The patch titled
genapic: optimize & fix APIC mode setup
has been added to the -mm tree. Its filename is
genapic-optimize-fix-apic-mode-setup-2.patch
See http://www.zip.com.au/~akpm/linux/patches/stuff/added-to-mm.txt to find
out what to do about this
------------------------------------------------------
Subject: genapic: optimize & fix APIC mode setup
From: Ingo Molnar <mingo@xxxxxxx>
Fix a couple of inconsistencies/problems I found while reviewing the x86_64
genapic code (when I was chasing mysterious eth0 timeouts that would only
trigger if CPU_HOTPLUG is enabled):
- AMD systems defaulted to the slower flat-physical mode instead
of the flat-logical mode. The only restriction on AMD systems
is that they should not use clustered APIC mode.
- removed the CPU hotplug hacks, switching the default for small
systems back from phys-flat to logical-flat. The switching to logical
flat mode on small systems fixed sporadic ethernet driver timeouts i
was getting on a dual-core Athlon64 system:
NETDEV WATCHDOG: eth0: transmit timed out
eth0: Transmit timeout, status 0c 0005 c07f media 80.
eth0: Tx queue start entry 32 dirty entry 28.
eth0: Tx descriptor 0 is 0008a04a. (queue head)
eth0: Tx descriptor 1 is 0008a04a.
eth0: Tx descriptor 2 is 0008a04a.
eth0: Tx descriptor 3 is 0008a04a.
eth0: link up, 100Mbps, full-duplex, lpa 0xC5E1
- The use of '<= 8' was a bug by itself (the valid APIC ids
for logical flat mode go from 0 to 7, not 0 to 8). The new logic
is to use logical flat mode on both AMD and Intel systems, and
to only switch to physical mode when logical mode cannot be used.
If CPU hotplug is racy wrt. APIC shutdown then CPU hotplug needs
fixing, not the whole IRQ system be made inconsistent and slowed
down.
- minor cleanups: simplified some code constructs
build & booted on a couple of AMD and Intel SMP systems.
Signed-off-by: Ingo Molnar <mingo@xxxxxxx>
Cc: Suresh Siddha <suresh.b.siddha@xxxxxxxxx>
Cc: Andi Kleen <ak@xxxxxxx>
Cc: "Li, Shaohua" <shaohua.li@xxxxxxxxx>
Cc: "Eric W. Biederman" <ebiederm@xxxxxxxxxxxx>
Signed-off-by: Andrew Morton <akpm@xxxxxxxx>
---
arch/x86_64/kernel/genapic.c | 54 ++++++++++++---------------------
1 files changed, 21 insertions(+), 33 deletions(-)
diff -puN arch/x86_64/kernel/genapic.c~genapic-optimize-fix-apic-mode-setup-2 arch/x86_64/kernel/genapic.c
--- a/arch/x86_64/kernel/genapic.c~genapic-optimize-fix-apic-mode-setup-2
+++ a/arch/x86_64/kernel/genapic.c
@@ -32,30 +32,26 @@ extern struct genapic apic_cluster;
extern struct genapic apic_flat;
extern struct genapic apic_physflat;
-struct genapic *genapic = &apic_flat;
-
+struct genapic __read_mostly *genapic = &apic_flat;
/*
* Check the APIC IDs in bios_cpu_apicid and choose the APIC mode.
*/
void __init clustered_apic_check(void)
{
- long i;
- u8 clusters, max_cluster;
- u8 id;
- u8 cluster_cnt[NUM_APIC_CLUSTERS];
- int max_apic = 0;
+ u8 id, clusters, max_cluster, cluster_cnt[NUM_APIC_CLUSTERS];
+ int i, max_apic = 0;
-#if defined(CONFIG_ACPI)
+#ifdef CONFIG_ACPI
/*
* Some x86_64 machines use physical APIC mode regardless of how many
* procs/clusters are present (x86_64 ES7000 is an example).
*/
- if (acpi_fadt.revision > FADT2_REVISION_ID)
- if (acpi_fadt.force_apic_physical_destination_mode) {
- genapic = &apic_cluster;
- goto print;
- }
+ if (acpi_fadt.revision > FADT2_REVISION_ID &&
+ acpi_fadt.force_apic_physical_destination_mode) {
+ genapic = &apic_cluster;
+ goto print;
+ }
#endif
memset(cluster_cnt, 0, sizeof(cluster_cnt));
@@ -68,20 +64,17 @@ void __init clustered_apic_check(void)
cluster_cnt[APIC_CLUSTERID(id)]++;
}
- /* Don't use clustered mode on AMD platforms. */
+ /*
+ * Don't use clustered mode on AMD platforms, default
+ * to flat logical mode.
