summaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* m68k: merge non-mmu and mmu versions of m68k_ksyms.cGreg Ungerer2011-05-243-64/+32
| | | | | | | | | | After cleaning up m68k_ksyms_no.c it is now strait forward to merge the non-mmu and mmu versions of m68k_ksyms.c. The need for the extra gcc functions is not strictly based on having an MMU or not. It is based on the family the processor belongs too, so use an appropriate conditional check. Signed-off-by: Greg Ungerer <gerg@uclinux.org>
* m68knommu: remove un-needed exporting of COLDFIRE symbolsGreg Ungerer2011-05-242-12/+3
| | | | | | | | | | There is no reason most of the symbols enclosed in a conditional on CONFIG_COLDFIRE need to be exported. And they sure don't need to be doing it in m68k_ksyms_no.c. Move the dma symbols export (which are currently needed) to the definitions of those, and remove the rest of the exporting here. Signed-off-by: Greg Ungerer <gerg@uclinux.org>
* m68knommu: move EXPORT of kernel_thread to function definitionGreg Ungerer2011-05-242-4/+1
| | | | | | | The EXPORT_SYMBOL(kernel_thread) belongs at the definition of that function, not in some other random code file. So move it there. Signed-off-by: Greg Ungerer <gerg@uclinux.org>
* m68knommu: move EXPORT of local checksumming functions to definitionsGreg Ungerer2011-05-242-5/+3
| | | | | | | The EXPORT_SYMBOL() of the local lib checksum functions belongs with the definitions, not in some other random code file. So move then there. Signed-off-by: Greg Ungerer <gerg@uclinux.org>
* m68knommu: move EXPORT of dump_fpu to function definitionGreg Ungerer2011-05-242-4/+1
| | | | | | | The EXPORT_SYMBOL(dump_fpu) belongs at the definition of the function, not in some other random code file. So move it there. Signed-off-by: Greg Ungerer <gerg@uclinux.org>
* m68knommu: clean up mm/init_no.cGreg Ungerer2011-05-241-47/+4
| | | | | | | | | | | | | The memory initialization code for m68knommu has grown a bit crufty, clean it up. . remove unused declaration for die_if_kernel() . remove un-needed declaration of free_initmem() . removed unused definitions of empty_bad_page and empty_bad_page_table . removed unused DEBUG code . make free_initmem() proper prototype Signed-off-by: Greg Ungerer <gerg@uclinux.org>
* m68k: merge the mmu and non-mmu mm/MakefileGreg Ungerer2011-05-243-18/+9
| | | | | | | Its is trivial to megre the mmu and non-mmu arch/m68k/mm/Makefile's back into a single file. So do it. Signed-off-by: Greg Ungerer <gerg@uclinux.org>
* m68k: mv kmap_mm.c to kmap.cGreg Ungerer2011-05-242-370/+365
| | | | | | | The non-mmu kmap_no.c has been removed. So we can move kmap_mm.c back to being the only kmap.c. Signed-off-by: Greg Ungerer <gerg@uclinux.org>
* m68knommu: remove stubs for __ioremap() and iounmap()Greg Ungerer2011-05-244-43/+6
| | | | | | | | The implementation of iounmap() and __ioremap() for non-mmu m68k is trivial. We can inline them in m68knommu headers and remove the trivial implementations. Signed-off-by: Greg Ungerer <gerg@uclinux.org>
* m68knommu: remove unused kernel_set_cachemode()Greg Ungerer2011-05-241-8/+0
| | | | | | | None of the m68knommu platforms will ever use kernel_set_cachemode(). And it is specific to a couple of m68k devices. So remove it. Signed-off-by: Greg Ungerer <gerg@uclinux.org>
* m68k: let Makefile sort out compiling mmu and non-mmu lib/checksum.cGreg Ungerer2011-05-242-8/+3
| | | | | | | | We don't need an arch/m68k/lib/checksum.c wrapper to include the correct mmu or non-mmu version of the checksum code. Let the Makefile just build the appropriate one. Signed-off-by: Greg Ungerer <gerg@uclinux.org>
* m68k: remove duplicate memcpy() implementationGreg Ungerer2011-05-244-131/+75
