_dma_initialize(drm_devic { struct list_head *list; - memset(dev_priv, 0, sizeof(drm_i810_private_t)); + memset(dev_priv, 0, sizeof(drm_i810_private_t)); list_for_each(list, &dev->maplist->head) { drm_map_list_t *r_list = list_entry(list, drm_map_list_t, head); @@ -355,51 +355,51 @@ static int i810_dma_initialize(drm_devic r_list->map->type == _DRM_SHM && r_list->map->flags & _DRM_CONTAINS_LOCK ) { dev_priv->sarea_map = r_list->map; - break; - } - } + break; + } + } if (!dev_priv->sarea_map) { dev->dev_private = (void *)dev_priv; - i810_dma_cleanup(dev); - DRM_ERROR("can not find sarea!\n"); - return -EINVAL; + i810_dma_cleanup(dev); + DRM_ERROR("can not find sarea!\n"); + return -EINVAL; } DRM_FIND_MAP( dev_priv->mmio_map, init->mmio_offset ); if (!dev_priv->mmio_map) { dev->dev_private = (void *)dev_priv; - i810_dma_cleanup(dev); - DRM_ERROR("can not find mmio map!\n"); - return -EINVAL; + i810_dma_cleanup(dev); + DRM_ERROR("can not find mmio map!\n"); + return -EINVAL; } DRM_FIND_MAP( dev_priv->buffer_map, init->buffers_offset ); if (!dev_priv->buffer_map) { dev->dev_private = (void *)dev_priv; - i810_dma_cleanup(dev); - DRM_ERROR("can not find dma buffer map!\n"); - return -EINVAL; + i810_dma_cleanup(dev); + DRM_ERROR("can not find dma buffer map!\n"); + return -EINVAL; } dev_priv->sarea_priv = (drm_i810_sarea_t *) ((u8 *)dev_priv->sarea_map->handle + init->sarea_priv_offset); - dev_priv->ring.Start = init->ring_start; - dev_priv->ring.End = init->ring_end; - dev_priv->ring.Size = init->ring_size; + dev_priv->ring.Start = init->ring_start; + dev_priv->ring.End = init->ring_end; + dev_priv->ring.Size = init->ring_size; - dev_priv->ring.virtual_start = DRM(ioremap)(dev->agp->base + + dev_priv->ring.virtual_start = DRM(ioremap)(dev->agp->base + init->ring_start, init->ring_size, dev); - if (dev_priv->ring.virtual_start == NULL) { + if (dev_priv->ring.virtual_start == NULL) { dev->dev_private = (void *) dev_priv; - i810_dma_cleanup(dev); - DRM_ERROR("can not ioremap virtual address for" + i810_dma_cleanup(dev); + DRM_ERROR("can not ioremap virtual address for" " ring buffer\n"); - return -ENOMEM; + return -ENOMEM; } - dev_priv->ring.tail_mask = dev_priv->ring.Size - 1; + dev_priv->ring.tail_mask = dev_priv->ring.Size - 1; dev_priv->w = init->w; dev_priv->h = init->h; @@ -415,33 +415,33 @@ static int i810_dma_initialize(drm_devic dev_priv->back_di1 = init->back_offset | init->pitch_bits; dev_priv->zi1 = init->depth_offset | init->pitch_bits; - /* Program Hardware Status Page */ - dev_priv->hw_status_page = + /* Program Hardware Status Page */ + dev_priv->hw_status_page = pci_alloc_consistent(dev->pdev, PAGE_SIZE, &dev_priv->dma_status_page); - if (!dev_priv->hw_status_page) { + if (!dev_priv->hw_status_page) { dev->dev_private = (void *)dev_priv; i810_dma_cleanup(dev); DRM_ERROR("Can not allocate hardware status page\n"); return -ENOMEM; } - memset(dev_priv->hw_status_page, 0, PAGE_SIZE); - DRM_DEBUG("hw status page @ %p\n", dev_priv->hw_status_page); + memset(dev_priv->hw_status_page, 0, PAGE_SIZE); + DRM_DEBUG("hw status page @ %p\n", dev_priv->hw_status_page); I810_WRITE(0x02080, dev_priv->dma_status_page); - DRM_DEBUG("Enabled hardware status page\n"); + DRM_DEBUG("Enabled hardware status page\n"); - /* Now we need to init our freelist */ + /* Now we need to init our freelist */ if (i810_freelist_init(dev, dev_priv) != 0) { dev->dev_private = (void *)dev_priv; - i810_dma_cleanup(dev); - DRM_ERROR("Not enough space in the status page for" + i810_dma_cleanup(dev); + DRM_ERROR("Not enough space in the status page for" " the freelist\n"); - return -ENOMEM; + return -ENOMEM; } dev->dev_private = (void *)dev_priv; - return 0; + return 0; } /* i810 DRM version 1.1 used a smaller init structure with different @@ -476,12 +476,12 @@ int i810_dma_init_compat(drm_i810_init_t /* This is a v1.1 client, fix the params */ DRM_INFO("Using PRE v1.2 init.\n"); - init->pitch_bits = init->h; - init->pitch = init->w; - init->h = init->overlay_physical; - init->w = init->overlay_offset; - init->overlay_physical = 0; - init->overlay_offset = 0; + init->pitch_bits = init->h; + init->pitch = init->w; + init->h = init->overlay_physical; + init->w = init->overlay_offset; + init->overlay_physical = 0; + init->overlay_offset = 0; } return 0; @@ -490,55 +490,55 @@ int i810_dma_init_compat(drm_i810_init_t int i810_dma_init(struct inode *inode, struct file *filp, unsigned int cmd, unsigned long arg) { - drm_file_t *priv = filp->private_data; - drm_device_t *dev = priv->dev; - drm_i810_private_t *dev_priv; - drm_i810_init_t init; - int retcode = 0; + drm_file_t *priv = filp->private_data; + drm_device_t *dev = priv->dev; + drm_i810_private_t *dev_priv; + drm_i810_init_t init; + int retcode = 0; /* Get only the init func */ if (copy_from_user(&init, (void *)arg, sizeof(drm_i810_init_func_t))) return -EFAULT; - switch(init.func) { - case I810_INIT_DMA: - /* This case is for backward compatibility. It + switch(init.func) { + case I810_INIT_DMA: + /* This case is for backward compatibility. It * handles XFree 4.1.0 and 4.2.0, and has to * do some parameter checking as described below. * It will someday go away. */ retcode = i810_dma_init_compat(&init, arg); - if (retcode) + if (retcode) return retcode; - dev_priv = DRM(alloc)(sizeof(drm_i810_private_t), + dev_priv = DRM(alloc)(sizeof(drm_i810_private_t), DRM_MEM_DRIVER); - if (dev_priv == NULL) - return -ENOMEM; - retcode = i810_dma_initialize(dev, dev_priv, &init); + if (dev_priv == NULL) + return -ENOMEM; + retcode = i810_dma_initialize(dev, dev_priv, &init); break; default: - case I810_INIT_DMA_1_4: + case I810_INIT_DMA_1_4: DRM_INFO("Using v1.4 init.\n"); - if (copy_from_user(&init, (drm_i810_init_t *)arg, + if (copy_from_user(&init, (drm_i810_init_t *)arg, sizeof(drm_i810_init_t))) { return -EFAULT; } - dev_priv = DRM(alloc)(sizeof(drm_i810_private_t), + dev_priv = DRM(alloc)(sizeof(drm_i810_private_t), DRM_MEM_DRIVER); if (dev_priv == NULL) return -ENOMEM; - retcode = i810_dma_initialize(dev, dev_priv, &init); + retcode = i810_dma_initialize(dev, dev_priv, &init); break; - case I810_CLEANUP_DMA: - DRM_INFO("DMA Cleanup\n"); - retcode = i810_dma_cleanup(dev); - break; + case I810_CLEANUP_DMA: + DRM_INFO("DMA Cleanup\n"); + retcode = i810_dma_cleanup(dev); + break; } - return retcode; + return retcode; } @@ -552,7 +552,7 @@ int i810_dma_init(struct inode *inode, s static void i810EmitContextVerified( drm_device_t *dev, volatile unsigned int *code ) { - drm_i810_private_t *dev_priv = dev->dev_private; + drm_i810_private_t *dev_priv = dev->dev_private; int i, j = 0; unsigned int tmp; RING_LOCALS; @@ -586,7 +586,7 @@ static void i810EmitContextVerified( drm static void i810EmitTexVerified( drm_device_t *dev, volatile unsigned int *code ) { - drm_i810_private_t *dev_priv = dev->dev_private; + drm_i810_private_t *dev_priv = dev->dev_private; int i, j = 0; unsigned int tmp; RING_LOCALS; @@ -622,7 +622,7 @@ static void i810EmitTexVerified( drm_dev static void i810EmitDestVerified( drm_device_t *dev, volatile unsigned int *code ) { - drm_i810_private_t *dev_priv = dev->dev_private; + drm_i810_private_t *dev_priv = dev->dev_private; unsigned int tmp; RING_LOCALS; @@ -658,8 +658,8 @@ static void i810EmitDestVerified( drm_de static void i810EmitState( drm_device_t *dev ) { - drm_i810_private_t *dev_priv = dev->dev_private; - drm_i810_sarea_t *sarea_priv = dev_priv->sarea_priv; + drm_i810_private_t *dev_priv = dev->dev_private; + drm_i810_sarea_t *sarea_priv = dev_priv->sarea_priv; unsigned int dirty = sarea_priv->dirty; DRM_DEBUG("%s %x\n", __FUNCTION__, dirty); @@ -693,8 +693,8 @@ static void i810_dma_dispatch_clear( drm unsigned int clear_color, unsigned int clear_zval ) { - drm_i810_private_t *dev_priv = dev->dev_private; - drm_i810_sarea_t *sarea_priv = dev_priv->sarea_priv; + drm_i810_private_t *dev_priv = dev->dev_private; + drm_i810_sarea_t *sarea_priv = dev_priv->sarea_priv; int nbox = sarea_priv->nbox; drm_clip_rect_t *pbox = sarea_priv->boxes; int pitch = dev_priv->pitch; @@ -703,17 +703,17 @@ static void i810_dma_dispatch_clear( drm RING_LOCALS; if ( dev_priv->current_page == 1 ) { - unsigned int tmp = flags; - + unsigned int tmp = flags; + flags &= ~(I810_FRONT | I810_BACK); if (tmp & I810_FRONT) flags |= I810_BACK; if (tmp & I810_BACK) flags |= I810_FRONT; } - i810_kernel_lost_context(dev); + i810_kernel_lost_context(dev); - if (nbox > I810_NR_SAREA_CLIPRECTS) - nbox = I810_NR_SAREA_CLIPRECTS; + if (nbox > I810_NR_SAREA_CLIPRECTS) + nbox = I810_NR_SAREA_CLIPRECTS; for (i = 0 ; i < nbox ; i++, pbox++) { unsigned int x = pbox->x1; @@ -728,7 +728,7 @@ static void i810_dma_dispatch_clear( drm pbox->y2 > dev_priv->h) continue; - if ( flags & I810_FRONT ) { + if ( flags & I810_FRONT ) { BEGIN_LP_RING( 6 ); OUT_RING( BR00_BITBLT_CLIENT | BR00_OP_COLOR_BLT | 0x3 ); @@ -768,8 +768,8 @@ static void i810_dma_dispatch_clear( drm static void i810_dma_dispatch_swap( drm_device_t *dev ) { - drm_i810_private_t *dev_priv = dev->dev_private; - drm_i810_sarea_t *sarea_priv = dev_priv->sarea_priv; + drm_i810_private_t *dev_priv = dev->dev_private; + drm_i810_sarea_t *sarea_priv = dev_priv->sarea_priv; int nbox = sarea_priv->nbox; drm_clip_rect_t *pbox = sarea_priv->boxes; int pitch = dev_priv->pitch; @@ -779,10 +779,10 @@ static void i810_dma_dispatch_swap( drm_ DRM_DEBUG("swapbuffers\n"); - i810_kernel_lost_context(dev); + i810_kernel_lost_context(dev); - if (nbox > I810_NR_SAREA_CLIPRECTS) - nbox = I810_NR_SAREA_CLIPRECTS; + if (nbox > I810_NR_SAREA_CLIPRECTS) + nbox = I810_NR_SAREA_CLIPRECTS; for (i = 0 ; i < nbox; i++, pbox++) { @@ -820,19 +820,19 @@ static void i810_dma_dispatch_vertex(drm int discard, int used) { - drm_i810_private_t *dev_priv = dev->dev_private; + drm_i810_private_t *dev_priv = dev->dev_private; drm_i810_buf_priv_t *buf_priv = buf->dev_private; - drm_i810_sarea_t *sarea_priv = dev_priv->sarea_priv; - drm_clip_rect_t *box = sarea_priv->boxes; - int nbox = sarea_priv->nbox; + drm_i810_sarea_t *sarea_priv = dev_priv->sarea_priv; + drm_clip_rect_t *box = sarea_priv->boxes; + int nbox = sarea_priv->nbox; unsigned long address = (unsigned long)buf->bus_address; unsigned long start = address - dev->agp->base; int i = 0; - RING_LOCALS; + RING_LOCALS; - i810_kernel_lost_context(dev); + i810_kernel_lost_context(dev); - if (nbox > I810_NR_SAREA_CLIPRECTS) + if (nbox > I810_NR_SAREA_CLIPRECTS) nbox = I810_NR_SAREA_CLIPRECTS; if (used > 4*1024) @@ -898,7 +898,7 @@ static void i810_dma_dispatch_vertex(drm static void i810_dma_dispatch_flip( drm_device_t *dev ) { - drm_i810_private_t *dev_priv = dev->dev_private; + drm_i810_private_t *dev_priv = dev->dev_private; int pitch = dev_priv->pitch; RING_LOCALS; @@ -907,10 +907,10 @@ static void i810_dma_dispatch_flip( drm_ dev_priv->current_page, dev_priv->sarea_priv->pf_current_page); - i810_kernel_lost_context(dev); + i810_kernel_lost_context(dev); BEGIN_LP_RING( 2 ); - OUT_RING( INST_PARSER_CLIENT | INST_OP_FLUSH | INST_FLUSH_MAP_CACHE ); + OUT_RING( INST_PARSER_CLIENT | INST_OP_FLUSH | INST_FLUSH_MAP_CACHE ); OUT_RING( 0 ); ADVANCE_LP_RING(); @@ -945,44 +945,44 @@ static void i810_dma_dispatch_flip( drm_ void i810_dma_quiescent(drm_device_t *dev) { - drm_i810_private_t *dev_priv = dev->dev_private; - RING_LOCALS; + drm_i810_private_t *dev_priv = dev->dev_private; + RING_LOCALS; -/* printk("%s\n", __FUNCTION__); */ +/* printk("%s\n", __FUNCTION__); */ - i810_kernel_lost_context(dev); + i810_kernel_lost_context(dev); - BEGIN_LP_RING(4); - OUT_RING( INST_PARSER_CLIENT | INST_OP_FLUSH | INST_FLUSH_MAP_CACHE ); - OUT_RING( CMD_REPORT_HEAD ); - OUT_RING( 0 ); - OUT_RING( 0 ); - ADVANCE_LP_RING(); + BEGIN_LP_RING(4); + OUT_RING( INST_PARSER_CLIENT | INST_OP_FLUSH | INST_FLUSH_MAP_CACHE ); + OUT_RING( CMD_REPORT_HEAD ); + OUT_RING( 0 ); + OUT_RING( 0 ); + ADVANCE_LP_RING(); i810_wait_ring( dev, dev_priv->ring.Size - 8 ); } static int i810_flush_queue(drm_device_t *dev) { - drm_i810_private_t *dev_priv = dev->dev_private; + drm_i810_private_t *dev_priv = dev->dev_private; drm_device_dma_t *dma = dev->dma; - int i, ret = 0; - RING_LOCALS; + int i, ret = 0; + RING_LOCALS; -/* printk("%s\n", __FUNCTION__); */ +/* printk("%s\n", __FUNCTION__); */ - i810_kernel_lost_context(dev); + i810_kernel_lost_context(dev); - BEGIN_LP_RING(2); - OUT_RING( CMD_REPORT_HEAD ); - OUT_RING( 0 ); - ADVANCE_LP_RING(); + BEGIN_LP_RING(2); + OUT_RING( CMD_REPORT_HEAD ); + OUT_RING( 0 ); + ADVANCE_LP_RING(); i810_wait_ring( dev, dev_priv->ring.Size - 8 ); - for (i = 0; i < dma->buf_count; i++) { - drm_buf_t *buf = dma->buflist[ i ]; - drm_i810_buf_priv_t *buf_priv = buf->dev_private; + for (i = 0; i < dma->buf_count; i++) { + drm_buf_t *buf = dma->buflist[ i ]; + drm_i810_buf_priv_t *buf_priv = buf->dev_private; int used = cmpxchg(buf_priv->in_use, I810_BUF_HARDWARE, I810_BUF_FREE); @@ -993,7 +993,7 @@ static int i810_flush_queue(drm_device_t DRM_DEBUG("still on client\n"); } - return ret; + return ret; } /* Must be called with the lock held */ @@ -1005,14 +1005,14 @@ void i810_reclaim_buffers(struct file *f int i; if (!dma) return; - if (!dev->dev_private) return; + if (!dev->dev_private) return; if (!dma->buflist) return; - i810_flush_queue(dev); + i810_flush_queue(dev); for (i = 0; i < dma->buf_count; i++) { - drm_buf_t *buf = dma->buflist[ i ]; - drm_i810_buf_priv_t *buf_priv = buf->dev_private; + drm_buf_t *buf = dma->buflist[ i ]; + drm_i810_buf_priv_t *buf_priv = buf->dev_private; if (buf->filp == filp && buf_priv) { int used = cmpxchg(buf_priv->in_use, I810_BUF_CLIENT, @@ -1021,7 +1021,7 @@ void i810_reclaim_buffers(struct file *f if (used == I810_BUF_CLIENT) DRM_DEBUG("reclaimed from client\n"); if (buf_priv->currently_mapped == I810_BUF_MAPPED) - buf_priv->currently_mapped = I810_BUF_UNMAPPED; + buf_priv->currently_mapped = I810_BUF_UNMAPPED; } } } @@ -1029,16 +1029,16 @@ void i810_reclaim_buffers(struct file *f int i810_flush_ioctl(struct inode *inode, struct file *filp, unsigned int cmd, unsigned long arg) { - drm_file_t *priv = filp->private_data; - drm_device_t *dev = priv->dev; + drm_file_t *priv = filp->private_data; + drm_device_t *dev = priv->dev; if (!_DRM_LOCK_IS_HELD(dev->lock.hw_lock->lock)) { DRM_ERROR("i810_flush_ioctl called without lock held\n"); return -EINVAL; } - i810_flush_queue(dev); - return 0; + i810_flush_queue(dev); + return 0; } @@ -1048,10 +1048,10 @@ int i810_dma_vertex(struct inode *inode, drm_file_t *priv = filp->private_data; drm_device_t *dev = priv->dev; drm_device_dma_t *dma = dev->dma; - drm_i810_private_t *dev_priv = (drm_i810_private_t *)dev->dev_private; - u32 *hw_status = dev_priv->hw_status_page; - drm_i810_sarea_t *sarea_priv = (drm_i810_sarea_t *) - dev_priv->sarea_priv; + drm_i810_private_t *dev_priv = (drm_i810_private_t *)dev->dev_private; + u32 *hw_status = dev_priv->hw_status_page; + drm_i810_sarea_t *sarea_priv = (drm_i810_sarea_t *) + dev_priv->sarea_priv; drm_i810_vertex_t vertex; if (copy_from_user(&vertex, (drm_i810_vertex_t *)arg, sizeof(vertex))) @@ -1072,10 +1072,10 @@ int i810_dma_vertex(struct inode *inode, dma->buflist[ vertex.idx ], vertex.discard, vertex.used ); - atomic_add(vertex.used, &dev->counts[_DRM_STAT_SECONDARY]); + atomic_add(vertex.used, &dev->counts[_DRM_STAT_SECONDARY]); atomic_inc(&dev->counts[_DRM_STAT_DMA]); sarea_priv->last_enqueue = dev_priv->counter-1; - sarea_priv->last_dispatch = (int) hw_status[5]; + sarea_priv->last_dispatch = (int) hw_status[5]; return 0; } @@ -1089,7 +1089,7 @@ int i810_clear_bufs(struct inode *inode, drm_device_t *dev = priv->dev; drm_i810_clear_t clear; - if (copy_from_user(&clear, (drm_i810_clear_t *)arg, sizeof(clear))) + if (copy_from_user(&clear, (drm_i810_clear_t *)arg, sizeof(clear))) return -EFAULT; if (!