Messages in this thread | | | Date | Tue, 28 Aug 2001 13:22:59 +0200 | From | Stephan von Krawczynski <> | Subject | Re: [PATCH][RFC] simpler __alloc_pages{_limit} |
| |
On Sat, 25 Aug 2001 21:24:40 +0200 Roger Larsson <roger.larsson@skelleftea.mail.telia.com> wrote:
> I would like to see from were these higher order allocs comes from. > Please insert something like this: > > printk("pid=%d; __alloc_pages(gfp=0x%x, order=%ld, ...)\n", > current->pid, gfp_mask, order); > show_trace(NULL);
Ok, Roger (and all the others involved). I am back with new tests for memory problems. Current test setup is:
software: 2.4.10-pre1 with Marco Tosatti's latest mem patch. hardware: P3 1GHz 1 GB RAM, Adaptec SCSI, IBM 36GB HD 100MBit Tulip (roughly :-)
I copy files from via NFS to the machine on a reiserfs partition and read a CD (with xcdroast) to the same partition (reading an image). A note on that: NFS is exported with option no_subtree_check, because otherwise it fails on low mem condition and truncates files. This effect _only_ shows when exporting a reiserfs filesystem, on ext2 works well. Thanks to Neil Brown for helping me find this out.
meminfo before test:
total: used: free: shared: buffers: cached: Mem: 921726976 92602368 829124608 0 6684672 37040128 Swap: 271392768 0 271392768 MemTotal: 900124 kB MemFree: 809692 kB MemShared: 0 kB Buffers: 6528 kB Cached: 36172 kB SwapCached: 0 kB Active: 3672 kB Inact_dirty: 39028 kB Inact_clean: 0 kB Inact_target: 704 kB HighTotal: 0 kB HighFree: 0 kB LowTotal: 900124 kB LowFree: 809692 kB SwapTotal: 265032 kB SwapFree: 265032 kB
First experience during test: CPU load is pretty high (varying around 6 - 8), sometimes results from "cat /proc/meminfo" take half a minute to show up. Anyway, everything runs ok (besides the fact that it looks slow) until reading the CD is done. Then this shows up:
Aug 28 13:08:20 admin kernel: __alloc_pages: 3-order allocation failed. Aug 28 13:08:20 admin kernel: pid=1139; __alloc_pages(gfp=0x20, order=3, ...) Aug 28 13:08:20 admin kernel: Call Trace: [_alloc_pages+22/24] [__get_free_pages+10/24] [<fdcec826>] [<fdcec8f5>] [<fdceb7d7>] Aug 28 13:08:20 admin kernel: [<fdcec0f5>] [<fdcea589>] [_alloc_pages+22/24] [__get_free_pages+10/24] [<fdcec826>] [<fdcec8f5>] Aug 28 13:08:20 admin kernel: [<fdceb6bd>] [filemap_nopage+171/1008] [do_no_page+90/244] [handle_mm_fault+97/192] [<fdce94aa>] [do_page_fault+0/1164] Aug 28 13:08:20 admin kernel: [dentry_open+189/316] [filp_open+82/92] [do_fcntl+370/712] [sys_ioctl+443/532] [system_call+51/56] Aug 28 13:08:20 admin kernel: __alloc_pages: 3-order allocation failed. Aug 28 13:08:20 admin kernel: pid=1139; __alloc_pages(gfp=0x20, order=3, ...) Aug 28 13:08:20 admin kernel: Call Trace: [_alloc_pages+22/24] [__get_free_pages+10/24] [<fdcec826>] [<fdcec8f5>] [<fdceb7d7>] Aug 28 13:08:20 admin kernel: [<fdcec0f5>] [<fdcea589>] [_alloc_pages+22/24] [__get_free_pages+10/24] [<fdcec826>] [<fdcec8f5>] Aug 28 13:08:20 admin kernel: [<fdceb6bd>] [filemap_nopage+171/1008] [do_no_page+90/244] [handle_mm_fault+97/192] [<fdce94aa>] [do_page_fault+0/1164] Aug 