Re: Recent update to page_alloc.c causing a crash
On Mon 2 July 2012 09:25:08 Prabhakar Lad wrote: Hi, Recently when I updated my driver to 3.5 from 3.3, I am observing that my driver is failing for dma_alloc_coherent, when I traced it, I see lots of changes for MM in 3.5 release any Ideas why this could be happening? Here is the case my driver works fine for 7 runs of the application for 8th time it crashes for MMAP buffers below is the crash log. Hi Prabhakar, I've got to ask the obvious question: have you checked that the memory allocated in the first 7 runs is also freed? In other words, aren't you just running out of memory? The other option is memory fragmentation, making it impossible to allocate the contiguous memory that you need. I actually suspect that this is the reason for you running out of memory (it's not a crash, it's just telling you that it can't allocate the memory you need). The only correct way to solve this problem is to use CMA. As far as I can tell you can just configure the amount of CMA memory as a config option or as a kernel parameter (see Documentation/kernel-parameters.txt, search for cma). That should ensure that the amount of memory you need is always available for the buffers. Regards, Hans Thx, --Prabhakar Lad vpif_mmap_loopb: page allocation failure: order:10, mode:0xd0 Backtrace: [c000c608] (dump_backtrace+0x0/0x114) from [c033cfc0] (dump_stack+0x18/0x1c) r6: r5:00d0 r4:0001 r3:c048a060 [c033cfa8] (dump_stack+0x0/0x1c) from [c0061c00] (warn_alloc_failed+0xf4/0x118) [c0061b0c] (warn_alloc_failed+0x0/0x118) from [c00627f4] (__alloc_pages_nodemask+0x544/0x590) r3:c2cfbb64 r2: r7:c04b2f94 r6:c2cfa000 r5:000a r4:00d0 [c00622b0] (__alloc_pages_nodemask+0x0/0x590) from [c000e084] (arm_dma_alloc+0x138/0x358) [c000df4c] (arm_dma_alloc+0x0/0x358) from [c024ca94] (vb2_dma_contig_alloc+0x78/0xfc) [c024ca1c] (vb2_dma_contig_alloc+0x0/0xfc) from [c024a6cc] (__vb2_queue_alloc+0xd4/0x338) r6:c36d5400 r5: r4:c361a190 [c024a5f8] (__vb2_queue_alloc+0x0/0x338) from [c024adfc] (vb2_reqbufs+0x22c/0x308) [c024abd0] (vb2_reqbufs+0x0/0x308) from [c025399c] (vpif_reqbufs+0x114/0x148) r7:0001 r6:c2cfbe20 r5:c361a000 r4:c2d09400 [c0253888] (vpif_reqbufs+0x0/0x148) from [c023a470] (__video_do_ioctl+0x1a14/0x4914) r8:c0145608 r7:c03601e8 r6: r5:c3563000 r4:c2cfbe20 r3: [c0238a5c] (__video_do_ioctl+0x0/0x4914) from [c0238890] (video_usercopy+0x368/0x4b4) [c0238528] (video_usercopy+0x0/0x4b4) from [c02389f0] (video_ioctl2+0x14/0x1c) [c02389dc] (video_ioctl2+0x0/0x1c) from [c02367e8] (v4l2_ioctl+0xac/0x158) [c023673c] (v4l2_ioctl+0x0/0x158) from [c009e67c] (vfs_ioctl+0x28/0x40) r8:c00095a4 r7:0003 r6:c37992b0 r5:c2c77e80 r4:bed27a24 r3:c023673c [c009e654] (vfs_ioctl+0x0/0x40) from [c009edd8] (do_vfs_ioctl+0x52c/0x588) [c009e8ac] (do_vfs_ioctl+0x0/0x588) from [c009ee74] (sys_ioctl+0x40/0x64) r9:c2cfa000 r8:c00095a4 r7:0003 r6:c0145608 r5:bed27a24 r4:c2c77e80 [c009ee34] (sys_ioctl+0x0/0x64) from [c0009420] (ret_fast_syscall+0x0/0x2c) r7:0036 r6:8710 r5: r4:b308 Mem-info: DMA per-cpu: CPU0: hi: 18, btch: 3 usd: 3 active_anon:315 inactive_anon:18 isolated_anon:0 active_file:3 inactive_file:112 isolated_file:0 unevictable:0 dirty:0 writeback:0 unstable:0 free:12196 slab_reclaimable:236 slab_unreclaimable:569 