| | | | | | | | | | | Merging the mmu and non-mmu directories we ended up with duplicate implementations of memcpy(). One is a little more optimized for the >= 68020 case, but that can easily be inserted into a single implementation of memcpy(). Clean up the exporting of this symbol too, otherwise we end up exporting it twice on a no-mmu build. Signed-off-by: Greg Ungerer <gerg@uclinux.org> Acked-by: Geert Uytterhoeven <geert@linux-m68k.org>
* m68k: remove duplicate memset() implementationGreg Ungerer2011-05-244-107/+74
| | | | | | | | | | | Merging the mmu and non-mmu directories we ended up with duplicate implementations of memset(). One is a little more optimized for the >= 68020 case, but that can easily be inserted into a single implementation of memset(). Clean up the exporting of this symbol too, otherwise we end up exporting it twice on a no-mmu build. Signed-off-by: Greg Ungerer <gerg@uclinux.org> Acked-by: Geert Uytterhoeven <geert@linux-m68k.org>
* m68k: remove duplicate memmove() implementationGreg Ungerer2011-05-243-99/+2
| | | | | | | | Merging the mmu and non-mmu directories we ended up with duplicate (and identical) implementations of memmove(). Remove one of them. Signed-off-by: Greg Ungerer <gerg@uclinux.org> Acked-by: Geert Uytterhoeven <geert@linux-m68k.org>
* m68k: merge mmu and non-mmu versions of lib/MakefileGreg Ungerer2011-05-243-15/+11
| | | | | | | | | We can easily support the slight differences in libs needed by the mmu and non-mmu builds in a single Makefile, so merge them back into a single file again. Signed-off-by: Greg Ungerer <gerg@uclinux.org> Acked-by: Geert Uytterhoeven <geert@linux-m68k.org>
* m68k: merge mmu and non-mmu versions of muldi3Greg Ungerer2011-05-243-152/+96
| | | | | | | | | | | The implementation of gcc's muldi3 support function differs only in the use of the machine's 64 bit sized mul or not. (It isn't based on using an MMU or not). Merge the current mmu and non-mmu versions of arc/m68k/lib/muldi3 and use the appropriate pre-processor conditionals to get the right version for all m68k processor types. Signed-off-by: Greg Ungerer <gerg@uclinux.org> Acked-by: Geert Uytterhoeven <geert@linux-m68k.org>
* Merge branch 'x86-apic-for-linus' of ↵Linus Torvalds2011-05-2320-563/+539
|\ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip * 'x86-apic-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip: x86, apic: Include module.h header in apic_flat_64.c x86, apic: Make apic drivers static x86, apic: Clean up bigsmp apic selection code x86, apic: Use .apicdrivers section for the apic drivers list x86, apic: Introduce .apicdrivers section to find the list of apic drivers x86, x2apic: Move the common bits to x2apic.h x86, x2apic: Minimize IPI register writes using cluster groups x86, x2apic: Track the x2apic cluster sibling map x86, x2apic: Remove duplicate code for IPI mask routines x86, apic: Use probe routines to simplify apic selection x86, ioapic: Consolidate mp_ioapic_routing[] into 'struct ioapic' x86, ioapic: Consolidate gsi routing info into 'struct ioapic' x86, ioapic: Consolidate mp_ioapics[] into 'struct ioapic' x86, ioapic: Consolidate ioapic_saved_data[] into 'struct ioapic' x86, ioapic: Add struct ioapic x86, ioapic: Remove duplicate code for saving/restoring RTEs x86, ioapic: Use ioapic_saved_data while enabling intr-remapping x86, ioapic: Allocate ioapic_saved_data early x86, ioapic: Fix potential resume deadlock
| * x86, apic: Include module.h header in apic_flat_64.cRandy Dunlap2011-05-231-0/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | apic_flat_64.c needs to include module.h because it uses EXPORT_SYMBOL_GPL(). This fixes these warnings on some !SMP randconfigs: arch/x86/kernel/apic/apic_flat_64.c:31: warning: data definition has no type or storage class arch/x86/kernel/apic/apic_flat_64.c:31: warning: type defaults to 'int' in declaration of 'EXPORT_SYMBOL_GPL' arch/x86/kernel/apic/apic_flat_64.c:31: warning: parameter names (without types) in function declaration Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com> Cc: Stephen Rothwell <sfr@canb.auug.org.au> Link: http://lkml.kernel.org/r/20110523104300.dd532a99.randy.dunlap@oracle.com Signed-off-by: Ingo Molnar <mingo@elte.hu>
