1. 27 12月, 2007 2 次提交
  2. 24 12月, 2007 1 次提交
  3. 21 12月, 2007 2 次提交
  4. 20 12月, 2007 3 次提交
  5. 19 12月, 2007 5 次提交
  6. 18 12月, 2007 16 次提交
    • A
      block: let elv_register() return void · 2fdd82bd
      Adrian Bunk 提交于
      elv_register() always returns 0, and there isn't anything it does where
      it should return an error (the only error condition is so grave that
      it's handled with a BUG_ON).
      Signed-off-by: NAdrian Bunk <bunk@kernel.org>
      Signed-off-by: NJens Axboe <jens.axboe@oracle.com>
      2fdd82bd
    • C
      quicklist: Set tlb->need_flush if pages are remaining in quicklist 0 · 421d9919
      Christoph Lameter 提交于
      This ensures that the quicklists are drained. Otherwise draining may only
      occur when the processor reaches an idle state.
      
      Fixes fatal leakage of pgd_t's on 2.6.22 and later.
      Signed-off-by: NChristoph Lameter <clameter@sgi.com>
      Reported-by: NDhaval Giani <dhaval@linux.vnet.ibm.com>
      Cc: <stable@kernel.org>
      Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
      Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>
      421d9919
    • N
      Revert "hugetlb: Add hugetlb_dynamic_pool sysctl" · 368d2c63
      Nishanth Aravamudan 提交于
      This reverts commit 54f9f80d ("hugetlb:
      Add hugetlb_dynamic_pool sysctl")
      
      Given the new sysctl nr_overcommit_hugepages, the boolean dynamic pool
      sysctl is not needed, as its semantics can be expressed by 0 in the
      overcommit sysctl (no dynamic pool) and non-0 in the overcommit sysctl
      (pool enabled).
      
      (Needed in 2.6.24 since it reverts a post-2.6.23 userspace-visible change)
      Signed-off-by: NNishanth Aravamudan <nacc@us.ibm.com>
      Acked-by: NAdam Litke <agl@us.ibm.com>
      Cc: William Lee Irwin III <wli@holomorphy.com>
      Cc: Dave Hansen <haveblue@us.ibm.com>
      Cc: David Gibson <david@gibson.dropbear.id.au>
      Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
      Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>
      368d2c63
    • N
      hugetlb: introduce nr_overcommit_hugepages sysctl · d1c3fb1f
      Nishanth Aravamudan 提交于
      hugetlb: introduce nr_overcommit_hugepages sysctl
      
      While examining the code to support /proc/sys/vm/hugetlb_dynamic_pool, I
      became convinced that having a boolean sysctl was insufficient:
      
      1) To support per-node control of hugepages, I have previously submitted
      patches to add a sysfs attribute related to nr_hugepages. However, with
      a boolean global value and per-mount quota enforcement constraining the
      dynamic pool, adding corresponding control of the dynamic pool on a
      per-node basis seems inconsistent to me.
      
      2) Administration of the hugetlb dynamic pool with multiple hugetlbfs
      mount points is, arguably, more arduous than it needs to be. Each quota
      would need to be set separately, and the sum would need to be monitored.
      
      To ease the administration, and to help make the way for per-node
      control of the static & dynamic hugepage pool, I added a separate
      sysctl, nr_overcommit_hugepages. This value serves as a high watermark
      for the overall hugepage pool, while nr_hugepages serves as a low
      watermark. The boolean sysctl can then be removed, as the condition
      
      	nr_overcommit_hugepages > 0
      
      indicates the same administrative setting as
      
      	hugetlb_dynamic_pool == 1
      
      Quotas still serve as local enforcement of the size of the pool on a
      per-mount basis.
      
      A few caveats:
      
      1) There is a race whereby the global surplus huge page counter is
      incremented before a hugepage has allocated. Another process could then
      try grow the pool, and fail to convert a surplus huge page to a normal
      huge page and instead allocate a fresh huge page. I believe this is
      benign, as no memory is leaked (the actual pages are still tracked
      correctly) and the counters won't go out of sync.
      
      2) Shrinking the static pool while a surplus is in effect will allow the
      number of surplus huge pages to exceed the overcommit value. As long as
      this condition holds, however, no more surplus huge pages will be
      allowed on the system until one of the two sysctls are increased
      sufficiently, or the surplus huge pages go out of use and are freed.
      