mapped:31 shmem:22 pagetables:34 bounce:0 DMA free:48712kB min:984kB low:1228kB high:1476kB active_anon:1260kB inactive_anon:72kB active_file:12kB inactive_file:560kB unevictable:0kB isolated(anon):0kB isolated(file):0kB present:609o lowmem_reserve[]: 0 0 0 DMA: 69*4kB 108*8kB 115*16kB 53*32kB 44*64kB 19*128kB 13*256kB 9*512kB 8*1024kB 11*2048kB 0*4096kB = 48580kB 217 total pagecache pages 0 pages in swap cache Swap cache stats: add 0, delete 0, find 0/0 Free swap = 0kB Total swap = 0kB 15360 pages of RAM 12199 free pages 1335 reserved pages 669 slab pages 155 pages shared 0 pages swap cached vpif_capture vpif_capture: dma_alloc_coherent of size 2622464 failed cannot allocate memory : Cannot allocate memory -- To unsubscribe from this list: send the line unsubscribe linux-media in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
Re: Recent update to page_alloc.c causing a crash
Hi Hans, On Mon, Jul 2, 2012 at 1:19 PM, Hans Verkuil hverk...@xs4all.nl wrote: On Mon 2 July 2012 09:25:08 Prabhakar Lad wrote: Hi, Recently when I updated my driver to 3.5 from 3.3, I am observing that my driver is failing for dma_alloc_coherent, when I traced it, I see lots of changes for MM in 3.5 release any Ideas why this could be happening? Here is the case my driver works fine for 7 runs of the application for 8th time it crashes for MMAP buffers below is the crash log. Hi Prabhakar, I've got to ask the obvious question: have you checked that the memory allocated in the first 7 runs is also freed? In other words, aren't you just running out of memory? Yes Of course, I tested the same application on 3.3 and it works fine. Yes I am running out of memory. The other option is memory fragmentation, making it impossible to allocate the contiguous memory that you need. I actually suspect that this is the reason for you running out of memory (it's not a crash, it's just telling you that it can't allocate the memory you need). Yes its soft crash. The only correct way to solve this problem is to use CMA. As far as I can tell you can just configure the amount of CMA memory as a config option or as a kernel parameter (see Documentation/kernel-parameters.txt, search for cma). That should ensure that the amount of memory you need is always available for the buffers. I tried with cma=60M and its still the same case :( a soft crash. Thx, --Prabhakar Lad Regards, Hans Thx, --Prabhakar Lad vpif_mmap_loopb: page allocation failure: order:10, mode:0xd0 Backtrace: [c000c608] (dump_backtrace+0x0/0x114) from [c033cfc0] (dump_stack+0x18/0x1c) r6: r5:00d0 r4:0001 r3:c048a060 [c033cfa8] (dump_stack+0x0/0x1c) from [c0061c00] (warn_alloc_failed+0xf4/0x118) [c0061b0c] (warn_alloc_failed+0x0/0x118) from [c00627f4] (__alloc_pages_nodemask+0x544/0x590) r3:c2cfbb64 r2: r7:c04b2f94 r6:c2cfa000 r5:000a r4:00d0 [c00622b0] (__alloc_pages_nodemask+0x0/0x590) from [c000e084] (arm_dma_alloc+0x138/0x358) [c000df4c] (arm_dma_alloc+0x0/0x358) from [c024ca94] (vb2_dma_contig_alloc+0x78/0xfc) [c024ca1c] (vb2_dma_contig_alloc+0x0/0xfc) from [c024a6cc] (__vb2_queue_alloc+0xd4/0x338) r6:c36d5400 r5: r4:c361a190 [c024a5f8] (__vb2_queue_alloc+0x0/0x338) from [c024adfc] (vb2_reqbufs+0x22c/0x308) [c024abd0] (vb2_reqbufs+0x0/0x308) from [c025399c] (vpif_reqbufs+0x114/0x148) r7:0001 