commit d25790532370e7448e3d3bd25a17e1e9f1299816 Author: Stephen Rothwell Date: Tue May 22 11:10:19 2007 +1000 Fix headers check fallout commit e8edc6e03a5c8562dc70a6d969f732bdb355a7e7 added an include of linux/jiffies.h in linux/smb_fs.h outside the ifdef __KERNEL__. Signed-off-by: Stephen Rothwell Signed-off-by: Linus Torvalds commit df5f6311f171cc249e39ccd06321b68823ac68f0 Author: David Woodhouse Date: Mon May 21 19:53:35 2007 -0400 Make 'headerscheck' stop immediately on an error This should make it stop immediately after printing the _helpful_ error message, rather than continuing to spit out many pages more of 'CHECK include/linux/foo.h' before eventually coming to a halt with something less obvious. Now I get this... CHECK include/linux/smb_fs.h /shiny/git/linux-2.6/usr/include/linux/smb_fs.h requires linux/jiffies.h, which does not exist in exported headers make[2]: *** [/shiny/git/linux-2.6/usr/include/linux/.check.smb_fs.h] Error 1 make[1]: *** [linux] Error 2 make: *** [headers_check] Error 2 Signed-off-by-if-Sam-says-so: David Woodhouse [ Sam had better say so! This made me waste way too much time. - Linus] Signed-off-by: Linus Torvalds commit 72dd9ca59944f117c719a1cc3fc9010bce0486f2 Author: Jeff Garzik Date: Tue May 22 00:28:11 2007 -0400 partitions/LDM: build fix This from a "tested" patch... Signed-off-by: Jeff Garzik Cc: Anton Altaparmakov Signed-off-by: Linus Torvalds commit efa5bf1dd2cf3cdee0bfe97cfd76ff2296179ae4 Author: Linus Torvalds Date: Mon May 21 18:41:17 2007 -0700 Revert "kbuild: make better section mismatch reports on i386, arm and mips" This reverts commit f892b7d480eec809a5dfbd6e65742b3f3155e50e, which totally broke the build on x86 with CONFIG_RELOCATABLE (which, as far as I can tell, is the only case where it should even matter!) due to a SIGSEGV in modpost. Cc: Sam Ravnborg Cc: Atsushi Nemoto Signed-off-by: Linus Torvalds commit d6f2fe98ebc2dd8f3e9926673e0d8acd2c764e71 Merge: 04fc5fd... 2053ed0... Author: Linus Torvalds Date: Mon May 21 17:45:25 2007 -0700 Merge branch 'upstream-linus' of master.kernel.org:/pub/scm/linux/kernel/git/jgarzik/netdev-2.6 * 'upstream-linus' of master.kernel.org:/pub/scm/linux/kernel/git/jgarzik/netdev-2.6: Add constant for FCS/CRC length (frame check sequence) declance: Remove a dangling spin_unlock_irq() thingy e1000: Don't enable polling in open() (was: e1000: assertion hit in e1000_clean(), kernel 2.6.21.1) commit 04fc5fd39e09c460fd478ea84b34c445dd376759 Merge: 8aee74c... 8bc3fc4... Author: Linus Torvalds Date: Mon May 21 17:44:34 2007 -0700 Merge branch 'upstream-linus' of master.kernel.org:/pub/scm/linux/kernel/git/jgarzik/libata-dev * 'upstream-linus' of master.kernel.org:/pub/scm/linux/kernel/git/jgarzik/libata-dev: libata: bump versions libata: Trim trailing whitespace libata: Kiss post_set_mode goodbye ata_piix: clean up pata_hpt366: Enable bits are unreliable so don't use them libata: Add Seagate STT20000A to DMA blacklist. ahci: disable 64bit dma on sb600 commit 8bc3fc470eb25af4d70c72485cbcb130cc657691 Author: Jeff Garzik Date: Mon May 21 20:26:38 2007 -0400 libata: bump versions Signed-off-by: Jeff Garzik commit a617c09f6d646b60f31efc8afd9f81b752bf21b7 Author: Jeff Garzik Date: Mon May 21 20:14:23 2007 -0400 libata: Trim trailing whitespace Signed-off-by: Jeff Garzik commit bc9a8a7eaa58d802a2b26c75edcc95cb92835db3 Author: Alan Cox Date: Mon May 21 15:20:33 2007 +0100 libata: Kiss post_set_mode goodbye As of the -mm tree we don't have post_set_mode users any more. Signed-off-by: Alan Cox Signed-off-by: Jeff Garzik commit 2f91d81db3060f6d218e32d935b9a18ffca4dde2 Author: Alan Cox Date: Mon May 21 15:15:51 2007 +0100 ata_piix: clean up With cable methods in place we don't need a custom error handler for SATA so get rid of it Signed-off-by: Alan Cox Signed-off-by: Jeff Garzik commit 4349eebf67b2b06eab36f3dad651ac36619ae986 Author: Alan Cox Date: Mon May 21 14:52:49 2007 +0100 pata_hpt366: Enable bits are unreliable so don't use them Various people had problems with both old and new IDE when hpt366 enable bits started getting honoured. It turns out they are not reliable so don't rely on them Signed-off-by: Alan Cox Signed-off-by: Jeff Garzik commit 39f198868dfd4216ff221d741c72a009d995a7ba Author: Dave Jones Date: Mon May 21 14:31:03 2007 -0400 libata: Add Seagate STT20000A to DMA blacklist. http://bugzilla.kernel.org/show_bug.cgi?id=1044 points out an additional hard disk that doesn't handle DMA transfers correctly. This patch is the libata variant of the earlier patch to drivers/ide/ Signed-off-by: Dave Jones Signed-off-by: Jeff Garzik commit c7a42156d99bcea7f8173ba7a6034bbaa2ecb77c Author: Tejun Heo Date: Fri May 18 16:23:19 2007 +0200 ahci: disable 64bit dma on sb600 SB600 claims it can do 64bit DMA but it can't. Disable it. Signed-off-by: Tejun Heo Signed-off-by: Jeff Garzik commit 2053ed02a62f2c66d9cb9b51aea0836a7d8dbabf Author: Auke Kok Date: Tue May 15 15:29:16 2007 -0700 Add constant for FCS/CRC length (frame check sequence) About a dozen drivers that have some form of crc checksumming or offloading use this constant, warranting a global define for it. Signed-off-by: Auke Kok Signed-off-by: Jeff Garzik commit f1e9a4eaea5c22cfb557e58a8cd0ca3bde5160a5 Author: Maciej W. Rozycki Date: Mon May 21 14:33:21 2007 +0100 declance: Remove a dangling spin_unlock_irq() thingy The spin_unlock_irq() invocation in lance_start_xmit() has no matching locking request. The call is already protected by netif_tx_lock, so remove the statement. Signed-off-by: Maciej W. Rozycki Signed-off-by: Jeff Garzik commit 3e1657c8ef53e1cd541cc1e420f3230dc075949b Author: Auke Kok Date: Mon May 21 14:51:35 2007 -0700 e1000: Don't enable polling in open() (was: e1000: assertion hit in e1000_clean(), kernel 2.6.21.1) Herbert Xu wrote: "netif_poll_enable can only be called if you've previously called netif_poll_disable. Otherwise a poll might already be in action and you may get a crash like this." Removing the call to netif_poll_enable in e1000_open should fix this issue, the only other call to netif_poll_enable is in e1000_up() which is only reached after a device reset or resume. Bugzilla: http://bugzilla.kernel.org/show_bug.cgi?id=8455 https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=240339 Tested by Doug Chapman Signed-off-by: Auke Kok Acked-by: Herbert Xu Signed-off-by: Jeff Garzik commit 8aee74c8ee875448cc6d1cf995c9469eb60ae515 Merge: 080e892... 9f81036... Author: Linus Torvalds Date: Mon May 21 16:19:32 2007 -0700 Merge branch 'for-linus' of master.kernel.org:/pub/scm/linux/kernel/git/roland/infiniband * 'for-linus' of master.kernel.org:/pub/scm/linux/kernel/git/roland/infiniband: IB/cm: Improve local id allocation IPoIB/cm: Fix SRQ WR leak IB/ipoib: Fix typos in error messages IB/mlx4: Check if SRQ is full when posting receive IB/mlx4: Pass send queue sizes from userspace to kernel IB/mlx4: Fix check of opcode in mlx4_ib_post_send() mlx4_core: Fix array overrun in dump_dev_cap_flags() IB/mlx4: Fix RESET to RESET and RESET to ERROR transitions IB/mthca: Fix RESET to ERROR transition IB/mlx4: Set GRH:HopLimit when sending globally routed MADs IB/mthca: Set GRH:HopLimit when building MLX headers IB/mlx4: Fix check of max_qp_dest_rdma in modify QP IB/mthca: Fix use-after-free on device restart IB/ehca: Return proper error code if register_mr fails IPoIB: Handle P_Key table reordering IB/core: Use start_port() and end_port() IB/core: Add helpers for uncached GID and P_Key searches IB/ipath: Fix potential deadlock with multicast spinlocks IB/core: Free umem when mm is