_DRM_LOCK_IS_HELD(dev->lock.hw_lock->lock)) { @@ -1097,15 +1097,15 @@ int i810_clear_bufs(struct inode *inode, return -EINVAL; } - /* GH: Someone's doing nasty things... */ - if (!dev->dev_private) { - return -EINVAL; - } + /* GH: Someone's doing nasty things... */ + if (!dev->dev_private) { + return -EINVAL; + } i810_dma_dispatch_clear( dev, clear.flags, clear.clear_color, clear.clear_depth ); - return 0; + return 0; } int i810_swap_bufs(struct inode *inode, struct file *filp, @@ -1122,20 +1122,20 @@ int i810_swap_bufs(struct inode *inode, } i810_dma_dispatch_swap( dev ); - return 0; + return 0; } int i810_getage(struct inode *inode, struct file *filp, unsigned int cmd, unsigned long arg) { - drm_file_t *priv = filp->private_data; + drm_file_t *priv = filp->private_data; drm_device_t *dev = priv->dev; - drm_i810_private_t *dev_priv = (drm_i810_private_t *)dev->dev_private; - u32 *hw_status = dev_priv->hw_status_page; - drm_i810_sarea_t *sarea_priv = (drm_i810_sarea_t *) - dev_priv->sarea_priv; + drm_i810_private_t *dev_priv = (drm_i810_private_t *)dev->dev_private; + u32 *hw_status = dev_priv->hw_status_page; + drm_i810_sarea_t *sarea_priv = (drm_i810_sarea_t *) + dev_priv->sarea_priv; - sarea_priv->last_dispatch = (int) hw_status[5]; + sarea_priv->last_dispatch = (int) hw_status[5]; return 0; } @@ -1146,12 +1146,12 @@ int i810_getbuf(struct inode *inode, str drm_device_t *dev = priv->dev; int retcode = 0; drm_i810_dma_t d; - drm_i810_private_t *dev_priv = (drm_i810_private_t *)dev->dev_private; - u32 *hw_status = dev_priv->hw_status_page; - drm_i810_sarea_t *sarea_priv = (drm_i810_sarea_t *) - dev_priv->sarea_priv; + drm_i810_private_t *dev_priv = (drm_i810_private_t *)dev->dev_private; + u32 *hw_status = dev_priv->hw_status_page; + drm_i810_sarea_t *sarea_priv = (drm_i810_sarea_t *) + dev_priv->sarea_priv; - if (copy_from_user(&d, (drm_i810_dma_t *)arg, sizeof(d))) + if (copy_from_user(&d, (drm_i810_dma_t *)arg, sizeof(d))) return -EFAULT; if (!_DRM_LOCK_IS_HELD(dev->lock.hw_lock->lock)) { @@ -1168,7 +1168,7 @@ int i810_getbuf(struct inode *inode, str if (copy_to_user((drm_dma_t *)arg, &d, sizeof(d))) return -EFAULT; - sarea_priv->last_dispatch = (int) hw_status[5]; + sarea_priv->last_dispatch = (int) hw_status[5]; return retcode; } @@ -1384,5 +1384,5 @@ int i810_flip_bufs(struct inode *inode, i810_do_init_pageflip( dev ); i810_dma_dispatch_flip( dev ); - return 0; + return 0; } --- diff/drivers/char/drm/i830.h 2003-05-21 11:50:14.000000000 +0100 +++ source/drivers/char/drm/i830.h 2004-03-01 15:42:16.000000000 +0000 @@ -77,6 +77,13 @@ [DRM_IOCTL_NR(DRM_IOCTL_I830_GETPARAM)] = { i830_getparam, 1, 0 }, \ [DRM_IOCTL_NR(DRM_IOCTL_I830_SETPARAM)] = { i830_setparam, 1, 0 } +#define DRIVER_PCI_IDS \ + {0x8086, 0x3577, 0, "Intel i830M GMCH"}, \ + {0x8086, 0x2562, 0, "Intel i845G GMCH"}, \ + {0x8086, 0x3582, 0, "Intel i852GM/i855GM GMCH"}, \ + {0x8086, 0x2572, 0, "Intel i865G GMCH"}, \ + {0, 0, 0, NULL} + #define __HAVE_COUNTERS 4 #define __HAVE_COUNTER6 _DRM_STAT_IRQ #define __HAVE_COUNTER7 _DRM_STAT_PRIMARY @@ -115,10 +122,10 @@ #define USE_IRQS 0 #if USE_IRQS -#define __HAVE_DMA_IRQ 1 +#define __HAVE_IRQ 1 #define __HAVE_SHARED_IRQ 1 #else -#define __HAVE_DMA_IRQ 0 +#define __HAVE_IRQ 0 #endif --- diff/drivers/char/drm/i830_dma.c 2003-06-09 14:18:18.000000000 +0100 +++ source/drivers/char/drm/i830_dma.c 2004-03-01 15:42:16.000000000 +0000 @@ -231,12 +231,12 @@ int i830_dma_cleanup(drm_device_t *dev) { drm_device_dma_t *dma = dev->dma; -#if _HAVE_DMA_IRQ +#if __HAVE_IRQ /* Make sure interrupts are disabled here because the uninstall ioctl * may not have been called from userspace and after dev_private * is freed, it's too late. */ - if (dev->irq) DRM(irq_uninstall)(dev); + if (dev->irq_enabled) DRM(irq_uninstall)(dev); #endif if (dev->dev_private) { @@ -1539,7 +1539,7 @@ int i830_getparam( struct inode *inode, switch( param.param ) { case I830_PARAM_IRQ_ACTIVE: - value = dev->irq ? 1 : 0; + value = dev->irq_enabled; break; default: return -EINVAL; --- diff/drivers/char/drm/i830_drv.c 2003-05-21 11:50:14.000000000 +0100 +++ source/drivers/char/drm/i830_drv.c 2004-03-01 15:42:16.000000000 +0000 @@ -50,6 +50,7 @@ #include "drm_fops.h" #include "drm_init.h" #include "drm_ioctl.h" +#include "drm_irq.h" #include "drm_lock.h" #include "drm_memory.h" #include "drm_proc.h" --- diff/drivers/char/drm/i830_irq.c 2003-09-30 15:46:12.000000000 +0100 +++ source/drivers/char/drm/i830_irq.c 2004-03-01 15:42:16.000000000 +0000 @@ -35,7 +35,7 @@ #include -irqreturn_t DRM(dma_service)( DRM_IRQ_ARGS ) +irqreturn_t DRM(irq_handler)( DRM_IRQ_ARGS ) { drm_device_t *dev = (drm_device_t *)arg; drm_i830_private_t *dev_priv = (drm_i830_private_t *)dev->dev_private; --- diff/drivers/char/drm/mga.h 2003-05-21 11:50:14.000000000 +0100 +++ source/drivers/char/drm/mga.h 2004-03-01 15:42:16.000000000 +0000 @@ -64,6 +64,12 @@ [DRM_IOCTL_NR(DRM_IOCTL_MGA_BLIT)] = { mga_dma_blit, 1, 0 }, \ [DRM_IOCTL_NR(DRM_IOCTL_MGA_GETPARAM)]= { mga_getparam, 1, 0 }, +#define DRIVER_PCI_IDS \ + {0x102b, 0x0521, 0, "Matrox G200 (AGP)"}, \ + {0x102b, 0x0525, 0, "Matrox G400/G450 (AGP)"}, \ + {0x102b, 0x2527, 0, "Matrox G550 (AGP)"}, \ + {0, 0, 0, NULL} + #define __HAVE_COUNTERS 3 #define __HAVE_COUNTER6 _DRM_STAT_IRQ #define __HAVE_COUNTER7 _DRM_STAT_PRIMARY @@ -78,7 +84,7 @@ /* DMA customization: */ #define __HAVE_DMA 1 -#define __HAVE_DMA_IRQ 1 +#define __HAVE_IRQ 1 #define __HAVE_VBL_IRQ 1 #define __HAVE_SHARED_IRQ 1 --- diff/drivers/char/drm/mga_dma.c 2003-06-09 14:18:18.000000000 +0100 +++ source/drivers/char/drm/mga_dma.c 2004-03-01 15:42:16.000000000 +0000 @@ -500,14 +500,6 @@ static int mga_do_init_dma( drm_device_t return DRM_ERR(EINVAL); } - DRM_FIND_MAP( dev_priv->fb, init->fb_offset ); - if(!dev_priv->fb) { - DRM_ERROR( "failed to find framebuffer!\n" ); - /* Assign dev_private so we can do cleanup. */ - dev->dev_private = (void *)dev_priv; - mga_do_cleanup_dma( dev ); - return DRM_ERR(EINVAL); - } DRM_FIND_MAP( dev_priv->mmio, init->mmio_offset ); if(!dev_priv->mmio) { DRM_ERROR( "failed to find mmio region!\n" ); @@ -639,12 +631,12 @@ int mga_do_cleanup_dma( drm_device_t *de { DRM_DEBUG( "\n" ); -#if _HAVE_DMA_IRQ +#if __HAVE_IRQ /* Make sure interrupts are disabled here because the uninstall ioctl * may not have been called from userspace and after dev_private * is freed, it's too late. */ - if ( dev->irq ) DRM(irq_uninstall)(dev); + if ( dev->irq_enabled ) DRM(irq_uninstall)(dev); #endif if ( dev->dev_private ) { --- diff/drivers/char/drm/mga_drv.c 2002-10-16 04:27:56.000000000 +0100 +++ source/drivers/char/drm/mga_drv.c 2004-03-01 15:42:16.000000000 +0000 @@ -45,6 +45,7 @@ #include "drm_fops.h" #include "drm_init.h" #include "drm_ioctl.h" +#include "drm_irq.h" #include "drm_lock.h" #include "drm_memory.h" #include "drm_proc.h" --- diff/drivers/char/drm/mga_drv.h 2003-06-09 14:18:18.000000000 +0100 +++ source/drivers/char/drm/mga_drv.h 2004-03-01 15:42:16.000000000 +0000 @@ -91,7 +91,6 @@ typedef struct drm_mga_private { unsigned int texture_size; drm_local_map_t *sarea; - drm_local_map_t *fb; drm_local_map_t *mmio; drm_local_map_t *status; drm_local_map_t *warp; --- diff/drivers/char/drm/mga_irq.c 2003-05-21 11:50:14.000000000 +0100 +++ source/drivers/char/drm/mga_irq.c 2004-03-01 15:42:16.000000000 +0000 @@ -36,7 +36,7 @@ #include "mga_drm.h" #include "mga_drv.h" -irqreturn_t mga_dma_service( DRM_IRQ_ARGS ) +irqreturn_t mga_irq_handler( DRM_IRQ_ARGS ) { drm_device_t *dev = (drm_device_t *) arg; drm_mga_private_t *dev_priv = --- diff/drivers/char/drm/r128.h 2003-08-26 10:00:52.000000000 +0100 +++ source/drivers/char/drm/r128.h 2004-03-01 15:42:16.000000000 +0000 @@ -79,6 +79,46 @@ [DRM_IOCTL_NR(DRM_IOCTL_R128_INDIRECT)] = { r128_cce_indirect, 1, 1 }, \ [DRM_IOCTL_NR(DRM_IOCTL_R128_GETPARAM)] = { r128_getparam, 1, 0 }, +#define DRIVER_PCI_IDS \ + {0x1002, 0x4c45, 0, "ATI Rage 128 Mobility LE (PCI)"}, \ + {0x1002, 0x4c46, 0, "ATI Rage 128 Mobility LF (AGP)"}, \ + {0x1002, 0x4d46, 0, "ATI Rage 128 Mobility MF (AGP)"}, \ + {0x1002, 0x4d4c, 0, "ATI Rage 128 Mobility ML (AGP)"}, \ + {0x1002, 0x5041, 0, "ATI Rage 128 Pro PA (PCI)"}, \ + {0x1002, 0x5042, 0, "ATI Rage 128 Pro PB (AGP)"}, \ + {0x1002, 0x5043, 0, "ATI Rage 128 Pro PC (AGP)"}, \ + {0x1002, 0x5044, 0, "ATI Rage 128 Pro PD (PCI)"}, \ + {0x1002, 0x5045, 0, "ATI Rage 128 Pro PE (AGP)"}, \ + {0x1002, 0x5046, 0, "ATI Rage 128 Pro PF (AGP)"}, \ + {0x1002, 0x5047, 0, "ATI Rage 128 Pro PG (PCI)"}, \ + {0x1002, 0x5048, 0, "ATI Rage 128 Pro PH (AGP)"}, \ + {0x1002, 0x5049, 0, "ATI Rage 128 Pro PI (AGP)"}, \ + {0x1002, 0x504A, 0, "ATI Rage 128 Pro PJ (PCI)"}, \ + {0x1002, 0x504B, 0, "ATI Rage 128 Pro PK (AGP)"}, \ + {0x1002, 0x504C, 0, "ATI Rage 128 Pro PL (AGP)"}, \ + {0x1002, 0x504D, 0, "ATI Rage 128 Pro PM (PCI)"}, \ + {0x1002, 0x504E, 0, "ATI Rage 128 Pro PN (AGP)"}, \ + {0x1002, 0x504F, 0, "ATI Rage 128 Pro PO (AGP)"}, \ + {0x1002, 0x5050, 0, "ATI Rage 128 Pro PP (PCI)"}, \ + {0x1002, 0x5051, 0, "ATI Rage 128 Pro PQ (AGP)"}, \ + {0x1002, 0x5052, 0, "ATI Rage 128 Pro PR (PCI)"}, \ + {0x1002, 0x5053, 0, "ATI Rage 128 Pro PS (PCI)"}, \ + {0x1002, 0x5054, 0, "ATI Rage 128 Pro PT (AGP)"}, \ + {0x1002, 0x5055, 0, "ATI Rage 128 Pro PU (AGP)"}, \ + {0x1002, 0x5056, 