Hello

The order-0 atomic allocation can be failed because of depletion of suitable free page.

Watermark check passes order-0 atomic allocation but it will be failed at finding a free page.

The  __rmqueue_smallest(zone, order, MIGRATE_HIGHATOMIC) can be used

only for highorder.

> > This means watermark min is not protected against non-atomic allocation.
> > The order-0 atomic allocation with ALLOC_HARDER unwantedly can be
> > failed. Additionally the __GFP_MEMALLOC allocation with
> > ALLOC_NO_WATERMARKS also can be failed.
> >
> > To avoid the problem, zone_watermark_fast should consider highatomic
> > reserve. If the actual size of high atomic free is counted accurately
> > like cma free, we may use it. On this patch just use
> > nr_reserved_highatomic. Additionally introduce
> > __zone_watermark_unusable_free to factor out common parts between
> > zone_watermark_fast and __zone_watermark_ok.
> >
> > This is trace log which shows GFP_HIGHUSER consumes free pages right
> > before ALLOC_NO_WATERMARKS.
> >
> >   <...>-22275 [006] ....   889.213383: mm_page_alloc: page=00000000d2be5665 pfn=970744 order=0 migratetype=0 nr_free=3650 gfp_flags=GFP_HIGHUSER|__GFP_ZERO
> >   <...>-22275 [006] ....   889.213385: mm_page_alloc: page=000000004b2335c2 pfn=970745 order=0 migratetype=0 nr_free=3650 gfp_flags=GFP_HIGHUSER|__GFP_ZERO
> >   <...>-22275 [006] ....   889.213387: mm_page_alloc: page=00000000017272e1 pfn=970278 order=0 migratetype=0 nr_free=3650 gfp_flags=GFP_HIGHUSER|__GFP_ZERO
> >   <...>-22275 [006] ....   889.213389: mm_page_alloc: page=00000000c4be79fb pfn=970279 order=0 migratetype=0 nr_free=3650 gfp_flags=GFP_HIGHUSER|__GFP_ZERO
> >   <...>-22275 [006] ....   889.213391: mm_page_alloc: page=00000000f8a51d4f pfn=970260 order=0 migratetype=0 nr_free=3650 gfp_flags=GFP_HIGHUSER|__GFP_ZERO
> >   <...>-22275 [006] ....   889.213393: mm_page_alloc: page=000000006ba8f5ac pfn=970261 order=0 migratetype=0 nr_free=3650 gfp_flags=GFP_HIGHUSER|__GFP_ZERO
> >   <...>-22275 [006] ....   889.213395: mm_page_alloc: page=00000000819f1cd3 pfn=970196 order=0 migratetype=0 nr_free=3650 gfp_flags=GFP_HIGHUSER|__GFP_ZERO
> >   <...>-22275 [006] ....   889.213396: mm_page_alloc: page=00000000f6b72a64 pfn=970197 order=0 migratetype=0 nr_free=3650 gfp_flags=GFP_HIGHUSER|__GFP_ZERO
> > kswapd0-1207  [005] ...1   889.213398: mm_page_alloc: page= (null) pfn=0 order=0 migratetype=1 nr_free=3650 gfp_flags=GFP_NOWAIT|__GFP_HIGHMEM|__GFP_NOWARN|__GFP_MOVABLE
> >
> > This is an example of ALLOC_HARDER allocation failure.
> >
> > <4>[ 6207.637280]  [3:  Binder:9343_3:22875] Binder:9343_3: page allocation failure: order:0, mode:0x480020(GFP_ATOMIC), nodemask=(null)
> > <4>[ 6207.637311]  [3:  Binder:9343_3:22875] Call trace:
> > <4>[ 6207.637346]  [3:  Binder:9343_3:22875] [<ffffff8008f40f8c>] dump_stack+0xb8/0xf0
> > <4>[ 6207.637356]  [3:  Binder:9343_3:22875] [<ffffff8008223320>] warn_alloc+0xd8/0x12c