| * x86, apic: Make apic drivers staticSuresh Siddha2011-05-2211-64/+22
| | | | | | | | | | | | | | | | | | | | | | | | | | | | Apic probe now looks at the apic drivers listed in the .apicdrivers section. Remove apic_probe[] and make each apic driver static. Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com> Tested-by: Cyrill Gorcunov <gorcunov@openvz.org> Cc: steiner@sgi.com Cc: gorcunov@openvz.org Cc: yinghai@kernel.org Link: http://lkml.kernel.org/r/20110521005526.341718626@sbsiddha-MOBL3.sc.intel.com Signed-off-by: Ingo Molnar <mingo@elte.hu>
| * x86, apic: Clean up bigsmp apic selection codeSuresh Siddha2011-05-223-30/+32
| | | | | | | | | | | | | | | | | | | | | | | | | | | | Make generic_bigsmp_probe() return struct apic *. This will avoid exporting apic_bigsmp, which will be consistent with others. Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com> Tested-by: Cyrill Gorcunov <gorcunov@openvz.org> Cc: steiner@sgi.com Cc: gorcunov@openvz.org Cc: yinghai@kernel.org Link: http://lkml.kernel.org/r/20110521005526.252703851@sbsiddha-MOBL3.sc.intel.com Signed-off-by: Ingo Molnar <mingo@elte.hu>
| * x86, apic: Use .apicdrivers section for the apic drivers listSuresh Siddha2011-05-222-31/+36
| | | | | | | | | | | | | | | | | | | | | | | | | | | | This will eliminate the need for apic_probe[], as the probing now will happen based on the apic drivers order in the .apcidrivers section. Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com> Tested-by: Cyrill Gorcunov <gorcunov@openvz.org> Cc: steiner@sgi.com Cc: gorcunov@openvz.org Cc: yinghai@kernel.org Link: http://lkml.kernel.org/r/20110521005526.164277071@sbsiddha-MOBL3.sc.intel.com Signed-off-by: Ingo Molnar <mingo@elte.hu>
| * x86, apic: Introduce .apicdrivers section to find the list of apic driversSuresh Siddha2011-05-2212-6/+63
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This will pave the way for each apic driver to be self-contained and eliminate the need for apic_probe[]. Order in which apic drivers are listed in the .apicdrivers section is important, as this determines the apic probe order. And this is enforced by the ordering of apic driver files in the Makefile and the macros apic_driver()/apic_drivers(). Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com> Tested-by: Cyrill Gorcunov <gorcunov@openvz.org> Cc: steiner@sgi.com Cc: gorcunov@openvz.org Cc: yinghai@kernel.org Link: http://lkml.kernel.org/r/20110521005526.068775085@sbsiddha-MOBL3.sc.intel.com Signed-off-by: Ingo Molnar <mingo@elte.hu>
| * x86, x2apic: Move the common bits to x2apic.hCyrill Gorcunov2011-05-203-124/+69
| | | | | | | | | | | | | | | | | | | | | | To eliminate code duplication. Signed-off-by: Cyrill Gorcunov <gorcunov@openvz.org> Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com> Cc: steiner@sgi.com Cc: yinghai@kernel.org Link: http://lkml.kernel.org/r/20110519234637.591426753@sbsiddha-MOBL3.sc.intel.com Signed-off-by: Ingo Molnar <mingo@elte.hu>
| * x86, x2apic: Minimize IPI register writes using cluster groupsCyrill Gorcunov2011-05-201-14/+44