      Successfully tested on x86_64 with the current libhugetlbfs snapshot,
      modified to use the new sysctl.
      Signed-off-by: NNishanth Aravamudan <nacc@us.ibm.com>
      Acked-by: NAdam Litke <agl@us.ibm.com>
      Cc: William Lee Irwin III <wli@holomorphy.com>
      Cc: Dave Hansen <haveblue@us.ibm.com>
      Cc: David Gibson <david@gibson.dropbear.id.au>
      Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
      Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>
      d1c3fb1f
    • A
      apm_event{,info}_t are userspace types · 8d936626
      Adam Jackson 提交于
      These types define the size of data read from /dev/apm_bios.  They should
      not be hidden behind #ifdef __KERNEL__.
      
      This is killing my xserver compile, apm_event_t is used in the xserver
      source.
      Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
      Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>
      8d936626
    • I
      alpha: build fixes · 9548b209
      Ivan Kokshaysky 提交于
      This fixes some of the alpha-specific build problems, except a) modpost
      warning about COMMON symbol "saved_config" and b) nasty final link
      failure with gcc-4.x, -Os and scsi-disk driver configured built-in
      (due to jump table in .rodata referencing discarded .exit.text).
      
      - build failure with gcc-4.2.x: fix up casts in cia_io* routines to avoid
        warnings ('discards qualifiers from pointer target type'), which are
        failures, thanks to -Werror;
      - modpost warnings: add missing __init qualifier for titan and marvel;
        for non-generic build, move machine vectors from .data to .data.init.refok
        section;
      - unbreak CPU-specific optimization: rearrange cpuflags-y assignments
        so that extended -mcpu value (ev56, pca56, ev67) overrides basic
        one (ev5, ev6) and not vice versa.
      Signed-off-by: NIvan Kokshaysky <ink@jurassic.park.msu.ru>
      Cc: Richard Henderson <rth@twiddle.net>
      Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
      Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>
      9548b209
    • A
      fix headers_install · 75527135
      Andrew Morton 提交于
      make[3]: *** No rule to make target `/usr/src/devel/include/linux/ticable.h', needed by `/usr/src/devel/usr/include/linux/ticable.h'.  Stop.
      Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
      Acked-by: NGreg Kroah-Hartman <gregkh@suse.de>
      Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>
      75527135
    • T
      libata: fix ATAPI draining · 140b5e59
      Tejun Heo 提交于
      With ATAPI transfer chunk size properly programmed, libata PIO HSM
      should be able to handle full spurious data chunks.  Also, it's a good
      idea to suppress trailing data warning for misc ATAPI commands as
      there can be many of them per command - for example, if the chunk size
      is 16 and the drive tries to transfer 510 bytes, there can be 31
      trailing data messages.
      
      This patch makes the following updates to libata ATAPI PIO HSM
      implementation.
      
      * Make it drain full spurious chunks.
      
      * Suppress trailing data warning message for misc commands.
      
      * Put limit on how many bytes can be drained.
      
      * If odd, round up consumed bytes and the number of bytes to be
        drained.  This gets the number of bytes to drain right for drivers
        which do 16bit PIO.
      
      This patch is partial backport of improve-ATAPI-data-xfer patchset
      pending for #upstream.
      Signed-off-by: NTejun Heo <htejun@gmail.com>
      Signed-off-by: NJeff Garzik <jeff@garzik.org>
      140b5e59
    • T
      libata-acpi: implement dev->gtf_cache and evaluate _GTF right after _STM during resume · 398e0782
      Tejun Heo 提交于
      On certain implementations, _GTF evaluation depends on preceding _STM
      and both can be pretty picky about the configuration.  Using _GTM
      result cached during controller initialization satisfies the most
      neurotic _STM implementation.  However, libata evaluates _GTF after
      reset during device configuration and the hardware state can be
      different from what _GTF expects and can cause evaluation failure.
      
      This patch adds dev->gtf_cache and updates ata_dev_get_GTF() such that
      it uses the cached value if available.  Cache is cleared with a call
      to ata_acpi_clear_gtf().
      
      Because for SATA ACPI nodes _GTF must be evaluated after _SDD which
      can't be done till IDENTIFY is complete, _GTF caching from
      ata_acpi_on_resume() is used only for IDE ACPI nodes.
      Signed-off-by: NTejun Heo <htejun@gmail.com>
      Signed-off-by: NJeff Garzik <jeff@garzik.org>
      398e0782
    • T
      libata-acpi: implement and use ata_acpi_init_gtm() · c05e6ff0
      Tejun Heo 提交于
      _GTM fetches currently configured transfer mode while _STM configures
      controller according to _GTM parameter and prepares transfer mode
      configuration TFs for _GTF.  In many cases _GTM and _STM
      implementations are quite brittle and can't cope with configuration
      changed by libata.
      
      libata does not depend on ATA ACPI to configure devices.  The only
      reason libata performs _GTM and _STM are to make _GTF evaluation
      succeed and libata also doesn't care about how _GTF TFs configure
      transfer mode.  It overrides that configuration anyway, so from
      libata's POV, it doesn't matter what value is feeded to _STM as long
      as evaluation succeeds for _STM and following _GTF.
      
