1. 19 Jun, 2007 6 commits
  2. 18 Jun, 2007 5 commits
    • Linus Torvalds's avatar
      Fix possible runqueue lock starvation in wait_task_inactive() · fa490cfd
      Linus Torvalds authored
      
      
      Miklos Szeredi reported very long pauses (several seconds, sometimes
      more) on his T60 (with a Core2Duo) which he managed to track down to
      wait_task_inactive()'s open-coded busy-loop.
      
      He observed that an interrupt on one core tries to acquire the
      runqueue-lock but does not succeed in doing so for a very long time -
      while wait_task_inactive() on the other core loops waiting for the first
      core to deschedule a task (which it wont do while spinning in an
      interrupt handler).
      
      This rewrites wait_task_inactive() to do all its waiting optimistically
      without any locks taken at all, and then just double-check the end
      result with the proper runqueue lock held over just a very short
      section.  If there were races in the optimistic wait, of a preemption
      event scheduled the process away, we simply re-synchronize, and start
      over.
      
      So the code now looks like this:
      
      	repeat:
      		/* Unlocked, optimistic looping! */
      		rq = task_rq(p);
      		while (task_running(rq, p))
      			cpu_relax();
      
      		/* Get the *real* values */
      		rq = task_rq_lock(p, &flags);
      		running = task_running(rq, p);
      		array = p->array;
      		task_rq_unlock(rq, &flags);
      
      		/* Check them.. */
      		if (unlikely(running)) {
      			cpu_relax();
      			goto repeat;
      		}
      
      		/* Preempted away? Yield if so.. */
      		if (unlikely(array)) {
      			yield();
      			goto repeat;
      		}
      
      Basically, that first "while()" loop is done entirely without any
      locking at all (and doesn't check for the case where the target process
      might have been preempted away), and so it's possibly "incorrect", but
      we don't really care.  Both the runqueue used, and the "task_running()"
      check might be the wrong tests, but they won't oops - they just mean
      that we could possibly get the wrong results due to lack of locking and
      exit the loop early in the case of a race condition.
      
      So once we've exited the loop, we then get the proper (and careful) rq
      lock, and check the running/runnable state _safely_.  And if it turns
      out that our quick-and-dirty and unsafe loop was wrong after all, we
      just go back and try it all again.
      
      (The patch also adds a lot of comments, which is the actual bulk of it
      all, to make it more obvious why we can do these things without holding
      the locks).
      
      Thanks to Miklos for all the testing and tracking it down.
      Tested-by: default avatarMiklos Szeredi <miklos@szeredi.hu>
      Acked-by: default avatarIngo Molnar <mingo@elte.hu>
      Signed-off-by: default avatarLinus Torvalds <torvalds@linux-foundation.org>
      fa490cfd
    • Ingo Molnar's avatar
      sched: fix SysRq-N (normalize RT tasks) · a0f98a1c
      Ingo Molnar authored
      
      
      Gene Heskett reported the following problem while testing CFS: SysRq-N
      is not always effective in normalizing tasks back to SCHED_OTHER.
      
      The reason for that turns out to be the following bug:
      
       - normalize_rt_tasks() uses for_each_process() to iterate through all
         tasks in the system.  The problem is, this method does not iterate
         through all tasks, it iterates through all thread groups.
      
      The proper mechanism to enumerate over all threads is to use a
      do_each_thread() + while_each_thread() loop.
      Reported-by: default avatarGene Heskett <gene.heskett@gmail.com>
      Signed-off-by: default avatarIngo Molnar <mingo@elte.hu>
      Signed-off-by: default avatarLinus Torvalds <torvalds@linux-foundation.org>
      a0f98a1c
    • Linus Torvalds's avatar
      Merge master.kernel.org:/pub/scm/linux/kernel/git/jejb/scsi-rc-fixes-2.6 · 4cc21505
      Linus Torvalds authored
      * master.kernel.org:/pub/scm/linux/kernel/git/jejb/scsi-rc-fixes-2.6:
        [SCSI] ESP: Don't forget to clear ESP_FLAG_RESETTING.
        [SCSI] fusion: fix for BZ 8426 - massive slowdown on SCSI CD/DVD drive
      4cc21505
    • Benjamin Herrenschmidt's avatar
      Fix signalfd interaction with thread-private signals · caec4e8d
      Benjamin Herrenschmidt authored
      
