Skip to content

L
linux
Switch branch/tag
  1. 22 Apr, 2008 12 commits
    • Hidetoshi Seto's avatar
      [IA64] disable interrupts on exit of ia64_trace_syscall · 38477ad7
      Hidetoshi Seto authored 
      While testing with CONFIG_VIRT_CPU_ACCOUNTING=y, I found that
I occasionally get very huge system time in some threads.

So I dug the issue and finally noticed that it was caused
because of an interrupt which interrupt in the following window:

> [arch/ia64/kernel/entry.S: (!CONFIG_PREEMPT && CONFIG_VIRT_CPU_ACCOUNTING)]
>
> ENTRY(ia64_leave_syscall)
>    :
> (pUStk) rsm psr.i
>         cmp.eq pLvSys,p0=r0,r0          // pLvSys=1: leave from syscall
> (pUStk) cmp.eq.unc p6,p0=r0,r0          // p6 <- pUStk
> .work_processed_syscall:
>         adds r2=PT(LOADRS)+16,r12
> (pUStk) mov.m r22=ar.itc                        // fetch time at leave
>         adds r18=TI_FLAGS+IA64_TASK_SIZE,r13
>         ;;
> <<< window: from here >>>
> (p6)    ld4 r31=[r18]  // load current_thread_info()->flags
>         ld8 r19=[r2],PT(B6)-PT(LOADRS)
>         adds r3=PT(AR_BSPSTORE)+16,r12
>         ;;
>         mov r16=ar.bsp
>         ld8 r18=[r2],PT(R9)-PT(B6)
> (p6)    and r15=TIF_WORK_MASK,r31  // any work other than TIF_SYSCALL_TRACE?
>         ;;
>         ld8 r23=[r3],PT(R11)-PT(AR_BSPSTORE)
> (p6)    cmp4.ne.unc p6,p0=r15, r0               // any special work pending?
> (p6)    br.cond.spnt .work_pending_syscall
>         ;;
>         ld8 r9=[r2],PT(CR_IPSR)-PT(R9)
>         ld8 r11=[r3],PT(CR_IIP)-PT(R11)
> (pNonSys) break 0 // bug check: we shouldn't be here if pNonSys is TRUE!
>         ;;
>         invala
> <<< window: to here >>>
>         rsm psr.i | psr.ic // turn off interrupts and interruption collection

If pUStk is true, it means we are going to return user mode, hence we fetch
ar.itc to get time at leave from system.
It seems that it is not possible to interrupt the window if pUStk is true,
because interrupts are disabled early.  And also disabling interrupt makes
sense because it is safe for referring current_thread_info()->flags.

However interrupting the window while pUStk is true was possible.
The route was:
ia64_trace_syscall
-> .work_pending_syscall_end
-> .work_processed_syscall
Only in case entering the window from this route, interrupts are enabled
during in the window even if pUStk is true.  I suppose interrupts must be
disabled here anyway if pUStk is true.
I'm not sure but afraid that what kind of bad effect were there, other
than crazy system time which I found.

FYI, there was a commit 6f6d7582 that
points out a bug at same point(exit of ia64_trace_syscall) in 2006.
It can be said that there was an another bug.
Signed-off-by: default avatarHidetoshi Seto <seto.hidetoshi@jp.fujitsu.com>
Signed-off-by: default avatarTony Luck <tony.luck@intel.com>
      38477ad7
    • Linus Torvalds's avatar
      Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/roland/infiniband · 80750147
      Linus Torvalds authored 
      * 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/roland/infiniband:
  IB/ipath: Correct capitalization "IntX" -> "INTx"
  IB/ipath: Remove tests of PCI_MSI in ipath_iba7220.c
  IB/ipath: Remove dependency on PCI_MSI || HT_IRQ
  IB/ipath: Build IBA7220 code unconditionally
  IB/ipath: Remove reference to dev->class_dev
  IB/ipath: Fix module parameter description for disable_sma
  RDMA/nes: Remove unneeded function declarations

Fix up conflict in drivers/infiniband/hw/ipath/ipath_verbs.c (by picking
the solution from the infiniband branch)
      80750147
    • Roland Dreier's avatar
      IB/ipath: Correct capitalization "IntX" -> "INTx" · bc751fe6
      Roland Dreier authored 
      Match what the PCI specification uses.
Signed-off-by: default avatarRoland Dreier <rolandd@cisco.com>
      bc751fe6
    • Roland Dreier's avatar
      IB/ipath: Remove tests of PCI_MSI in ipath_iba7220.c · 44957572
      Roland Dreier authored 
      The PCI MSI interface is stubbed out properly so that all the
functions just return failure if PCI_MSI=n, so there's no reason to
have "#ifdef CONFIG_PCI_MSI" blocks in ipath_iba7220.c.
Signed-off-by: default avatarRoland Dreier <rolandd@cisco.com>
      44957572
    • Roland Dreier's avatar
      IB/ipath: Remove dependency on PCI_MSI || HT_IRQ · 480f58e6
      Roland Dreier authored 
      Before IBA7220 support was added, the ipath driver didn't support any
hardware unless PCI_MSI and/or HT_IRQ was enabled.  However, the
IBA7220 can generate INTx interrupts, so it makes sense to allow the
driver to be build even if PCI_MSI=n and HT_IRQ=n.
Signed-off-by: default avatarRoland Dreier <rolandd@cisco.com>
      480f58e6
    • Roland Dreier's avatar
      IB/ipath: Build IBA7220 code unconditionally · 37a6ab52
      Roland Dreier authored 
      The new IBA7220 code added a call to ipath_init_iba7220_funcs() that
is compiled unconditionally, but only built the IBA7220 code if
PCI_MSI is enabled.  Fix this by building the IBA7220 file
unconditonally.

