Messages in this thread | | | Subject | Re: [drm/mgag200] 913ec479bb: vm-scalability.throughput 26.2% improvement | From | Thomas Zimmermann <> | Date | Sat, 29 Aug 2020 20:06:04 +0200 |
| |
Hi
Am 27.08.20 um 16:56 schrieb Mike Kravetz: > On 8/27/20 2:16 AM, Thomas Zimmermann wrote: >> Hi >> >> Am 26.08.20 um 10:58 schrieb kernel test robot: >>> Greeting, >>> >>> FYI, we noticed a 26.2% improvement of vm-scalability.throughput due to commit: >> >> I guess this resolves the once-measured performance penalty of similar >> magnitude. But do we really understand these tests? When I sent out >> patches to resolve the problem, nothing changed. And suddenly the >> performance is back to normal. >> >> Best regards >> Thomas >> >>> >>> >>> commit: 913ec479bb5cc27f99f24d5fd111b3ef29a4deb9 ("drm/mgag200: Replace VRAM helpers with SHMEM helpers") >>> https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git master >>> >>> >>> in testcase: vm-scalability >>> on test machine: 288 threads Intel(R) Xeon Phi(TM) CPU 7295 @ 1.50GHz with 80G memory >>> with following parameters: >>> >>> runtime: 300s >>> size: 8T >>> test: anon-cow-seq-hugetlb >>> cpufreq_governor: performance >>> ucode: 0x11 > > Hello Thomas, > > Did drm changes really impact anon-cow-seq-hugetlb performance? > > My change c0d0381ade79 ("hugetlbfs: use i_mmap_rwsem for more pmd sharing > synchronization") caused a -33.4% regression of anon-cow-seq-hugetlb. A > recent change 34ae204f185 (hugetlbfs: remove call to huge_pte_alloc without > i_mmap_rwsem) was tested by Zhengjun Xing and improved performance by 20 > something percent. That seems in line with this report/improvement.
Some of DRM's memory management might be affected by hugetable changes. While I cannot really point to a specific location, it's not impossible that there's a connection.
> > Perhaps the tooling is not always accurate in determining the commit which > causes the performance changes? > Perhaps I am misreading information in the reports? >
From what I remember, some of these tests print to the console, which has always been slow, and has generally been a bad idea for performance tests. I guess these tests are not very accurate.
Best regards Thomas
-- Thomas Zimmermann Graphics Driver Developer SUSE Software Solutions Germany GmbH Maxfeldstr. 5, 90409 Nürnberg, Germany (HRB 36809, AG Nürnberg) Geschäftsführer: Felix Imendörffer
[unhandled content-type:application/pgp-signature] | |