| lists.openwall.net | lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC | |
|
Open Source and information security mailing list archives
| ||
|
Message-ID: <9ea470500906180906o2d558a90hbcba11874ff8917b@mail.gmail.com> Date: Thu, 18 Jun 2009 18:06:34 +0200 From: Borislav Petkov <petkovbb@...glemail.com> To: Rainer Weikusat <rweikusat@...gmbh.com> Cc: linux-kernel@...r.kernel.org, Linux IDE mailing list <linux-ide@...r.kernel.org>, Bartlomiej Zolnierkiewicz <bzolnier@...il.com>, bruinjm@...all.nl Subject: Re: [PATCH] ide-cd: prevent null pointer deref via cdrom_newpc_intr Hi, On Thu, Jun 18, 2009 at 5:04 PM, Rainer Weikusat<rweikusat@...gmbh.com> wrote: > From: Rainer Weikusat <rweikusat@...gmbh.com> > > With 2.6.30, the error handling code in cdrom_newpc_intr was changed > to deal with partial request failures by normally completing the 'good' > parts of a request and only 'error' the last (and presumably, > incompletely transferred) bio associated with a particular > request. In order to do this, ide_complete_rq is called over > ide_cd_error_cmd() to partially complete the rq. The block layer > does partial completion only for requests with bio's and if the > rq doesn't have one (eg 'GPCMD_READ_DISC_INFO') the request is > completed as a whole and the drive->hwif->rq pointer set to NULL > afterwards. When calling ide_complete_rq again to report > the error, this null pointer is derefenced, resulting in a kernel > crash. Sorry, but still not good enough. Instead, I rediffed the change against current linux-ide branch and rewrote the commit message properly keeping your S-O-B. @Bart: please apply. -- From: Rainer Weikusat <rweikusat@...gmbh.com> Date: Thu, 18 Jun 2009 17:48:24 +0200 Subject: [PATCH] ide-cd: don't do partial completions on bio-less rqs The block layer completes bio-less requests totally instead of partially and this breaks cdrom drives which fragment packet command data in several DRQ turns (eg GPCMD_READ_DISC_INFO). More specifically, ide_complete_rq() is called over ide_cd_error_cmd() to partially complete the rq on error. This bio-less request is completed as a whole and when calling ide_complete_rq again to complete the request wholly, the rq is already vanished resulting in the following OOPS: BUG: unable to handle kernel NULL pointer dereference at 0000000000000048 IP: [<ffffffff804deb50>] ide_complete_rq+0x19/0x4d PGD 0 Thread overran stack, or stack corrupted Oops: 0000 [#1] SMP last sysfs file: CPU 0 Modules linked in: Pid: 0, comm: swapper Not tainted 2.6.30-rc8 #22 Precision WorkStation 380 RIP: 0010:[<ffffffff804deb50>] [<ffffffff804deb50>] ide_complete_rq+0x19/0x4d RSP: 0018:ffff880028022e18 EFLAGS: 00010096 RAX: 0000000000000002 RBX: ffff88011a84e000 RCX: 0000000000000200 RDX: 0000000000000200 RSI: 0000000000000000 RDI: ffff88011afc9000 RBP: ffff880028022e28 R08: 00000000fffffffb R09: 0000000000000000 R10: ffff8800280302e8 R11: ffff880028030310 R12: 0000000000000000 R13: 0000000000000000 R14: ffff88011afc9000 R15: ffff88011aff7140 FS: 0000000000000000(0000) GS:ffff88002801f000(0000) knlGS:0000000000000000 CS: 0010 DS: 0018 ES: 0018 CR0: 000000008005003b CR2: 0000000000000048 CR3: 0000000000201000 CR4: 00000000000026e0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400 Process swapper (pid: 0, threadinfo ffffffff809c0000, task ffffffff808f1360) Stack: ffff88011aecf840 ffff88011aff7140 ffff880028022eb8 ffffffff804ed0ef ffffffff8025a811 00ff880028022e70 ffffffff00000050 ffff88011a84e000 ffff88011a84e108 000000008025ef72 0000000000000000 ffff88011a84e000 Call Trace: <IRQ> <0> [<ffffffff804ed0ef>] cdrom_newpc_intr+0x20e/0xac5 [<ffffffff8025a811>] ? irq_exit+0x47/0x7d [<ffffffff804ecee1>] ? cdrom_newpc_intr+0x0/0xac5 [<ffffffff804de920>] ide_intr+0x1d2/0x220 [<ffffffff80284ee1>] handle_IRQ_event+0x3a/0xba [<ffffffff80286c38>] handle_edge_irq+0xce/0x133 [<ffffffff8022c0cd>] handle_irq+0x1d/0x29 [<ffffffff8022b8c7>] do_IRQ+0x5a/0xd5 [<ffffffff8022a013>] ret_from_intr+0x0/0xa <EOI> <0> [<ffffffff80230eee>] ? mwait_idle+0x60/0x6e [<ffffffff802282c2>] ? enter_idle+0x20/0x22 [<ffffffff802286ef>] ? cpu_idle+0x4a/0x8d [<ffffffff806ebb05>] ? rest_init+0x65/0x70 [<ffffffff809cdcef>] ? start_kernel+0x2da/0x3bb [<ffffffff809cd271>] ? x86_64_start_reservations+0x81/0xbc [<ffffffff809cd37b>] ? x86_64_start_kernel+0xcf/0xf1 Code: c3 48 25 ff ff ff fe 48 89 47 50 e8 a7 86 00 00 eb d7 55 48 89 e5 53 48 83 ec 08 41 89 f0 89 d1 48 8b 5f 40 48 8b b3 28 03 00 00 <f6> 46 48 0e 74 05 45 85 c0 7e 1e 44 89 c2 e8 85 ff ff ff 85 c0 RIP [<ffffffff804deb50>] ide_complete_rq+0x19/0x4d RSP <ffff880028022e18> CR2: 0000000000000048 ---[ end trace 6662ae44d700bf58 ]--- This fixes http://bugzilla.kernel.org/show_bug.cgi?id=13399. Tested-by: Hans de Bruin <bruinjm@...all.nl> Signed-Off-By: Rainer Weikusat <rweikusat@...gmbh.com> Signed-Off-By: Borislav Petkov <petkovbb@...il.com> --- drivers/ide/ide-cd.c | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/drivers/ide/ide-cd.c b/drivers/ide/ide-cd.c index 0b7645b..4a19686 100644 --- a/drivers/ide/ide-cd.c +++ b/drivers/ide/ide-cd.c @@ -667,7 +667,7 @@ out_end: rq->errors = -EIO; } - if (uptodate == 0) + if (uptodate == 0 && rq->bio) ide_cd_error_cmd(drive, cmd); /* make sure it's fully ended */ -- 1.6.3.1 -- Regards/Gruss, Boris -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@...r.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
Powered by blists - more mailing lists