Re: v3.2-rc2: kernel BUG at mm/migrate.c:578
From: Hillf Danton
Date: Sat Dec 03 2011 - 22:37:10 EST
Hello George
On Wed, Nov 30, 2011 at 10:07 PM, George Spelvin <linux@xxxxxxxxxxx> wrote:
> I just noticed the last one, but the preceding warnings were close enough I figured they might be relevant.
>
> Kernel is stock 3.2.0-rc2.
>
> Prcessor is 1st gen Core i3, 4G RAM, 64-bit kernel. ÂUserspace is 32-bit Debian/unstable.
[...]
>
> [1118868.292556] ------------[ cut here ]------------
> [1118868.292589] kernel BUG at mm/migrate.c:578!
> [1118868.292612] invalid opcode: 0000 [#1] SMP
> [1118868.292640] CPU 2
> [1118868.292652] Modules linked in: battery nfsd exportfs nfs lockd auth_rpcgss nfs_acl sunrpc fuse loop ftdi_sio usbserial r8169
> [1118868.292746]
> [1118868.292758] Pid: 3387, comm: firefox-bin Tainted: G Â Â Â ÂW Â Â3.2.0-rc2 #40 Gigabyte Technology Co., Ltd. H55M-UD2H/H55M-UD2H
> [1118868.292818] RIP: 0010:[<ffffffff8109ece2>] Â[<ffffffff8109ece2>] move_to_new_page+0x21/0x1b8
> [1118868.292868] RSP: 0000:ffff880105c25a38 ÂEFLAGS: 00010246
> [1118868.292895] RAX: 0000000000000000 RBX: ffffea00000dd940 RCX: 0000000000000001
> [1118868.292929] RDX: 0000000000000001 RSI: ffffea00000dd940 RDI: ffffea0001c763c0
> [1118868.292964] RBP: ffffea0001c763c0 R08: ffffea00000de280 R09: 000000000000000a
> [1118868.292999] R10: ffffea0001a08100 R11: ffffea0001a08100 R12: 0000000000000000
> [1118868.293033] R13: 0000000000000001 R14: 0000000000000001 R15: ffff880105c25be8
> [1118868.293068] FS: Â0000000000000000(0000) GS:ffff880117d00000(0063) knlGS:00000000f742f6e0
> [1118868.293106] CS: Â0010 DS: 002b ES: 002b CR0: 0000000080050033
> [1118868.293134] CR2: 000000006c000000 CR3: 00000001116a4000 CR4: 00000000000006e0
> [1118868.293169] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
> [1118868.293203] DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400
> [1118868.293237] Process firefox-bin (pid: 3387, threadinfo ffff880105c24000, task ffff880110bb0ca0)
> [1118868.293278] Stack:
> [1118868.293291] Âffff880117d0d870 ffffea00000dd940 0000000000000003 0000000000000000
> [1118868.293334] Â0000000000000001 0000000000000001 ffffea00000dd940 0000000000000003
> [1118868.293377] Â0000000000000000 0000000000000001 0000000000000001 ffffffff8109f145
> [1118868.293420] Call Trace:
> [1118868.293436] Â[<ffffffff8109f145>] ? migrate_pages+0x1e3/0x335
> [1118868.293466] Â[<ffffffff81099ef3>] ? pfn_valid.part.3+0x32/0x32
> [1118868.293496] Â[<ffffffff8109a5b7>] ? compact_zone+0x3f1/0x5bf
> [1118868.293526] Â[<ffffffff8100738f>] ? read_tsc+0x5/0x14
> [1118868.293552] Â[<ffffffff8109a8a6>] ? try_to_compact_pages+0x121/0x17e
> [1118868.293585] Â[<ffffffff81077f6b>] ? __alloc_pages_direct_compact+0xaa/0x161
> [1118868.293619] Â[<ffffffff8107868f>] ? __alloc_pages_nodemask+0x66d/0x6cc
> [1118868.293653] Â[<ffffffff8108ff77>] ? anon_vma_prepare+0x1e/0x126
> [1118868.293683] Â[<ffffffff810a1b5a>] ? do_huge_pmd_anonymous_page+0xa4/0x227
> [1118868.293718] Â[<ffffffff8101b13e>] ? do_page_fault+0x288/0x2ce
> [1118868.293749] Â[<ffffffff8108cef4>] ? mmap_region+0x353/0x44a
> [1118868.293780] Â[<ffffffff81333def>] ? page_fault+0x1f/0x30
> [1118868.293806] Code: 44 24 08 48 83 c4 18 5b 5d c3 41 56 41 88 ce 41 55 41 89 d5 41 54 55 48 89 fd 53 48 89 f3 48 83 ec 30 e8 03 f8 ff ff 85 c0 75 02 <0f> 0b 48 8b 43 10 48 89 45 10 48 8b 43 08 48 89 45 08 48 8b 03
> [1118868.294066] RIP Â[<ffffffff8109ece2>] move_to_new_page+0x21/0x1b8
> [1118868.294099] ÂRSP <ffff880105c25a38>
> [1118868.358440] ---[ end trace ceac60930a99a4ce ]---
> --
The following diff is prepared for spotting/catching the triggered bug,
would you please take another try with it included?
Thanks
Hillf
---
--- a/mm/compaction.c Thu Nov 24 20:23:28 2011
+++ b/mm/compaction.c Sun Dec 4 11:31:03 2011
@@ -322,6 +322,19 @@ static isolate_migrate_t isolate_migrate
if (PageBuddy(page))
continue;
+ if (!PageLRU(page))
+ continue;
+
+ /*
+ * PageLRU is set, and lru_lock excludes isolation,
+ * splitting and collapsing (collapsing has already
+ * happened if PageLRU is set).
+ */
+ if (PageTransHuge(page)) {
+ low_pfn += (1 << compound_order(page)) - 1;
+ continue;
+ }
+
/*
* For async migration, also only scan in MOVABLE blocks. Async
* migration is optimistic to see if the minimum amount of work
@@ -333,19 +346,6 @@ static isolate_migrate_t isolate_migrate
low_pfn += pageblock_nr_pages;
low_pfn = ALIGN(low_pfn, pageblock_nr_pages) - 1;
last_pageblock_nr = pageblock_nr;
- continue;
- }
-
- if (!PageLRU(page))
- continue;
-
- /*
- * PageLRU is set, and lru_lock excludes isolation,
- * splitting and collapsing (collapsing has already
- * happened if PageLRU is set).
- */
- if (PageTransHuge(page)) {
- low_pfn += (1 << compound_order(page)) - 1;
continue;
}
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/