Suspend/resume cycle testing may fail with X page allocation errors present

[greenfeld]

After one or more suspend/resume cyles with X page allocation failures, S/R testing may stop on its own with the XO powered on with X Windows running (and usually responsive).

Seen with on a B1 SKU 198 with 11.3.1 os10.

One example hang (note the two backtraces back-to-back):

rtcwake: wakeup from "mem" using /dev/rtc0 at Thu Jan  1 03:27:15 1970

[ 8813.775709] X: page allocation failure: order:11, mode:0x21
[ 8813.786840] [<c00359ac>] (unwind_backtrace+0x0/0x120) from [<c0090248>] (warn_alloc_failed+0xe0/0x104)
[ 8813.818695] [<c0090248>] (warn_alloc_failed+0xe0/0x104) from [<c00929a8>] (__alloc_pages_nodemask+0x528/0x580)
[ 8813.846312] [<c00929a8>] (__alloc_pages_nodemask+0x528/0x580) from [<c0092a78>] (__get_free_pages+0x10/0x3c)
[ 8813.869668] [<c0092a78>] (__get_free_pages+0x10/0x3c) from [<c01c16e4>] (gckOS_AllocatePagedMemoryEx+0x6c/0x188)
[ 8813.889579] [<c01c16e4>] (gckOS_AllocatePagedMemoryEx+0x6c/0x188) from [<c01c5b7c>] (gckVIDMEM_ConstructVirtual+0xc8/0x128)
[ 8813.919898] [<c01c5b7c>] (gckVIDMEM_ConstructVirtual+0xc8/0x128) from [<c01c34ec>] (_AllocateMemory+0x94/0x170)
[ 8813.939645] [<c01c34ec>] (_AllocateMemory+0x94/0x170) from [<c01c3a2c>] (gckKERNEL_Dispatch+0x290/0x518)
[ 8813.939704] [<c01c3a2c>] (gckKERNEL_Dispatch+0x290/0x518) from [<c01bff7c>] (drv_ioctl+0x1d0/0x2f0)
[ 8813.969664] [<c01bff7c>] (drv_ioctl+0x1d0/0x2f0) from [<c00c8420>] (do_vfs_ioctl+0x544/0x5b8)
[ 8813.978207] [<c00c8420>] (do_vfs_ioctl+0x544/0x5b8) from [<c00c84c8>] (sys_ioctl+0x34/0x54)
[ 8814.009770] [<c00c84c8>] (sys_ioctl+0x34/0x54) from [<c0030080>] (ret_fast_syscall+0x0/0x30)
[ 8814.009770] Mem-info:
[ 8814.029667] Normal per-cpu:
[ 8814.029690] CPU    0: hi:  186, btch:  31 usd: 168
[ 8814.032476] active_anon:5399 inactive_anon:114 isolated_anon:0
[ 8814.037246]  active_file:834 inactive_file:830 isolated_file:32
[ 8814.037255]  unevictable:8 dirty:0 writeback:12 unstable:0
[ 8814.037263]  free:91622 slab_reclaimable:538 slab_unreclaimable:2123
[ 8814.037280]  mapped:919 shmem:1433 pagetables:255 bounce:0
[ 8814.099929] Normal free:370828kB min:2672kB low:3340kB high:4008kB active_anon:21492kB inactive_anon:456kB active_file:1192kB inactive_file:1136k
B unevictable:32kB isolated(anon):104kB isolated(file):68kB present:446528kB mlocked:32kB dirty:0kB writeback:48kB mapped:2248kB shmem:5732kB slab_r
eclaimable:2136kB slab_unreclaimable:8492kB kernel_stack:1184kB pagetables:1020kB unstable:0kB bounce:0kB writeback_tmp:0kB pages_scanned:109 all_un
reclaimable? no
[ 8814.169974] lowmem_reserve[]: 0 0

