kernel oops with more zram devices

[GoogleCodeExporter]

What steps will reproduce the problem?
1. modprobe zram num_devices=2
2. echo $((128*1024*1024)) >/sys/block/zram0/disksize
3. mkswap /dev/zram0
4. swapon /dev/zram0
5. echo $((512*1024*1024)) >/sys/block/zram1/disksize
6. mkfs.ext4 /dev/zram1
7. kernel oops

What version of the product are you using? On what operating system?
kernel 2.6.35-gentoo-ck-r1 (BFS/compcache patches)
development compcache compiled kernel (checkout at 29.08.2010)

Please provide any additional information below.

[ 1586.483860] BUG: unable to handle kernel NULL pointer dereference at (null)
[ 1586.483868] IP: [<ffffffffa003d1de>] zram_free_page+0x18/0xf6 [zram]
[ 1586.483876] PGD 8136f067 PUD 9f799067 PMD 0 
[ 1586.483881] Oops: 0000 [#1] PREEMPT SMP 
[ 1586.483885] last sysfs file: /sys/devices/virtual/block/zram1/disksize
[ 1586.483888] CPU 1 
[ 1586.483890] Modules linked in: zram snd_pcm_oss snd_mixer_oss 
snd_hda_codec_intelhdmi lib80211_crypt_tkip wl(P) snd_hda_codec_analog 
snd_hda_intel snd_hda_codec snd_hwdep
 snd_pcm snd_timer snd_page_alloc sky2 lib80211 pcspkr sr_mod hp_wmi rfkill cdrom usb_storage evdev [last unloaded: zram]
[ 1586.483910] 
[ 1586.483915] Pid: 17194, comm: mkfs.ext4 Tainted: P            2.6.35-ck-r1 
#1 3072/HP ProBook 4510s
[ 1586.483918] RIP: 0010:[<ffffffffa003d1de>]  [<ffffffffa003d1de>] 
zram_free_page+0x18/0xf6 [zram]
[ 1586.483924] RSP: 0018:ffff880081389c28  EFLAGS: 00010246
[ 1586.483927] RAX: 0000000000000000 RBX: ffff880090a73dd8 RCX: ffff880001880000
[ 1586.483930] RDX: 0000000000000001 RSI: 0000000000000000 RDI: ffff880090a73dd8
[ 1586.483932] RBP: ffff880090a73dd8 R08: 0000000000000000 R09: ffffea00022d75e0
[ 1586.483935] R10: 0000000000001000 R11: dead000000100100 R12: 0000000000000000
[ 1586.483938] R13: 0000000020000000 R14: 0000000000100000 R15: 0000000000000000
[ 1586.483941] FS:  00007f2cf54ac760(0000) GS:ffff880001880000(0000) 
knlGS:0000000000000000
[ 1586.483944] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 1586.483947] CR2: 0000000000000000 CR3: 000000009b890000 CR4: 00000000000406e0
[ 1586.483950] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 1586.483953] DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400
[ 1586.483956] Process mkfs.ext4 (pid: 17194, threadinfo ffff880081388000, task 
ffff8800b851dc70)
[ 1586.483958] Stack:
[ 1586.483960]  ffff88009a28b6c0 ffff880090a73dd8 0000000000000008 
0000000020000000
[ 1586.483964] <0> 0000000000100000 ffffffffa003d50a ffffffff0fe00001 
0000000000000001
[ 1586.483968] <0> 00000001b9ca0d50 8000000000000000 ffff880081389ce4 
ffffffff81669580
[ 1586.483973] Call Trace:
[ 1586.483979]  [<ffffffffa003d50a>] ? zram_make_request+0x79/0x5db [zram]
[ 1586.483986]  [<ffffffff8106f59c>] ? __alloc_pages_nodemask+0xf8/0x6b3
[ 1586.483992]  [<ffffffff81151cde>] ? generic_make_request+0x17d/0x1de
[ 1586.483996]  [<ffffffff810be57b>] ? blkdev_open+0xa6/0xde
[ 1586.484000]  [<ffffffff81151de5>] ? submit_bio+0xa6/0xad
[ 1586.484005]  [<ffffffff81156f8d>] ? blkdev_issue_discard+0x16f/0x1f6
[ 1586.484010]  [<ffffffff810a4b6c>] ? do_filp_open+0x4ef/0x56e
[ 1586.484014]  [<ffffffff81157592>] ? blkdev_ioctl+0x2b2/0x634
[ 1586.484018]  [<ffffffff810bd219>] ? block_ioctl+0x32/0x36
[ 1586.484021]  [<ffffffff810a656c>] ? vfs_ioctl+0x23/0x93
[ 1586.484025]  [<ffffffff810a6a2e>] ? do_vfs_ioctl+0x3e0/0x417
[ 1586.484030]  [<ffffffff81381f08>] ? _raw_spin_unlock+0x1e/0x2b
[ 1586.484034]  [<ffffffff81093b19>] ? virt_to_head_page+0x9/0x2b
[ 1586.484037]  [<ffffffff810a6aa1>] ? sys_ioctl+0x3c/0x5c
[ 1586.484042]  [<ffffffff81001eeb>] ? system_call_fastpath+0x16/0x1b
[ 1586.484044] Code: e8 94 ff ff ff 85 c0 74 07 41 59 e9 94 3a 34 e1 41 58 c3 
41 56 89 f0 41 55 48 c1 e0 03 41 54 49 89 f4 55 53 48 89 fb 48 03 47 08 <4c> 8b 
30 41 f6 c6 01 74 1d 49 83 e6 fe 48 83 ca ff 4c 89 30 be 
[ 1586.484078] RIP  [<ffffffffa003d1de>] zram_free_page+0x18/0xf6 [zram]
[ 1586.484083]  RSP <ffff880081389c28>
[ 1586.484084] CR2: 0000000000000000
[ 1586.484087] ---[ end trace f65cb8d27f6dd15b ]---

Original issue reported on code.google.com by Slavomir...@gmail.com on 29 Aug 2010 at 2:10

[GoogleCodeExporter]

Sorry, tested once again with only 1 zram device, mkfs.ext4 crashes too

Original comment by Slavomir...@gmail.com on 29 Aug 2010 at 2:14

[GoogleCodeExporter]

Please pull the code now and try again.

Original comment by nitingupta910@gmail.com on 30 Aug 2010 at 11:49

[GoogleCodeExporter]

Thank You, seems to be resolved. Works fine.

Original comment by Slavomir...@gmail.com on 31 Aug 2010 at 4:06

[GoogleCodeExporter]

Thanks. Closing the Issue.

Original comment by nitingupta910@gmail.com on 31 Aug 2010 at 8:35

• Changed state: Fixed