already gone commit 9f81036c54ed1f860d2807c5a6aa4f2b30c21204 Author: Michael S. Tsirkin Date: Mon May 21 19:06:54 2007 +0300 IB/cm: Improve local id allocation The IB CM uses an idr for local id allocations, with a running counter as start_id. This fails to generate distinct ids if 1. An id is constantly created and destroyed 2. A chunk of ids just beyond the current next_id value is occupied This in turn leads to an increased chance of connection request being mis-detected as a duplicate, sometimes for several retries, until next_id gets past the block of allocated ids. This has been observed in practice. As a fix, remember the last id allocated and start immediately above it. This also fixes a problem with the old code, where next_id might overflow and become negative. Signed-off-by: Michael S. Tsirkin Acked-by: Sean Hefty Signed-off-by: Roland Dreier commit 518b1646f8a31904ca637b8df0c1e31c34a7a3c2 Author: Michael S. Tsirkin Date: Mon May 21 15:04:59 2007 +0300 IPoIB/cm: Fix SRQ WR leak SRQ WR leakage has been observed with IPoIB/CM: e.g. flipping ports on and off will, with time, leak out all WRs and then all connections will start getting RNR NAKs. Fix this in the way suggested by spec: move the QP being destroyed to the error state, wait for "Last WQE Reached" event and then post WR on a "drain QP" connected to the same CQ. Once we observe a completion on the drain QP, it's safe to call ib_destroy_qp. Signed-off-by: Michael S. Tsirkin Signed-off-by: Roland Dreier commit 24bd1e4e32e88cd3d0675482d15bea498a922ca8 Author: Michael S. Tsirkin Date: Fri May 18 16:12:54 2007 +0300 IB/ipoib: Fix typos in error messages Trivial error message fixups. Signed-off-by: Michael S. Tsirkin Signed-off-by: Roland Dreier commit 080e89270a7bfb7d01fac9a67050f8ac6d6cdd11 Merge: 4b85df0... 38bdc32... Author: Linus Torvalds Date: Mon May 21 12:03:04 2007 -0700 Merge git://git.kernel.org/pub/scm/linux/kernel/git/sam/kbuild-fix * git://git.kernel.org/pub/scm/linux/kernel/git/sam/kbuild-fix: mm/slab: fix section mismatch warning mm: fix section mismatch warnings init/main: use __init_refok to fix section mismatch kbuild: introduce __init_refok/__initdata_refok to supress section mismatch warnings all-archs: consolidate .data section definition in asm-generic all-archs: consolidate .text section definition in asm-generic kbuild: add "Section mismatch" warning whitelist for powerpc kbuild: make better section mismatch reports on i386, arm and mips kbuild: make modpost section warnings clearer kconfig: search harder for curses library in check-lxdialog.sh kbuild: include limits.h in sumversion.c for PATH_MAX powerpc: Fix the MODALIAS generation in modpost for of devices commit 4b85df0419d1fb135d9d845876e8c7ee1393f83c Merge: 6044ab3... be2864b... Author: Linus Torvalds Date: Mon May 21 10:01:36 2007 -0700 Merge branch 'for-linus' of git://git390.osdl.marist.edu/pub/scm/linux-2.6 * 'for-linus' of git://git390.osdl.marist.edu/pub/scm/linux-2.6: [S390] More verbose show_mem() like other architectures. [S390] Make use of kretprobe_assert. [S390] Wire up signald, timerfd and eventfd syscalls. [S390] Wire up sys_utimensat. [S390] cio: Update documentation. commit 6044ab324ce4c180bce17aa9be00157c611d7d48 Merge: d07b3c2... 9093bbb... Author: Linus Torvalds Date: Mon May 21 10:00:57 2007 -0700 Merge branch 'master' of master.kernel.org:/pub/scm/linux/kernel/git/davem/net-2.6 * 'master' of master.kernel.org:/pub/scm/linux/kernel/git/davem/net-2.6: [NET]: Fix race condition about network device name allocation. [IPV4]: icmp: fix crash with sysctl_icmp_errors_use_inbound_ifaddr [NETFILTER]: nf_conntrack_ipv4: fix incorrect #ifdef config name [NETFILTER]: nf_conntrack: fix use-after-free in helper destroy callback invocation [IPSEC] pfkey: Load specific algorithm in pfkey_add rather than all [TCP] FRTO: Prevent state inconsistency in corner cases [TCP] FRTO: Add missing ECN CWR sending to one of the responses [NET]: Fix net/core/skbuff.c gcc-3.2.3 compilation error [RFKILL]: Fix check for correct rfkill allocation [IPV6]: Add ip6_tunnel.h to headers_install commit d07b3c25327c5ae3792d0ed0c135dee4727200a1 Merge: dde3334... 6cbdc8c... Author: Linus Torvalds Date: Mon May 21 10:00:22 2007 -0700 Merge master.kernel.org:/home/rmk/linux-2.6-arm * master.kernel.org:/home/rmk/linux-2.6-arm: (22 commits) [ARM] spelling fixes [ARM] at91_adc parenthesis balance [ARM] 4400/1: S3C24XX: Add high-speed MMC device definition [ARM] 4399/2: S3C2443: Fix SMDK2443 nand timings [ARM] 4398/1: S3C2443: Fix watchdog IRQ number [ARM] 4397/1: S3C2443: remove SDI0/1 IRQ ambiguity [ARM] 4396/1: S3C2443: Add missing HCLK clocks [ARM] 4395/1: S3C24XX: add include of to relevant machines [ARM] 4388/1: no need for arm/mm mmap range checks for non-mmu [ARM] 4387/1: fix /proc/cpuinfo formatting for pre-ARM7 parts [ARM] ARMv6: add CPU_HAS_ASID configuration [ARM] integrator: fix pci_v3 compile error with DEBUG_LL [ARM] gic: Fix gic cascade irq handling [ARM] Silence OMAP kernel configuration warning [ARM] Update ARM syscalls [ARM] 4384/1: S3C2412/13 SPI registers offset correction [ARM] 4383/1: iop: fix usage of '__init' and 'inline' in iop files [ARM] 4382/1: iop13xx: fix msi support [ARM] Remove Integrator/CP SMP platform support [ARM] 4378/1: KS8695: Serial driver fix ... commit dde33348e53ecab687a9768bf5262f0b8f79b7f2 Author: Anton Altaparmakov Date: Mon May 21 09:37:42 2007 +0100 LDM: Fix for Windows Vista dynamic disks This fixes the LDM driver so that it works with Windows Vista dynamic disks which are subtly different to Windows 2000/XP ones. The patch was needed to get a Vista formatted dynamic disk to be recognized and parsed successfully. Thanks go to Chris Teachworth for the report and testing. Cc: Richard Russon Signed-off-by: Anton Altaparmakov Signed-off-by: Linus Torvalds commit 17304383ebc1ce68a88030ac4d18ea549d9578c7 Author: Brian Gerst Date: Mon May 21 14:31:53 2007 +0200 i386: fix PGE mask cr4 is a 32-bit register, so casting the mask to an unsigned char is wrong, as it clears more than the PGE bit. Signed-off-by: Brian Gerst Signed-off-by: Andi Kleen Signed-off-by: Linus Torvalds commit d0aff6e6f4e54f79f9c89d147d371bad384454e9 Author: john stultz Date: Mon May 21 14:31:52 2007 +0200 x86_64: vsyscall time() fix The vsyscall time() function basically returns the second portion of xtime directly. This however means that there is about a ticks worth of time each second where time() will return a second value less then what gettimeofday() does. Additionally, this window where vtime() is behind vgettimeofday() grows when dynticks is enabled, so its probably good to get this in before dynticks lands. Big thanks to Sripathi for noticing this issue and creating a test case to work with! This patch changes the vtime() implemenation to call vgettimeofday(), much as syscall time() implementation calls gettimeofday(). 