> > <4>[ 6207.637365]  [3:  Binder:9343_3:22875] [<ffffff80082245e4>] __alloc_pages_nodemask+0x120c/0x1250
> > <4>[ 6207.637374]  [3:  Binder:9343_3:22875] [<ffffff800827f6e8>] new_slab+0x128/0x604
> > <4>[ 6207.637381]  [3:  Binder:9343_3:22875] [<ffffff800827b0cc>] ___slab_alloc+0x508/0x670
> > <4>[ 6207.637387]  [3:  Binder:9343_3:22875] [<ffffff800827ba00>] __kmalloc+0x2f8/0x310
> > <4>[ 6207.637396]  [3:  Binder:9343_3:22875] [<ffffff80084ac3e0>] context_struct_to_string+0x104/0x1cc
> > <4>[ 6207.637404]  [3:  Binder:9343_3:22875] [<ffffff80084ad8fc>] security_sid_to_context_core+0x74/0x144
> > <4>[ 6207.637412]  [3:  Binder:9343_3:22875] [<ffffff80084ad880>] security_sid_to_context+0x10/0x18
> > <4>[ 6207.637421]  [3:  Binder:9343_3:22875] [<ffffff800849bd80>] selinux_secid_to_secctx+0x20/0x28
> > <4>[ 6207.637430]  [3:  Binder:9343_3:22875] [<ffffff800849109c>] security_secid_to_secctx+0x3c/0x70
> > <4>[ 6207.637442]  [3:  Binder:9343_3:22875] [<ffffff8008bfe118>] binder_transaction+0xe68/0x454c
> > <4>[ 6207.637569]  [3:  Binder:9343_3:22875] Mem-Info:
> > <4>[ 6207.637595]  [3:  Binder:9343_3:22875] active_anon:102061 inactive_anon:81551 isolated_anon:0
> > <4>[ 6207.637595]  [3:  Binder:9343_3:22875]  active_file:59102 inactive_file:68924 isolated_file:64
> > <4>[ 6207.637595]  [3:  Binder:9343_3:22875]  unevictable:611 dirty:63 writeback:0 unstable:0
> > <4>[ 6207.637595]  [3:  Binder:9343_3:22875]  slab_reclaimable:13324 slab_unreclaimable:44354
> > <4>[ 6207.637595]  [3:  Binder:9343_3:22875]  mapped:83015 shmem:4858 pagetables:26316 bounce:0
> > <4>[ 6207.637595]  [3:  Binder:9343_3:22875]  free:2727 free_pcp:1035 free_cma:178
> > <4>[ 6207.637616]  [3:  Binder:9343_3:22875] Node 0 active_anon:408244kB inactive_anon:326204kB active_file:236408kB inactive_file:275696kB unevictable:2444kB isolated(anon):0kB isolated(file):256kB mapped:332060kB dirty:252kB writeback:0kB shmem:19432kB writeback_tmp:0kB unstable:0kB all_unreclaimable? no
> > <4>[ 6207.637627]  [3:  Binder:9343_3:22875] Normal free:10908kB min:6192kB low:44388kB high:47060kB active_anon:409160kB inactive_anon:325924kB active_file:235820kB inactive_file:276628kB unevictable:2444kB writepending:252kB present:3076096kB managed:2673676kB mlocked:2444kB kernel_stack:62512kB pagetables:105264kB bounce:0kB free_pcp:4140kB local_pcp:40kB free_cma:712kB
> > <4>[ 6207.637632]  [3:  Binder:9343_3:22875] lowmem_reserve[]: 0 0
> > <4>[ 6207.637637]  [3:  Binder:9343_3:22875] Normal: 505*4kB (H) 357*8kB (H) 201*16kB (H) 65*32kB (H) 1*64kB (H) 0*128kB 0*256kB 0*512kB 0*1024kB 0*2048kB 0*4096kB = 10236kB
>
> OK this shows we are well above min watermark, and indeed only free pages are
> highatomic. Why doesn't the rmqueue() part quoted above work as expected and
> allow this allocation to use those highatomic blocks?
>

Because highatomic free is reserved free only for high atomic, actually ALLOC_HARDER.

