- 04 Oct, 2013 18 commits
-
-
Ingo Molnar authored
Since we are changing wait.h profoundly, use the opportunity to: - add a sentence to explain what this file is about - remove whitespace noise - prettify weird looking line break fixup attempts - standardize type definition and initialization sequences - use consistent style details No code is changed. Acked-by:
Peter Zijlstra <peterz@infradead.org> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Andrew Morton <akpm@linux-foundation.org> Link: http://lkml.kernel.org/n/tip-O8dIie5swnctqpupakatvqyq@git.kernel.orgSigned-off-by:
Ingo Molnar <mingo@kernel.org>
-
Peter Zijlstra authored
Took a while to sort out these bits and we'd like to be Cc:-ed on future modifications to the waitqueue APIs and all that. Signed-off-by:
-
Peter Zijlstra authored
Change all __wait_event*() implementations to match the corresponding wait_event*() signature for convenience. In particular this does away with the weird 'ret' logic. Since there are __wait_event*() users this requires we update them too. Reviewed-by:
Oleg Nesterov <oleg@redhat.com> Signed-off-by:
-
Peter Zijlstra authored
While not a whole-sale replacement like the others we can still reduce the size of __wait_event_hrtimeout() considerably by noting that the actual core of __wait_event_hrtimeout() is identical to what ___wait_event() generates. Reviewed-by:
-
Peter Zijlstra authored
Reduce macro complexity by using the new ___wait_event() helper. No change in behaviour, identical generated code. Reviewed-by:
-
Peter Zijlstra authored
Reduce macro complexity by using the new ___wait_event() helper. No change in behaviour, identical generated code. Reviewed-by:
-
Peter Zijlstra authored
Reduce macro complexity by using the new ___wait_event() helper. No change in behaviour, identical generated code. Reviewed-by:
-
Peter Zijlstra authored
Reduce macro complexity by using the new ___wait_event() helper. No change in behaviour, identical generated code. Reviewed-by:
-
Peter Zijlstra authored
Reduce macro complexity by using the new ___wait_event() helper. No change in behaviour, identical generated code. Reviewed-by:
-
Peter Zijlstra authored
Reduce macro complexity by using the new ___wait_event() helper. No change in behaviour, identical generated code. Reviewed-by:
-
Peter Zijlstra authored
Reduce macro complexity by using the new ___wait_event() helper. No change in behaviour, identical generated code. Reviewed-by:
-
Peter Zijlstra authored
Reduce macro complexity by using the new ___wait_event() helper. No change in behaviour, identical generated code. Reviewed-by:
-
Peter Zijlstra authored
Reduce macro complexity by using the new ___wait_event() helper. No change in behaviour, identical generated code. Reviewed-by:
-
Peter Zijlstra authored
Reduce macro complexity by using the new ___wait_event() helper. No change in behaviour, identical generated code. Reviewed-by:
-
Peter Zijlstra authored
There's far too much duplication in the __wait_event macros; in order to fix this introduce ___wait_event() a macro with the capability to replace most other macros. With the previous patches changing the various __wait_event*() implementations to be more uniform; we can now collapse the lot without also changing generated code. Reviewed-by:
-
Peter Zijlstra authored
Purely a preparatory patch; it changes the control flow to match what will soon be generated by generic code so that that patch can be a unity transform. Reviewed-by:
-
Peter Zijlstra authored
Commit 4c663cfc ("wait: fix false timeouts when using wait_event_timeout()") introduced an additional condition check after a timeout but there's a few issues; - it forgot one site - it put the check after the main loop; not at the actual timeout check. Cure both; by wrapping the condition (as suggested by Oleg), this avoids double evaluation of 'condition' which could be quite big. Reviewed-by:
-
Peter Zijlstra authored
There's two patterns to check signals in the __wait_event*() macros: if (!signal_pending(current)) { schedule(); continue; } ret = -ERESTARTSYS; break; And the more natural: if (signal_pending(current)) { ret = -ERESTARTSYS; break; } schedule(); Change them all into the latter form. Reviewed-by:
-
- 28 Sep, 2013 1 commit
-
-
Peter Zijlstra authored
Yuanhan reported a serious throughput regression in his pigz benchmark. Using the ftrace patch I found that several idle paths need more TLC before we can switch the generic need_resched() over to preempt_need_resched. The preemption paths benefit most from preempt_need_resched and do indeed use it; all other need_resched() users don't really care that much so reverting need_resched() back to tif_need_resched() is the simple and safe solution. Reported-by:
Yuanhan Liu <yuanhan.liu@linux.intel.com> Signed-off-by:
-
- 25 Sep, 2013 13 commits
-
-
Peter Zijlstra authored
Remove the bloat of the C calling convention out of the preempt_enable() sites by creating an ASM wrapper which allows us to do an asm("call ___preempt_schedule") instead. calling.h bits by Andi Kleen Suggested-by:Linus Torvalds <torvalds@linux-foundation.org> Signed-off-by:
-
Peter Zijlstra authored