+ */
if (boot_cpu_data.x86_vendor == X86_VENDOR_AMD) {
- genapic = &apic_physflat;
-#ifndef CONFIG_HOTPLUG_CPU
- /* In the CPU hotplug case we cannot use broadcast mode
- because that opens a race when a CPU is removed.
- Stay at physflat mode in this case.
- It is bad to do this unconditionally though. Once
- we have ACPI platform support for CPU hotplug
- we should detect hotplug capablity from ACPI tables and
- only do this when really needed. -AK */
- if (max_apic <= 8)
- genapic = &apic_flat;
-#endif
+ /*
+ * Switch to physical flat mode if more than 8 APICs
+ * (In the case of 8 CPUs APIC ID goes from 0 to 7):
+ */
+ if (max_apic >= 8)
+ genapic = &apic_physflat;
goto print;
}
@@ -103,14 +96,9 @@ void __init clustered_apic_check(void)
* (We don't use lowest priority delivery + HW APIC IRQ steering, so
* can ignore the clustered logical case and go straight to physical.)
*/
- if (clusters <= 1 && max_cluster <= 8 && cluster_cnt[0] == max_cluster) {
-#ifdef CONFIG_HOTPLUG_CPU
- /* Don't use APIC shortcuts in CPU hotplug to avoid races */
- genapic = &apic_physflat;
-#else
+ if (clusters <= 1 && max_cluster <= 8 && cluster_cnt[0] == max_cluster)
genapic = &apic_flat;
-#endif
- } else
+ else
genapic = &apic_cluster;
print:
_
Patches currently in -mm which might be from mingo@xxxxxxx are
add-bottom_half.h.patch
git-scsi-misc.patch
revert-x86_64-mm-add-genapic_force.patch
fix-x86_64-mm-fix-the-irqbalance-quirk-for-e7320-e7520-e7525.patch
cleanup-arch-i386-kernel-smpbootcsmp_tune_scheduling.patch
convert-i386-pda-code-to-use-%fs.patch
x86_64-check-vector-in-setup_ioapic_dest-to-verify-if-need-setup_io_apic_irq.patch
i386-kernel-smpc-dont-use-set_irq_regs.patch
genapic-optimize-fix-apic-mode-setup-2.patch
genapic-always-use-physical-delivery-mode-on-8-cpus.patch
genapic-remove-es7000-workaround.patch
genapic-remove-clustered-apic-mode.patch
genapic-default-to-physical-mode-on-hotplug-cpu-kernels.patch
convert-pm_sem-to-a-mutex.patch
convert-pm_sem-to-a-mutex-fix.patch
lockdep-annotate-nfs-nfsd-in-kernel-sockets.patch
lockdep-annotate-nfs-nfsd-in-kernel-sockets-tidy.patch
sysrq-x-show-blocked-tasks.patch
lockdep-internal-locking-fixes.patch
lockdep-misc-fixes-in-lockdepc.patch
cpuset-remove-sched-domain-hooks-from-cpusets.patch
binfmt_elf-randomize-pie-binaries.patch
disable-clone_child_cleartid-for-abnormal-exit.patch
lockdep-annotate-bcsp-driver.patch
lockdep-print-current-locks-on-in_atomic-warnings.patch
lockdep-name-some-old-style-locks.patch
sleep-profiling.patch
sleep-profiling-fixes.patch
sleep-profiling-fix.patch
debug-workqueue-locking-sanity.patch
debug-workqueue-locking-sanity-fix.patch
retries-in-ext3_prepare_write-violate-ordering-requirements.patch
ktime-fix-signed--unsigned-mismatch-in-ktime_to_ns.patch
kprobes-enable-booster-on-the-preemptible-kernel.patch
declare-smp_call_function_single-in-generic-code.patch
up-smp_call_function_single-should-disable-interrupts.patch
up-smp_call_function_single-should-disable-interrupts-fix.patch
smp_call_function_single-check-that-local-interrupts-are-enabled.patch
hotplug-cpu-clean-up-hotcpu_notifier-use.patch
remove-kernel-lockdepclockdep_internal.patch
lockdep-more-chains.patch
lockdep-show-more-details-about-self-test-failures.patch
softirq-remove-bug_ons-which-can-incorrectly-trigger.patch
futex-remove-unneeded-barrier.patch
remove-the-old-bd_mutex-lockdep-annotation.patch
new-bd_mutex-lockdep-annotation.patch
remove-lock_key-approach-to-managing-nested-bd_mutex-locks.patch
simplify-some-aspects-of-bd_mutex-nesting.patch
use-mutex_lock_nested-for-bd_mutex-to-avoid-lockdep-warning.patch
avoid-lockdep-warning-in-md.patch
fix-generic-warn_on-message.patch
schedc-correct-comment-for-this_rq_lock-routine.patch
sched-fix-migration-cost-estimator.patch
sched-domain-move-sched-group-allocations-to-percpu-area.patch