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | In the case of x2apic cluster mode we can group IPI register writes based on the cluster group instead of individual per-cpu destination messages. This reduces the apic register writes and reduces the amount of IPI messages (in the best case we can reduce it by a factor of 16). With this change, the cost of flush_tlb_others(), with the flush tlb IPI being sent from a cpu in the socket-1 to all the logical cpus in socket-2 (on a Westmere-EX system that has 20 logical cpus in a socket) is 3x times better now (compared to the former 'send one-by-one' algorithm). Signed-off-by: Cyrill Gorcunov <gorcunov@openvz.org> Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com> Cc: steiner@sgi.com Cc: yinghai@kernel.org Link: http://lkml.kernel.org/r/20110519234637.512271057@sbsiddha-MOBL3.sc.intel.com Signed-off-by: Ingo Molnar <mingo@elte.hu>
| * x86, x2apic: Track the x2apic cluster sibling mapCyrill Gorcunov2011-05-201-2/+70
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | In the case of x2apic cluster mode, we can group IPI register writes based on the cluster group instead of individual per-cpu destination messages. For this purpose, track the cpu's that belong to the same x2apic cluster. Signed-off-by: Cyrill Gorcunov <gorcunov@openvz.org> Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com> Cc: steiner@sgi.com Cc: yinghai@kernel.org Link: http://lkml.kernel.org/r/20110519234637.421800999@sbsiddha-MOBL3.sc.intel.com Signed-off-by: Ingo Molnar <mingo@elte.hu>
| * x86, x2apic: Remove duplicate code for IPI mask routinesSuresh Siddha2011-05-202-62/+31
| | | | | | | | | | | | | | | | | | Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com> Acked-by: Cyrill Gorcunov <gorcunov@openvz.org> Cc: steiner@sgi.com Cc: yinghai@kernel.org Link: http://lkml.kernel.org/r/20110519234637.337024125@sbsiddha-MOBL3.sc.intel.com Signed-off-by: Ingo Molnar <mingo@elte.hu>
| * x86, apic: Use probe routines to simplify apic selectionSuresh Siddha2011-05-205-19/+37
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Use the unused probe routine in the apic driver to finalize the apic model selection. This cleans up the default_setup_apic_routing() and this probe routine in future can also be used for doing any apic model specific initialisation. Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com> Acked-by: Cyrill Gorcunov <gorcunov@openvz.org> Cc: steiner@sgi.com Cc: yinghai@kernel.org Link: http://lkml.kernel.org/r/20110519234637.247458931@sbsiddha-MOBL3.sc.intel.com Signed-off-by: Ingo Molnar <mingo@elte.hu>
| * x86, ioapic: Consolidate mp_ioapic_routing[] into 'struct ioapic'Suresh Siddha2011-05-201-6/+3
| | | | | | | | | | | | | | Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com> Cc: daniel.blueman@gmail.com Link: http://lkml.kernel.org/r/20110518233158.089978277@sbsiddha-MOBL3.sc.intel.com Signed-off-by: Ingo Molnar <mingo@elte.hu>
| * x86, ioapic: Consolidate gsi routing info into 'struct ioapic'Suresh Siddha2011-05-203-14/+29
| | | | | | | | | | | | | | Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com> Cc: daniel.blueman@gmail.com Link: http://lkml.kernel.org/r/20110518233157.994002011@sbsiddha-MOBL3.sc.intel.com Signed-off-by: Ingo Molnar <mingo@elte.hu>
| * x86, ioapic: Consolidate mp_ioapics[] into 'struct ioapic'Suresh Siddha2011-05-205-50/+62
| | | | | | | | | | | | | | Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com> Cc: daniel.blueman@gmail.com Link: http://lkml.kernel.org/r/20110518233157.909013179@sbsiddha-MOBL3.sc.intel.com Signed-off-by: Ingo Molnar <mingo@elte.hu>
| * x86, ioapic: Consolidate ioapic_saved_data[] into 'struct ioapic'Suresh Siddha2011-05-201-14/+13
| | | | | | | | | | | | | | Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com> Cc: daniel.blueman@gmail.com Link: http://lkml.kernel.org/r/20110518233157.830697056@sbsiddha-MOBL3.sc.intel.com Signed-off-by: Ingo Molnar <mingo@elte.hu>
| * x86, ioapic: Add struct ioapicSuresh Siddha2011-05-202-16/+17
| | | | | | | | | | | | | | | | | | | | | | | | Introduce struct ioapic with nr_registers field. This will pave way for consolidating different MAX_IO_APICS arrays into it. Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com> Cc: daniel.blueman@gmail.com Link: http://lkml.kernel.org/r/20110518233157.744315519@sbsiddha-MOBL3.sc.intel.com Signed-off-by: Ingo Molnar <mingo@elte.hu>