Order-0 atomic allocation cannot use the highatomic free.

> > <4>[ 6207.637655]  [3:  Binder:9343_3:22875] 138826 total pagecache pages
> > <4>[ 6207.637663]  [3:  Binder:9343_3:22875] 5460 pages in swap cache
> > <4>[ 6207.637668]  [3:  Binder:9343_3:22875] Swap cache stats: add 8273090, delete 8267506, find 1004381/4060142
> >
> > This is an example of ALLOC_NO_WATERMARKS allocation failure.
> >
> > <6>[  156.701551]  [4:        kswapd0: 1209] kswapd0 cpuset=/ mems_allowed=0
> > <4>[  156.701563]  [4:        kswapd0: 1209] CPU: 4 PID: 1209 Comm: kswapd0 Tainted: G        W       4.14.113-18113966 #1
> > <4>[  156.701572]  [4:        kswapd0: 1209] Call trace:
> > <4>[  156.701605]  [4:        kswapd0: 1209] [<0000000000000000>] dump_stack+0x68/0x90
> > <4>[  156.701612]  [4:        kswapd0: 1209] [<0000000000000000>] warn_alloc+0x104/0x198
> > <4>[  156.701617]  [4:        kswapd0: 1209] [<0000000000000000>] __alloc_pages_nodemask+0xdc0/0xdf0
> > <4>[  156.701623]  [4:        kswapd0: 1209] [<0000000000000000>] zs_malloc+0x148/0x3d0
> > <4>[  156.701630]  [4:        kswapd0: 1209] [<0000000000000000>] zram_bvec_rw+0x250/0x568
> > <4>[  156.701634]  [4:        kswapd0: 1209] [<0000000000000000>] zram_rw_page+0x8c/0xe0
> > <4>[  156.701640]  [4:        kswapd0: 1209] [<0000000000000000>] bdev_write_page+0x70/0xbc
> > <4>[  156.701645]  [4:        kswapd0: 1209] [<0000000000000000>] __swap_writepage+0x58/0x37c
> > <4>[  156.701649]  [4:        kswapd0: 1209] [<0000000000000000>] swap_writepage+0x40/0x4c
> > <4>[  156.701654]  [4:        kswapd0: 1209] [<0000000000000000>] shrink_page_list+0xc3c/0xf54
> > <4>[  156.701659]  [4:        kswapd0: 1209] [<0000000000000000>] shrink_inactive_list+0x2b0/0x61c
> > <4>[  156.701664]  [4:        kswapd0: 1209] [<0000000000000000>] shrink_node_memcg+0x23c/0x618
> > <4>[  156.701668]  [4:        kswapd0: 1209] [<0000000000000000>] shrink_node+0x1c8/0x304
> > <4>[  156.701673]  [4:        kswapd0: 1209] [<0000000000000000>] kswapd+0x680/0x7c4
> > <4>[  156.701679]  [4:        kswapd0: 1209] [<0000000000000000>] kthread+0x110/0x120
> > <4>[  156.701684]  [4:        kswapd0: 1209] [<0000000000000000>] ret_from_fork+0x10/0x18
> > <4>[  156.701689]  [4:        kswapd0: 1209] Mem-Info:
> > <4>[  156.701712]  [4:        kswapd0: 1209] active_anon:88690 inactive_anon:88630 isolated_anon:0
> > <4>[  156.701712]  [4:        kswapd0: 1209]  active_file:99173 inactive_file:169305 isolated_file:32
> > <4>[  156.701712]  [4:        kswapd0: 1209]  unevictable:48292 dirty:538 writeback:38 unstable:0
> > <4>[  156.701712]  [4:        kswapd0: 1209]  slab_reclaimable:15131 slab_unreclaimable:47762
> > <4>[  156.701712]  [4:        kswapd0: 1209]  mapped:274654 shmem:2824 pagetables:25088 bounce:0
> > <4>[  156.701712]  [4:        kswapd0: 1209]  free:2489 free_pcp:444 free_cma:3
> > <4>[  156.701728]  [4:        kswapd0: 1209] Node 0 active_anon:354760kB inactive_anon:354520kB active_file:396692kB inactive_file:677220kB unevictable:193168kB isolated(anon):0kB isolated(file):128kB mapped:1098616kB dirty:2152kB writeback:152kB shmem:11296kB writeback_tmp:0kB unstable:0kB all_unreclaimable? no
> > <4>[  156.701738]  [4:        kswapd0: 1209] Normal free:9956kB min:7428kB low:93440kB high:97032kB active_anon:355176kB inactive_anon:354580kB active_file:396196kB inactive_file:677284kB unevictable:193168kB writepending:2304kB present:4081664kB managed:3593324kB mlocked:193168kB kernel_stack:55008kB pagetables:100352kB bounce:0kB free_pcp:1776kB local_pcp:656kB free_cma:12kB
> > <4>[  156.701741]  [4:        kswapd0: 1209] lowmem_reserve[]: 0 0
> > <4>[  156.701747]  [4:        kswapd0: 1209] Normal: 196*4kB (H) 141*8kB (H) 109*16kB (H) 63*32kB (H) 20*64kB (H) 8*128kB (H) 2*256kB (H) 1*512kB (H) 0*1024kB 0*2048kB 0*4096kB = 9000kB
>
> Same here, although here AFAICS ALLOC_NO_WATERMARKS doesn't imply ALLOC_HARDER,
> so the rmqueue() check wouldn't kick in? That would be something to fix... but
> doesn't explain the GFP_ATOMIC case above.