      This patch adds dev->__acpi_init_gtm and store initial _GTM values on
      host initialization before modified by reset and mode configuration.
      If the field is valid, ata_acpi_init_gtm() returns pointer to the
      saved _GTM structure; otherwise, NULL.
      
      This saved value is used for _STM during resume and peek at
      BIOS/firmware programmed initial timing for later use.  The accessor
      is there to make building w/o ACPI easy as dev->__acpi_init doesn't
      exist if ACPI is not enabled.
      
      On driver detach, the initial BIOS configuration is restored by
      executing _STM with the initial _GTM values such that the next driver
      can also use the initial BIOS configured values.
      Signed-off-by: NTejun Heo <htejun@gmail.com>
      Signed-off-by: NJeff Garzik <jeff@garzik.org>
      c05e6ff0
    • T
      libata: add more opcodes to ata.h · ce2e0abb
      Tejun Heo 提交于
      Add constants for DEVICE CONFIGURATION OVERLAY and SET_MAX to
      include/linux/ata.h.
      Signed-off-by: NTejun Heo <htejun@gmail.com>
      Signed-off-by: NJeff Garzik <jeff@garzik.org>
      ce2e0abb
    • T
      libata: update ata_*_printk() macros such that level can be a variable · c2e366a1
      Tejun Heo 提交于
      Make prink helpers format @lv together rather than prepending to the
      format string as constant.
      Signed-off-by: NTejun Heo <htejun@gmail.com>
      Signed-off-by: NJeff Garzik <jeff@garzik.org>
      c2e366a1
    • T
      libata-acpi: adjust constness in ata_acpi_gtm/stm() parameters · 0d02f0b2
      Tejun Heo 提交于
      * No internal function uses const ata_port.  Drop const from @ap.
      
      * Make ata_acpi_stm() copy @stm before using it and change @stm to
        const.
      Signed-off-by: NTejun Heo <htejun@gmail.com>
      Signed-off-by: NJeff Garzik <jeff@garzik.org>
      0d02f0b2
    • R
      usb.h: fix kernel-doc warning · f88ed90d
      Randy Dunlap 提交于
      Fix kernel-doc warning in usb.h:
      Warning(linux-2.6.24-rc3-git7//include/linux/usb.h:166): No description found for parameter 'sysfs_files_created'
      Signed-off-by: NRandy Dunlap <randy.dunlap@oracle.com>
      Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>
      f88ed90d
    • D
      usb-storage: Fix devices that cannot handle 32k transfers · 33abc04f
      Doug Maxey 提交于
      When a device cannot handle the smallest previously limited transfer
      size (64 blocks) without stalling, limit the device to the amount of
      packets that fit in a platform native page.
      
      The lowest possible limit is PAGE_CACHE_SIZE, so if the device is ever
      used on a platform that has larger than 8K pages, you lose unless you
      can convince the device firmware folks to fix the issue.
      
      Cc: Mathew Dharm <mdharm-scsi@one-eyed-alien.net>
      Cc: Alan Stern <stern@rowland.harvard.edu>
      Cc: Pete Zaitcev <zaitcev@redhat.com>
      Signed-off-by: NDoug Maxey <dwm@austin.ibm.com>
      Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>
      33abc04f
    • R
      tipar: remove obsolete module · cb8c9b6d
      Romain Liévin 提交于
      tipar: remove obsolete module
      
      The tipar character driver was used to implement bit-banging access
      to Texas Instruments parallel link cable. A user-land method now 
      exists thru PPDEV & PARPORT.
      Signed-off-by: NRomain Liévin <roms@lpg.ticalc.org>
      Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>
      cb8c9b6d
  7. 17 12月, 2007 1 次提交
    • M
      [S390] pud_present/pmd_present bug. · 0d017923
      Martin Schwidefsky 提交于
      Git commit 3610cce8 (yeah my own :-/)
      introduced a bug in regard to pud/pmd table entries.
      If the address of the page table refered to by a pud/pmd value happens
      to have zeroes in the lower 32 bits, pud_present and pmd_present return
      false. The obvious effect is that this triggers the BUG_ON in exit_mmap
      because some ptes will not get released on process end.  Worse is that
      the next fault for memory covered by that pud/pmd will allocate another
      pmd/pte table and populate the pud/pmd entry. The old page table
      entries hanging below this entry are lost!
      
