On Mon, Jun 17, 2024 at 3:12 PM Barry Song <21cnbao@xxxxxxxxx> wrote: > > > > 在 2024年6月17日星期一，Huang, Ying <ying.huang@xxxxxxxxx> 写道： >> >> >> Barry Song <21cnbao@xxxxxxxxx> writes: >> >> > On Sat, Jun 15, 2024 at 2:59 PM Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx> wrote: >> >> >> >> On Fri, 14 Jun 2024 19:51:11 -0700 Chris Li <chrisl@xxxxxxxxxx> wrote: >> >> >> >> > > I'm having trouble understanding the overall impact of this on users. >> >> > > We fail the mTHP swap allocation and fall back, but things continue to >> >> > > operate OK? >> >> > >> >> > Continue to operate OK in the sense that the mTHP will have to split >> >> > into 4K pages before the swap out, aka the fall back. The swap out and >> >> > swap in can continue to work as 4K pages, not as the mTHP. Due to the >> >> > fallback, the mTHP based zsmalloc compression with 64K buffer will not >> >> > happen. That is the effect of the fallback. But mTHP swap out and swap >> >> > in is relatively new, it is not really a regression. >> >> >> >> Sure, but it's pretty bad to merge a new feature only to have it >> >> ineffective after a few hours use. >> >> >> >> > > >> >> > > > There is some test number in the V1 thread of this series: >> >> > > > https://lore.kernel.org/r/20240524-swap-allocator-v1-0-47861b423b26@xxxxxxxxxx >> >> > > >> >> > > Well, please let's get the latest numbers into the latest patchset. >> >> > > Along with a higher-level (and quantitative) description of the user impact. >> >> > >> >> > I will need Barray's help to collect the number. I don't have the >> >> > setup to reproduce his test result. >> >> > Maybe a follow up commit message amendment for the test number when I get it? >> > >> > Although the issue may seem complex at a systemic level, even a small program can >> > demonstrate the problem and highlight how Chris's patch has improved the >> > situation. >> > >> > To demonstrate this, I designed a basic test program that maximally allocates >> > two memory blocks: >> > >> > * A memory block of up to 60MB, recommended for HUGEPAGE usage >> > * A memory block of up to 1MB, recommended for NOHUGEPAGE usage >> > >> > In the system configuration, I enabled 64KB mTHP and 64MB zRAM, providing more than >> > enough space for both the 60MB and 1MB allocations in the worst case. This setup >> > allows us to assess two effects: >> > >> > 1. When we don't enable mem2 (small folios), we consistently allocate and free >> > swap slots aligned with 64KB. whether there is a risk of failure to obtain >> > swap slots even though the zRAM has sufficient free space? >> > 2. When we enable mem2 (small folios), the presence of small folios may lead >> > to fragmentation of clusters, potentially impacting the swapout process for >> > large folios negatively. >> > >> >> IIUC, the test results are based on not-yet-merged patchset [1] (mm: >> support large folios swap-in)? > > > no. this data is based on mm-unstable. > > the visible impact is that swapping out mthp will have 14% regression if > fallback againest swapping out nr_pages small folios regardless if mthp swapin is there. Ryan initially reported 14% swapout regression without mTHP swapout. then he reported 46.3% improvement if mTHP can be swapped out as a whole[1]. so we will drop 60%+ performance if fallback. but the 14% regression against pure small folios are unacceptable considering more than 2/3 memory can be swapped out on mobile devices. So I am hoping we can find some way to merge Chris' patchset soon. though the WARN_ONCE still indicates some BUG in v2. Hopefully, Chris can fix it in v3. [1] https://lore.kernel.org/all/20240408183946.2991168-1-ryan.roberts@xxxxxxx/ > > >> >> [1] https://lore.kernel.org/linux-mm/20240304081348.197341-1-21cnbao@xxxxxxxxx/ >> >> If so, do we have any visible effect without that? If not, should we >> wait for patchset [1] (or something similar) to be merged firstly? >> >> -- >> Best Regards, >> Huang, Ying >> Thanks Barry