I might miss some log above, AFAIK it was order-0 allocation.

For an order-0 allocation with ALLOC_NO_WATERMARKS, rmqueue will search free

a page but it cannot find one because other direct reclaimable allocation already took

the suitable free without direct reclaimatiion.

>
> ...
>
> > @@ -3598,9 +3604,12 @@ static inline bool zone_watermark_fast(struct zone *z, unsigned int order,
>         /*
>          * Fast check for order-0 only. If this fails then the reserves
>          * need to be calculated. There is a corner case where the check
>          * passes but only the high-order atomic reserve are free. If
> >        * the caller is !atomic then it'll uselessly search the free
> >        * list. That corner case is then slower but it is harmless.
> >        */
>
> The comment stops being true after this patch? It also suggests that Mel
> anticipated this corner case, but that it should only cause a false positive
> zone_watermark_fast() and then rmqueue() fails for !ALLOC_HARDER as it cannot
> use MIGRATE_HIGHATOMIC blocks. It expects atomic order-0 still works. So what's
> going on?

As Mel also agreed with me in v1 mail thread, this highatomic reserved should

be considered even in watermark fast.

The comment, I think, may need to be changed. Prior to this patch, non highatomic

allocation may do useless search, but it also can take ALL non highatomic free.

With this patch, non highatomic allocation will NOT do useless search. Rather,

it may be required direct reclamation even when there are some non high atomic free.

i.e)

In following situation, watermark check fails (9MB - 8MB < 4MB) though there are

enough free (9MB - 4MB > 4MB). If this is really matter, we need to count highatomic

free accurately.

min : 4MB, 

highatomic reserved : 8MB

Total free : 9MB

  actual highatomic free : 4MB

  non highatomic free : 5MB

>
> > -     if (!order && (free_pages - cma_pages) >
> > -                             mark + z->lowmem_reserve[highest_zoneidx])
> > -             return true;
> > +     if (!order) {
> > +             long fast_free = free_pages - unusable_free;
> > +
> > +             if (fast_free > mark + z->lowmem_reserve[highest_zoneidx])
> > +                     return true;
> > +     }
> >
> >       return __zone_watermark_ok(z, order, mark, highest_zoneidx, alloc_flags,
> >                                       free_pages);
> >
>