28 13:08:20 admin kernel: [dentry_open+189/316] [filp_open+82/92] [do_fcntl+370/712] [sys_ioctl+443/532] [system_call+51/56] Aug 28 13:08:20 admin kernel: __alloc_pages: 3-order allocation failed. Aug 28 13:08:20 admin kernel: pid=1139; __alloc_pages(gfp=0x20, order=3, ...) Aug 28 13:08:20 admin kernel: Call Trace: [_alloc_pages+22/24] [__get_free_pages+10/24] [<fdcec826>] [<fdcec8f5>] [<fdceb7d7>] Aug 28 13:08:20 admin kernel: [<fdcec0f5>] [<fdcea589>] [_alloc_pages+22/24] [__get_free_pages+10/24] [<fdcec826>] [<fdcec8f5>] Aug 28 13:08:20 admin kernel: [<fdceb6bd>] [filemap_nopage+171/1008] [do_no_page+90/244] [handle_mm_fault+97/192] [<fdce94aa>] [do_page_fault+0/1164] Aug 28 13:08:20 admin kernel: [dentry_open+189/316] [filp_open+82/92] [do_fcntl+370/712] [sys_ioctl+443/532] [system_call+51/56] Aug 28 13:08:20 admin kernel: __alloc_pages: 3-order allocation failed. Aug 28 13:08:20 admin kernel: pid=1139; __alloc_pages(gfp=0x20, order=3, ...) Aug 28 13:08:20 admin kernel: Call Trace: [_alloc_pages+22/24] [__get_free_pages+10/24] [<fdcec826>] [<fdcec8f5>] [<fdceb7d7>] Aug 28 13:08:20 admin kernel: [<fdcec0f5>] [<fdcea589>] [_alloc_pages+22/24] [__get_free_pages+10/24] [<fdcec826>] [<fdcec8f5>] Aug 28 13:08:20 admin kernel: [<fdceb6bd>] [filemap_nopage+171/1008] [do_no_page+90/244] [handle_mm_fault+97/192] [<fdce94aa>] [do_page_fault+0/1164] Aug 28 13:08:20 admin kernel: [dentry_open+189/316] [filp_open+82/92] [do_fcntl+370/712] [sys_ioctl+443/532] [system_call+51/56] Aug 28 13:08:20 admin kernel: __alloc_pages: 3-order allocation failed. Aug 28 13:08:20 admin kernel: pid=1139; __alloc_pages(gfp=0x20, order=3, ...) Aug 28 13:08:20 admin kernel: Call Trace: [_alloc_pages+22/24] [__get_free_pages+10/24] [<fdcec826>] [<fdcec8f5>] [<fdceb7d7>] Aug 28 13:08:20 admin kernel: [<fdcec0f5>] [<fdcea589>] [_alloc_pages+22/24] [__get_free_pages+10/24] [<fdcec826>] [<fdcec8f5>] Aug 28 13:08:20 admin kernel: [<fdceb6bd>] [filemap_nopage+171/1008] [do_no_page+90/244] [handle_mm_fault+97/192] [<fdce94aa>] [do_page_fault+0/1164] Aug 28 13:08:20 admin kernel: [dentry_open+189/316] [filp_open+82/92] [do_fcntl+370/712] [sys_ioctl+443/532] [system_call+51/56]
meminfo shows:
total: used: free: shared: buffers: cached: Mem: 921726976 918597632 3129344 0 26300416 783720448 Swap: 271392768 0 271392768 MemTotal: 900124 kB MemFree: 3056 kB MemShared: 0 kB Buffers: 25684 kB Cached: 765352 kB SwapCached: 0 kB Active: 66372 kB Inact_dirty: 723020 kB Inact_clean: 1644 kB Inact_target: 5708 kB HighTotal: 0 kB HighFree: 0 kB LowTotal: 900124 kB LowFree: 3056 kB SwapTotal: 265032 kB SwapFree: 265032 kB
I was a bit surprised by the kernel errors. I just wanted to start the CD compare to image, but these problems showed up before that.
If I can do any additional debugging, tell me. There are some unresolved addresses above, maybe I made some mistake with System.map (name System.map-2.4.10-pre1 should be ok, isn't it?)
Regards, Stephan
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
| |