0, "ATI Rage 128 Pro PV (PCI)"}, \ + {0x1002, 0x5057, 0, "ATI Rage 128 Pro PW (AGP)"}, \ + {0x1002, 0x5058, 0, "ATI Rage 128 Pro PX (AGP)"}, \ + {0x1002, 0x5245, 0, "ATI Rage 128 RE (PCI)"}, \ + {0x1002, 0x5246, 0, "ATI Rage 128 RF (AGP)"}, \ + {0x1002, 0x5247, 0, "ATI Rage 128 RG (AGP)"}, \ + {0x1002, 0x524b, 0, "ATI Rage 128 RK (PCI)"}, \ + {0x1002, 0x524c, 0, "ATI Rage 128 RL (AGP)"}, \ + {0x1002, 0x534d, 0, "ATI Rage 128 SM (AGP)"}, \ + {0x1002, 0x5446, 0, "ATI Rage 128 Pro Ultra TF (AGP)"}, \ + {0x1002, 0x544C, 0, "ATI Rage 128 Pro Ultra TL (AGP)"}, \ + {0x1002, 0x5452, 0, "ATI Rage 128 Pro Ultra TR (AGP)"}, \ + {0, 0, 0, NULL} + /* Driver customization: */ #define DRIVER_PRERELEASE() do { \ @@ -97,7 +137,7 @@ /* DMA customization: */ #define __HAVE_DMA 1 -#define __HAVE_DMA_IRQ 1 +#define __HAVE_IRQ 1 #define __HAVE_VBL_IRQ 1 #define __HAVE_SHARED_IRQ 1 --- diff/drivers/char/drm/r128_cce.c 2003-09-30 15:46:12.000000000 +0100 +++ source/drivers/char/drm/r128_cce.c 2004-03-01 15:42:16.000000000 +0000 @@ -212,7 +212,7 @@ int r128_do_cce_idle( drm_r128_private_t int i; for ( i = 0 ; i < dev_priv->usec_timeout ; i++ ) { - if ( GET_RING_HEAD( &dev_priv->ring ) == dev_priv->ring.tail ) { + if ( GET_RING_HEAD( dev_priv ) == dev_priv->ring.tail ) { int pm4stat = R128_READ( R128_PM4_STAT ); if ( ( (pm4stat & R128_PM4_FIFOCNT_MASK) >= dev_priv->cce_fifo_size ) && @@ -238,7 +238,8 @@ static void r128_do_cce_start( drm_r128_ r128_do_wait_for_idle( dev_priv ); R128_WRITE( R128_PM4_BUFFER_CNTL, - dev_priv->cce_mode | dev_priv->ring.size_l2qw ); + dev_priv->cce_mode | dev_priv->ring.size_l2qw + | R128_PM4_BUFFER_CNTL_NOUPDATE ); R128_READ( R128_PM4_BUFFER_ADDR ); /* as per the sample code */ R128_WRITE( R128_PM4_MICRO_CNTL, R128_PM4_MICRO_FREERUN ); @@ -253,7 +254,6 @@ static void r128_do_cce_reset( drm_r128_ { R128_WRITE( R128_PM4_BUFFER_DL_WPTR, 0 ); R128_WRITE( R128_PM4_BUFFER_DL_RPTR, 0 ); - SET_RING_HEAD( &dev_priv->ring, 0 ); dev_priv->ring.tail = 0; } @@ -264,7 +264,8 @@ static void r128_do_cce_reset( drm_r128_ static void r128_do_cce_stop( drm_r128_private_t *dev_priv ) { R128_WRITE( R128_PM4_MICRO_CNTL, 0 ); - R128_WRITE( R128_PM4_BUFFER_CNTL, R128_PM4_NONPM4 ); + R128_WRITE( R128_PM4_BUFFER_CNTL, + R128_PM4_NONPM4 | R128_PM4_BUFFER_CNTL_NOUPDATE ); dev_priv->cce_running = 0; } @@ -333,26 +334,6 @@ static void r128_cce_init_ring_buffer( d R128_WRITE( R128_PM4_BUFFER_DL_WPTR, 0 ); R128_WRITE( R128_PM4_BUFFER_DL_RPTR, 0 ); - /* DL_RPTR_ADDR is a physical address in AGP space. */ - SET_RING_HEAD( &dev_priv->ring, 0 ); - - if ( !dev_priv->is_pci ) { - R128_WRITE( R128_PM4_BUFFER_DL_RPTR_ADDR, - dev_priv->ring_rptr->offset ); - } else { - drm_sg_mem_t *entry = dev->sg; - unsigned long tmp_ofs, page_ofs; - - tmp_ofs = dev_priv->ring_rptr->offset - dev->sg->handle; - page_ofs = tmp_ofs >> PAGE_SHIFT; - - R128_WRITE( R128_PM4_BUFFER_DL_RPTR_ADDR, - entry->busaddr[page_ofs]); - DRM_DEBUG( "ring rptr: offset=0x%08lx handle=0x%08lx\n", - (unsigned long) entry->busaddr[page_ofs], - entry->handle + tmp_ofs ); - } - /* Set watermark control */ R128_WRITE( R128_PM4_BUFFER_WM_CNTL, ((R128_WATERMARK_L/4) << R128_WMA_SHIFT) @@ -486,13 +467,6 @@ static int r128_do_init_cce( drm_device_ return DRM_ERR(EINVAL); } - DRM_FIND_MAP( dev_priv->fb, init->fb_offset ); - if(!dev_priv->fb) { - DRM_ERROR("could not find framebuffer!\n"); - dev->dev_private = (void *)dev_priv; - r128_do_cleanup_cce( dev ); - return DRM_ERR(EINVAL); - } DRM_FIND_MAP( dev_priv->mmio, init->mmio_offset ); if(!dev_priv->mmio) { DRM_ERROR("could not find mmio region!\n"); @@ -567,9 +541,6 @@ static int r128_do_init_cce( drm_device_ #endif dev_priv->cce_buffers_offset = dev->sg->handle; - dev_priv->ring.head = ((__volatile__ u32 *) - dev_priv->ring_rptr->handle); - dev_priv->ring.start = (u32 *)dev_priv->cce_ring->handle; dev_priv->ring.end = ((u32 *)dev_priv->cce_ring->handle + init->ring_size / sizeof(u32)); @@ -580,7 +551,6 @@ static int r128_do_init_cce( drm_device_ (dev_priv->ring.size / sizeof(u32)) - 1; dev_priv->ring.high_mark = 128; - dev_priv->ring.ring_rptr = dev_priv->ring_rptr; dev_priv->sarea_priv->last_frame = 0; R128_WRITE( R128_LAST_FRAME_REG, dev_priv->sarea_priv->last_frame ); @@ -589,8 +559,9 @@ static int r128_do_init_cce( drm_device_ R128_WRITE( R128_LAST_DISPATCH_REG, dev_priv->sarea_priv->last_dispatch ); -#if __REALLY_HAVE_SG +#if __REALLY_HAVE_AGP if ( dev_priv->is_pci ) { +#endif if (!DRM(ati_pcigart_init)( dev, &dev_priv->phys_pci_gart, &dev_priv->bus_pci_gart) ) { DRM_ERROR( "failed to init PCI GART!\n" ); @@ -599,6 +570,7 @@ static int r128_do_init_cce( drm_device_ return DRM_ERR(ENOMEM); } R128_WRITE( R128_PCI_GART_PAGE, dev_priv->bus_pci_gart ); +#if __REALLY_HAVE_AGP } #endif @@ -615,12 +587,12 @@ static int r128_do_init_cce( drm_device_ int r128_do_cleanup_cce( drm_device_t *dev ) { -#if _HAVE_DMA_IRQ +#if __HAVE_IRQ /* Make sure interrupts are disabled here because the uninstall ioctl * may not have been called from userspace and after dev_private * is freed, it's too late. */ - if ( dev->irq ) DRM(irq_uninstall)(dev); + if ( dev->irq_enabled ) DRM(irq_uninstall)(dev); #endif if ( dev->dev_private ) { @@ -901,7 +873,7 @@ int r128_wait_ring( drm_r128_private_t * int i; for ( i = 0 ; i < dev_priv->usec_timeout ; i++ ) { - r128_update_ring_snapshot( ring ); + r128_update_ring_snapshot( dev_priv ); if ( ring->space >= n ) return 0; DRM_UDELAY( 1 ); --- diff/drivers/char/drm/r128_drv.c 2002-10-16 04:27:48.000000000 +0100 +++ source/drivers/char/drm/r128_drv.c 2004-03-01 15:42:16.000000000 +0000 @@ -47,6 +47,7 @@ #include "drm_fops.h" #include "drm_init.h" #include "drm_ioctl.h" +#include "drm_irq.h" #include "drm_lock.h" #include "drm_memory.h" #include "drm_proc.h" --- diff/drivers/char/drm/r128_drv.h 2003-08-20 14:16:27.000000000 +0100 +++ source/drivers/char/drm/r128_drv.h 2004-03-01 15:42:16.000000000 +0000 @@ -34,8 +34,7 @@ #ifndef __R128_DRV_H__ #define __R128_DRV_H__ -#define GET_RING_HEAD(ring) DRM_READ32( (ring)->ring_rptr, 0 ) /* (ring)->head */ -#define SET_RING_HEAD(ring,val) DRM_WRITE32( (ring)->ring_rptr, 0, (val) ) /* (ring)->head */ +#define GET_RING_HEAD(dev_priv) R128_READ( R128_PM4_BUFFER_DL_RPTR ) typedef struct drm_r128_freelist { unsigned int age; @@ -50,13 +49,11 @@ typedef struct drm_r128_ring_buffer { int size; int size_l2qw; - volatile u32 *head; u32 tail; u32 tail_mask; int space; int high_mark; - drm_local_map_t *ring_rptr; } drm_r128_ring_buffer_t; typedef struct drm_r128_private { @@ -100,7 +97,6 @@ typedef struct drm_r128_private { u32 span_pitch_offset_c; drm_local_map_t *sarea; - drm_local_map_t *fb; drm_local_map_t *mmio; drm_local_map_t *cce_ring; drm_local_map_t *ring_rptr; @@ -132,14 +128,6 @@ extern drm_buf_t *r128_freelist_get( drm extern int r128_wait_ring( drm_r128_private_t *dev_priv, int n ); -static __inline__ void -r128_update_ring_snapshot( drm_r128_ring_buffer_t *ring ) -{ - ring->space = (GET_RING_HEAD( ring ) - ring->tail) * sizeof(u32); - if ( ring->space <= 0 ) - ring->space += ring->size; -} - extern int r128_do_cce_idle( drm_r128_private_t *dev_priv ); extern int r128_do_cleanup_cce( drm_device_t *dev ); extern int r128_do_cleanup_pageflip( drm_device_t *dev ); @@ -279,6 +267,7 @@ extern int r128_cce_indirect( DRM_IOCTL_ # define R128_PM4_64PIO_64VCBM_64INDBM (7 << 28) # define R128_PM4_64BM_64VCBM_64INDBM (8 << 28) # define R128_PM4_64PIO_64VCPIO_64INDPIO (15 << 28) +# define R128_PM4_BUFFER_CNTL_NOUPDATE (1 << 27) #define R128_PM4_BUFFER_WM_CNTL 0x0708 # define R128_WMA_SHIFT 0 @@ -403,6 +392,15 @@ extern int R128_READ_PLL(drm_device_t *d (pkt) | ((n) << 16)) +static __inline__ void +r128_update_ring_snapshot( drm_r128_private_t *dev_priv ) +{ + drm_r128_ring_buffer_t *ring = &dev_priv->ring; + ring->space = (GET_RING_HEAD( dev_priv ) - ring->tail) * sizeof(u32); + if ( ring->space <= 0 ) + ring->space += ring->size; +} + /* ================================================================ * Misc helper macros */ @@ -412,7 +410,7 @@ do { \ drm_r128_ring_buffer_t *ring = &dev_priv->ring; int i; \ if ( ring->space < ring->high_mark ) { \ for ( i = 0 ; i < dev_priv->usec_timeout ; i++ ) { \ - r128_update_ring_snapshot( ring ); \ + r128_update_ring_snapshot( dev_priv ); \ if ( ring->space >= ring->high_mark ) \ goto __ring_space_done; \ DRM_UDELAY(1); \ @@ -445,17 +443,10 @@ do { \ * Ring control */ -#if defined(__powerpc__) -#define r128_flush_write_combine() (void) GET_RING_HEAD( &dev_priv->ring ) -#else -#define r128_flush_write_combine() DRM_WRITEMEMORYBARRIER() -#endif - - #define R128_VERBOSE 0 #define RING_LOCALS \ - int write; unsigned int tail_mask; volatile u32 *ring; + int write, _nr; unsigned int tail_mask; volatile u32 *ring; #define BEGIN_RING( n ) do { \ if ( R128_VERBOSE ) { \ @@ -463,9 +454,10 @@ do { \ (n), __FUNCTION__ ); \ } \ if ( dev_priv->ring.space <= (n) * sizeof(u32) ) { \ + COMMIT_RING(); \ r128_wait_ring( dev_priv, (n) * sizeof(u32) ); \ } \ - dev_priv->ring.space -= (n) * sizeof(u32); \ + _nr = n; dev_priv->ring.space -= (n) * sizeof(u32); \ ring = dev_priv->ring.start; \ write = dev_priv->ring.tail; \ tail_mask = dev_priv->ring.tail_mask; \ @@ -488,9 +480,23 @@ do { \ dev_priv->ring.start, \ write * sizeof(u32) ); \ } \ - r128_flush_write_combine(); \ - dev_priv->ring.tail = write; \ - R128_WRITE( R128_PM4_BUFFER_DL_WPTR, write ); \ + if (((dev_priv->ring.tail + _nr) & tail_mask) != write) { \ + DRM_ERROR( \ + "ADVANCE_RING(): mismatch: nr: %x write: %x line: %d\n", \ + ((dev_priv->ring.tail + _nr) & tail_mask), \ + write, __LINE__); \ + } else \ + dev_priv->ring.tail = write; \ +} while (0) + +#define COMMIT_RING() do { \ + if ( R128_VERBOSE ) { \ + DRM_INFO( "COMMIT_RING() tail=0x%06x\n", \ + dev_priv->ring.tail ); \ + } \ + DRM_MEMORYBARRIER(); \ + R128_WRITE( R128_PM4_BUFFER_DL_WPTR, dev_priv->ring.tail ); \ + R128_READ( R128_PM4_BUFFER_DL_WPTR ); \ } while (0) #define OUT_RING( x ) do { \ --- diff/drivers/char/drm/r128_irq.c 2003-05-21 11:50:14.000000000 +0100 +++ source/drivers/char/drm/r128_irq.c 2004-03-01 15:42:16.000000000 +0000 @@ -36,7 +36,7 @@ #include "r128_drm.h" #include "r128_drv.h" -irqreturn_t r128_dma_service( DRM_IRQ_ARGS ) +irqreturn_t r128_irq_handler( DRM_IRQ_ARGS ) { drm_device_t *dev = (drm_device_t *) arg; drm_r128_private_t *dev_priv = --- diff/drivers/char/drm/r128_state.c 2003-08-20 14:16:27.000000000 +0100 +++ source/drivers/char/drm/r128_state.c 2004-03-01 15:42:16.000000000 +0000 @@ -45,7 +45,7 @@ static void r128_emit_clip_rects( drm_r1 RING_LOCALS; DRM_DEBUG( " %s\n", __FUNCTION__ ); - BEGIN_RING( 17 ); + BEGIN_RING( (count < 3? count: 3) * 5 + 2 ); if ( count >= 1 ) { OUT_RING( CCE_PACKET0( R128_AUX1_SC_LEFT, 3 ) ); @@ -1269,6 +1269,7 @@ int r128_cce_clear( DRM_IOCTL_ARGS ) sarea_priv->nbox = R128_NR_SAREA_CLIPRECTS; r128_cce_dispatch_clear( dev, &clear ); + COMMIT_RING(); /* Make sure we restore the 3D state next time. */ @@ -1304,8 +1305,10 @@ int r128_do_cleanup_pageflip( drm_device R128_WRITE( R128_CRTC_OFFSET, dev_priv->crtc_offset ); R128_WRITE( R128_CRTC_OFFSET_CNTL, dev_priv->crtc_offset_cntl ); - if (dev_priv->current_page != 0) + if (dev_priv->current_page != 0) { r128_cce_dispatch_flip( dev ); + COMMIT_RING(); + } dev_priv->page_flipping = 0; return 0; @@ -1330,6 +1333,7 @@ int r128_cce_flip( DRM_IOCTL_ARGS ) r128_cce_dispatch_flip( dev ); + COMMIT_RING(); return 0; } @@ -1351,6 +1355,7 @@ int r128_cce_swap( DRM_IOCTL_ARGS ) dev_priv->sarea_priv->dirty |= (R128_UPLOAD_CONTEXT | R128_UPLOAD_MASKS); + COMMIT_RING(); return 0; } @@ -1410,6 +1415,7 @@ int r128_cce_vertex( DRM_IOCTL_ARGS ) r128_cce_dispatch_vertex( dev, buf ); + COMMIT_RING(); return 0; } @@ -1481,6 +1487,7 @@ int r128_cce_indices( DRM_IOCTL_ARGS ) r128_cce_dispatch_indices( dev, buf, elts.start, elts.end, count ); + COMMIT_RING(); return 0; } @@ -1490,6 +1497,7 @@ int r128_cce_blit( DRM_IOCTL_ARGS ) drm_device_dma_t *dma = dev->dma; drm_r128_private_t *dev_priv = dev->dev_private; drm_r128_blit_t blit; + int ret; LOCK_TEST_WITH_RETURN( dev, filp ); @@ -1507,7 +1515,10 @@ int r128_cce_blit( DRM_IOCTL_ARGS ) RING_SPACE_TEST_WITH_RETURN( dev_priv ); VB_AGE_TEST_WITH_RETURN( dev_priv ); - return r128_cce_dispatch_blit( filp, dev, &blit ); + ret = r128_cce_dispatch_blit( filp, dev, &blit ); + + COMMIT_RING(); + return ret; } int r128_cce_depth( DRM_IOCTL_ARGS ) @@ -1515,6 +1526,7 @@ int r128_cce_depth( DRM_IOCTL_ARGS ) DRM_DEVICE; drm_r128_private_t *dev_priv = dev->dev_private; drm_r128_depth_t depth; + int ret; LOCK_TEST_WITH_RETURN( dev, filp ); @@ -1523,18 +1535,20 @@ int r128_cce_depth( DRM_IOCTL_ARGS ) RING_SPACE_TEST_WITH_RETURN( dev_priv ); + ret = DRM_ERR(EINVAL); switch ( depth.func ) { case R128_WRITE_SPAN: - return r128_cce_dispatch_write_span( dev, &depth ); + ret = r128_cce_dispatch_write_span( dev, &depth ); case R128_WRITE_PIXELS: - return r128_cce_dispatch_write_pixels( dev, &depth ); + ret = r128_cce_dispatch_write_pixels( dev, &depth ); case R128_READ_SPAN: - return r128_cce_dispatch_read_span( dev, &depth ); + ret = r128_cce_dispatch_read_span( dev, &depth ); case R128_READ_PIXELS: - return r128_cce_dispatch_read_pixels( dev, &depth ); + ret = r128_cce_dispatch_read_pixels( dev, &depth ); } - return DRM_ERR(EINVAL); + COMMIT_RING(); + return ret; } int r128_cce_stipple( DRM_IOCTL_ARGS ) @@ -1557,6 +1571,7 @@ int r128_cce_stipple( DRM_IOCTL_ARGS ) r128_cce_dispatch_stipple( dev, mask ); + COMMIT_RING(); return 0; } @@ -1632,6 +1647,7 @@ int r128_cce_indirect( DRM_IOCTL_ARGS ) */ r128_cce_dispatch_indirect( dev, buf, indirect.start, indirect.end ); + COMMIT_RING(); return 0; } --- diff/drivers/char/drm/radeon.h 2003-09-30 15:46:12.000000000 +0100 +++ source/drivers/char/drm/radeon.h 2004-03-01 15:42:16.000000000 +0000 @@ -51,7 +51,7 @@ #define DRIVER_DATE "20020828" #define DRIVER_MAJOR 1 -#define DRIVER_MINOR 9 +#define DRIVER_MINOR 10 #define DRIVER_PATCHLEVEL 0 /* Interface history: @@ -81,6 +81,9 @@ * Add 'GET' queries for starting additional clients on different VT's. * 1.9 - Add DRM_IOCTL_RADEON_CP_RESUME ioctl. * Add texture rectangle support for r100. + * 1.10- Add SETPARAM ioctl; first parameter to set is FB_LOCATION, which + * clients use to tell the DRM where they think the framebuffer is + * located in the card's address space */ #define DRIVER_IOCTLS \ [DRM_IOCTL_NR(DRM_IOCTL_DMA)] = { radeon_cp_buffers, 1, 0 }, \ @@ -106,10 +109,82 @@ [DRM_IOCTL_NR(DRM_IOCTL_RADEON_ALLOC)] = { radeon_mem_alloc, 1, 0 }, \ [DRM_IOCTL_NR(DRM_IOCTL_RADEON_FREE)] = { radeon_mem_free, 1, 0 }, \ [DRM_IOCTL_NR(DRM_IOCTL_RADEON_INIT_HEAP)] = { radeon_mem_init_heap, 1, 1 }, \ - [DRM_IOCTL_NR(DRM_IOCTL_RADEON_IRQ_EMIT)] = { radeon_irq_emit, 1, 0 }, \ - [DRM_IOCTL_NR(DRM_IOCTL_RADEON_IRQ_WAIT)] = { radeon_irq_wait, 1, 0 }, + [DRM_IOCTL_NR(DRM_IOCTL_RADEON_IRQ_EMIT)] = { radeon_irq_emit, 1, 0 }, \ + [DRM_IOCTL_NR(DRM_IOCTL_RADEON_IRQ_WAIT)] = { radeon_irq_wait, 1, 0 }, \ + [DRM_IOCTL_NR(DRM_IOCTL_RADEON_SETPARAM)] = { radeon_cp_setparam, 1, 0 }, \ + +#define DRIVER_PCI_IDS \ + {0x1002, 0x4136, 0, "ATI Radeon RS100 IGP 320M"}, \ + {0x1002, 0x4137, 0, "ATI Radeon RS200 IGP"}, \ + {0x1002, 0x4237, 0, "ATI Radeon RS250 IGP"}, \ + {0x1002, 0x4242, 0, "ATI Radeon BB R200 AIW 8500DV"}, \ + {0x1002, 0x4242, 0, "ATI Radeon BC R200"}, \ + {0x1002, 0x4336, 0, "ATI Radeon RS100 Mobility U1"}, \ + {0x1002, 0x4337, 0, "ATI Radeon RS200 Mobility IGP 340M"}, \ + {0x1002, 0x4437, 0, "ATI Radeon RS250 Mobility IGP"}, \ + {0x1002, 0x4964, 0, "ATI Radeon Id R250 9000"}, \ + {0x1002, 0x4965, 0, "ATI Radeon Ie R250 9000"}, \ + {0x1002, 0x4966, 0, "ATI Radeon If R250 9000"}, \ + {0x1002, 0x4967, 0, "ATI Radeon Ig R250 9000"}, \ + {0x1002, 0x4C57, 0, "ATI Radeon LW Mobility 7500 M7"}, \ + {0x1002, 0x4C58, 0, "ATI Radeon LX RV200 Mobility FireGL 7800 M7"}, \ + {0x1002, 0x4C59, 0, "ATI Radeon LY Mobility M6"}, \ + {0x1002, 0x4C5A, 0, "ATI Radeon LZ Mobility M6"}, \ + {0x1002, 0x4C64, 0, "ATI Radeon Ld R250 Mobility 9000 M9"}, \ + {0x1002, 0x4C65, 0, "ATI Radeon Le R250 Mobility 9000 M9"}, \ + {0x1002, 0x4C66, 0, "ATI Radeon Lf R250 Mobility 9000 M9"}, \ + {0x1002, 0x4C67, 0, "ATI Radeon Lg R250 Mobility 9000 M9"}, \ + {0x1002, 0x5144, 0, "ATI Radeon QD R100"}, \ + {0x1002, 0x5145, 0, "ATI Radeon QE R100"}, \ + {0x1002, 0x5146, 0, "ATI Radeon QF R100"}, \ + {0x1002, 0x5147, 0, "ATI Radeon QG R100"}, \ + {0x1002, 0x5148, 0, "ATI Radeon QH R200 8500"}, \ + {0x1002, 0x5149, 0, "ATI Radeon QI R200"}, \ + {0x1002, 0x514A, 0, "ATI Radeon QJ R200"}, \ + {0x1002, 0x514B, 0, "ATI Radeon QK R200"}, \ + {0x1002, 0x514C, 0, "ATI Radeon QL R200 8500 LE"}, \ + {0x1002, 0x514D, 0, "ATI Radeon QM R200 9100"}, \ + {0x1002, 0x514E, 