| * x86, ioapic: Remove duplicate code for saving/restoring RTEsSuresh Siddha2011-05-201-31/+5
| | | | | | | | | | | | | | | | | | | | | | Code flow for enabling interrupt-remapping has its own routines for saving and restoring io-apic RTE's. ioapic suspend/resume code flow also has similar routines. Remove the duplicate code. Tested-by: Daniel J Blueman <daniel.blueman@gmail.com> Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com> Link: http://lkml.kernel.org/r/20110518233157.673130611@sbsiddha-MOBL3.sc.intel.com Signed-off-by: Ingo Molnar <mingo@elte.hu>
| * x86, ioapic: Use ioapic_saved_data while enabling intr-remappingSuresh Siddha2011-05-203-111/+37
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Code flow for enabling interrupt-remapping was allocating/freeing buffers for saving/restoring io-apic RTE's. ioapic suspend/resume code uses boot time allocated ioapic_saved_data that is a perfect match for reuse here. This will remove the unnecessary allocation/free of the temporary buffers during suspend/resume of interrupt-remapping enabled platforms aswell as paving the way for further code consolidation. Tested-by: Daniel J Blueman <daniel.blueman@gmail.com> Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com> Link: http://lkml.kernel.org/r/20110518233157.574469296@sbsiddha-MOBL3.sc.intel.com Signed-off-by: Ingo Molnar <mingo@elte.hu>
| * x86, ioapic: Allocate ioapic_saved_data earlySuresh Siddha2011-05-201-14/+13
| | | | | | | | | | | | | | | | | | | | | | | | This allows re-using this buffer for enabling interrupt-remapping during boot and resume. And thus allow for consolidating the code between ioapic suspend/resume and interrupt-remapping. Tested-by: Daniel J Blueman <daniel.blueman@gmail.com> Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com> Link: http://lkml.kernel.org/r/20110518233157.481404505@sbsiddha-MOBL3.sc.intel.com Signed-off-by: Ingo Molnar <mingo@elte.hu>
| * x86, ioapic: Fix potential resume deadlockDaniel J Blueman2011-05-201-2/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Fix a potential deadlock when resuming; here the calling function has disabled interrupts, so we cannot sleep. Change the memory allocation flag from GFP_KERNEL to GFP_ATOMIC. TODO: We can do away with this memory allocation during resume by reusing the ioapic suspend/resume code that uses boot time allocated buffers, but we want to keep this -stable patch simple. Signed-off-by: Daniel J Blueman <daniel.blueman@gmail.com> Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com> Cc: <stable@kernel.org> # v2.6.38/39 Link: http://lkml.kernel.org/r/20110518233157.385970138@sbsiddha-MOBL3.sc.intel.com Signed-off-by: Ingo Molnar <mingo@elte.hu>
* | Merge branch 'sched-core-for-linus' of ↵Linus Torvalds2011-05-233-43/+82
|\ \ | | | | | | | | | | | | | | | | | | | | | | | | git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip * 'sched-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip: sched: Increase SCHED_LOAD_SCALE resolution sched: Introduce SCHED_POWER_SCALE to scale cpu_power calculations sched: Cleanup set_load_weight()
| * | sched: Increase SCHED_LOAD_SCALE resolutionNikhil Rao2011-05-202-10/+41