r6:c2cfbe20 r5:c361a000 r4:c2d09400 [c0253888] (vpif_reqbufs+0x0/0x148) from [c023a470] (__video_do_ioctl+0x1a14/0x4914) r8:c0145608 r7:c03601e8 r6: r5:c3563000 r4:c2cfbe20 r3: [c0238a5c] (__video_do_ioctl+0x0/0x4914) from [c0238890] (video_usercopy+0x368/0x4b4) [c0238528] (video_usercopy+0x0/0x4b4) from [c02389f0] (video_ioctl2+0x14/0x1c) [c02389dc] (video_ioctl2+0x0/0x1c) from [c02367e8] (v4l2_ioctl+0xac/0x158) [c023673c] (v4l2_ioctl+0x0/0x158) from [c009e67c] (vfs_ioctl+0x28/0x40) r8:c00095a4 r7:0003 r6:c37992b0 r5:c2c77e80 r4:bed27a24 r3:c023673c [c009e654] (vfs_ioctl+0x0/0x40) from [c009edd8] (do_vfs_ioctl+0x52c/0x588) [c009e8ac] (do_vfs_ioctl+0x0/0x588) from [c009ee74] (sys_ioctl+0x40/0x64) r9:c2cfa000 r8:c00095a4 r7:0003 r6:c0145608 r5:bed27a24 r4:c2c77e80 [c009ee34] (sys_ioctl+0x0/0x64) from [c0009420] (ret_fast_syscall+0x0/0x2c) r7:0036 r6:8710 r5: r4:b308 Mem-info: DMA per-cpu: CPU0: hi: 18, btch: 3 usd: 3 active_anon:315 inactive_anon:18 isolated_anon:0 active_file:3 inactive_file:112 isolated_file:0 unevictable:0 dirty:0 writeback:0 unstable:0 free:12196 slab_reclaimable:236 slab_unreclaimable:569 mapped:31 shmem:22 pagetables:34 bounce:0 DMA free:48712kB min:984kB low:1228kB high:1476kB active_anon:1260kB inactive_anon:72kB active_file:12kB inactive_file:560kB unevictable:0kB isolated(anon):0kB isolated(file):0kB present:609o lowmem_reserve[]: 0 0 0 DMA: 69*4kB 108*8kB 115*16kB 53*32kB 44*64kB 19*128kB 13*256kB 9*512kB 8*1024kB 11*2048kB 0*4096kB = 48580kB 217 total pagecache pages 0 pages in swap cache Swap cache stats: add 0, delete 0, find 0/0 Free swap = 0kB Total swap = 0kB 15360 pages of RAM 12199 free pages 1335 reserved pages 669 slab pages 155 pages shared 0 pages swap cached vpif_capture vpif_capture: dma_alloc_coherent of size 2622464 failed cannot allocate memory : Cannot allocate memory -- To unsubscribe from this list: send the line unsubscribe linux-media in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
Re: Recent update to page_alloc.c causing a crash
Hi Hans, On Mon, Jul 2, 2012 at 1:37 PM, Prabhakar Lad prabhakar.cse...@gmail.com wrote: Hi Hans, On Mon, Jul 2, 2012 at 1:19 PM, Hans Verkuil hverk...@xs4all.nl wrote: On Mon 2 July 2012 09:25:08 Prabhakar Lad wrote: Hi, Recently when I updated my driver to 3.5 from 3.3, I am observing that my driver is failing for dma_alloc_coherent, when I traced it, I see lots of changes for MM in 3.5 release any Ideas why this could be happening? Here is the case my driver works fine for 7 runs of the application for 8th time it crashes for MMAP buffers below is the crash log. Hi Prabhakar, I've got to ask the obvious question: have you checked that the memory allocated in the first 7 runs is also freed? In other words, aren't you just running out of memory? Yes Of course, I tested the same application on 3.3 and it works fine. Yes I am running out of memory. The other option is memory fragmentation, making it impossible to allocate the contiguous memory that you need. I actually suspect that this is the reason for you running out of memory (it's not a crash, it's just telling you that it can't allocate the memory you need). Yes its soft crash. The only correct way to solve this problem is to use CMA. As far as I can tell you can