0, "ATI Radeon QN R200 8500 LE"}, \ + {0x1002, 0x514F, 0, "ATI Radeon QO R200 8500 LE"}, \ + {0x1002, 0x5157, 0, "ATI Radeon QW RV200 7500"}, \ + {0x1002, 0x5158, 0, "ATI Radeon QX RV200 7500"}, \ + {0x1002, 0x5159, 0, "ATI Radeon QY RV100 7000/VE"}, \ + {0x1002, 0x515A, 0, "ATI Radeon QZ RV100 7000/VE"}, \ + {0x1002, 0x5168, 0, "ATI Radeon Qh R200"}, \ + {0x1002, 0x5169, 0, "ATI Radeon Qi R200"}, \ + {0x1002, 0x516A, 0, "ATI Radeon Qj R200"}, \ + {0x1002, 0x516B, 0, "ATI Radeon Qk R200"}, \ + {0x1002, 0x516C, 0, "ATI Radeon Ql R200"}, \ + {0x1002, 0x5834, 0, "ATI Radeon RS300 IGP"}, \ + {0x1002, 0x5835, 0, "ATI Radeon RS300 Mobility IGP"}, \ + {0x1002, 0x5836, 0, "ATI Radeon RS300 IGP"}, \ + {0x1002, 0x5837, 0, "ATI Radeon RS300 IGP"}, \ + {0x1002, 0x5960, 0, "ATI Radeon RV280 9200"}, \ + {0x1002, 0x5961, 0, "ATI Radeon RV280 9200 SE"}, \ + {0x1002, 0x5962, 0, "ATI Radeon RV280 9200"}, \ + {0x1002, 0x5963, 0, "ATI Radeon RV280 9200"}, \ + {0x1002, 0x5964, 0, "ATI Radeon RV280 9200 SE"}, \ + {0x1002, 0x5968, 0, "ATI Radeon RV280 9200"}, \ + {0x1002, 0x5969, 0, "ATI Radeon RV280 9200"}, \ + {0x1002, 0x596A, 0, "ATI Radeon RV280 9200"}, \ + {0x1002, 0x596B, 0, "ATI Radeon RV280 9200"}, \ + {0x1002, 0x5c61, 0, "ATI Radeon RV280 Mobility"}, \ + {0x1002, 0x5c62, 0, "ATI Radeon RV280"}, \ + {0x1002, 0x5c63, 0, "ATI Radeon RV280 Mobility"}, \ + {0x1002, 0x5c64, 0, "ATI Radeon RV280"}, \ + {0, 0, 0, NULL} +#define DRIVER_FILE_FIELDS \ + int64_t radeon_fb_delta; \ +#define DRIVER_OPEN_HELPER( filp_priv, dev ) \ +do { \ + drm_radeon_private_t *dev_priv = dev->dev_private; \ + if ( dev_priv ) \ + filp_priv->radeon_fb_delta = dev_priv->fb_location; \ + else \ + filp_priv->radeon_fb_delta = 0; \ +} while( 0 ) /* When a client dies: * - Check for and clean up flipped page state @@ -125,7 +200,7 @@ do { \ radeon_do_cleanup_pageflip( dev ); \ } \ radeon_mem_release( filp, dev_priv->gart_heap ); \ - radeon_mem_release( filp, dev_priv->fb_heap ); \ + radeon_mem_release( filp, dev_priv->fb_heap ); \ } \ } while (0) @@ -142,7 +217,7 @@ do { \ /* DMA customization: */ #define __HAVE_DMA 1 -#define __HAVE_DMA_IRQ 1 +#define __HAVE_IRQ 1 #define __HAVE_VBL_IRQ 1 #define __HAVE_SHARED_IRQ 1 --- diff/drivers/char/drm/radeon_cp.c 2003-09-30 15:46:12.000000000 +0100 +++ source/drivers/char/drm/radeon_cp.c 2004-03-01 15:42:16.000000000 +0000 @@ -855,7 +855,8 @@ static void radeon_cp_init_ring_buffer( /* Initialize the memory controller */ RADEON_WRITE( RADEON_MC_FB_LOCATION, - (dev_priv->gart_vm_start - 1) & 0xffff0000 ); + ( ( dev_priv->gart_vm_start - 1 ) & 0xffff0000 ) + | ( dev_priv->fb_location >> 16 ) ); #if __REALLY_HAVE_AGP if ( !dev_priv->is_pci ) { @@ -1071,13 +1072,6 @@ static int radeon_do_init_cp( drm_device dev_priv->depth_offset = init->depth_offset; dev_priv->depth_pitch = init->depth_pitch; - dev_priv->front_pitch_offset = (((dev_priv->front_pitch/64) << 22) | - (dev_priv->front_offset >> 10)); - dev_priv->back_pitch_offset = (((dev_priv->back_pitch/64) << 22) | - (dev_priv->back_offset >> 10)); - dev_priv->depth_pitch_offset = (((dev_priv->depth_pitch/64) << 22) | - (dev_priv->depth_offset >> 10)); - /* Hardware state for depth clears. Remove this if/when we no * longer clear the depth buffer with a 3D rectangle. Hard-code * all values to prevent unwanted 3D state from slipping through @@ -1124,13 +1118,6 @@ static int radeon_do_init_cp( drm_device return DRM_ERR(EINVAL); } - DRM_FIND_MAP( dev_priv->fb, init->fb_offset ); - if(!dev_priv->fb) { - DRM_ERROR("could not find framebuffer!\n"); - dev->dev_private = (void *)dev_priv; - radeon_do_cleanup_cp(dev); - return DRM_ERR(EINVAL); - } DRM_FIND_MAP( dev_priv->mmio, init->mmio_offset ); if(!dev_priv->mmio) { DRM_ERROR("could not find mmio region!\n"); @@ -1204,9 +1191,26 @@ static int radeon_do_init_cp( drm_device dev_priv->buffers->handle ); } + dev_priv->fb_location = ( RADEON_READ( RADEON_MC_FB_LOCATION ) + & 0xffff ) << 16; + + dev_priv->front_pitch_offset = (((dev_priv->front_pitch/64) << 22) | + ( ( dev_priv->front_offset + + dev_priv->fb_location ) >> 10 ) ); + + dev_priv->back_pitch_offset = (((dev_priv->back_pitch/64) << 22) | + ( ( dev_priv->back_offset + + dev_priv->fb_location ) >> 10 ) ); + + dev_priv->depth_pitch_offset = (((dev_priv->depth_pitch/64) << 22) | + ( ( dev_priv->depth_offset + + dev_priv->fb_location ) >> 10 ) ); + dev_priv->gart_size = init->gart_size; - dev_priv->gart_vm_start = RADEON_READ( RADEON_CONFIG_APER_SIZE ); + dev_priv->gart_vm_start = dev_priv->fb_location + + RADEON_READ( RADEON_CONFIG_APER_SIZE ); + #if __REALLY_HAVE_AGP if ( !dev_priv->is_pci ) dev_priv->gart_buffers_offset = (dev_priv->buffers->offset @@ -1271,12 +1275,12 @@ int radeon_do_cleanup_cp( drm_device_t * { DRM_DEBUG( "\n" ); -#if _HAVE_DMA_IRQ +#if __HAVE_IRQ /* Make sure interrupts are disabled here because the uninstall ioctl * may not have been called from userspace and after dev_private * is freed, it's too late. */ - if ( dev->irq ) DRM(irq_uninstall)(dev); + if ( dev->irq_enabled ) DRM(irq_uninstall)(dev); #endif if ( dev->dev_private ) { --- diff/drivers/char/drm/radeon_drm.h 2003-09-30 15:46:12.000000000 +0100 +++ source/drivers/char/drm/radeon_drm.h 2004-03-01 15:42:16.000000000 +0000 @@ -390,6 +390,7 @@ typedef struct { #define DRM_IOCTL_RADEON_IRQ_WAIT DRM_IOW( 0x57, drm_radeon_irq_wait_t) /* added by Charl P. Botha - see radeon_cp.c for details */ #define DRM_IOCTL_RADEON_CP_RESUME DRM_IO(0x58) +#define DRM_IOCTL_RADEON_SETPARAM DRM_IOW(0x59, drm_radeon_setparam_t) typedef struct drm_radeon_init { enum { @@ -502,7 +503,7 @@ typedef struct drm_radeon_tex_image { } drm_radeon_tex_image_t; typedef struct drm_radeon_texture { - int offset; + unsigned int offset; int pitch; int format; int width; /* Texture image coordinates */ @@ -537,6 +538,7 @@ typedef struct drm_radeon_indirect { #define RADEON_PARAM_STATUS_HANDLE 8 #define RADEON_PARAM_SAREA_HANDLE 9 #define RADEON_PARAM_GART_TEX_HANDLE 10 +#define RADEON_PARAM_SCRATCH_OFFSET 11 typedef struct drm_radeon_getparam { int param; @@ -578,4 +580,16 @@ typedef struct drm_radeon_irq_wait { } drm_radeon_irq_wait_t; +/* 1.10: Clients tell the DRM where they think the framebuffer is located in + * the card's address space, via a new generic ioctl to set parameters + */ + +typedef struct drm_radeon_setparam { + unsigned int param; + int64_t value; +} drm_radeon_setparam_t; + +#define RADEON_SETPARAM_FB_LOCATION 1 /* determined framebuffer location */ + + #endif --- diff/drivers/char/drm/radeon_drv.c 2003-07-22 18:54:27.000000000 +0100 +++ source/drivers/char/drm/radeon_drv.c 2004-03-01 15:42:16.000000000 +0000 @@ -48,6 +48,7 @@ #include "drm_fops.h" #include "drm_init.h" #include "drm_ioctl.h" +#include "drm_irq.h" #include "drm_lock.h" #include "drm_memory.h" #include "drm_proc.h" --- diff/drivers/char/drm/radeon_drv.h 2003-09-30 15:46:12.000000000 +0100 +++ source/drivers/char/drm/radeon_drv.h 2004-03-01 15:42:16.000000000 +0000 @@ -73,6 +73,8 @@ typedef struct drm_radeon_private { drm_radeon_ring_buffer_t ring; drm_radeon_sarea_t *sarea_priv; + u32 fb_location; + int gart_size; u32 gart_vm_start; unsigned long gart_buffers_offset; @@ -133,7 +135,6 @@ typedef struct drm_radeon_private { unsigned long gart_textures_offset; drm_local_map_t *sarea; - drm_local_map_t *fb; drm_local_map_t *mmio; drm_local_map_t *cp_ring; drm_local_map_t *ring_rptr; @@ -184,6 +185,7 @@ extern int radeon_cp_indirect( DRM_IOCTL extern int radeon_cp_vertex2( DRM_IOCTL_ARGS ); extern int radeon_cp_cmdbuf( DRM_IOCTL_ARGS ); extern int radeon_cp_getparam( DRM_IOCTL_ARGS ); +extern int radeon_cp_setparam( DRM_IOCTL_ARGS ); extern int radeon_cp_flip( DRM_IOCTL_ARGS ); extern int radeon_mem_alloc( DRM_IOCTL_ARGS ); @@ -239,6 +241,7 @@ extern void radeon_do_release(drm_device #define RADEON_CRTC2_OFFSET 0x0324 #define RADEON_CRTC2_OFFSET_CNTL 0x0328 +#define RADEON_RB3D_COLOROFFSET 0x1c40 #define RADEON_RB3D_COLORPITCH 0x1c48 #define RADEON_DP_GUI_MASTER_CNTL 0x146c @@ -332,6 +335,7 @@ extern void radeon_do_release(drm_device #define RADEON_PP_MISC 0x1c14 #define RADEON_PP_ROT_MATRIX_0 0x1d58 #define RADEON_PP_TXFILTER_0 0x1c54 +#define RADEON_PP_TXOFFSET_0 0x1c5c #define RADEON_PP_TXFILTER_1 0x1c6c #define RADEON_PP_TXFILTER_2 0x1c84 --- diff/drivers/char/drm/radeon_irq.c 2003-05-21 11:50:14.000000000 +0100 +++ source/drivers/char/drm/radeon_irq.c 2004-03-01 15:42:16.000000000 +0000 @@ -54,7 +54,7 @@ * tied to dma at all, this is just a hangover from dri prehistory. */ -irqreturn_t DRM(dma_service)( DRM_IRQ_ARGS ) +irqreturn_t DRM(irq_handler)( DRM_IRQ_ARGS ) { drm_device_t *dev = (drm_device_t *) arg; drm_radeon_private_t *dev_priv = --- diff/drivers/char/drm/radeon_state.c 2004-02-18 08:54:09.000000000 +0000 +++ source/drivers/char/drm/radeon_state.c 2004-03-01 15:42:16.000000000 +0000 @@ -36,6 +36,151 @@ /* ================================================================ + * Helper functions for client state checking and fixup + */ + +static __inline__ int radeon_check_and_fixup_offset( drm_radeon_private_t *dev_priv, + drm_file_t *filp_priv, + u32 *offset ) { + u32 off = *offset; + + if ( off >= dev_priv->fb_location && + off < ( dev_priv->gart_vm_start + dev_priv->gart_size ) ) + return 0; + + off += filp_priv->radeon_fb_delta; + + DRM_DEBUG( "offset fixed up to 0x%x\n", off ); + + if ( off < dev_priv->fb_location || + off >= ( dev_priv->gart_vm_start + dev_priv->gart_size ) ) + return DRM_ERR( EINVAL ); + + *offset = off; + + return 0; +} + +static __inline__ int radeon_check_and_fixup_offset_user( drm_radeon_private_t *dev_priv, + drm_file_t *filp_priv, + u32 *offset ) { + u32 off; + + DRM_GET_USER_UNCHECKED( off, offset ); + + if ( radeon_check_and_fixup_offset( dev_priv, filp_priv, &off ) ) + return DRM_ERR( EINVAL ); + + DRM_PUT_USER_UNCHECKED( offset, off ); + + return 0; +} + +static __inline__ int radeon_check_and_fixup_packets( drm_radeon_private_t *dev_priv, + drm_file_t *filp_priv, + int id, + u32 *data ) { + if ( id == RADEON_EMIT_PP_MISC && + radeon_check_and_fixup_offset_user( dev_priv, filp_priv, + &data[( RADEON_RB3D_DEPTHOFFSET + - RADEON_PP_MISC ) / 4] ) ) { + DRM_ERROR( "Invalid depth buffer offset\n" ); + return DRM_ERR( EINVAL ); + } else if ( id == RADEON_EMIT_PP_CNTL && + radeon_check_and_fixup_offset_user( dev_priv, filp_priv, + &data[( RADEON_RB3D_COLOROFFSET + - RADEON_PP_CNTL ) / 4] ) ) { + DRM_ERROR( "Invalid colour buffer offset\n" ); + return DRM_ERR( EINVAL ); + } else if ( id >= R200_EMIT_PP_TXOFFSET_0 && + id <= R200_EMIT_PP_TXOFFSET_5 && + radeon_check_and_fixup_offset_user( dev_priv, filp_priv, + &data[0] ) ) { + DRM_ERROR( "Invalid R200 texture offset\n" ); + return DRM_ERR( EINVAL ); + } else if ( ( id == RADEON_EMIT_PP_TXFILTER_0 || id == RADEON_EMIT_PP_TXFILTER_1 || + id == RADEON_EMIT_PP_TXFILTER_2 /*|| id == RADEON_EMIT_PP_TXFILTER_3 || + id == RADEON_EMIT_PP_TXFILTER_4 || id == RADEON_EMIT_PP_TXFILTER_5*/ ) && + radeon_check_and_fixup_offset_user( dev_priv, filp_priv, + &data[( RADEON_PP_TXOFFSET_0 + - RADEON_PP_TXFILTER_0 ) / 4] ) ) { + DRM_ERROR( "Invalid R100 texture offset\n" ); + return DRM_ERR( EINVAL ); + } else if ( id == R200_PP_CUBIC_OFFSET_F1_0 || id == R200_PP_CUBIC_OFFSET_F1_1 || + id == R200_PP_CUBIC_OFFSET_F1_2 || id == R200_PP_CUBIC_OFFSET_F1_3 || + id == R200_PP_CUBIC_OFFSET_F1_4 || id == R200_PP_CUBIC_OFFSET_F1_5 ) { + int i; + for ( i = 0; i < 6; i++ ) { + if ( radeon_check_and_fixup_offset_user( dev_priv, + filp_priv, + &data[i] ) ) { + DRM_ERROR( "Invalid R200 cubic texture offset\n" ); + return DRM_ERR( EINVAL ); + } + } + } + + return 0; +} + +static __inline__ int radeon_check_and_fixup_packet3( drm_radeon_private_t *dev_priv, + drm_file_t *filp_priv, + drm_radeon_cmd_buffer_t *cmdbuf, + unsigned int *cmdsz ) { + u32 tmp[4], *cmd = ( u32* )cmdbuf->buf; + + if ( DRM_COPY_FROM_USER_UNCHECKED( tmp, cmd, sizeof( tmp ) ) ) { + DRM_ERROR( "Failed to copy data from user space\n" ); + return DRM_ERR( EFAULT ); + } + + *cmdsz = 2 + ( ( tmp[0] & RADEON_CP_PACKET_COUNT_MASK ) >> 16 ); + + if ( ( tmp[0] & 0xc0000000 ) != RADEON_CP_PACKET3 ) { + DRM_ERROR( "Not a type 3 packet\n" ); + return DRM_ERR( EINVAL ); + } + + if ( 4 * *cmdsz > cmdbuf->bufsz ) { + DRM_ERROR( "Packet size larger than size of data provided\n" ); + return DRM_ERR( EINVAL ); + } + + /* Check client state and fix it up if necessary */ + if ( tmp[0] & 0x8000 ) { /* MSB of opcode: next DWORD GUI_CNTL */ + u32 offset; + + if ( tmp[1] & ( RADEON_GMC_SRC_PITCH_OFFSET_CNTL + | RADEON_GMC_DST_PITCH_OFFSET_CNTL ) ) { + offset = tmp[2] << 10; + if ( radeon_check_and_fixup_offset( dev_priv, filp_priv, &offset ) ) { + DRM_ERROR( "Invalid first packet offset\n" ); + return DRM_ERR( EINVAL ); + } + tmp[2] = ( tmp[2] & 0xffc00000 ) | offset >> 10; + } + + if ( ( tmp[1] & RADEON_GMC_SRC_PITCH_OFFSET_CNTL ) && + ( tmp[1] & RADEON_GMC_DST_PITCH_OFFSET_CNTL ) ) { + offset = tmp[3] << 10; + if ( radeon_check_and_fixup_offset( dev_priv, filp_priv, &offset ) ) { + DRM_ERROR( "Invalid second packet offset\n" ); + return DRM_ERR( EINVAL ); + } + tmp[3] = ( tmp[3] & 0xffc00000 ) | offset >> 10; + } + + if ( DRM_COPY_TO_USER_UNCHECKED( cmd, tmp, sizeof( tmp ) ) ) { + DRM_ERROR( "Failed to copy data to user space\n" ); + return DRM_ERR( EFAULT ); + } + } + + return 0; +} + + +/* ================================================================ * CP hardware state programming functions */ @@ -57,15 +202,28 @@ static __inline__ void radeon_emit_clip_ /* Emit 1.1 state */ -static void radeon_emit_state( drm_radeon_private_t *dev_priv, - drm_radeon_context_regs_t *ctx, - drm_radeon_texture_regs_t *tex, - unsigned int dirty ) +static int radeon_emit_state( drm_radeon_private_t *dev_priv, + drm_file_t *filp_priv, + drm_radeon_context_regs_t *ctx, + drm_radeon_texture_regs_t *tex, + unsigned int dirty ) { RING_LOCALS; DRM_DEBUG( "dirty=0x%08x\n", dirty ); if ( dirty & RADEON_UPLOAD_CONTEXT ) { + if ( radeon_check_and_fixup_offset( dev_priv, filp_priv, + &ctx->rb3d_depthoffset ) ) { + DRM_ERROR( "Invalid depth buffer offset\n" ); + return DRM_ERR( EINVAL ); + } + + if ( radeon_check_and_fixup_offset( dev_priv, filp_priv, + &ctx->rb3d_coloroffset ) ) { + DRM_ERROR( "Invalid depth buffer offset\n" ); + return DRM_ERR( EINVAL ); + } + BEGIN_RING( 14 ); OUT_RING( CP_PACKET0( RADEON_PP_MISC, 6 ) ); OUT_RING( ctx->pp_misc ); @@ -149,6 +307,12 @@ static void radeon_emit_state( drm_radeo } if ( dirty & RADEON_UPLOAD_TEX0 ) { + if ( radeon_check_and_fixup_offset( dev_priv, filp_priv, + &tex[0].pp_txoffset ) ) { + DRM_ERROR( "Invalid texture offset for unit 0\n" ); + return DRM_ERR( EINVAL ); + } + BEGIN_RING( 9 ); OUT_RING( CP_PACKET0( RADEON_PP_TXFILTER_0, 5 ) ); OUT_RING( tex[0].pp_txfilter ); @@ -163,6 +327,12 @@ static void radeon_emit_state( drm_radeo } if ( dirty & RADEON_UPLOAD_TEX1 ) { + if ( radeon_check_and_fixup_offset( dev_priv, filp_priv, + &tex[1].pp_txoffset ) ) { + DRM_ERROR( "Invalid texture offset for unit 1\n" ); + return DRM_ERR( EINVAL ); + } + BEGIN_RING( 9 ); OUT_RING( CP_PACKET0( RADEON_PP_TXFILTER_1, 5 ) ); OUT_RING( tex[1].pp_txfilter ); @@ -177,6 +347,12 @@ static void radeon_emit_state( drm_radeo } if ( dirty & RADEON_UPLOAD_TEX2 ) { + if ( radeon_check_and_fixup_offset( dev_priv, filp_priv, + &tex[2].pp_txoffset ) ) { + DRM_ERROR( "Invalid texture offset for unit 2\n" ); + return DRM_ERR( EINVAL ); + } + BEGIN_RING( 9 ); OUT_RING( CP_PACKET0( RADEON_PP_TXFILTER_2, 5 ) ); OUT_RING( tex[2].pp_txfilter ); @@ -189,12 +365,15 @@ static void radeon_emit_state( drm_radeo OUT_RING( tex[2].pp_border_color ); ADVANCE_RING(); } + + return 0; } /* Emit 1.2 state */ -static void radeon_emit_state2( drm_radeon_private_t *dev_priv, - drm_radeon_state_t *state ) +static int radeon_emit_state2( drm_radeon_private_t *dev_priv, + drm_file_t *filp_priv, + drm_radeon_state_t *state ) { RING_LOCALS; @@ -206,7 +385,7 @@ static void radeon_emit_state2( drm_rade ADVANCE_RING(); } - radeon_emit_state( dev_priv, &state->context, + return radeon_emit_state( dev_priv, filp_priv, &state->context, state->tex, state->dirty ); } @@ -1065,6 +1244,7 @@ static int radeon_cp_dispatch_texture( D drm_radeon_tex_image_t *image ) { drm_radeon_private_t *dev_priv = dev->dev_private; + drm_file_t *filp_priv; drm_buf_t *buf; u32 format; u32 *buffer; @@ -1074,6 +1254,13 @@ static int radeon_cp_dispatch_texture( D int i; RING_LOCALS; + DRM_GET_PRIV_WITH_RETURN( filp_priv, filp ); + + if ( radeon_check_and_fixup_offset( dev_priv, filp_priv, &tex->offset ) ) { + DRM_ERROR( "Invalid destination offset\n" ); + return DRM_ERR( EINVAL ); + } + dev_priv->stats.boxes |= RADEON_BOX_TEXTURE_LOAD; /* Flush the pixel