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Introduce SCHED_LOAD_RESOLUTION, which scales is added to SCHED_LOAD_SHIFT and increases the resolution of SCHED_LOAD_SCALE. This patch sets the value of SCHED_LOAD_RESOLUTION to 10, scaling up the weights for all sched entities by a factor of 1024. With this extra resolution, we can handle deeper cgroup hiearchies and the scheduler can do better shares distribution and load load balancing on larger systems (especially for low weight task groups). This does not change the existing user interface, the scaled weights are only used internally. We do not modify prio_to_weight values or inverses, but use the original weights when calculating the inverse which is used to scale execution time delta in calc_delta_mine(). This ensures we do not lose accuracy when accounting time to the sched entities. Thanks to Nikunj Dadhania for fixing an bug in c_d_m() that broken fairness. Below is some analysis of the performance costs/improvements of this patch. 1. Micro-arch performance costs: Experiment was to run Ingo's pipe_test_100k 200 times with the task pinned to one cpu. I measured instruction, cycles and stalled-cycles for the runs. See: http://thread.gmane.org/gmane.linux.kernel/1129232/focus=1129389 for more info. -tip (baseline): Performance counter stats for '/root/load-scale/pipe-test-100k' (200 runs): 964,991,769 instructions # 0.82 insns per cycle # 0.33 stalled cycles per insn # ( +- 0.05% ) 1,171,186,635 cycles # 0.000 GHz ( +- 0.08% ) 306,373,664 stalled-cycles-backend # 26.16% backend cycles idle ( +- 0.28% ) 314,933,621 stalled-cycles-frontend # 26.89% frontend cycles idle ( +- 0.34% ) 1.122405684 seconds time elapsed ( +- 0.05% ) -tip+patches: Performance counter stats for './load-scale/pipe-test-100k' (200 runs): 963,624,821 instructions # 0.82 insns per cycle # 0.33 stalled cycles per insn # ( +- 0.04% ) 1,175,215,649 cycles # 0.000 GHz ( +- 0.08% ) 315,321,126 stalled-cycles-backend # 26.83% backend cycles idle ( +- 0.28% ) 316,835,873 stalled-cycles-frontend # 26.96% frontend cycles idle ( +- 0.29% ) 1.122238659 seconds time elapsed ( +- 0.06% ) With this patch, instructions decrease by ~0.10% and cycles increase by 0.27%. This doesn't look statistically significant. The number of stalled cycles in the backend increased from 26.16% to 26.83%. This can be attributed to the shifts we do in c_d_m() and other places. The fraction of stalled cycles in the frontend remains about the same, at 26.96% compared to 26.89% in -tip. 2. Balancing low-weight task groups Test setup: run 50 tasks with random sleep/busy times (biased around 100ms) in a low weight container (with cpu.shares = 2). Measure %idle as reported by mpstat over a 10s window. -tip (baseline): 06:47:48 PM CPU %usr %nice %sys %iowait %irq %soft %steal %guest %idle intr/s 06:47:49 PM all 94.32 0.00 0.06 0.00 0.00 0.00 0.00 0.00 5.62 15888.00 06:47:50 PM all 94.57 0.00 0.62 0.00 0.00 0.00 0.00 0.00 4.81 16180.00 06:47:51 PM all 94.69 0.00 0.06 0.00 0.00 0.00 0.00 0.00 5.25 15966.00 06:47:52 PM all 95.81 0.00 0.00 0.00 0.00 0.00 0.00 0.00 4.19 16053.00 06:47:53 PM all 94.88 0.06 0.00 0.00 0.00 0.00 0.00 0.00 5.06 15984.00 06:47:54 PM all 93.31 0.00 0.00 0.00 0.00 0.00 0.00 0.00 6.69 15806.00 06:47:55 PM all 94.19 0.00 0.06 0.00 0.00 0.00 0.00 0.00 5.75 15896.00 06:47:56 PM all 92.87 0.00 0.00 0.00 0.00 0.00 0.00 0.00 7.13 15716.00 06:47:57 PM all 94.88 0.00 0.00 0.00 0.00 0.00 0.00 0.00 5.12 15982.00 06:47:58 PM all 95.44 0.00 0.00 0.00 0.00 0.00 0.00 0.00 4.56 16075.00 Average: all 94.49 0.01 0.08 0.00 0.00 0.00 0.00 0.00 5.42 15954.60 -tip+patches: 06:47:03 PM CPU %usr %nice %sys %iowait %irq %soft %steal %guest %idle intr/s 06:47:04 PM all 100.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 16630.00 06:47:05 PM all 99.69 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.31 16580.20 06:47:06 PM all 99.69 0.00 0.06 0.00 0.00 0.00 0.00 0.00 0.25 16596.00 06:47:07 PM all 99.20 0.00 0.74 0.00 0.00 0.06 0.00 0.00 0.00 17838.61 06:47:08 PM all 100.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 16540.00 06:47:09 PM all 100.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 16575.00 06:47:10 PM all 100.