Convert x86 to use a per-cpu preemption count. The reason for doing so is that accessing per-cpu variables is a lot cheaper than accessing thread_info variables. We still need to save/restore the actual preemption count due to PREEMPT_ACTIVE so we place the per-cpu __preempt_count variable in the same cache-line as the other hot __switch_to() variables such as current_task. NOTE: this save/restore is required even for !PREEMPT kernels as cond_resched() also relies on preempt_count's PREEMPT_ACTIVE to ignore task_struct::state. Also rename thread_info::preempt_count to ensure nobody is 'accidentally' still poking at it. Suggested-by:
-
Peter Zijlstra authored
When using per-cpu preempt_count variables we need to save/restore the preempt_count on context switch (into per task storage; for instance the old thread_info::preempt_count variable) because of PREEMPT_ACTIVE. However, this means that on fork() the preempt_count value of the last context switch gets copied and if we had a PREEMPT_ACTIVE switch right before cloning a child task the child task will now too have PREEMPT_ACTIVE set and start its life with an extra PREEMPT_ACTIVE count. Therefore we need to make init_task_preempt_count() unconditional; this resets whatever preempt_count we inherited from our parent process. Doing so for !per-cpu implementations is harmless. For !PREEMPT_COUNT kernels we need to be careful not to start life with an increased preempt_count. Signed-off-by:
-
Peter Zijlstra authored
Rewrite the preempt_count macros in order to extract the 3 basic preempt_count value modifiers: __preempt_count_add() __preempt_count_sub() and the new: __preempt_count_dec_and_test() And since we're at it anyway, replace the unconventional $op_preempt_count names with the more conventional preempt_count_$op. Since these basic operators are equivalent to the previous _notrace() variants, do away with the _notrace() versions. Signed-off-by:
-
Peter Zijlstra authored
We need a few special preempt_count accessors: - task_preempt_count() for when we're interested in the preemption count of another (non-running) task. - init_task_preempt_count() for properly initializing the preemption count. - init_idle_preempt_count() a special case of the above for the idle threads. With these no generic code ever touches thread_info::preempt_count anymore and architectures could choose to remove it. Signed-off-by:
-
Peter Zijlstra authored
In order to prepare to per-arch implementations of preempt_count move the required bits into an asm-generic header and use this for all archs. Signed-off-by:
-
Peter Zijlstra authored
In order to combine the preemption and need_resched test we need to fold the need_resched information into the preempt_count value. Since the NEED_RESCHED flag is set across CPUs this needs to be an atomic operation, however we very much want to avoid making preempt_count atomic, therefore we keep the existing TIF_NEED_RESCHED infrastructure in place but at 3 sites test it and fold its value into preempt_count; namely: - resched_task() when setting TIF_NEED_RESCHED on the current task - scheduler_ipi() when resched_task() sets TIF_NEED_RESCHED on a remote task it follows it up with a reschedule IPI and we can modify the cpu local preempt_count from there. - cpu_idle_loop() for when resched_task() found tsk_is_polling(). We use an inverted bitmask to indicate need_resched so that a 0 means both need_resched and !atomic. Also remove the barrier() in preempt_enable() between preempt_enable_no_resched() and preempt_check_resched() to avoid having to reload the preemption value and allow the compiler to use the flags of the previuos decrement. I couldn't come up with any sane reason for this barrier() to be there as preempt_enable_no_resched() already has a barrier() before doing the decrement. Suggested-by: -
Peter Zijlstra authored
Replace the single preempt_count() 'function' that's an lvalue with two proper functions: preempt_count() - returns the preempt_count value as rvalue preempt_count_set() - Allows setting the preempt-count value Also provide preempt_count_ptr() as a convenience wrapper to implement all modifying operations. Signed-off-by:
-
Peter Zijlstra authored
Mike reported that commit 7d1a9417 ("x86: Use generic idle loop") regressed several workloads and caused excessive reschedule interrupts. The patch in question failed to notice that the x86 code had an inverted sense of the polling state versus the new generic code (x86: default polling, generic: default !polling). Fix the two prominent x86 mwait based idle drivers and introduce a few new generic polling helpers (fixing the wrong smp_mb__after_clear_bit usage). Also switch the idle routines to using tif_need_resched() which is an immediate TIF_NEED_RESCHED test as opposed to need_resched which will end up being slightly different. Reported-by:
Mike Galbraith <bitbucket@online.de> Signed-off-by:
-
Peter Zijlstra authored
Preemption semantics are going to change which mandate a change. All DRM usage sites are already broken and will not be affected (much) by this change. DRM people are aware and will remove the last few stragglers. For now, leave an empty stub that generates a warning, once all users are gone we can remove this. Signed-off-by:
-
Peter Zijlstra authored
We're going to deprecate and remove set_need_resched() for it will do the wrong thing. Make an exception for RCU and allow it to use resched_cpu() which will do the right thing. Signed-off-by:
-
Peter Zijlstra authored
Linus suggested using asm goto to get rid of the typical SETcc + TEST instruction pair -- which also clobbers an extra register -- for our typical modify_and_test() functions. Because asm goto doesn't allow output fields it has to include an unconditinal memory clobber when it changes a memory variable to force a reload. Luckily all atomic ops already imply a compiler barrier to go along with their memory barrier semantics. Suggested-by:
-
Michael S. Tsirkin authored
We always know the rq used, let's just pass it around. This seems to cut the size of scheduler core down a tiny bit: Before: [linux]$ size kernel/sched/core.o.orig text data bss dec hex filename 62760 16130 3876 82766 1434e kernel/sched/core.o.orig After: [linux]$ size kernel/sched/core.o.patched text data bss dec hex filename 62566 16130 3876 82572 1428c kernel/sched/core.o.patched Probably speeds it up as well. Signed-off-by:Michael S. Tsirkin <mst@redhat.com> Signed-off-by:
-
- 20 Sep, 2013 8 commits
-
-
Jason Low authored
This patch builds on patch 2 and periodically decays that max value to do idle balancing per sched domain by approximately 1% per second. Also decay the rq's max_idle_balance_cost value. Signed-off-by:
Jason Low <jason.low2@hp.com> Signed-off-by:
-
Jason Low authored
In this patch, we keep track of the max cost we spend doing idle load balancing for each sched domain. If the avg time the CPU remains idle is less then the time we have already spent on idle balancing + the max cost of idle balancing in the sched domain, then we don't continue to attempt the balance. We also keep a per rq variable, max_idle_balance_cost, which keeps track of the max time spent on newidle load balances throughout all its domains so that we can determine the avg_idle's max value. By using the max, we avoid overrunning the average. This further reduces the chance we attempt balancing when the CPU is not idle for longer than the cost to balance. Signed-off-by:
-
Jason Low authored
When updating avg_idle, if the delta exceeds some max value, then avg_idle gets set to the max, regardless of what the previous avg was. This can cause avg_idle to often be overestimated. This patch modifies the way we update avg_idle by always updating it with the function call to update_avg() first. Then, if avg_idle exceeds the max, we set it to the max. Signed-off-by:
Rik van Riel <riel@redhat.com> Reviewed-by:
Srikar Dronamraju <srikar@linux.vnet.ibm.com> Signed-off-by:
-
Vladimir Davydov authored
Currently new_dst_cpu is prevented from being reselected actually, not dst_cpu. This can result in attempting to pull tasks to this_cpu twice. Signed-off-by:
Vladimir Davydov <vdavydov@parallels.com> Signed-off-by:
-
Ingo Molnar authored
Merge in the latest fixes before applying a dependent patch. Signed-off-by: -
Vladimir Davydov authored
Patch a003a2 (sched: Consider runnable load average in move_tasks()) sets all top-level cfs_rqs' h_load to rq->avg.load_avg_contrib, which is always 0. This mistype leads to all tasks having weight 0 when load balancing in a cpu-cgroup enabled setup. There obviously should be sum of weights of all runnable tasks there instead. Fix it. Signed-off-by:
Paul Turner <pjt@google.com> Signed-off-by:
-
Vladimir Davydov authored
In busiest->group_imb case we can come to fix_small_imbalance() with local->avg_load > busiest->avg_load. This can result in wrong imbalance fix-up, because there is the following check there where all the members are unsigned: if (busiest->avg_load - local->avg_load + scaled_busy_load_per_task >= (scaled_busy_load_per_task * imbn)) { env->imbalance = busiest->load_per_task; return; } As a result we can end up constantly bouncing tasks from one cpu to another if there are pinned tasks. Fix it by substituting the subtraction with an equivalent addition in the check. [ The bug can be caught by running 2*N cpuhogs pinned to two logical cpus belonging to different cores on an HT-enabled machine with N logical cpus: just look at se.nr_migrations growth. ] Signed-off-by: -
Vladimir Davydov authored
In busiest->group_imb case we can come to calculate_imbalance() with local->avg_load >= busiest->avg_load >= sds->avg_load. This can result in imbalance overflow, because it is calculated as follows env->imbalance = min( max_pull * busiest->group_power, (sds->avg_load - local->avg_load) * local->group_power) / SCHED_POWER_SCALE; As a result we can end up constantly bouncing tasks from one cpu to another if there are pinned tasks. Fix this by skipping the assignment and assuming imbalance=0 in case local->avg_load > sds->avg_load. [ The bug can be caught by running 2*N cpuhogs pinned to two logical cpus belonging to different cores on an HT-enabled machine with N logical cpus: just look at se.nr_migrations growth. ] Signed-off-by:
-