cache. This ensures no pixel data gets mixed @@ -1377,6 +1564,7 @@ int radeon_cp_vertex( DRM_IOCTL_ARGS ) { DRM_DEVICE; drm_radeon_private_t *dev_priv = dev->dev_private; + drm_file_t *filp_priv; drm_radeon_sarea_t *sarea_priv = dev_priv->sarea_priv; drm_device_dma_t *dma = dev->dma; drm_buf_t *buf; @@ -1390,6 +1578,8 @@ int radeon_cp_vertex( DRM_IOCTL_ARGS ) return DRM_ERR(EINVAL); } + DRM_GET_PRIV_WITH_RETURN( filp_priv, filp ); + DRM_COPY_FROM_USER_IOCTL( vertex, (drm_radeon_vertex_t *)data, sizeof(vertex) ); @@ -1429,11 +1619,14 @@ int radeon_cp_vertex( DRM_IOCTL_ARGS ) buf->used = vertex.count; /* not used? */ if ( sarea_priv->dirty & ~RADEON_UPLOAD_CLIPRECTS ) { - radeon_emit_state( dev_priv, - &sarea_priv->context_state, - sarea_priv->tex_state, - sarea_priv->dirty ); - + if ( radeon_emit_state( dev_priv, filp_priv, + &sarea_priv->context_state, + sarea_priv->tex_state, + sarea_priv->dirty ) ) { + DRM_ERROR( "radeon_emit_state failed\n" ); + return DRM_ERR( EINVAL ); + } + sarea_priv->dirty &= ~(RADEON_UPLOAD_TEX0IMAGES | RADEON_UPLOAD_TEX1IMAGES | RADEON_UPLOAD_TEX2IMAGES | @@ -1461,6 +1654,7 @@ int radeon_cp_indices( DRM_IOCTL_ARGS ) { DRM_DEVICE; drm_radeon_private_t *dev_priv = dev->dev_private; + drm_file_t *filp_priv; drm_radeon_sarea_t *sarea_priv = dev_priv->sarea_priv; drm_device_dma_t *dma = dev->dma; drm_buf_t *buf; @@ -1475,6 +1669,8 @@ int radeon_cp_indices( DRM_IOCTL_ARGS ) return DRM_ERR(EINVAL); } + DRM_GET_PRIV_WITH_RETURN( filp_priv, filp ); + DRM_COPY_FROM_USER_IOCTL( elts, (drm_radeon_indices_t *)data, sizeof(elts) ); @@ -1523,10 +1719,13 @@ int radeon_cp_indices( DRM_IOCTL_ARGS ) buf->used = elts.end; if ( sarea_priv->dirty & ~RADEON_UPLOAD_CLIPRECTS ) { - radeon_emit_state( dev_priv, - &sarea_priv->context_state, - sarea_priv->tex_state, - sarea_priv->dirty ); + if ( radeon_emit_state( dev_priv, filp_priv, + &sarea_priv->context_state, + sarea_priv->tex_state, + sarea_priv->dirty ) ) { + DRM_ERROR( "radeon_emit_state failed\n" ); + return DRM_ERR( EINVAL ); + } sarea_priv->dirty &= ~(RADEON_UPLOAD_TEX0IMAGES | RADEON_UPLOAD_TEX1IMAGES | @@ -1686,6 +1885,7 @@ int radeon_cp_vertex2( DRM_IOCTL_ARGS ) { DRM_DEVICE; drm_radeon_private_t *dev_priv = dev->dev_private; + drm_file_t *filp_priv; drm_radeon_sarea_t *sarea_priv = dev_priv->sarea_priv; drm_device_dma_t *dma = dev->dma; drm_buf_t *buf; @@ -1700,6 +1900,8 @@ int radeon_cp_vertex2( DRM_IOCTL_ARGS ) return DRM_ERR(EINVAL); } + DRM_GET_PRIV_WITH_RETURN( filp_priv, filp ); + DRM_COPY_FROM_USER_IOCTL( vertex, (drm_radeon_vertex2_t *)data, sizeof(vertex) ); @@ -1747,7 +1949,10 @@ int radeon_cp_vertex2( DRM_IOCTL_ARGS ) sizeof(state) ) ) return DRM_ERR(EFAULT); - radeon_emit_state2( dev_priv, &state ); + if ( radeon_emit_state2( dev_priv, filp_priv, &state ) ) { + DRM_ERROR( "radeon_emit_state2 failed\n" ); + return DRM_ERR( EINVAL ); + } laststate = prim.stateidx; } @@ -1784,6 +1989,7 @@ int radeon_cp_vertex2( DRM_IOCTL_ARGS ) static int radeon_emit_packets( drm_radeon_private_t *dev_priv, + drm_file_t *filp_priv, drm_radeon_cmd_header_t header, drm_radeon_cmd_buffer_t *cmdbuf ) { @@ -1798,8 +2004,15 @@ static int radeon_emit_packets( sz = packet[id].len; reg = packet[id].start; - if (sz * sizeof(int) > cmdbuf->bufsz) + if (sz * sizeof(int) > cmdbuf->bufsz) { + DRM_ERROR( "Packet size provided larger than data provided\n" ); return DRM_ERR(EINVAL); + } + + if ( radeon_check_and_fixup_packets( dev_priv, filp_priv, id, data ) ) { + DRM_ERROR( "Packet verification failed\n" ); + return DRM_ERR( EINVAL ); + } BEGIN_RING(sz+1); OUT_RING( CP_PACKET0( reg, (sz-1) ) ); @@ -1882,24 +2095,21 @@ static __inline__ int radeon_emit_vector static int radeon_emit_packet3( drm_device_t *dev, + drm_file_t *filp_priv, drm_radeon_cmd_buffer_t *cmdbuf ) { drm_radeon_private_t *dev_priv = dev->dev_private; - int cmdsz, tmp; - int *cmd = (int *)cmdbuf->buf; + unsigned int cmdsz; + int *cmd = (int *)cmdbuf->buf, ret; RING_LOCALS; - DRM_DEBUG("\n"); - if (DRM_GET_USER_UNCHECKED( tmp, &cmd[0])) - return DRM_ERR(EFAULT); - - cmdsz = 2 + ((tmp & RADEON_CP_PACKET_COUNT_MASK) >> 16); - - if ((tmp & 0xc0000000) != RADEON_CP_PACKET3 || - cmdsz * 4 > cmdbuf->bufsz) - return DRM_ERR(EINVAL); + if ( ( ret = radeon_check_and_fixup_packet3( dev_priv, filp_priv, + cmdbuf, &cmdsz ) ) ) { + DRM_ERROR( "Packet verification failed\n" ); + return ret; + } BEGIN_RING( cmdsz ); OUT_RING_USER_TABLE( cmd, cmdsz ); @@ -1912,27 +2122,25 @@ static int radeon_emit_packet3( drm_devi static int radeon_emit_packet3_cliprect( drm_device_t *dev, + drm_file_t *filp_priv, drm_radeon_cmd_buffer_t *cmdbuf, int orig_nbox ) { drm_radeon_private_t *dev_priv = dev->dev_private; drm_clip_rect_t box; - int cmdsz, tmp; - int *cmd = (int *)cmdbuf->buf; + unsigned int cmdsz; + int *cmd = (int *)cmdbuf->buf, ret; drm_clip_rect_t *boxes = cmdbuf->boxes; int i = 0; RING_LOCALS; DRM_DEBUG("\n"); - if (DRM_GET_USER_UNCHECKED( tmp, &cmd[0])) - return DRM_ERR(EFAULT); - - cmdsz = 2 + ((tmp & RADEON_CP_PACKET_COUNT_MASK) >> 16); - - if ((tmp & 0xc0000000) != RADEON_CP_PACKET3 || - cmdsz * 4 > cmdbuf->bufsz) - return DRM_ERR(EINVAL); + if ( ( ret = radeon_check_and_fixup_packet3( dev_priv, filp_priv, + cmdbuf, &cmdsz ) ) ) { + DRM_ERROR( "Packet verification failed\n" ); + return ret; + } if (!orig_nbox) goto out; @@ -2009,6 +2217,7 @@ int radeon_cp_cmdbuf( DRM_IOCTL_ARGS ) { DRM_DEVICE; drm_radeon_private_t *dev_priv = dev->dev_private; + drm_file_t *filp_priv; drm_device_dma_t *dma = dev->dma; drm_buf_t *buf = 0; int idx; @@ -2023,6 +2232,8 @@ int radeon_cp_cmdbuf( DRM_IOCTL_ARGS ) return DRM_ERR(EINVAL); } + DRM_GET_PRIV_WITH_RETURN( filp_priv, filp ); + DRM_COPY_FROM_USER_IOCTL( cmdbuf, (drm_radeon_cmd_buffer_t *)data, sizeof(cmdbuf) ); @@ -2053,7 +2264,7 @@ int radeon_cp_cmdbuf( DRM_IOCTL_ARGS ) switch (header.header.cmd_type) { case RADEON_CMD_PACKET: DRM_DEBUG("RADEON_CMD_PACKET\n"); - if (radeon_emit_packets( dev_priv, header, &cmdbuf )) { + if (radeon_emit_packets( dev_priv, filp_priv, header, &cmdbuf )) { DRM_ERROR("radeon_emit_packets failed\n"); return DRM_ERR(EINVAL); } @@ -2096,7 +2307,7 @@ int radeon_cp_cmdbuf( DRM_IOCTL_ARGS ) case RADEON_CMD_PACKET3: DRM_DEBUG("RADEON_CMD_PACKET3\n"); - if (radeon_emit_packet3( dev, &cmdbuf )) { + if (radeon_emit_packet3( dev, filp_priv, &cmdbuf )) { DRM_ERROR("radeon_emit_packet3 failed\n"); return DRM_ERR(EINVAL); } @@ -2104,7 +2315,7 @@ int radeon_cp_cmdbuf( DRM_IOCTL_ARGS ) case RADEON_CMD_PACKET3_CLIP: DRM_DEBUG("RADEON_CMD_PACKET3_CLIP\n"); - if (radeon_emit_packet3_cliprect( dev, &cmdbuf, orig_nbox )) { + if (radeon_emit_packet3_cliprect( dev, filp_priv, &cmdbuf, orig_nbox )) { DRM_ERROR("radeon_emit_packet3_clip failed\n"); return DRM_ERR(EINVAL); } @@ -2214,3 +2425,31 @@ int radeon_cp_getparam( DRM_IOCTL_ARGS ) return 0; } + +int radeon_cp_setparam( DRM_IOCTL_ARGS ) { + DRM_DEVICE; + drm_radeon_private_t *dev_priv = dev->dev_private; + drm_file_t *filp_priv; + drm_radeon_setparam_t sp; + + if ( !dev_priv ) { + DRM_ERROR( "%s called with no initialization\n", __FUNCTION__ ); + return DRM_ERR( EINVAL ); + } + + DRM_GET_PRIV_WITH_RETURN( filp_priv, filp ); + + DRM_COPY_FROM_USER_IOCTL( sp, ( drm_radeon_setparam_t* )data, + sizeof( sp ) ); + + switch( sp.param ) { + case RADEON_SETPARAM_FB_LOCATION: + filp_priv->radeon_fb_delta = dev_priv->fb_location - sp.value; + break; + default: + DRM_DEBUG( "Invalid parameter %d\n", sp.param ); + return DRM_ERR( EINVAL ); + } + + return 0; +} --- diff/drivers/char/drm/sis.h 2003-09-30 15:46:12.000000000 +0100 +++ source/drivers/char/drm/sis.h 2004-03-01 15:42:16.000000000 +0000 @@ -62,6 +62,13 @@ [DRM_IOCTL_NR(DRM_IOCTL_SIS_AGP_FREE)] = { sis_ioctl_agp_free, 1, 0 }, \ [DRM_IOCTL_NR(DRM_IOCTL_SIS_FB_INIT)] = { sis_fb_init, 1, 1 } +#define DRIVER_PCI_IDS \ + {0x1039, 0x0300, 0, "SiS 300/305"}, \ + {0x1039, 0x5300, 0, "SiS 540"}, \ + {0x1039, 0x6300, 0, "SiS 630"}, \ + {0x1039, 0x7300, 0, "SiS 730"}, \ + {0, 0, 0, NULL} + #define __HAVE_COUNTERS 5 /* Buffer customization: --- diff/drivers/char/drm/sis_mm.c 2003-09-30 15:46:12.000000000 +0100 +++ source/drivers/char/drm/sis_mm.c 2004-03-01 15:42:16.000000000 +0000 @@ -34,7 +34,11 @@ #include "sis_drv.h" #include "sis_ds.h" #if defined(__linux__) && defined(CONFIG_FB_SIS) +#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,0) #include