      The fix is simple, properly check against 0. The check is added for
      pud_none/pmd_none as well even if these two functions work because
      the invalid bit is in the lower 32 bits.
      Signed-off-by: NMartin Schwidefsky <schwidefsky@de.ibm.com>
      0d017923
  8. 15 12月, 2007 3 次提交
  9. 14 12月, 2007 1 次提交
  10. 13 12月, 2007 4 次提交
    • D
      [SPARC64]: Fix two kernel linear mapping setup bugs. · 8f361453
      David S. Miller 提交于
      This was caught and identified by Greg Onufer.
      
      Since we setup the 256M/4M bitmap table after taking over the trap
      table, it's possible for some 4M mapping to get loaded in the TLB
      beforhand which later will be 256M mappings.
      
      This can cause illegal TLB multiple-match conditions.  Fix this by
      setting up the bitmap before we take over the trap table.
      
      Next, __flush_tlb_all() was not doing anything on hypervisor
      platforms.  Fix by adding sun4v_mmu_demap_all() and calling it.
      Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      8f361453
    • B
      ide: DMA reporting and validity checking fixes (take 3) · 3ab7efe8
      Bartlomiej Zolnierkiewicz 提交于
      * ide_xfer_verbose() fixups:
        - beautify returned mode names
        - fix PIO5 reporting
        - make it return 'const char *'
      
      * Change printk() level from KERN_DEBUG to KERN_INFO in ide_find_dma_mode().
      
      * Add ide_id_dma_bug() helper based on ide_dma_verbose() to check for invalid
        DMA info in identify block.
      
      * Use ide_id_dma_bug() in ide_tune_dma() and ide_driveid_update().
      
        As a result DMA won't be tuned or will be disabled after tuning if device
        reports inconsistent info about enabled DMA mode (ide_dma_verbose() does the
        same checks while the IDE device is probed by ide-{cd,disk} device driver).
      
      * Remove no longer needed ide_dma_verbose().
      
      This patch should fix the following problem with out-of-sync IDE messages
      reported by Nick Warne:
      
             hdd: ATAPI 48X DVD-ROM DVD-R-RAM CD-R/RW drive, 2048kB Cache<7>hdd:
             skipping word 93 validity check
              , UDMA(66)
      
      and later debugged by Mark Lord to be caused by:
      
              ide_dma_verbose()
                      printk( ... "2048kB Cache");
              eighty_ninty_three()
                      printk(KERN_DEBUG "%s: skipping word 93 validity check\n");
              ide_dma_verbose()
                      printk(", UDMA(66)"
      
      Please note that as a result ide-{cd,disk} device drivers won't report the
      DMA speed used but this is intended since now DMA mode being used is always
      reported by IDE core code.
      
      v2:
      * fixes suggested by Randy:
        - use KERN_CONT for printk()-s in ide-{cd,disk}.c
        - don't remove argument name from ide_xfer_verbose() declaration
      
      v3:
      * Remove incorrect check for (id->field_valid & 1) from ide_id_dma_bug()
        (spotted by Sergei).
      
      * "XFER SLOW" -> "PIO SLOW" in ide_xfer_verbose() (suggested by Sergei).
      
      * Fix ide_find_dma_mode() to report the correct mode ('mode' after being
        limited by 'req_mode').
      
      Cc: Sergei Shtylyov <sshtylyov@ru.mvista.com>
      Cc: Nick Warne <nick@ukfsn.org>
      Cc: Mark Lord <lkml@rtr.ca>
      Cc: Randy Dunlap <randy.dunlap@oracle.com>
      Signed-off-by: NBartlomiej Zolnierkiewicz <bzolnier@gmail.com>
      3ab7efe8
    • N
      mmc: remove unused 'mode' from the mmc_host structure · cc3000e4
      Nicolas Pitre 提交于
      This field and corresponding defines are simply never used anywhere
      in the code.  But its mere presence is enough to confuse some host
      driver authors who attempt to rely on it.  Let's eliminate the
      possibility for confusion and remove it entirely.
      Signed-off-by: NNicolas Pitre <nico@cam.org>
      Signed-off-by: NPierre Ossman <drzeus@drzeus.cx>
      cc3000e4
    • P
      sdhci: support JMicron JMB38x chips · 84c46a53
      Pierre Ossman 提交于
      The JMicron JMB38x chip doesn't support transfers that aren't 32-bit
      aligned (both size and start address). It also doesn't like switching
      between PIO and DMA mode, so it needs to be reset after each request.
      Signed-off-by: NPierre Ossman <drzeus@drzeus.cx>
      84c46a53
  11. 12 12月, 2007 1 次提交
  12. 11 12月, 2007 1 次提交
反馈
建议
客服 返回
顶部