This fixes build breakage when PCI_MSI=n, HT_IRQ=y and
INFINIBAND_IPATH=y reported by Ingo Molnar <mingo@elte.hu>:

 drivers/built-in.o: In function `ipath_init_one':
 ipath_driver.c:(.devinit.text+0x1e5bc): undefined reference to `ipath_init_iba7220_funcs'
Signed-off-by: default avatarRoland Dreier <rolandd@cisco.com>
      37a6ab52
    • Roland Dreier's avatar
      IB/ipath: Remove reference to dev->class_dev ... · 88a8317b
      Roland Dreier authored 
      IB/ipath: Remove reference to dev->class_dev                                                        
                                                                                                    
Commit 124b4dcb ("IB/ipath: add calls to new 7220 code and enable in                                
build") inadvertently added core to set dev->class_dev.dev back into                                
ib_ipath.  This is completely redundant since commit 1912ffbb ("IB: Set                             
class_dev->dev in core for nice device symlink"), which removed                                     
class_dev setting from low-level drivers, and also will break the build
when class_dev is removed completely from struct ib_device.
Signed-off-by: default avatarRoland Dreier <rolandd@cisco.com>
      88a8317b
    • Paul Bolle's avatar
      IB/ipath: Fix module parameter description for disable_sma · 9862874d
      Paul Bolle authored 
      Describe disable_sma parameter with its name rather than the internal
ib_ipath_disable_sma variable name, so that the description shows up
properly in modinfo.
Signed-off-by: default avatarPaul Bolle <pebolle@tiscali.nl>
Acked-by: default avatarRalph Campbell <ralph.campbell@qlogic.com>
Signed-off-by: default avatarRoland Dreier <rolandd@cisco.com>
      9862874d
    • Roland Dreier's avatar
      RDMA/nes: Remove unneeded function declarations · 6a5546e7
      Roland Dreier authored 
      Remove redundant static declarations of functions that are defined
before they are used in the source.
Signed-off-by: default avatarRoland Dreier <rolandd@cisco.com>
      6a5546e7
    • Stephen Rothwell's avatar
      infiniband: class_device fallout · 28480770
      Stephen Rothwell authored 
      Signed-off-by: default avatarStephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: default avatarLinus Torvalds <torvalds@linux-foundation.org>
      28480770
    • Linus Torvalds's avatar
      Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/drzeus/mmc · 135cedad
      Linus Torvalds authored 
      * 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/drzeus/mmc: (26 commits)
  mmc: sdio_ops.c should #include "sdio_ops.h"
  mmc: proper prototypes for mmc_attach_*()
  mmc: make __mmc_release_bus() static
  sdhci: improve no card, no reset quirk
  MMC: OMAP: Do not busy wait for end of command for ever
  MMC: OMAP: Start new commands from work queue instead of irq
  MMC: OMAP: Lazy clock shutdown
  MMC: OMAP: Move failing command abortion to workqueue
  MMC: OMAP: Use tasklet instead of workqueue for cover switch notification
  MMC: OMAP: Check the get_cover_state function pointer if not set
  MMC: OMAP: Using setup_timer instead of init_timer
  MMC: OMAP: Abort stuck commands
  MMC: OMAP: General cleanup for MMC multislot support
  MMC: OMAP: Power functions modified to MMC multislot support
  MMC: OMAP: Fix timeout calculation for MMC multislot support
  MMC: OMAP: New release dma and abort xfer functions
  MMC: OMAP: Add back cover switch support
  MMC: OMAP: Introduce new multislot structure and change driver to use it
  MMC: OMAP: Remove cover switch handling to allow adding multislot support
  MMC: OMAP: Fix the BYTEBLOCK capability removal
  ...
      135cedad
    • Linus Torvalds's avatar
      Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/sparc-2.6 · 8a322726
      Linus Torvalds authored 
      * git://git.kernel.org/pub/scm/linux/kernel/git/davem/sparc-2.6:
  [SPARC]: Remove SunOS and Solaris binary support.
      8a322726
  3. 21 Apr, 2008 28 commits
GitLab Nexedi Edition | About GitLab | About Nexedi | 沪ICP备2021021310号-2 | 沪ICP备2021021310号-7