      
      Don't let signalfd dequeue private signals off other threads (in the
      case of things like SIGILL or SIGSEGV, trying to do so would result
      in undefined behaviour on who actually gets the signal, since they
      are force unblocked).
      Signed-off-by: default avatarBenjamin Herrenschmidt <benh@kernel.crashing.org>
      Acked-by: default avatarDavide Libenzi <davidel@xmailserver.org>
      Signed-off-by: default avatarLinus Torvalds <torvalds@linux-foundation.org>
      caec4e8d
    • Thomas Gleixner's avatar
      Revert "futex_requeue_pi optimization" · bd197234
      Thomas Gleixner authored
      This reverts commit d0aa7a70.
      
      It not only introduced user space visible changes to the futex syscall,
      it is also non-functional and there is no way to fix it proper before
      the 2.6.22 release.
      
      The breakage report ( http://lkml.org/lkml/2007/5/12/17
      
       ) went
      unanswered, and unfortunately it turned out that the concept is not
      feasible at all.  It violates the rtmutex semantics badly by introducing
      a virtual owner, which hacks around the coupling of the user-space
      pi_futex and the kernel internal rt_mutex representation.
      
      At the moment the only safe option is to remove it fully as it contains
      user-space visible changes to broken kernel code, which we do not want
      to expose in the 2.6.22 release.
      
      The patch reverts the original patch mostly 1:1, but contains a couple
      of trivial manual cleanups which were necessary due to patches, which
      touched the same area of code later.
      
      Verified against the glibc tests and my own PI futex tests.
      Signed-off-by: default avatarThomas Gleixner <tglx@linutronix.de>
      Acked-by: default avatarIngo Molnar <mingo@elte.hu>
      Acked-by: default avatarUlrich Drepper <drepper@redhat.com>
      Cc: Pierre Peiffer <pierre.peiffer@bull.net>
      Signed-off-by: default avatarLinus Torvalds <torvalds@linux-foundation.org>
      bd197234
  3. 17 Jun, 2007 1 commit
  4. 16 Jun, 2007 27 commits
  5. 15 Jun, 2007 1 commit
    • Paul Mundt's avatar
      mm: Fix memory/cpu hotplug section mismatch and oops. · d09c6b80
      Paul Mundt authored
      
      
      When building with memory hotplug enabled and cpu hotplug disabled, we
      end up with the following section mismatch:
      
      WARNING: mm/built-in.o(.text+0x4e58): Section mismatch: reference to
      .init.text: (between 'free_area_init_node' and '__build_all_zonelists')
      
      This happens as a result of:
      
              -> free_area_init_node()
                -> free_area_init_core()
                  -> zone_pcp_init() <-- all __meminit up to this point
                    -> zone_batchsize() <-- marked as __cpuinit                     fo
      
      This happens because CONFIG_HOTPLUG_CPU=n sets __cpuinit to __init, but
      CONFIG_MEMORY_HOTPLUG=y unsets __meminit.
      
      Changing zone_batchsize() to __devinit fixes this.
      
      __devinit is the only thing that is common between CONFIG_HOTPLUG_CPU=y and
      CONFIG_MEMORY_HOTPLUG=y. In the long run, perhaps this should be moved to
      another section identifier completely. Without this, memory hot-add
      of offline nodes (via hotadd_new_pgdat()) will oops if CPU hotplug is
      not also enabled.
      Signed-off-by: default avatarPaul Mundt <lethal@linux-sh.org>
      Acked-by: default avatarYasunori Goto <y-goto@jp.fujitsu.com>
      Signed-off-by: default avatarLinus Torvalds <torvalds@linux-foundation.org>
      
      --
      
       mm/page_alloc.c |    2 +-
       1 file changed, 1 insertion(+), 1 deletion(-)
      d09c6b80