move_task_off_dead_cpu-should-be-called-with-disabled-ints.patch
sched-domain-increase-the-smt-busy-rebalance-interval.patch
sched-avoid-taking-rq-lock-in-wake_priority_sleeper.patch
sched-remove-staggering-of-load-balancing.patch
sched-disable-interrupts-for-locking-in-load_balance.patch
sched-extract-load-calculation-from-rebalance_tick.patch
sched-move-idle-status-calculation-into-rebalance_tick.patch
sched-use-softirq-for-load-balancing.patch
sched-call-tasklet-less-frequently.patch
sched-add-option-to-serialize-load-balancing.patch
sched-add-option-to-serialize-load-balancing-fix.patch
sched-improve-migration-accuracy.patch
sched-improve-migration-accuracy-tidy.patch
sched-decrease-number-of-load-balances.patch
mm-only-sched-add-a-few-scheduler-event-counters.patch
sched-optimize-activate_task-for-rt-task.patch
kernel-schedc-whitespace-cleanups.patch
kernel-schedc-whitespace-cleanups-more.patch
sched-add-above-background-load-function.patch
mm-implement-swap-prefetching.patch
mm-implement-swap-prefetching-use-ctl_unnumbered.patch
sched-cleanup-remove-task_t-convert-to-struct-task_struct-prefetch.patch
gtod-exponential-update_wall_time.patch
gtod-persistent-clock-support-core.patch
gtod-persistent-clock-support-i386.patch
time-uninline-jiffiesh.patch
time-uninline-jiffiesh-fix.patch
time-fix-msecs_to_jiffies-bug.patch
time-fix-timeout-overflow.patch
cleanup-uninline-irq_enter-and-move-it-into-a-function.patch
dynticks-extend-next_timer_interrupt-to-use-a-reference-jiffie.patch
dynticks-extend-next_timer_interrupt-to-use-a-reference-jiffie-remove-incorrect-warning-in-kernel-timerc.patch
hrtimers-namespace-and-enum-cleanup.patch
hrtimers-clean-up-locking.patch
hrtimers-clean-up-locking-fix.patch
updated-hrtimers-state-tracking.patch
updated-hrtimers-clean-up-callback-tracking.patch
updated-hrtimers-move-and-add-documentation.patch
updated-add-a-framework-to-manage-clock-event-devices.patch
updated-acpi-include-apich.patch
updated-acpi-keep-track-of-timer-broadcast.patch
updated-acpi-add-state-propagation-for-dynamic-broadcasting.patch
updated-i386-cleanup-apic-code.patch
updated-i386-convert-to-clock-event-devices.patch
updated-pm_timer-allow-early-access-and-move-externs-to-a-header-file.patch
updated-i386-rework-local-apic-calibration.patch
updated-high-res-timers-core.patch
updated-high-res-timers-core-high-res-timers-do-itimer-rearming-in-process-context.patch
updated-gtod-mark-tsc-unusable-for-highres-timers.patch
high-res-timers-utilize-tsc-clocksource-again.patch
high-res-timers-utilize-tsc-clocksource-again-fix.patch
updated-dynticks-core-code.patch
updated-dynticks-core-code-fix-resume-bug.patch
updated-dyntick-add-nohz-stats-to-proc-stat.patch
updated-dynticks-i386-arch-code.patch
updated-dynticks-fix-nmi-watchdog.patch
updated-high-res-timers-dynticks-enable-i386-support.patch
updated-debugging-feature-timer-stats.patch
clockevents-core-check-for-clock-event-device-handler-being-non-null-before-calling-it.patch
round_jiffies-infrastructure.patch
round_jiffies-infrastructure-fix.patch
clocksource-add-usage-of-config_sysfs.patch
clocksource-small-cleanup-2.patch
clocksource-small-cleanup-2-fix.patch
clocksource-small-acpi_pm-cleanup.patch
kvm-amd-svm-implementation-more-i386-fixes.patch
detect-atomic-counter-underflows.patch
debug-shared-irqs.patch
make-frame_pointer-default=y.patch
mutex-subsystem-synchro-test-module.patch
vdso-print-fatal-signals.patch
vdso-improve-print_fatal_signals-support-by-adding-memory-maps.patch
vdso-print-fatal-signals-use-ctl_unnumbered.patch
lockdep-show-held-locks-when-showing-a-stackdump.patch
lockdep-show-held-locks-when-showing-a-stackdump-fix.patch
lockdep-show-held-locks-when-showing-a-stackdump-fix-2.patch
-
To unsubscribe from this list: send the line "unsubscribe mm-commits" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
