kvm_alloc_stage2_pgd has to do an order 9 allocation, ie. 512 contiguous pages I think. This often leads to problems running qemu when memory is relatively low -- eg. if you have one VM running, a healthy number of host applications, and perhaps "just" 4GB free; then you decide to run the libguestfs test suite. Any suggestions how to deal with this? Rich. NB: this is not the memory leak discussed previously -- that was fixed by applying: 4f853a714bf16338ff5261128e6c7ae2569e9505 arm/arm64: KVM: Fix and refactor unmap_range ---------------------------------------------------------------------- [268578.291005] qemu-system-aar: page allocation failure: order:9, mode:0xd0 [268578.297817] CPU: 3 PID: 26187 Comm: qemu-system-aar Not tainted 3.16.0-0.rc7.git4.1.rwmj4.fc22.aarch64 #1 [268578.307451] Call trace: [268578.309982] [<ffffffc000088d10>] dump_backtrace+0x0/0x174 [268578.315446] [<ffffffc000088ea0>] show_stack+0x1c/0x28 [268578.320589] [<ffffffc0007d4ae0>] dump_stack+0x80/0xac [268578.325736] [<ffffffc0001b4744>] warn_alloc_failed+0xd8/0x140 [268578.331551] [<ffffffc0001b9730>] __alloc_pages_nodemask+0x914/0xb90 [268578.337900] [<ffffffc0001b99e0>] __get_free_pages+0x34/0xac [268578.343541] [<ffffffc0000a1a50>] kvm_alloc_stage2_pgd+0x2c/0x88 [268578.349549] [<ffffffc00009eb94>] kvm_arch_init_vm+0x28/0xa0 [268578.355189] [<ffffffc00009b550>] kvm_dev_ioctl+0xfc/0x4e8 [268578.360675] [<ffffffc00023a390>] do_vfs_ioctl+0x364/0x5a4 [268578.366170] [<ffffffc00023a65c>] SyS_ioctl+0x8c/0xa4 [268578.371198] Mem-Info: [268578.373546] DMA per-cpu: [268578.376177] CPU 0: hi: 186, btch: 31 usd: 0 [268578.381034] CPU 1: hi: 186, btch: 31 usd: 0 [268578.385911] CPU 2: hi: 186, btch: 31 usd: 19 [268578.390767] CPU 3: hi: 186, btch: 31 usd: 0 [268578.395643] CPU 4: hi: 186, btch: 31 usd: 0 [268578.400499] CPU 5: hi: 186, btch: 31 usd: 0 [268578.405353] CPU 6: hi: 186, btch: 31 usd: 0 [268578.410231] CPU 7: hi: 186, btch: 31 usd: 0 [268578.415091] Normal per-cpu: [268578.417983] CPU 0: hi: 186, btch: 31 usd: 0 [268578.422837] CPU 1: hi: 186, btch: 31 usd: 0 [268578.427724] CPU 2: hi: 186, btch: 31 usd: 42 [268578.432581] CPU 3: hi: 186, btch: 31 usd: 0 [268578.437458] CPU 4: hi: 186, btch: 31 usd: 0 [268578.442315] CPU 5: hi: 186, btch: 31 usd: 0 [268578.447196] CPU 6: hi: 186, btch: 31 usd: 0 [268578.452053] CPU 7: hi: 186, btch: 31 usd: 0 [268578.456936] active_anon:905569 inactive_anon:179367 isolated_anon:0 active_file:1343354 inactive_file:1343154 isolated_file:0 unevictable:0 dirty:41358 writeback:0 unstable:0 free:25107 slab_reclaimable:262491 slab_unreclaimable:31308 mapped:22452 shmem:74 pagetables:2859 bounce:0 free_cma:3995 [268578.490794] DMA free:67996kB min:4100kB low:5124kB high:6148kB active_anon:851792kB inactive_anon:261100kB active_file:1333008kB inactive_file:1332204kB unevictable:0kB isolated(anon):0kB isolated(file):0kB present:4192256kB managed:4163884kB mlocked:0kB dirty:43304kB writeback:0kB mapped:24516kB shmem:100kB slab_reclaimable:268248kB slab_unreclaimable:32288kB kernel_stack:672kB pagetables:2308kB unstable:0kB bounce:0kB free_cma:15980kB writeback_tmp:0kB pages_scanned:0 all_unreclaimable? no [268578.534199] lowmem_reserve[]: 0 12036 12036 [268578.538534] Normal free:32168kB min:12136kB low:15168kB high:18204kB active_anon:2770484kB inactive_anon:456368kB active_file:4040668kB inactive_file:4040412kB unevictable:0kB isolated(anon):0kB isolated(file):0kB present:12582912kB managed:12325816kB mlocked:0kB dirty:122128kB writeback:0kB mapped:65292kB shmem:196kB slab_reclaimable:781716kB slab_unreclaimable:92944kB kernel_stack:2768kB pagetables:9128kB unstable:0kB bounce:0kB free_cma:0kB writeback_tmp:0kB pages_scanned:0 all_unreclaimable? no [268578.582548] lowmem_reserve[]: 0 0 0 [268578.586178] DMA: 2083*4kB (UEMC) 646*8kB (UEMR) 612*16kB (UEMR) 287*32kB (UEMRC) 148*64kB (UMR) 63*128kB (UM) 7*256kB (UM) 2*512kB (RC) 1*1024kB (C) 1*2048kB (C) 3*4096kB (C) = 68188kB [268578.602896] Normal: 2836*4kB (UEM) 121*8kB (UEM) 121*16kB (UEMR) 89*32kB (UEMR) 39*64kB (UEMR) 23*128kB (UEMR) 13*256kB (UEM) 5*512kB (UM) 2*1024kB (MR) 1*2048kB (R) 0*4096kB = 32520kB [268578.619619] Node 0 hugepages_total=0 hugepages_free=0 hugepages_surp=0 hugepages_size=2048kB [268578.628148] 2687070 total pagecache pages [268578.632228] 123 pages in swap cache [268578.635798] Swap cache stats: add 591, delete 468, find 98/134 [268578.641690] Free swap = 16775380kB [268578.645245] Total swap = 16777212kB [268578.648834] 4193792 pages RAM [268578.651878] 0 pages HighMem/MovableOnly [268578.655805] 64274 pages reserved -- Richard Jones, Virtualization Group, Red Hat http://people.redhat.com/~rjones Read my programming and virtualization blog: http://rwmj.wordpress.com virt-df lists disk usage of guests without needing to install any software inside the virtual machine. Supports Linux and Windows. http://people.redhat.com/~rjones/virt-df/ _______________________________________________ kvmarm mailing list kvmarm@xxxxxxxxxxxxxxxxxxxxx https://lists.cs.columbia.edu/mailman/listinfo/kvmarm
