davidlt/kernel-ark
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity
dc1829a4c3
kernel-ark/arch/i386
History
Ingo Molnar dc1829a4c3 [PATCH] i386/x86_64: ACPI cpu_idle_wait() fix 
The scheduler on Andreas Friedrich's hyperthreading system stopped
working properly: the scheduler would never move tasks to another CPU!
The lask known working kernel was 2.6.8.

After a couple of attempts to corner the bug, the following smoking gun
was found:

  BIOS reported wrong ACPI idfor the processor
  CPU#1: set_cpus_allowed(), swapper:1, 3 -> 2
   [<c0103bbe>] show_trace_log_lvl+0x34/0x4a
   [<c0103ceb>] show_trace+0x2c/0x2e
   [<c01045f8>] dump_stack+0x2b/0x2d
   [<c0116a77>] set_cpus_allowed+0x52/0xec
   [<c0101d86>] cpu_idle_wait+0x2e/0x100
   [<c0259c57>] acpi_processor_power_exit+0x45/0x58
   [<c0259752>] acpi_processor_remove+0x46/0xea
   [<c025c6fb>] acpi_start_single_object+0x47/0x54
   [<c025cee5>] acpi_bus_register_driver+0xa4/0xd3
   [<c04ab2d7>] acpi_processor_init+0x57/0x77
   [<c01004d7>] init+0x146/0x2fd
   [<c0103a87>] kernel_thread_helper+0x7/0x10

a quick look at cpu_idle_wait() shows how broken that code is
on i386: it changes the init task's affinity map but never
restores it ...

and because all userspace tasks get forked by init, they all
inherited that single-CPU affinity mask. x86_64 cloned this
bug too.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Cc: Andreas Friedrich <andreas.friedrich@fujitsu-siemens.com>
Cc: Wolfgang Erig <Wolfgang.Erig@fujitsu-siemens.com>
Cc: Andrew Morton <akpm@osdl.org>
Cc: Adrian Bunk <bunk@stusta.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-11-17 08:20:09 -08:00
..
boot [PATCH] Some config.h removals 2006-10-01 00:39:34 -07:00
crypto [CRYPTO] api: Get rid of flags argument to setkey 2006-09-21 11:41:02 +10:00
kernel [PATCH] i386/x86_64: ACPI cpu_idle_wait() fix 2006-11-17 08:20:09 -08:00
lib [PATCH] separate bdi congestion functions from queue congestion functions 2006-10-20 10:26:35 -07:00
mach-default [PATCH] irq-flags: i386: Use the new IRQF_ constants 2006-07-02 13:58:47 -07:00
mach-es7000
mach-generic [PATCH] i386: Replace i386 open-coded cmdline parsing with 2006-09-26 10:52:32 +02:00
mach-visws [PATCH] visws build fix 2006-10-28 11:30:52 -07:00
mach-voyager [VOYAGER] fix up ptregs removal mess 2006-10-12 22:25:03 -05:00
math-emu
mm [PATCH] mm: use symbolic names instead of indices for zone initialisation 2006-10-11 11:14:14 -07:00
oprofile [PATCH] oprofile: ppro: need to enable/disable all the counters 2006-09-29 09:18:11 -07:00
pci [PATCH] fix via586 irq routing for pirq 5 2006-11-14 09:09:27 -08:00
power [PATCH] Change the name of pagedir_nosave 2006-09-26 08:49:01 -07:00
defconfig [PATCH] i386: Update defconfig 2006-10-21 18:37:01 +02:00
Kconfig Attack of "the the"s in arch 2006-10-03 22:21:02 +02:00
Kconfig.cpu [PATCH] uml: fix processor selection to exclude unsupported processors and features 2006-10-11 11:14:20 -07:00
Kconfig.debug [PATCH] lockdep: irqtrace subsystem, i386 support 2006-07-03 15:27:03 -07:00
Makefile [PATCH] x86: Use -maccumulate-outgoing-args 2006-10-21 18:37:01 +02:00
Makefile.cpu