linux

iv/linux

History

Chaitanya Kulkarni 59c157433f nvme-core: check bdev value for NULL

The nvme-core sets the bdev to NULL when admin comamnd is issued from
IOCTL in the following path e.g. nvme list :-

block_ioctl()
 blkdev_ioctl()
  nvme_ioctl()
   nvme_user_cmd()
    nvme_submit_user_cmd()

The commit 309dca309fc3 ("block: store a block_device pointer in struct bio")
now uses bdev unconditionally in the macro bio_set_dev() and assumes
that bdev value is not NULL which results in the following crash in
since thats where bdev is actually accessed :-

void bio_associate_blkg_from_css(struct bio *bio,
				 struct cgroup_subsys_state *css)
{
	if (bio->bi_blkg)
		blkg_put(bio->bi_blkg);

	if (css && css->parent) {
		bio->bi_blkg = blkg_tryget_closest(bio, css);
	} else {
-------------->	blkg_get(bio->bi_bdev->bd_disk->queue->root_blkg);
		bio->bi_blkg = bio->bi_bdev->bd_disk->queue->root_blkg;
	}
}
EXPORT_SYMBOL_GPL(bio_associate_blkg_from_css);

[  345.385947] BUG: kernel NULL pointer dereference, address: 0000000000000690
[  345.387103] #PF: supervisor read access in kernel mode
[  345.387894] #PF: error_code(0x0000) - not-present page
[  345.388756] PGD 162a2b067 P4D 162a2b067 PUD 1633eb067 PMD 0
[  345.389625] Oops: 0000 [#1] SMP NOPTI
[  345.390206] CPU: 15 PID: 4100 Comm: nvme Tainted: G           OE     5.11.0-rc5blk+ #141
[  345.391377] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.12.0-59-gc9ba52764
[  345.393074] RIP: 0010:bio_associate_blkg_from_css.cold.47+0x58/0x21f

[  345.396362] RSP: 0018:ffffc90000dbbce8 EFLAGS: 00010246
[  345.397078] RAX: 0000000000000000 RBX: 0000000000000000 RCX: 0000000000000027
[  345.398114] RDX: 0000000000000000 RSI: ffff888813be91f0 RDI: ffff888813be91f8
[  345.399039] RBP: ffffc90000dbbd30 R08: 0000000000000001 R09: 0000000000000001
[  345.399950] R10: 0000000064c66670 R11: 00000000ef955201 R12: ffff888812d32800
[  345.401031] R13: 0000000000000000 R14: ffff888113e51540 R15: ffff888113e51540
[  345.401976] FS:  00007f3747f1d780(0000) GS:ffff888813a00000(0000) knlGS:0000000000000000
[  345.402997] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[  345.403737] CR2: 0000000000000690 CR3: 000000081a4bc000 CR4: 00000000003506e0
[  345.404685] Call Trace:
[  345.405031]  bio_associate_blkg+0x71/0x1c0
[  345.405649]  nvme_submit_user_cmd+0x1aa/0x38e [nvme_core]
[  345.406348]  nvme_user_cmd.isra.73.cold.98+0x54/0x92 [nvme_core]
[  345.407117]  nvme_ioctl+0x226/0x260 [nvme_core]
[  345.407707]  blkdev_ioctl+0x1c8/0x2b0
[  345.408183]  block_ioctl+0x3f/0x50
[  345.408627]  __x64_sys_ioctl+0x84/0xc0
[  345.409117]  do_syscall_64+0x33/0x40
[  345.409592]  entry_SYSCALL_64_after_hwframe+0x44/0xa9
[  345.410233] RIP: 0033:0x7f3747632107

[  345.413125] RSP: 002b:00007ffe461b6648 EFLAGS: 00000206 ORIG_RAX: 0000000000000010
[  345.414086] RAX: ffffffffffffffda RBX: 00000000007b7fd0 RCX: 00007f3747632107
[  345.414998] RDX: 00007ffe461b6650 RSI: 00000000c0484e41 RDI: 0000000000000004
[  345.415966] RBP: 0000000000000004 R08: 00000000007b7fe8 R09: 00000000007b9080
[  345.416883] R10: 00007ffe461b62c0 R11: 0000000000000206 R12: 00000000007b7fd0
[  345.417808] R13: 0000000000000000 R14: 0000000000000003 R15: 0000000000000000

Add a NULL check before we set the bdev for bio.

This issue is found on block/for-next tree.

Fixes: 309dca309fc3 ("block: store a block_device pointer in struct bio")
Signed-off-by: Chaitanya Kulkarni <chaitanya.kulkarni@wdc.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>

2021-01-27 10:10:15 -07:00

core.c

nvme-core: check bdev value for NULL

2021-01-27 10:10:15 -07:00

fabrics.c

nvme-fabrics: reject I/O to offline device

2020-12-01 20:36:37 +01:00

fabrics.h

nvme-fabrics: reject I/O to offline device

2020-12-01 20:36:37 +01:00

fault_inject.c

nvme: enable to inject errors into admin commands