[ 8814.173997] Normal: 3780*4kB 2975*8kB 2258*16kB 1483*32kB 1061*64kB 580*128kB 219*256kB 44*512kB 5*1024kB 4*2048kB 2*4096kB 1*8192kB 0*16384kB 0*
32768kB = 372936kB
[ 8814.189919] 1557 total pagecache pages
[ 8814.193907] 0 pages in swap cache
[ 8814.193918] Swap cache stats: add 0, delete 0, find 0/0
[ 8814.202468] Free swap  = 0kB
[ 8814.205435] Total swap = 0kB
[ 8814.225929] 129024 pages of RAM
[ 8814.225929] 95578 free pages
[ 8814.231992] 23007 reserved pages
[ 8814.236054] 2653 slab pages
[ 8814.236071] 651 pages shared
[ 8814.241880] 0 pages swap cached
[ 8814.399823] X: page allocation failure: order:11, mode:0x21
[ 8814.405444] [<c00359ac>] (unwind_backtrace+0x0/0x120) from [<c0090248>] (warn_alloc_failed+0xe0/0x104)
[ 8814.427847] [<c0090248>] (warn_alloc_failed+0xe0/0x104) from [<c00929a8>] (__alloc_pages_nodemask+0x528/0x580)
[ 8814.438408] [<c00929a8>] (__alloc_pages_nodemask+0x528/0x580) from [<c0092a78>] (__get_free_pages+0x10/0x3c)
[ 8814.448568] [<c0092a78>] (__get_free_pages+0x10/0x3c) from [<c01c16e4>] (gckOS_AllocatePagedMemoryEx+0x6c/0x188)
[ 8814.458850] [<c01c16e4>] (gckOS_AllocatePagedMemoryEx+0x6c/0x188) from [<c01c5b7c>] (gckVIDMEM_ConstructVirtual+0xc8/0x128)
[ 8814.470071] [<c01c5b7c>] (gckVIDMEM_ConstructVirtual+0xc8/0x128) from [<c01c34ec>] (_AllocateMemory+0x94/0x170)
[ 8814.480398] [<c01c34ec>] (_AllocateMemory+0x94/0x170) from [<c01c3a2c>] (gckKERNEL_Dispatch+0x290/0x518)
[ 8814.490559] [<c01c3a2c>] (gckKERNEL_Dispatch+0x290/0x518) from [<c01bff7c>] (drv_ioctl+0x1d0/0x2f0)
[ 8814.499864] [<c01bff7c>] (drv_ioctl+0x1d0/0x2f0) from [<c00c8420>] (do_vfs_ioctl+0x544/0x5b8)
[ 8814.508527] [<c00c8420>] (do_vfs_ioctl+0x544/0x5b8) from [<c00c84c8>] (sys_ioctl+0x34/0x54)
[ 8814.516966] [<c00c84c8>] (sys_ioctl+0x34/0x54) from [<c0030080>] (ret_fast_syscall+0x0/0x30)
[ 8814.525520] Mem-info:
[ 8814.527778] Normal per-cpu:
[ 8814.531604] CPU    0: hi:  186, btch:  31 usd:  53
[ 8814.536664] active_anon:5400 inactive_anon:113 isolated_anon:0
[ 8814.536699]  active_file:30 inactive_file:133 isolated_file:0
[ 8814.536699]  unevictable:8 dirty:0 writeback:2 unstable:0
[ 8814.536708]  free:93265 slab_reclaimable:530 slab_unreclaimable:2123
[ 8814.536725]  mapped:46 shmem:1433 pagetables:255 bounce:0
[ 8814.565952] Normal free:373060kB min:2672kB low:3340kB high:4008kB active_anon:21600kB inactive_anon:452kB active_file:120kB inactive_file:700kB unevictable:32kB isolated(anon):0kB isolated(file):0kB present:446528kB mlocked:32kB dirty:0kB writeback:8kB mapped:188kB shmem:5732kB slab_reclaimable:2120kB slab_unreclaimable:8492kB kernel_stack:1184kB pagetables:1020kB unstable:0kB bounce:0kB writeback_tmp:0kB pages_scanned:221 all_unreclaimable? no
[ 8814.605609] lowmem_reserve[]: 0 0
[ 8814.609278] Normal: 3715*4kB 2959*8kB 2250*16kB 1477*32kB 1052*64kB 579*128kB 223*256kB 45*512kB 5*1024kB 4*2048kB 2*4096kB 1*8192kB 0*16384kB 0*32768kB = 373060kB
[ 8814.624766] 1659 total pagecache pages
[ 8814.629612] 0 pages in swap cache
[ 8814.633323] Swap cache stats: add 0, delete 0, find 0/0
[ 8814.638752] Free swap  = 0kB
[ 8814.645218] Total swap = 0kB
[ 8814.664925] 129024 pages of RAM
[ 8814.670819] 95484 free pages
[ 8814.674217] 23007 reserved pages
[ 8814.678100] 2653 slab pages
[ 8814.681984] 641 pages shared
[ 8814.685405] 0 pages swap cached
[ 8814.705432] PM: Syncing filesystems ... done.
[ 8828.349661] dcon_freeze_store: 0
[ 8828.353481] dcon_freeze_store: no DCON ram, ignoring
[ 8886.194686] ADDRCONF(NETDEV_CHANGE): eth0: link becomes ready

[greenfeld]

Full test log up to a failure to suspend with X errors present

[carrott]

I am seeing similar errors unrelated to suspend resume, see #11478

[martin.langhoff]

Recent builds - OS17 OS18 don't show this problem.