On 22.03.24 10:12:04, kernel test robot wrote:
> kernel test robot noticed a -43.6% regression of vm-scalability.throughput on:
>
>
> commit: 09f84589a8ed1162c0274a1b5b3dadb98ee3155e ("[PATCH v2 3/3] ACPI/NUMA: Remove architecture dependent remainings")
> url: https://github.com/intel-lab-lkp/linux/commits/Robert-Richter/x86-numa-Fix-SRAT-lookup-of-CFMWS-ranges-with-numa_fill_memblks/20240319-200345
> base: https://git.kernel.org/cgit/linux/kernel/git/rafael/linux-pm.git linux-next
> patch link: https://lore.kernel.org/all/20240319120026.2246389-4-rrichter@xxxxxxx/
> patch subject: [PATCH v2 3/3] ACPI/NUMA: Remove architecture dependent remainings
I have root cause the regression. After refactoring of the function an
error in acpi_numa_memory_affinity_init() is no longer ignored
(e.g. if a mem range is disabled) which finally causes
acpi_numa_init() to fail too.
Will be fixed in v3.
Thanks to the kernel test robot team.
-Robert
