nvme: check for valid nvme_identify_ns() before using it
authorEwan D. Milne <emilne@redhat.com>
Mon, 27 Nov 2023 20:56:57 +0000 (15:56 -0500)
committerGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Fri, 8 Dec 2023 07:51:15 +0000 (08:51 +0100)
commit d8b90d600aff181936457f032d116dbd8534db06 upstream.

When scanning namespaces, it is possible to get valid data from the first
call to nvme_identify_ns() in nvme_alloc_ns(), but not from the second
call in nvme_update_ns_info_block().  In particular, if the NSID becomes
inactive between the two commands, a storage device may return a buffer
filled with zero as per 4.1.5.1.  In this case, we can get a kernel crash
due to a divide-by-zero in blk_stack_limits() because ns->lba_shift will
be set to zero.

PID: 326      TASK: ffff95fec3cd8000  CPU: 29   COMMAND: "kworker/u98:10"
 #0 [ffffad8f8702f9e0] machine_kexec at ffffffff91c76ec7
 #1 [ffffad8f8702fa38] __crash_kexec at ffffffff91dea4fa
 #2 [ffffad8f8702faf8] crash_kexec at ffffffff91deb788
 #3 [ffffad8f8702fb00] oops_end at ffffffff91c2e4bb
 #4 [ffffad8f8702fb20] do_trap at ffffffff91c2a4ce
 #5 [ffffad8f8702fb70] do_error_trap at ffffffff91c2a595
 #6 [ffffad8f8702fbb0] exc_divide_error at ffffffff928506e6
 #7 [ffffad8f8702fbd0] asm_exc_divide_error at ffffffff92a00926
    [exception RIP: blk_stack_limits+434]
    RIP: ffffffff92191872  RSP: ffffad8f8702fc80  RFLAGS: 00010246
    RAX: 0000000000000000  RBX: ffff95efa0c91800  RCX: 0000000000000001
    RDX: 0000000000000000  RSI: 0000000000000001  RDI: 0000000000000001
    RBP: 00000000ffffffff   R8: ffff95fec7df35a8   R9: 0000000000000000
    R10: 0000000000000000  R11: 0000000000000001  R12: 0000000000000000
    R13: 0000000000000000  R14: 0000000000000000  R15: ffff95fed33c09a8
    ORIG_RAX: ffffffffffffffff  CS: 0010  SS: 0018
 #8 [ffffad8f8702fce0] nvme_update_ns_info_block at ffffffffc06d3533 [nvme_core]
 #9 [ffffad8f8702fd18] nvme_scan_ns at ffffffffc06d6fa7 [nvme_core]

This happened when the check for valid data was moved out of nvme_identify_ns()
into one of the callers.  Fix this by checking in both callers.

Link: https://bugzilla.kernel.org/show_bug.cgi?id=218186
Fixes: 0dd6fff2aad4 ("nvme: bring back auto-removal of deleted namespaces during sequential scan")
Cc: stable@vger.kernel.org
Signed-off-by: Ewan D. Milne <emilne@redhat.com>
Signed-off-by: Keith Busch <kbusch@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
drivers/nvme/host/core.c

index 25ddfab..5b156c5 100644 (file)
@@ -2058,6 +2058,13 @@ static int nvme_update_ns_info_block(struct nvme_ns *ns,
        if (ret)
                return ret;
 
+       if (id->ncap == 0) {
+               /* namespace not allocated or attached */
+               info->is_removed = true;
+               ret = -ENODEV;
+               goto error;
+       }
+
        blk_mq_freeze_queue(ns->disk->queue);
        lbaf = nvme_lbaf_index(id->flbas);
        ns->lba_shift = id->lbaf[lbaf].ds;
@@ -2107,6 +2114,8 @@ out:
                set_bit(NVME_NS_READY, &ns->flags);
                ret = 0;
        }
+
+error:
        kfree(id);
        return ret;
 }