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 16614.00 06:47:11 PM all 99.94 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.06 16588.00 06:47:12 PM all 99.94 0.00 0.06 0.00 0.00 0.00 0.00 0.00 0.00 16593.00 06:47:13 PM all 99.94 0.00 0.06 0.00 0.00 0.00 0.00 0.00 0.00 16551.00 Average: all 99.84 0.00 0.09 0.00 0.00 0.01 0.00 0.00 0.06 16711.58 We see an improvement in idle% on the system (drops from 5.42% on -tip to 0.06% with the patches). We see an improvement in idle% on the system (drops from 5.42% on -tip to 0.06% with the patches). Signed-off-by: Nikhil Rao <ncrao@google.com> Acked-by: Peter Zijlstra <peterz@infradead.org> Cc: Nikunj A. Dadhania <nikunj@linux.vnet.ibm.com> Cc: Srivatsa Vaddagiri <vatsa@linux.vnet.ibm.com> Cc: Stephan Barwolf <stephan.baerwolf@tu-ilmenau.de> Cc: Mike Galbraith <efault@gmx.de> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Andrew Morton <akpm@linux-foundation.org> Link: http://lkml.kernel.org/r/1305754668-18792-1-git-send-email-ncrao@google.com Signed-off-by: Ingo Molnar <mingo@elte.hu>
| * | sched: Introduce SCHED_POWER_SCALE to scale cpu_power calculationsNikhil Rao2011-05-203-32/+37
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | SCHED_LOAD_SCALE is used to increase nice resolution and to scale cpu_power calculations in the scheduler. This patch introduces SCHED_POWER_SCALE and converts all uses of SCHED_LOAD_SCALE for scaling cpu_power to use SCHED_POWER_SCALE instead. This is a preparatory patch for increasing the resolution of SCHED_LOAD_SCALE, and there is no need to increase resolution for cpu_power calculations. Signed-off-by: Nikhil Rao <ncrao@google.com> Acked-by: Peter Zijlstra <peterz@infradead.org> Cc: Nikunj A. Dadhania <nikunj@linux.vnet.ibm.com> Cc: Srivatsa Vaddagiri <vatsa@linux.vnet.ibm.com> Cc: Stephan Barwolf <stephan.baerwolf@tu-ilmenau.de> Cc: Mike Galbraith <efault@gmx.de> Link: http://lkml.kernel.org/r/1305738580-9924-3-git-send-email-ncrao@google.com Signed-off-by: Ingo Molnar <mingo@elte.hu>
| * | sched: Cleanup set_load_weight()Nikhil Rao2011-05-201-4/+7
| |/ | | | | | | | | | | | | | | | | | | | | | | | | | | Avoid using long repetitious names; make this simpler and nicer to read. No functional change introduced in this patch. Signed-off-by: Nikhil Rao <ncrao@google.com> Acked-by: Peter Zijlstra <peterz@infradead.org> Cc: Nikunj A. Dadhania <nikunj@linux.vnet.ibm.com> Cc: Srivatsa Vaddagiri <vatsa@linux.vnet.ibm.com> Cc: Stephan Barwolf <stephan.baerwolf@tu-ilmenau.de> Cc: Mike Galbraith <efault@gmx.de> Link: http://lkml.kernel.org/r/1305738580-9924-2-git-send-email-ncrao@google.com Signed-off-by: Ingo Molnar <mingo@elte.hu>
* | Merge branch 'staging-next' of ↵Linus Torvalds2011-05-23822-171853/+68284
|\ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/staging-2.6 * 'staging-next' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/staging-2.6: (970 commits) staging: usbip: replace usbip_u{dbg,err,info} and printk with dev_ and pr_ staging:iio: Trivial kconfig reorganization and uniformity improvements. staging:iio:documenation partial update. staging:iio: use pollfunc allocation helpers in remaining drivers. staging:iio:max1363 misc cleanups and use of for_each_bit_set to simplify event code spitting out. staging:iio: implement an iio_info structure to take some of the constant elements out of iio_dev. staging:iio:meter:ade7758: Use private data space from iio_allocate_device staging:iio:accel:lis3l02dq make write_reg_8 take value not a pointer to value. staging:iio: ring core cleanups + check if read_last available in lis3l02dq staging:iio:core cleanup: squash tiny wrappers and use dev_set_name to handle creation of event interface name. staging:iio: poll func allocation clean up. staging:iio:ad7780 trivial unused header cleanup. staging:iio:adc: AD7780: Use private data space from iio_allocate_device + trivial fixes staging:iio:adc:AD7780: Convert to new channel