2.6.21 stable candidate too Signed-off-by: John Stultz Signed-off-by: Andi Kleen Signed-off-by: Linus Torvalds commit d8902bfcacde6001e1b11bb06137c3bae3ae52d0 Author: Yinghai Lu Date: Mon May 21 14:31:51 2007 +0200 x86_64: early_print kernel console should send CRLF not LFCR In commit d358788f3f30113e49882187d794832905e42592 Author: Russell King Date: Mon Mar 20 20:00:09 2006 +0000 Glen Turner reported that writing LFCR rather than the more traditional CRLF causes issues with some terminals. Since this afflicts many serial drivers, extract the common code to a library function (uart_console_write) and arrange for each driver to supply a "putchar" function. but early_printk is left out. Signed-off-by: Yinghai Lu Cc: Russell King Signed-off-by: Andi Kleen Signed-off-by: Linus Torvalds commit 39427d6e595ebee38fdd77bcf55d6b13d7a4324a Author: Andi Kleen Date: Mon May 21 14:31:50 2007 +0200 i386: Enable CX8/PGE CPUID bits early on VIA C3 Fix boot failures with the early CPUID checking on VIA C3 Includes fixes from Christian Volkmann Signed-off-by: Andi Kleen Signed-off-by: Linus Torvalds commit 4c1f59d8be7e5da75d9380da23671005b363c45c Author: Christian Volkmann Date: Mon May 21 14:31:48 2007 +0200 i386: Fix wrong CPU error message in early boot path - boot/setup.S did not print "PANIC: CPU too old for this kernel" ( not visible, also the message did not match ) - I add "# missed before: set ds" => somebody should check if I am right with the way to set. => seems to be a generic error in setup.S not to set "ds" for error messages. AK: extracted patch out of other changes AK: also couldn't find any other case where ds is wrong Signed-off-by: Andi Kleen Signed-off-by: Linus Torvalds commit c12ceb766eb5ac975c1145d51236fcdcf81a6578 Author: Andi Kleen Date: Mon May 21 14:31:47 2007 +0200 i386: Clear MCE flag on AMD K6 It reports machine check capability in CPUID, but doesn't actually implement all the necessary MSRs of the standard Intel machine check architecture. This fixes a boot failure on K6s recently introduced. Signed-off-by: Andi Kleen Signed-off-by: Linus Torvalds commit 21124a82bb82e100369846ff2044dd5ea65fc934 Author: Andi Kleen Date: Mon May 21 14:31:46 2007 +0200 x86_64: Support gcc 5 properly The ifdef tests were broken. Assume it acts like gcc 4 Signed-off-by: Andi Kleen Signed-off-by: Linus Torvalds commit 6c977aad03a18019015035958c65b6729cd0574c Author: Andi Kleen Date: Mon May 21 14:31:45 2007 +0200 i386: Fix K8/core2 oprofile on multiple CPUs Only try to allocate MSRs once instead of for every CPU. This assumes the MSRs are the same on all CPUs which is currently true. P4-HT is a special case for different SMT threads, but the code always saves/restores all MSRs so it works identical. Signed-off-by: Andi Kleen Signed-off-by: Linus Torvalds commit 4c87b948b258829f0bdffe51baa72fba5c42d359 Author: Andi Kleen Date: Mon May 21 14:31:44 2007 +0200 x86_64: Support x86_64 in make buildtar Signed-off-by: Andi Kleen Signed-off-by: Linus Torvalds commit 20c3a3d0dd5a91ee8986b94071da0cae8c057af5 Author: Andi Kleen Date: Mon May 21 14:31:43 2007 +0200 i386: Update defconfig Signed-off-by: Andi Kleen Signed-off-by: Linus Torvalds commit 5e200c28955802b26d831f098a67402fa9d20ffc Author: Andi Kleen Date: Mon May 21 14:31:42 2007 +0200 x86_64: Update defconfig Signed-off-by: Andi Kleen Signed-off-by: Linus Torvalds commit 88b4036934b5a6d8cda810487503d2fc4c46d448 Author: Bryan Wu Date: Mon May 21 18:32:16 2007 +0800 Blackfin SPI: cleanup according to David Brownell's review a) platorm_driver_probe(...) instead of platform_driver_register(&driver); b) set bfin_spi_enable and bfin_spi_disable static c) Why is the width flag a u32? d) maybe use dev_dbg() instead of pr_debug() Signed-off-by: Bryan Wu Signed-off-by: Linus Torvalds commit 2ac5ee4738cbd9c146ccda53f02006eaf6a43352 Author: Mike Frysinger Date: Mon May 21 18:09:39 2007 +0800 Blackfin serial driver: implement support for ignoring parity/break errors properly setting up and respecting the read_status_mask / ignore_status_mask fields of the serial core Signed-off-by: Mike Frysinger Signed-off-by: Bryan Wu Signed-off-by: Linus Torvalds commit 5c4e472b0af57553f9584e0b33c491b168ac1dff Author: Aubrey Li Date: Mon May 21 18:09:38 2007 +0800 Blackfin serial driver: fix overhead issue Signed-off-by: Aubrey Li Signed-off-by: Bryan Wu Signed-off-by: Linus Torvalds commit 9d7b6677ac35fdfee93e8f50589e04bcc95649c1 Author: Bryan Wu Date: Mon May 21 18:09:37 2007 +0800 Blackfin arch: update pm.c according to power management API change. Signed-off-by: Bryan Wu Signed-off-by: Linus Torvalds commit c4d1889f7be0a7f34d29e5c4ce00a2b4b78ebb26 Author: Bryan Wu Date: Mon May 21 18:09:36 2007 +0800 Blackfin arch: update defconfig files Signed-off-by: Bryan Wu Signed-off-by: Linus Torvalds commit ca9cb6fd4405a1e2276e17c895297004dd81734d Author: Mike Frysinger Date: Mon May 21 18:09:35 2007 +0800 Blackfin arch: add board default configs to blackfin arch Signed-off-by: Mike Frysinger Signed-off-by: Bryan Wu Signed-off-by: Linus Torvalds commit bde7db8608cab31c6d92982b00e8064bd492a85b Author: Sonic Zhang Date: Mon May 21 18:09:34 2007 +0800 Blackfin arch: Change NO_ACCESS_CHECK to ACCESS_CHECK Signed-off-by: Sonic Zhang Signed-off-by: Bryan Wu Signed-off-by: Linus Torvalds commit c824498db47285aeff430c007685a193b673c949 Author: Bernd Schmidt Date: Mon May 21 18:09:33 2007 +0800 Blackfin arch: fix signal handling bug There's a forum thread at https://blackfin.uclinux.org/gf/project/uclinux-dist/forum/?action=ForumBrowse&_forum_action=MessageReply&message_id=24741 which has a testcase involving signal handling that crashes quite readily. Inspecting the code I believe what happens is that signal handling can become confused when it is invoked on return from an interrupt, if the contents of P0 and R0 at the time of the interrupt happen to be such that P0 is larger than zero (indicating to the signal code that we're in a syscall), and R0 happens to have a value of something like -EINTR or -ERESTARTSYS. Fixed by setting orig_p0 to -1 if we're returning from an interrupt. The testcase now seems to run without problems. Signed-off-by: Bernd Schmidt Signed-off-by: Bryan Wu Signed-off-by: Linus Torvalds commit a9c59c2746c7e773839d51027c0e16ccf41f8fef Author: Mike Frysinger Date: Mon May 21 18:09:32 2007 +0800 Blackfin arch: cache SWRST value at bootup so other things like watchdog can non-destructively query it Signed-off-by: Mike Frysinger Signed-off-by: Bryan Wu Signed-off-by: Linus Torvalds commit 19381f024b01413d83cec1655c3fc4c9c09ae274 Author: Bryan Wu Date: Mon May 21 18:09:31 2007 +0800 Blackfin arch: update blackfin header files to latest one in VDSP. a) add new processor BF52x/BF54x header files b) update blackfin BF533/BF537/BF561 header files to latest one in VDSP. c) scrub watchdog/rtc masks from headers as we dont need/want them (too generic and the drivers dont use them) Signed-off-by: Mike Frysinger Signed-off-by: Roy Huang Signed-off-by: Sonic Zhang Signed-off-by: Bryan Wu Signed-off-by: Linus Torvalds commit c09c4e006590210001ced90d59e62182bfd396f9 Author: Mike Frysinger Date: Mon May 21 18:09:30 2007 +0800 Blackfin arch: make sure we use local labels Signed-off-by: Mike Frysinger Signed-off-by: Bryan Wu Signed-off-by: Linus Torvalds commit ac57ac39e4f92f327451ad4084c4d5f60bcfa999 Author: Mike Frysinger Date: Mon May 21 18:09:29 2007 +0800 Blackfin arch: dont clear the bit that tells coreb to start booting Signed-off-by: Mike Frysinger Signed-off-by: Bryan Wu Signed-off-by: Linus Torvalds commit 5f9a3e899897201a49965cd9e1bb2570b3bde2b5 Author: Aubrey Li Date: Mon May 21 18:09:28 2007 +0800 Blackfin arch: dma_memcpy borken for > 64K Signed-off-by: Aubrey Li Signed-off-by: Bryan Wu Signed-off-by: Linus Torvalds commit 5079df993c3ab45af10dc9a45b3b9e712fd56f23 Author: Mike Frysinger Date: Mon May 21 18:09:27 2007 +0800 Blackfin arch: document why we have to touch the UART peripheral in our boot up code Signed-off-by: Mike Frysinger Signed-off-by: Bryan Wu Signed-off-by: Linus Torvalds commit ef9256d2831df0896566c3823cd2bdf0e55df984 Author: Mike Frysinger Date: Mon May 21 18:09:26 2007 +0800 Blackfin arch: issue reset via SWRST so we dont clobber the watchdog state