linux/include
Tony Luck f2edc1fb9c ACPI: APEI: GHES: Improve ghes_notify_nmi() status check
ghes_notify_nmi() is called for every NMI and must check whether the NMI was
generated because an error was signalled by platform firmware.

This check is very expensive as for each registered GHES NMI source it reads
from the acpi generic address attached to this error source to get the physical
address of the acpi_hest_generic_status block.  It then checks the "block_status"
to see if an error was logged.

The ACPI/APEI code must create virtual mappings for each of those physical
addresses, and tear them down afterwards. On an Icelake system this takes around
15,000 TSC cycles. Enough to disturb efforts to profile system performance.

If that were not bad enough, there are some atomic accesses in the code path
that will cause cache line bounces between CPUs. A problem that gets worse as
the core count increases.

But BIOS changes neither the acpi generic address nor the physical address of
the acpi_hest_generic_status block. So this walk can be done once when the NMI is
registered to save the virtual address (unmapping if the NMI is ever unregistered).
The "block_status" can be checked directly in the NMI handler. This can be done
without any atomic accesses.

Resulting time to check that there is not an error record is around 900 cycles.

Reported-by: Andi Kleen <andi.kleen@intel.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
Tested-by: Tony Luck <tony.luck@intel.com>
Signed-off-by: Shuai Xue <xueshuai@linux.alibaba.com>
Reviewed-by: Hanjun Guo <guohanjun@huawei.com>
Link: https://patch.msgid.link/20260112032239.30023-2-xueshuai@linux.alibaba.com
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2026-01-14 17:05:05 +01:00
..
acpi ACPI: APEI: GHES: Improve ghes_notify_nmi() status check 2026-01-14 17:05:05 +01:00
asm-generic hyperv-next for v6.19 2025-12-09 06:10:17 +09:00
clocksource
crypto This update includes the following changes: 2025-12-03 11:28:38 -08:00
cxl
drm drm/pagemap, drm/xe: Ensure that the devmem allocation is idle before use 2025-12-23 10:11:59 +01:00
dt-bindings This pull request is entirely SoC clk drivers, not for lack of trying to modify 2025-12-08 09:38:52 +09:00
hyperv mshv: Add definitions for MSHV sleep state configuration 2025-12-05 23:24:57 +00:00
keys
kunit kunit: Enforce task execution in {soft,hard}irq contexts 2025-12-22 12:20:08 -08:00
kvm
linux EFI/CPER: don't go past the ARM processor CPER record buffer 2026-01-14 17:04:21 +01:00
math-emu
media
memory
misc
net net: dsa: properly keep track of conduit reference 2025-12-23 10:32:08 +01:00
pcmcia
ras Significant patch series in this merge are as follows: 2025-12-05 13:52:43 -08:00
rdma
rv rv: Fix compilation if !CONFIG_RV_REACTORS 2025-12-02 12:33:37 -05:00
scsi
soc This pull request is entirely SoC clk drivers, not for lack of trying to modify 2025-12-08 09:38:52 +09:00
sound ASoC: soc-acpi / SOF: Add best_effort flag to get_function_tplg_files op 2025-12-15 23:08:35 +09:00
target
trace Miscellaneous x86 fixes: 2025-12-21 14:41:29 -08:00
uapi RDMA v6.19 first rc request 2026-01-02 12:25:47 -08:00
ufs
vdso
video
xen
Kbuild