On Mon, May 09, 2022 at 05:45:51PM +0800, Dong Aisheng wrote:
> This reverts commit a4efc174b382fcdb62e2d90d39e78a274a975e38 which
> introduced a regression issue that when there're multiple processes
> allocating dma memory in parallel by calling dma_alloc_coherent(), it
> may fail sometimes as follows:
>
> Error log:
> cma: cma_alloc: linux,cma: alloc failed, req-size: 148 pages, ret: -16
> cma: number of available pages:
> 3@125+20@172+12@236+4@380+32@736+17@2287+23@2473+20@36076+99@40477+108@40852+44@41108+20@41196+108@41364+108@41620+
> 108@42900+108@43156+483@44061+1763@45341+1440@47712+20@49324+20@49388+5076@49452+2304@55040+35@58141+20@58220+20@58284+
> 7188@58348+84@66220+7276@66452+227@74525+6371@75549=> 33161 free of 81920 total pages
>
> When issue happened, we saw there were still 33161 pages (129M) free CMA
> memory and a lot available free slots for 148 pages in CMA bitmap that we
> want to allocate.
>
> When dumping memory info, we found that there was also ~342M normal memory,
> but only 1352K CMA memory left in buddy system while a lot of pageblocks
> were isolated.
>
> Memory info log:
> Normal free:351096kB min:30000kB low:37500kB high:45000kB reserved_highatomic:0KB
> active_anon:98060kB inactive_anon:98948kB active_file:60864kB inactive_file:31776kB
> unevictable:0kB writepending:0kB present:1048576kB managed:1018328kB mlocked:0kB
> bounce:0kB free_pcp:220kB local_pcp:192kB free_cma:1352kB lowmem_reserve[]: 0 0 0
> Normal: 78*4kB (UECI) 1772*8kB (UMECI) 1335*16kB (UMECI) 360*32kB (UMECI) 65*64kB (UMCI)
> 36*128kB (UMECI) 16*256kB (UMCI) 6*512kB (EI) 8*1024kB (UEI) 4*2048kB (MI) 8*4096kB (EI)
> 8*8192kB (UI) 3*16384kB (EI) 8*32768kB (M) = 489288kB
>
> The root cause of this issue is that since commit a4efc174b382
> ("mm/cma.c: remove redundant cma_mutex lock"), CMA supports concurrent
> memory allocation. It's possible that the memory range process A trying
> to alloc has already been isolated by the allocation of process B during
> memory migration.
>
> The problem here is that the memory range isolated during one allocation
> by start_isolate_page_range() could be much bigger than the real size we
> want to alloc due to the range is aligned to MAX_ORDER_NR_PAGES.
>
> Taking an ARMv7 platform with 1G memory as an example, when MAX_ORDER_NR_PAGES
> is big (e.g. 32M with max_order 14) and CMA memory is relatively small
> (e.g. 128M), there're only 4 MAX_ORDER slot, then it's very easy that
> all CMA memory may have already been isolated by other processes when
> one trying to allocate memory using dma_alloc_coherent().
> Since current CMA code will only scan one time of whole available CMA
> memory, then dma_alloc_coherent() may easy fail due to contention with
> other processes.
>
> This patch simply falls back to the original method that using cma_mutex
> to make alloc_contig_range() run sequentially to avoid the issue.
>
Link: https://lore.kernel.org/all/20220315144521.3810298-2-aisheng.dong@xxxxxxx/
Adding a link would be helpful why decided to revert.
> Cc: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>
> Cc: Marek Szyprowski <m.szyprowski@xxxxxxxxxxx>
> Cc: Lecopzer Chen <lecopzer.chen@xxxxxxxxxxxx>
> Cc: David Hildenbrand <david@xxxxxxxxxx>
> Cc: Vlastimil Babka <vbabka@xxxxxxx>
> Cc: Minchan Kim <minchan@xxxxxxxxxx>
> CC: stable@xxxxxxxxxxxxxxx # 5.11+
> Fixes: a4efc174b382 ("mm/cma.c: remove redundant cma_mutex lock")
> Signed-off-by: Dong Aisheng <aisheng.dong@xxxxxxx>
Acked-by: Minchan Kim <minchan@xxxxxxxxxx>