Signed-off-by: Mike Frysinger Signed-off-by: Bryan Wu Signed-off-by: Linus Torvalds commit c0fc525dcc407a516132fc11af82375319ebdadb Author: Mike Frysinger Date: Mon May 21 18:09:25 2007 +0800 Blackfin arch: move board specific setup out of common init code and into the board specific init code Signed-off-by: Mike Frysinger Signed-off-by: Bryan Wu Signed-off-by: Linus Torvalds commit 95e493c00ac0d2371c3f627fdb99d776d29a8166 Author: Mike Frysinger Date: Mon May 21 18:09:24 2007 +0800 Blackfin arch: finish removing p* volatile defines for MMRs Signed-off-by: Mike Frysinger Signed-off-by: Bryan Wu Signed-off-by: Linus Torvalds commit 30870b93cd76bdfa7385f8d471345f5f6269fb00 Author: Mike Frysinger Date: Mon May 21 18:09:23 2007 +0800 Blackfin arch: dont clear status register bits in SWRST so we can actually use it Signed-off-by: Mike Frysinger Signed-off-by: Bryan Wu Signed-off-by: Linus Torvalds commit eb8d5f6c0322e641ec32ed632dd3b02586205673 Author: Mike Frysinger Date: Mon May 21 18:09:22 2007 +0800 Blackfin arch: make sure we declare the revid functions as pure (since they are) Signed-off-by: Mike Frysinger Signed-off-by: Bryan Wu Signed-off-by: Linus Torvalds commit 0a290593a053752d9629fb1276392ce35c26b348 Author: Mike Frysinger Date: Mon May 21 18:09:21 2007 +0800 Blackfin arch: initial tepla-bf561 board support Signed-off-by: Mike Frysinger Signed-off-by: Bryan Wu Signed-off-by: Linus Torvalds commit 6e537e9329d133526a576b741f85b3d48edc4ac1 Author: Michael Hennerich Date: Mon May 21 18:09:20 2007 +0800 Blackfin arch: Fix bug using usb keyboard crashes kernel Without conswitchp preset, we have the following situation: - During initcalls: con_init is called, and returns because of !display_desc. - At this point there is no memory allocated for vc_cons[].d A bit later vty_init calls kbd_init. - From now on events are passed to kbd_event which will then call kbd_keycode. - kbd_keycode will oops on a NULL pointer dereference on vc->vc_tty Signed-off-by: Michael Hennerich Cc: Pekka Enberg Signed-off-by: Bryan Wu [ Added commit description based on email thread. - Linus ] Signed-off-by: Linus Torvalds commit 8a0e6656030ffe9bcb81b725e956917bafc7522d Author: Mike Frysinger Date: Mon May 21 18:09:19 2007 +0800 Blackfin arch: fix trace output for FLAT binaries Signed-off-by: Mike Frysinger Signed-off-by: Bryan Wu Signed-off-by: Linus Torvalds commit 678402709602a5e62bb79454f0f41bd952373300 Author: Michael Hennerich Date: Mon May 21 18:09:18 2007 +0800 Blackfin arch: GPIO fix some defines Signed-off-by: Michael Hennerich Signed-off-by: Bryan Wu Signed-off-by: Linus Torvalds commit 23ee968d96ef5ad564c0f0b8a30412fb93137ebb Author: Michael Hennerich Date: Mon May 21 18:09:17 2007 +0800 Blackfin arch: DMA operation cleanup 1) Disable Interrupts during DMA memcpy to avoid raise conditions. 2) Mark MDMA channel 0 as reserved, since were using it internally. 3) Add DMA based equivalents for insX and outsX. 4) Our insX and outsX only handles len <= 2^16. Signed-off-by: Michael Hennerich Signed-off-by: Bryan Wu Signed-off-by: Linus Torvalds commit d1b945fdb62f3e6852fea5a8779c0151ac1e55a8 Author: Michael Hennerich Date: Mon May 21 18:09:16 2007 +0800 Blackfin arch: Move write to VR_CTL closer to IDLE Signed-off-by: Michael Hennerich Signed-off-by: Bryan Wu Signed-off-by: Linus Torvalds commit 9ae246cd85e05d3fb31b9770588c1fe0914872fe Author: Michael Hennerich Date: Mon May 21 18:09:15 2007 +0800 Blackfin arch: Fix reserved map after we changed PORT_H definition Signed-off-by: Michael Hennerich Signed-off-by: Bryan Wu Signed-off-by: Linus Torvalds commit a161bb0510d793a450127c12160309907af3d3f7 Author: Mike Frysinger Date: Mon May 21 18:09:14 2007 +0800 Blackfin arch: mark a bunch of local functions as static Signed-off-by: Mike Frysinger Signed-off-by: Bryan Wu Signed-off-by: Linus Torvalds commit 140a9aec22f454120d1953f24cd2b5338b6141e2 Author: Michael Hennerich Date: Mon May 21 18:09:13 2007 +0800 Blackfin arch: Add configuration data for ISP176x on BF561 Signed-off-by: Michael Hennerich Signed-off-by: Bryan Wu Signed-off-by: Linus Torvalds commit 6dae211411448dd9a12895b802bdedca8296f23d Author: Mike Frysinger Date: Mon May 21 18:09:12 2007 +0800 Blackfin arch: fix a few random warnings Signed-off-by: Mike Frysinger Signed-off-by: Bryan Wu Signed-off-by: Linus Torvalds commit e3defffe915961cf9c6e1f16486ae83fb7212f29 Author: Aubrey Li Date: Mon May 21 18:09:11 2007 +0800 Blackfin arch: ISP1761 doesn't work for USB flash disk ZONE_DMA is required for some drivers subsystem, such as USB/SCSI. Signed-off-by: Aubrey Li Signed-off-by: Bryan Wu Signed-off-by: Linus Torvalds commit ed2fbf57ea8d2ec1c0f6758bd5f9f98cef519049 Author: Bryan Wu Date: Mon May 21 18:09:10 2007 +0800 Blackfin arch: add SPI MMC driver support on bf533-stamp, tested on STAMP-BF533 Signed-off-by: Bryan Wu Signed-off-by: Linus Torvalds commit 8af10b79877ff5dfd5c7b1928d57cf38483e819d Author: Michael Hennerich Date: Mon May 21 18:09:09 2007 +0800 Blackfin arch: Add Workaround for ANOMALY 05000257 Signed-off-by: Michael Hennerich Signed-off-by: Bryan Wu Signed-off-by: Linus Torvalds commit e8edc6e03a5c8562dc70a6d969f732bdb355a7e7 Author: Alexey Dobriyan Date: Mon May 21 01:22:52 2007 +0400 Detach sched.h from mm.h First thing mm.h does is including sched.h solely for can_do_mlock() inline function which has "current" dereference inside. By dealing with can_do_mlock() mm.h can be detached from sched.h which is good. See below, why. This patch a) removes unconditional inclusion of sched.h from mm.h b) makes can_do_mlock() normal function in mm/mlock.c c) exports can_do_mlock() to not break compilation d) adds sched.h inclusions back to files that were getting it indirectly. e) adds less bloated headers to some files (asm/signal.h, jiffies.h) that were getting them indirectly Net result is: a) mm.h users would get less code to open, read, preprocess, parse, ... if they don't need sched.h b) sched.h stops being dependency for significant number of files: on x86_64 allmodconfig touching sched.h results in recompile of 4083 files, after patch it's only 3744 (-8.3%). Cross-compile tested on all arm defconfigs, all mips defconfigs, all powerpc defconfigs, alpha alpha-up arm i386 i386-up i386-defconfig i386-allnoconfig ia64 ia64-up m68k mips parisc parisc-up powerpc powerpc-up s390 s390-up sparc sparc-up sparc64 sparc64-up um-x86_64 x86_64 x86_64-up x86_64-defconfig x86_64-allnoconfig as well as my two usual configs. Signed-off-by: Alexey Dobriyan Signed-off-by: Linus Torvalds commit ff1be9ad61e3e17ba83702d8ed0b534e5b8ee15c Author: OGAWA Hirofumi Date: Sun May 20 23:39:40 2007 +0900 Fix "fs: convert core functions to zero_user_page" The bug was introduced by 01f2705daf5a36208e69d7cf95db9c330f843af6. It misses to convert the first argument, it should be "new_page". This became a cause of fatfs corruption. Cc: Nate Diller Signed-off-by: OGAWA Hirofumi Signed-off-by: Linus Torvalds commit 5f663404b27fa72546a592c5c932b0605b9c93a6 Merge: 55b637c... da7cd59... Author: Linus Torvalds Date: Mon May 21 09:00:25 2007 -0700 Merge master.kernel.org:/pub/scm/linux/kernel/git/herbert/crypto-2.6 * master.kernel.org:/pub/scm/linux/kernel/git/herbert/crypto-2.6: [CRYPTO] api: Read module pointer before freeing algorithm commit be2864b5ee46e0d5ed626de6cbfeb9abbd9c2e6f Author: Heiko Carstens Date: Mon May 21 11:25:23 2007 +0200 [S390] More verbose show_mem() like other architectures. Signed-off-by: Heiko Carstens Signed-off-by: Martin Schwidefsky commit a5a60a2beea07c23e242ad8e7398e761935d669a Author: Heiko