registration method staging:iio:adc: AD7606: Drop dev_data in favour of iio_priv() staging:iio:adc: AD7606: Consitently use indio_dev staging:iio: Rip out helper for software rings. staging:iio:adc:AD7298: Use private data space from iio_allocate_device staging:iio: rationalization of different buffer implementation hooks. staging:iio:imu:adis16400 avoid allocating rx, tx, and state separately from iio_dev. ... Fix up trivial conflicts in - drivers/staging/intel_sst/intelmid.c: patches applied in both branches - drivers/staging/rt2860/common/cmm_data_{pci,usb}.c: removed vs spelling - drivers/staging/usbip/vhci_sysfs.c: trivial header file inclusion
| * | staging: usbip: replace usbip_u{dbg,err,info} and printk with dev_ and pr_matt mooney2011-05-2011-306/+220
| | | | | | | | | | | | | | | | | | | | | | | | | | | This switches all of the usbip_u{dbg,err,info} and printk statements to dev_<level>, if possible, or pr_<level> macros. And removes a few unnecessary debug statements. Signed-off-by: matt mooney <mfm@muteddisk.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
| * | staging:iio: Trivial kconfig reorganization and uniformity improvements.Jonathan Cameron2011-05-203-41/+41
| | | | | | | | | | | | | | | Signed-off-by: Jonathan Cameron <jic23@cam.ac.uk> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
| * | staging:iio:documenation partial update.Jonathan Cameron2011-05-205-78/+91
| | | | | | | | | | | | | | | | | | | | | | | | | | | More to be added, but this brings the docs in line with the current code. Now they are hopefully just uninformative rather than actually incorrect. Signed-off-by: Jonathan Cameron <jic23@cam.ac.uk> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
| * | staging:iio: use pollfunc allocation helpers in remaining drivers.Jonathan Cameron2011-05-209-129/+79
| | | | | | | | | | | | | | | | | | | | | Some didn't get converted the first time around. Signed-off-by: Jonathan Cameron <jic23@cam.ac.uk> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
| * | staging:iio:max1363 misc cleanups and use of for_each_bit_set to simplify ↵Jonathan Cameron2011-05-203-116/+63
| | | | | | | | | | | | | | | | | | | | | event code spitting out. Signed-off-by: Jonathan Cameron <jic23@cam.ac.uk> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
| * | staging:iio: implement an iio_info structure to take some of the constant ↵Jonathan Cameron2011-05-2061-669/+964
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | elements out of iio_dev. This was suggested by Arnd Bergmann, Other elements may well move in here in future, but it definitely makes sense for these. Signed-off-by: Jonathan Cameron <jic23@cam.ac.uk> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
| * | staging:iio:meter:ade7758: Use private data space from iio_allocate_deviceMichael Hennerich2011-05-204-70/+62
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Use private data space from iio_allocate_device. Drop dev_data in favor of iio_priv(). Fix indention issues from previous patches. Signed-off-by: Michael Hennerich <michael.hennerich@analog.com> Acked-by: Jonathan Cameron <jic23@cam.ac.uk> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
| * | staging:iio:accel:lis3l02dq make write_reg_8 take value not a pointer to value.Jonathan Cameron2011-05-203-21/+21
| | | | | | | | | | | | | | | | | | | | | Silliness that has been there a long time. Signed-off-by: Jonathan Cameron <jic23@cam.ac.uk> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
| * | staging:iio: ring core cleanups + check if read_last available in lis3l02dqJonathan Cameron2011-05-203-233/+136
| | | | | | | | | | | | | | | Signed-off-by: Jonathan Cameron <jic23@cam.ac.uk> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
generated by cgit v1.2.3 (git 2.39.1) at 2024-12-18 04:43:16 +0100