just configure the amount of CMA memory as a config option or as a kernel parameter (see Documentation/kernel-parameters.txt, search for cma). That should ensure that the amount of memory you need is always available for the buffers. I tried with cma=60M and its still the same case :( a soft crash. Got it fixed problem was with application :) Thx, --Prabhakar Lad Thx, --Prabhakar Lad Regards, Hans Thx, --Prabhakar Lad vpif_mmap_loopb: page allocation failure: order:10, mode:0xd0 Backtrace: [c000c608] (dump_backtrace+0x0/0x114) from [c033cfc0] (dump_stack+0x18/0x1c) r6: r5:00d0 r4:0001 r3:c048a060 [c033cfa8] (dump_stack+0x0/0x1c) from [c0061c00] (warn_alloc_failed+0xf4/0x118) [c0061b0c] (warn_alloc_failed+0x0/0x118) from [c00627f4] (__alloc_pages_nodemask+0x544/0x590) r3:c2cfbb64 r2: r7:c04b2f94 r6:c2cfa000 r5:000a r4:00d0 [c00622b0] (__alloc_pages_nodemask+0x0/0x590) from [c000e084] (arm_dma_alloc+0x138/0x358) [c000df4c] (arm_dma_alloc+0x0/0x358) from [c024ca94] (vb2_dma_contig_alloc+0x78/0xfc) [c024ca1c] (vb2_dma_contig_alloc+0x0/0xfc) from [c024a6cc] (__vb2_queue_alloc+0xd4/0x338) r6:c36d5400 r5: r4:c361a190 [c024a5f8] (__vb2_queue_alloc+0x0/0x338) from [c024adfc] (vb2_reqbufs+0x22c/0x308) [c024abd0] (vb2_reqbufs+0x0/0x308) from [c025399c] (vpif_reqbufs+0x114/0x148) r7:0001 r6:c2cfbe20 r5:c361a000 r4:c2d09400 [c0253888] (vpif_reqbufs+0x0/0x148) from [c023a470] (__video_do_ioctl+0x1a14/0x4914) r8:c0145608 r7:c03601e8 r6: r5:c3563000 r4:c2cfbe20 r3: [c0238a5c] (__video_do_ioctl+0x0/0x4914) from [c0238890] (video_usercopy+0x368/0x4b4) [c0238528] (video_usercopy+0x0/0x4b4) from [c02389f0] (video_ioctl2+0x14/0x1c) [c02389dc] (video_ioctl2+0x0/0x1c) from [c02367e8] (v4l2_ioctl+0xac/0x158) [c023673c] (v4l2_ioctl+0x0/0x158) from [c009e67c] (vfs_ioctl+0x28/0x40) r8:c00095a4 r7:0003 r6:c37992b0 r5:c2c77e80 r4:bed27a24 r3:c023673c [c009e654] (vfs_ioctl+0x0/0x40) from [c009edd8] (do_vfs_ioctl+0x52c/0x588) [c009e8ac] (do_vfs_ioctl+0x0/0x588) from [c009ee74] (sys_ioctl+0x40/0x64) r9:c2cfa000 r8:c00095a4 r7:0003 r6:c0145608 r5:bed27a24 r4:c2c77e80 [c009ee34] (sys_ioctl+0x0/0x64) from [c0009420] (ret_fast_syscall+0x0/0x2c) r7:0036 r6:8710 r5: r4:b308 Mem-info: DMA per-cpu: CPU0: hi: 18, btch: 3 usd: 3 active_anon:315 inactive_anon:18 isolated_anon:0 active_file:3 inactive_file:112 isolated_file:0 unevictable:0 dirty:0 writeback:0 unstable:0 free:12196 slab_reclaimable:236 slab_unreclaimable:569 mapped:31 shmem:22 pagetables:34 bounce:0 DMA free:48712kB min:984kB low:1228kB high:1476kB active_anon:1260kB inactive_anon:72kB active_file:12kB inactive_file:560kB unevictable:0kB isolated(anon):0kB isolated(file):0kB present:609o lowmem_reserve[]: 0 0 0 DMA: 69*4kB 108*8kB 115*16kB 53*32kB 44*64kB 19*128kB 13*256kB 9*512kB 8*1024kB 11*2048kB 0*4096kB = 48580kB 217 total pagecache pages 0 pages in swap cache Swap cache stats: add 0, delete 0, find 0/0 Free swap = 0kB Total swap = 0kB 15360 pages of RAM 12199 free pages 1335 reserved pages 669 slab pages 155 pages shared 0 pages swap cached vpif_capture vpif_capture: dma_alloc_coherent of size 2622464 failed cannot allocate memory : Cannot allocate memory -- To unsubscribe from this list: send the line unsubscribe linux-media in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