Carstens Date: Mon May 21 11:25:22 2007 +0200 [S390] Make use of kretprobe_assert. s390 change for git commit 0f95b7fc839bc3272b1bf2325d8748a649bd3534. That is print kprobes debug data before BUG(). Cc: Ananth N Mavinakayanahalli Cc: Maneesh Soni Signed-off-by: Heiko Carstens Signed-off-by: Martin Schwidefsky commit 913f11ee48edfa78b531e0db9385b640fbf9f6d7 Author: Heiko Carstens Date: Mon May 21 11:25:21 2007 +0200 [S390] Wire up signald, timerfd and eventfd syscalls. Signed-off-by: Heiko Carstens Signed-off-by: Martin Schwidefsky commit fd5d8bf03a9843ec3cdb10c820f409aa04048b52 Author: Heiko Carstens Date: Mon May 21 11:25:20 2007 +0200 [S390] Wire up sys_utimensat. Wire up sys_utimensat, reserve syscall number for sys_fallocate and add a couple of syscalls to the ignore list to get rid of warings. Signed-off-by: Heiko Carstens Signed-off-by: Martin Schwidefsky commit 85ee32d02901bfd2e287ac669709758e04d73c4d Author: Cornelia Huck Date: Mon May 21 11:25:19 2007 +0200 [S390] cio: Update documentation. - read_dev_chars()/read_conf_data() are deprecated. Don't document them, but advise to issue the channel program from the driver itself. - Remove some really obsolete and incorrect stuff. Signed-off-by: Cornelia Huck Signed-off-by: Martin Schwidefsky commit 56a8c8b6ac4d6edba5153d17730aaf96ba8f1f8c Author: Roland Dreier Date: Sun May 20 20:19:24 2007 -0700 IB/mlx4: Check if SRQ is full when posting receive Make mlx4_post_srq_recv() fail if the SRQ is full (head == tail). Signed-off-by: Roland Dreier commit 6cbdc8c5357276307a77deeada3f04626ff17da6 Author: Simon Arlott Date: Fri May 11 20:40:30 2007 +0100 [ARM] spelling fixes Spelling fixes in arch/arm/. Signed-off-by: Simon Arlott Signed-off-by: Russell King commit fc432e1952a3899ce35e84b417e5d60f74cb901b Author: Mariusz Kozlowski Date: Sat May 19 00:39:36 2007 +0200 [ARM] at91_adc parenthesis balance Trivial unbalanced parenthesis macro fix. Signed-off-by: Mariusz Kozlowski Signed-off-by: Russell King commit 69e9c93d0d7d9dfa6bdf5144274dbe65a627b63a Author: Ben Dooks Date: Sun May 20 18:13:32 2007 +0100 [ARM] 4400/1: S3C24XX: Add high-speed MMC device definition Add definition for high-speed MMC/SD device and add to SMDK2443 device list. Signed-off-by: Ben Dooks Signed-off-by: Russell King commit 42482e3c77bcb310affdf13e92b93f077d44a7fc Author: Ben Dooks Date: Sun May 20 19:58:10 2007 +0100 [ARM] 4399/2: S3C2443: Fix SMDK2443 nand timings Reduce the Twrph0 timing slightly to fit on an SMDK2443. This should still produce valid timings for the NAND devices as it is still over the smallest device fitted to these boards. Signed-off-by: Ben Dooks <(address hidden)> Signed-off-by: Russell King commit 7e966f3c392ffb5297a6ba0ea701d6a20d1d7292 Author: Ben Dooks Date: Sun May 20 18:02:50 2007 +0100 [ARM] 4398/1: S3C2443: Fix watchdog IRQ number Fix the IRQ number for watchdog on S3C2443 Signed-off-by: Ben Dooks Signed-off-by: Russell King commit 5f2ef21703564d3e6f18fd92f2e72f723a5c3402 Author: Ben Dooks Date: Sun May 20 17:32:39 2007 +0100 [ARM] 4397/1: S3C2443: remove SDI0/1 IRQ ambiguity Change the name of the S3C2443_SDI1 to S3C2443_HSMMC to ensure that it is correctly identified. Signed-off-by: Ben Dooks Signed-off-by: Russell King commit 67364334681503655fbff6c2f2dacfb17f04cadc Author: Ben Dooks Date: Sun May 20 17:17:32 2007 +0100 [ARM] 4396/1: S3C2443: Add missing HCLK clocks Add the clocks missing form HCLKCON back into the set of clocks being registered at initalisation time. Signed-off-by: Ben Dooks Signed-off-by: Russell King commit 333a42e1f4c5c5c2ed7c652254c35b9e98d5c789 Author: Ben Dooks Date: Sun May 20 11:55:53 2007 +0100 [ARM] 4395/1: S3C24XX: add include of to relevant machines Include in any machines that use the PM functions which require struct sys_device. Signed-off-by: Ben Dooks Signed-off-by: Russell King commit 95ba71f76f1cce550cb2e2b345a204d4a926d29c Author: Greg Ungerer Date: Thu May 17 06:22:41 2007 +0100 [ARM] 4388/1: no need for arm/mm mmap range checks for non-mmu We don't need valid_phys_addr_range() or valid_mmap_phys_addr_range() for the !CONFIG_MMU case. Signed-off-by: Greg Ungerer Signed-off-by: Russell King commit 1d28bff7c4ea138032b44d514351b7caceb9fba5 Author: Greg Ungerer Date: Thu May 17 06:12:22 2007 +0100 [ARM] 4387/1: fix /proc/cpuinfo formatting for pre-ARM7 parts Fix the formating of the "CPU part" field to be consistent with the other fields for pre-ARM7 parts. One tab to many for them to all line up. Signed-off-by: Greg Ungerer Signed-off-by: Russell King commit 2446304dd687488c054d0437f2aeef1ef2bfbd02 Author: Eli Cohen Date: Thu May 17 10:32:41 2007 +0300 IB/mlx4: Pass send queue sizes from userspace to kernel Pass the number of WQEs for the send queue and their size from userspace to the kernel to avoid having to keep the QP size calculations in sync between the kernel driver and libmlx4. This fixes a bug seen with the current mlx4_ib driver and current libmlx4 caused by a difference in the calculated sizes for SQ WQEs. Also, this gives more flexibility for userspace to experiment with using multiple WQE BBs for a single SQ WQE. Signed-off-by: Eli Cohen Signed-off-by: Roland Dreier commit 9093bbb2d96d0184f037cea9b4e952a44ebe7c32 Author: Stephen Hemminger Date: Sat May 19 15:39:25 2007 -0700 [NET]: Fix race condition about network device name allocation. Kenji Kaneshige found this race between device removal and registration. On unregister it is possible for the old device to exist, because sysfs file is still open. A new device with 'eth%d' will select the same name, but sysfs kobject register will fial. The following changes the shutdown order slightly. It hold a removes the sysfs entries earlier (on unregister_netdevice), but holds a kobject reference. Then when todo runs the actual last put free happens. Signed-off-by: Stephen Hemminger Signed-off-by: David S. Miller commit d8cf27287ac7fb5cbfcc4139917a997c39d841ca Author: Patrick McHardy Date: Sat May 19 14:44:15 2007 -0700 [IPV4]: icmp: fix crash with sysctl_icmp_errors_use_inbound_ifaddr When icmp_send is called on the local output path before the packet hits ip_output, skb->dev is not set, causing a crash when sysctl_icmp_errors_use_inbound_ifaddr is set. This can happen with the netfilter REJECT target or IPsec tunnels. Let routing decide the ICMP source address in that case, since the packet is locally generated there is no inbound interface and the sysctl should not apply. The option actually seems to be unfixable broken, on the path after ip_output() skb->dev points to the outgoing device and we don't know the incoming device anymore, so its going to do the absolute wrong thing and pick the address of the outgoing interface. Add a comment about this. Reported by Curtis Doty . Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller commit 3ad2a6fb6bcc2f464cdde093a76b76b90b90c66c Author: Patrick McHardy Date: Sat May 19 14:24:16 2007 -0700 [NETFILTER]: nf_conntrack_ipv4: fix incorrect #ifdef config name The option is named CONFIG_NF_NAT not CONFIG_IP_NF_NAT. Remove the ifdef completely since helpers also expect defragmented packet even without NAT. Noticed by Robert P. J. Day Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller commit 5397e97d7533a03b28a7b8aeee648cbb36a8afc6 Author: Patrick McHardy Date: Sat May 19 14:23:52 2007 -0700 [NETFILTER]: nf_conntrack: fix use-after-free in helper destroy callback invocation When the helper module is removed for a master connection that has a fulfilled expectation, but has already timed out and got removed from the hash tables, nf_conntrack_helper_unregister can't find the master connection to unset the helper, causing a use-after-free when the expected connection is destroyed and releases the last reference to the master. The helper destroy callback was introduced for the PPtP helper to clean up expectations and expected connections when the master connection times out, but doing this from destroy_conntrack only works for unfulfilled expectations since expected connections hold a reference to the master, preventing its destruction. Move the destroy callback to the timeout function, which fixes both problems. Reported/tested by Gabor Burjan . Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller commit c92b3a2f1f11655ecf6774b745017a414241d07c Author: Herbert Xu Date: Sat May 19 14:21:18 2007 -0700 [IPSEC] pfkey: Load specific algorithm in pfkey_add rather than all This is a natural extension of the changeset [XFRM]: Probe selected algorithm only. which only removed the probe call for xfrm_user. This patch does exactly the same thing for af_key. In other words, we load the algorithm requested by the user rather than everything when adding xfrm states in af_key. Signed-off-by: Herbert Xu Signed-off-by: David S. Miller commit 580e572a4a1bfea2f42af63ba4785ac7dfbcb45d Author: Ilpo Järvinen Date: Sat May 19 13:56:57 2007 -0700 [TCP] FRTO: Prevent state inconsistency in corner cases State could become inconsistent in two cases: 1) Userspace disabled FRTO by tuning sysctl when one of the TCP flows was in the middle of FRTO algorithm (and then RTO is again triggered) 2) SACK reneging occurs during FRTO algorithm A simple solution is just to abort the previous FRTO when such obscure condition occurs... Signed-off-by: Ilpo Järvinen Signed-off-by: David S. Miller commit 463236557db4b5d4de9eb3fafa2e7d7905ac65ab Author: Ilpo Järvinen Date: Sat May 19 13:56:23 2007 -0700 [TCP] FRTO: Add missing ECN CWR sending to one of the responses The conservative spurious RTO response did not queue CWR even though the sending rate was lowered. Whenever reduction happens regardless of reason, CWR should be sent (forgetting to send it is not very fatal though). A better approach would be to queue CWR when one of the sending rate reducing responses (rate-halving one or this conservative response) is used already at RTO. Doing that would allow CWR to be sent along with the two new data segments that are sent during FRTO. However, it's a bit "racy" because userland could tune the response sysctl to a more aggressive one in between. Signed-off-by: Ilpo Järvinen Signed-off-by: David S. Miller commit b6ccc67d8e42e38936df330b26ee6d022dda8a64 Author: Mikael Pettersson Date: Sat May 19 13:55:25 2007 -0700 [NET]: Fix net/core/skbuff.c gcc-3.2.3 compilation error Compiling 2.6.22-rc1 with gcc-3.2.3 for i486 fails with: gcc -m32 -Wp,-MD,net/core/.skbuff.o.d -nostdinc -isystem /home/mikpe/pkgs/linux-x86/gnu/lib/gcc-lib/i486-pc-linux-gnu/3.2.3/include -D__KERNEL__ -Iinclude -include include/linux/autoconf.h -Wall -Wundef -Wstrict-prototypes -Wno-trigraphs -fno-strict-aliasing -fno-common -O2 -pipe -msoft-float -mregparm=3 -freg-struct-return -mpreferred-stack-boundary=4 -march=i486 -ffreestanding -maccumulate-outgoing-args -DCONFIG_AS_CFI=1 -Iinclude/asm-i386/mach-default -fomit-frame-pointer -D"KBUILD_STR(s)=#s" -D"KBUILD_BASENAME=KBUILD_STR(skbuff)" -D"KBUILD_MODNAME=KBUILD_STR(skbuff)" -c -o net/core/skbuff.o net/core/skbuff.c net/core/skbuff.c:648:1: directives may not be used inside a macro argument net/core/skbuff.c:647:39: unterminated argument list invoking macro "memcpy" net/core/skbuff.c: In function `pskb_expand_head': net/core/skbuff.c:651: `memcpy' undeclared (first use in this function) net/core/skbuff.c:651: (Each undeclared identifier is reported only once net/core/skbuff.c:651: for each function it appears in.) net/core/skbuff.c:651: syntax error before "skb" make[2]: *** [net/core/skbuff.o] Error 1 make[1]: *** [net/core] Error 2 make: *** [net] Error 2 The patch below implements a simple workaround which is to clone the offending memcpy() call and specialise it for the two different scenarios. Other workarounds are of course possible: e.g. bind the varying parameter in a local variable, or use a macro or inline function to perform the varying computation. Signed-off-by: Mikael Pettersson Signed-off-by: David S. Miller commit d007da1fa6f0ad5e01ceae4a1f60cdbb23ecd706 Author: Ivo van Doorn Date: Sat May 19 12:24:39 2007 -0700 [RFKILL]: Fix check for correct rfkill allocation coverity has spotted a bug in rfkill.c (bug id #1627), in rfkill_allocate() NULL was returns if the kzalloc() works, and deref the NULL pointer if it fails, Signed-off-by: Ivo van Doorn Signed-off-by: David S. Miller commit c34b19fb4e243b26fa90a322bf32238356493674 Author: Deepak Saxena Date: Sat May 19 12:00:11 2007 -0700 [IPV6]: Add ip6_tunnel.h to headers_install The Mobile IPv6 package (http://www.mobile-ipv6.org/software/) needs this header file to build the tunnelctl component. The header already looks sanitized so is safe to export. Signed-off-by: Deepak Saxena Signed-off-by: David S. Miller commit 59b0ed121297b57abb2352bdc8313959e7cb5635 Author: Roland Dreier Date: Sat May 19 08:51:58 2007 -0700 IB/mlx4: Fix check of opcode in mlx4_ib_post_send() wr->opcode is invalid if it's >= ARRAY_SIZE(mlx4_ib_opcode), not just strictly >. This was spotted by the Coverity checker (CID 1643). Signed-off-by: Roland Dreier commit 23c15c21d34a4b4b4d7b9a95ce498991c5339c77 Author: Roland Dreier Date: Sat May 19 08:51:57 2007 -0700 mlx4_core: Fix array overrun in dump_dev_cap_flags() Don't overrun fname[] array when decoding device flags. This was spotted by the Coverity checker (CID 1642). Signed-off-by: Roland Dreier commit 65adfa911a3522c1e40e55afd472dd571dc2431b Author: Michael S. Tsirkin Date: Mon May 14 07:26:51 2007 +0300 IB/mlx4: Fix RESET to RESET and RESET to ERROR transitions According to the IB spec, a QP can be moved from RESET back to RESET or to the ERROR state, but mlx4 firmware does not support this and returns an error if we try. Fix the RESET to RESET transition by just returning 0 without doing anything, and fix RESET to ERROR by moving the QP from RESET to INIT with dummy parameters and then transitioning from INIT to ERROR. Signed-off-by: Michael S. Tsirkin Signed-off-by: Roland Dreier commit b18aad7150c85cc86a66be8a1c744b63b41b36e0 Author: Michael S. Tsirkin Date: Mon May 14 07:26:51 2007 +0300 IB/mthca: Fix RESET to ERROR transition According to the IB spec, a QP can be moved from RESET to the ERROR state, but mthca firmware does not support this and returns an error if we try. Work around this FW limitation by moving the QP from RESET to INIT with dummy parameters and then transitioning from INIT to ERROR. Signed-off-by: Michael S. Tsirkin Signed-off-by: Roland Dreier commit 1526130351b31c792ced90c6c5ee08df955696c1 Author: Roland Dreier Date: Sat May 19 08:51:57 2007 -0700 IB/mlx4: Set GRH:HopLimit when sending globally routed MADs This is the same issue discovered in mthca by Rolf Manderscheid . Signed-off-by: Roland Dreier commit 3f37cae6944de9d380c83f717f73d91ec6255d98 Author: Rolf Manderscheid Date: Thu May 17 09:45:48 2007 -0600 IB/mthca: Set GRH:HopLimit when building MLX headers Global CM packets used by rmda_cm were being sent with a GRH:hopLimit of zero, causing them to be dropped by the router. The problem is a missing initialization of the hop_limit field in mthca_read_ah(), which was called by build_mlx_header() when sending a MAD on QP1. Signed-off-by: Rolf Manderscheid Signed-off-by: Roland Dreier commit 1f8f7b7a7b885a0041a21b3d93c507269baf57c8 Author: Eli Cohen Date: Thu May 17 16:32:39 2007 +0300 IB/mlx4: Fix check of max_qp_dest_rdma in modify QP max_qp_dest_rdma is already in natural units - no need to shift. This was discovered by a test that deliberately requests more outstanding atomic operation than the device supports. Found by Sagi Rotem at Mellanox. Signed-off-by: Eli Cohen Signed-off-by: Roland Dreier commit de57c9f102ad7bdc8afa5a1560748cf4f1c18b8e Author: Ali Ayoub Date: Thu May 17 20:58:30 2007 +0300 IB/mthca: Fix use-after-free on device restart Signed-off-by: Michael S. Tsirkin Signed-off-by: Roland Dreier commit bd5a6ccc0e6d8eed3047b4af0e5c1e7168869cd8 Author: Hoang-Nam Nguyen Date: Wed May 16 14:50:55 2007 +0200 IB/ehca: Return proper error code if register_mr fails Set the return code of ehca_register_mr() to ENOMEM if the corresponding firmware call fails due to out of resources. Some other error codes were explicitly mapped to EINVAL -- just remove those cases so they get mapped to the default case, which already returns EINVAL anyway. Signed-off-by: Hoang-Nam Nguyen Signed-off-by: Roland Dreier commit 26bbf13ce1ca21ec69175bcc4b995cb8ffdf8593 Author: Yosef Etigin Date: Sat May 19 08:51:54 2007 -0700 IPoIB: Handle P_Key table reordering SM reconfiguration or failover possibly causes a shuffling of the values in the P_Key table. Right now, IPoIB only queries for the P_Key index once when it creates the device QP, and hence there are problems if the index of a P_Key value changes. Fix this by using the PKEY_CHANGE event to trigger a recheck of the P_Key index. Signed-off-by: Yosef Etigin Acked-by: Michael S. Tsirkin Signed-off-by: Roland Dreier commit 1af4c435f3ab9cdf72ce86c35a455c8bef1d6536 Author: Roland Dreier Date: Sat May 19 08:51:54 2007 -0700 IB/core: Use start_port() and end_port() Clean up ib_query_port() and ib_modify_port() slightly by using the just-added start_port() and end_port() helpers. Signed-off-by: Roland Dreier commit 5eb620c81ce35aa0c533131bf4d06c4c8c2bfadf Author: Yosef Etigin Date: Mon May 14 07:26:51 2007 +0300 IB/core: Add helpers for uncached GID and P_Key searches Add ib_find_gid() and ib_find_pkey() functions that use uncached device queries. The calls might block but the returns are always up-to-date. Cache P_Key and GID table lengths in core to avoid extra port info queries. Signed-off-by: Yosef Etigin Acked-by: Michael S. Tsirkin Signed-off-by: Roland Dreier commit 8b8c8bca3a63073bac20f0fca178e00fdf7f5a09 Author: Roland Dreier Date: Sat May 19 08:51:53 2007 -0700 IB/ipath: Fix potential deadlock with multicast spinlocks Lockdep found the following potential deadlock between mcast_lock and n_mcast_grps_lock: mcast_lock is taken from both interrupt context and process context, so spin_lock_irqsave() must be used to take it. n_mcast_grps_lock is only taken from process context, so at first it seems safe to take it with plain spin_lock(); however, it also nests inside mcast_lock, and hence we could deadlock: cpu A cpu B ipath_mcast_add(): spin_lock_irq(&mcast_lock); ipath_mcast_detach(): spin_lock(&n_mcast_grps_lock); ipath_mcast_find(): spin_lock_irqsave(&mcast_lock); spin_lock(&n_mcast_grps_lock); Fix this by using spin_lock_irq() to take n_mcast_grps_lock. Signed-off-by: Roland Dreier commit 7b82cd8ee7374f803a3daf9a6cbc6eb4bbb10a63 Author: Eli Cohen Date: Mon May 14 11:35:43 2007 +0300 IB/core: Free umem when mm is already gone Free umem when task's mm is already destroyed by the time ib_umem_release gets called. Found by Dotan Barak at Mellanox. Signed-off-by: Eli Cohen Signed-off-by: Roland Dreier commit 38bdc32af442b6ab09ed61b8b669072098c95dd2 Author: Sam Ravnborg Date: Thu May 17 23:48:19 2007 +0200 mm/slab: fix section mismatch warning Use the new __init_refok marker to avoid the section mismatch warning from slab.c Signed-off-by: Sam Ravnborg commit 577a32f620271416d05f852477151fb51c790bc6 Author: Sam Ravnborg Date: Thu May 17 23:29:25 2007 +0200 mm: fix section mismatch warnings modpost had two cases hardcoded for mm/ Shift over to __init_refok and kill the hardcoded function names in modpost. This has the drawback that the functions will always be kept no matter configuration. With previous code the function were placed in init section if configuration allowed it. Signed-off-by: Sam Ravnborg commit 92080309df1975729a9f8b45fd56528817e34db8 Author: Sam Ravnborg Date: Thu May 17 20:43:54 2007 +0200 init/main: use __init_refok to fix section mismatch Kill a special case in modpost by introducing the __init_refok marker. Signed-off-by: Sam Ravnborg commit 0e0d314e6a01bb14d303e35e6f7ba24b17020044 Author: Sam Ravnborg Date: Thu May 17 20:14:48 2007 +0200 kbuild: introduce __init_refok/__initdata_refok to supress section mismatch warnings Throughout the kernel there are a few legitimite references to init or exit sections. Most of these are covered by the patterns included in modpost but a few nees special attention. To avoid hardcoding a lot of function names in modpost introduce a marker so relevant function/data can be marked. When modpost see a reference to a init/exit function from a function/data marked no warning will be issued. Idea from: Andrew Morton Signed-off-by: Sam Ravnborg Cc: Andrew Morton commit ca967258b69eb65dcb07bbab90fdf964c6d2ec45 Author: Sam Ravnborg Date: Thu May 17 13:38:44 2007 +0200 all-archs: consolidate .data section definition in asm-generic With this consolidation we can now modify the .data section definition in one spot for all archs. Signed-off-by: Sam Ravnborg commit 7664709b44a13e2e0b545e2dd8e7b8797a1748dc Author: Sam Ravnborg Date: Sun May 13 00:31:33 2007 +0200 all-archs: consolidate .text section definition in asm-generic Move definition of .text section to asm-generic. Signed-off-by: Sam Ravnborg commit cd5477911fc9f5cc64678e2b95cdd606c59a11b5 Author: Li Yang Date: Mon May 14 18:04:28 2007 +0800 kbuild: add "Section mismatch" warning whitelist for powerpc This patch fixes the following class of "Section mismatch" warnings when building powerpc platforms. WARNING: arch/powerpc/kernel/built-in.o - Section mismatch: reference to .init.data:.got2 from prom_entry (offset 0x0) WARNING: arch/powerpc/platforms/built-in.o - Section mismatch: reference to .init.text:mpc8313_rdb_probe from .machine.desc after 'mach_mpc8313_rdb' (at offset 0x4) .... Signed-off-by: Li Yang Signed-off-by: Sam Ravnborg commit f892b7d480eec809a5dfbd6e65742b3f3155e50e Author: Atsushi Nemoto Date: Thu May 17 01:14:38 2007 +0900 kbuild: make better section mismatch reports on i386, arm and mips On i386, ARM and MIPS, warn_sec_mismatch() sometimes fails to show usefull symbol name. This is because empty 'refsym' due to 0 r_addend value. This patch is to adjust r_addend value, consulting with apply_relocate() routine in kernel code. Without this patch: MODPOST vmlinux WARNING: init/built-in.o - Section mismatch: reference to .init.text: from .text between 'rest_init' (at offset 0xf4) and 'try_name' WARNING: mm/built-in.o - Section mismatch: reference to .init.text: from .text between 'kmem_cache_create' (at offset 0x18a39) and 'cache_reap' WARNING: mm/built-in.o - Section mismatch: reference to .init.text: from .text between 'kmem_cache_create' (at offset 0x18a6b) and 'cache_reap' With this patch: MODPOST vmlinux WARNING: mm/built-in.o - Section mismatch: reference to .init.text:set_up_list3s from .text between 'kmem_cache_create' (at offset 0x18a39) and 'cache_reap' WARNING: mm/built-in.o - Section mismatch: reference to .init.text:set_up_list3s from .text between 'kmem_cache_create' (at offset 0x18a6b) and 'cache_reap' Now modpost can detect "kernel_init" name (and whitelist it) and show "set_up_list3s" name. Signed-off-by: Atsushi Nemoto Signed-off-by: Sam Ravnborg commit 2560120997403581dd824e5bd2389c719edcbf12 Author: Russell King Date: Thu May 10 23:03:25 2007 +0100 kbuild: make modpost section warnings clearer Change modpost section mismatch warnings to be less confusing; model them on the binutils linker warnings which we all know how to interpret. Also, fix the wrong ordering of arguments for the final case - fromsec and refsymname were reversed. Signed-off-by: Russell King Acked-by: Acked-by: David S. Miller Signed-off-by: Sam Ravnborg commit 03c9587d752669a12fd553b0cbd835f77b176607 Author: Mike Frysinger Date: Thu May 17 15:06:31 2007 -0400 kconfig: search harder for curses library in check-lxdialog.sh The check-lxdialog.sh script searches for "libFOO.so" which fails on OS X, due to their special naming of libraries like "libfoo.dylib". This patch turns the curses lib search into extensible loops and adds dylib as a valid extension. Signed-off-by: Mike Frysinger Signed-off-by: Sam Ravnborg commit fc31c7716355a226b8ed4e16f4581e5c8fa53570 Author: Mike Frysinger Date: Thu May 17 14:57:20 2007 -0400 kbuild: include limits.h in sumversion.c for PATH_MAX POSIX says limits.h defines PATH_MAX so we should include it (which fixes compiling on some systems like OS X). Signed-off-by: Mike Frysinger Signed-off-by: Sam Ravnborg commit d1ab423502e787e264b4797a5fa200d804c4fd63 Author: Sylvain Munaut Date: Tue May 8 19:59:29 2007 +1000 powerpc: Fix the MODALIAS generation in modpost for of devices Since the devices may have multiple (or none) compatible properties, the uevent generated internally by the kernel may have multiple "C..." entries. So the MODALIAS stored in the module must have wilcard before and after the compatible entry. Also, if the 'compatible' field is not used for matching, there will be no 'C' and that must handled as well. The previous code handled all those case incorrectly and it "mostly" worked ... but not always. Signed-off-by: Sylvain Munaut Signed-off-by: Paul Mackerras Signed-off-by: Sam Ravnborg commit da7cd59ab9c8ed233df4809f6c8c90c636f676c7 Author: Herbert Xu Date: Sat May 19 14:51:00 2007 +1000 [CRYPTO] api: Read module pointer before freeing algorithm The function crypto_mod_put first frees the algorithm and then drops the reference to its module. Unfortunately we read the module pointer which after freeing the algorithm and that pointer sits inside the object that we just freed. So this patch reads the module pointer out before we free the object. Thanks to Luca Tettamanti for reporting this. Signed-off-by: Herbert Xu commit 516793c61b3db1f60e0b0d0e3c382bcca9ae84fd Author: Russell King Date: Thu May 17 10:19:23 2007 +0100 [ARM] ARMv6: add CPU_HAS_ASID configuration Presently, we check for the minimum ARM architecture that we're building for to determine whether we need ASID support. This is wrong - if we're going to support a range of CPUs which include ARMv6 or higher, we need the ASID. Convert the checks to use a new configuration symbol, and arrange for ARMv6 and higher CPU entries to select it. Signed-off-by: Russell King commit c6af66b9fe93990c70aaee53ce3ce7e53a83676a Author: Russell King Date: Thu May 17 10:16:55 2007 +0100 [ARM] integrator: fix pci_v3 compile error with DEBUG_LL If DEBUG_LL is enabled, we want to use get_irq_regs(), but this causes a build error due to the inline function missing. Add the necessary header file. Signed-off-by: Russell King commit 0f347bb9136f55ff575d55441a29e92c16e87fb0 Author: Russell King Date: Thu May 17 10:11:34 2007 +0100 [ARM] gic: Fix gic cascade irq handling No need for the cascade irq function to have a "fastcall" annotation. Fix the range checking for valid IRQ numbers - comparing the value returned by the GIC with NR_IRQS is meaningless since we translate the GIC irq number to a Linux IRQ number afterwards. Check the GIC returned IRQ number is within limits first, then add the IRQ offset, and only then compare with NR_IRQS. Signed-off-by: Russell King commit f3270f6ef7d60251617f9d4c08f76ec01c647aa4 Author: Russell King Date: Wed May 16 17:36:17 2007 +0100 [ARM] Silence OMAP kernel configuration warning arch/arm/mach-omap1/Kconfig:41:warning: 'select' used by config symbol 'MACH_OMAP_H3' refers to undefined symbol 'GPIOEXPANDER_OMAP' Signed-off-by: Russell King commit c05107911a38235fec87892d83dc54aba9aaf3bd Author: Russell King Date: Wed May 16 12:41:15 2007 +0100 [ARM] Update ARM syscalls Add utimensat, signalfd, timerfd, eventfd syscalls. Add ignore defines for sync_file_range and fadvise64_64 which we implement differently. Signed-off-by: Russell King commit e903382ceae1dd85e650ffc7e98facdd59cc7a3f Author: Sandeep Sanjay Patil Date: Wed May 16 10:51:45 2007 +0100 [ARM] 4384/1: S3C2412/13 SPI registers offset correction Change the SPI Channel 1 register offset in s3c_spi1_resource[], and s3c2412_dma_mappings[]. Offset has to be 0x100 in s3c2412/13's case. Also, total SPI memory resource size changed to 0x24 for s3c2412/13. Signed-off-by: Sandeep Patil Signed-off-by: Ben Dooks Signed-off-by: Russell King commit d73d8011779292788def2cd2520d6f39d9b406de Author: Dan Williams Date: Tue May 15 01:03:36 2007 +0100 [ARM] 4383/1: iop: fix usage of '__init' and 'inline' in iop files WARNING: arch/arm/mach-iop13xx/built-in.o - Section mismatch: reference to .init.text:iop13xx_pcie_map_irq from .text between 'iop13xx_pci_setup' (at offset 0x7fc) and 'iop13xx_map_pci_memory' While fixing this warning I also recalled Adrian Bunk's recommendation to not use inline in .c files, as 'iop13xx_map_pci_memory' is needlessly inlined. Removing 'inline' uncovered some dead code so that is cleaned up as well. Signed-off-by: Dan Williams Signed-off-by: Russell King commit e702a7155d14a6e11645e17d829217ae98fd45bb Author: Dan Williams Date: Tue May 15 01:03:31 2007 +0100 [ARM] 4382/1: iop13xx: fix msi support updates iop13xx msi support for a msi api change in 2.6.22 rev7: * update for Michael Ellerman's "MSI: arch must connect the irq and the msi_desc" patch Signed-off-by: Dan Williams Signed-off-by: Russell King commit 06ba255571b0a2f1fb7e49e7406364e4194f4e10 Author: Russell King Date: Mon May 14 22:56:32 2007 +0100 [ARM] Remove Integrator/CP SMP platform support The Integrator/CP SMP platform support was never fully merged, and now it's causing build breakage. Remove it. Signed-off-by: Russell King commit e45c7a43825b0feb161ef2ebe5324cd517dd4388 Author: Andrew Victor Date: Mon May 14 14:32:43 2007 +0100 [ARM] 4378/1: KS8695: Serial driver fix A fix for a really stupid typo in the KS8695 serial driver. Signed-off-by: Andrew Victor Signed-off-by: Russell King commit 8858e9afdf31bab05247b1680b16c65fe7edbbe1 Author: Milan Svoboda Date: Mon May 14 07:50:42 2007 +0100 [ARM] 4376/1: Selects GENERIC_GPIO for ARCH_IXP4XX in Kconfig Selects GENERIC_GPIO for ARCH_IXP4XX in Kconfig. IXP4XX has generic GPIO support; however, ARCH_IXP4XX Kconfig entry currently does not select GENERIC_GPIO like other arch entries. Signed-off-by: Milan Svoboda Signed-off-by: Russell King commit aceb6f0b6da20a86516d59f73ecdc7ea3c4a23da Author: Richard Purdie Date: Sun May 13 13:32:23 2007 +0100 [ARM] 4375/1: sharpsl_pm: Fix compile warnings Fix compile warnings from sharpsl_pm.c. Attribute registration failure doesn't stop the driver working so just warn about it. Signed-off-by: Richard Purdie Signed-off-by: Russell King