Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]
Groups > linux.kernel > #1704458
| From | Greg Kroah-Hartman <gregkh@linuxfoundation.org> |
|---|---|
| Newsgroups | linux.kernel |
| Subject | [PATCH 4.4 22/91] xfs: dont BUG() on mixed direct and mapped I/O |
| Date | 2017-08-05 02:00 +0200 |
| Message-ID | <uaUFC-5lm-69@gated-at.bofh.it> (permalink) |
| References | <uaUcx-5a1-3@gated-at.bofh.it> |
| Organization | linux.* mail to news gateway |
4.4-stable review patch. If anyone has any objections, please let me know.
------------------
From: Brian Foster <bfoster@redhat.com>
commit 04197b341f23b908193308b8d63d17ff23232598 upstream.
We've had reports of generic/095 causing XFS to BUG() in
__xfs_get_blocks() due to the existence of delalloc blocks on a
direct I/O read. generic/095 issues a mix of various types of I/O,
including direct and memory mapped I/O to a single file. This is
clearly not supported behavior and is known to lead to such
problems. E.g., the lack of exclusion between the direct I/O and
write fault paths means that a write fault can allocate delalloc
blocks in a region of a file that was previously a hole after the
direct read has attempted to flush/inval the file range, but before
it actually reads the block mapping. In turn, the direct read
discovers a delalloc extent and cannot proceed.
While the appropriate solution here is to not mix direct and memory
mapped I/O to the same regions of the same file, the current
BUG_ON() behavior is probably overkill as it can crash the entire
system. Instead, localize the failure to the I/O in question by
returning an error for a direct I/O that cannot be handled safely
due to delalloc blocks. Be careful to allow the case of a direct
write to post-eof delalloc blocks. This can occur due to speculative
preallocation and is safe as post-eof blocks are not accompanied by
dirty pages in pagecache (conversely, preallocation within eof must
have been zeroed, and thus dirtied, before the inode size could have
been increased beyond said blocks).
Finally, provide an additional warning if a direct I/O write occurs
while the file is memory mapped. This may not catch all problematic
scenarios, but provides a hint that some known-to-be-problematic I/O
methods are in use.
Signed-off-by: Brian Foster <bfoster@redhat.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
Signed-off-by: Dave Chinner <david@fromorbit.com>
Signed-off-by: Nikolay Borisov <nborisov@suse.com>
Acked-by: Darrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
---
fs/xfs/xfs_aops.c | 21 ++++++++++++++++++++-
1 file changed, 20 insertions(+), 1 deletion(-)
--- a/fs/xfs/xfs_aops.c
+++ b/fs/xfs/xfs_aops.c
@@ -1426,6 +1426,26 @@ __xfs_get_blocks(
if (error)
goto out_unlock;
+ /*
+ * The only time we can ever safely find delalloc blocks on direct I/O
+ * is a dio write to post-eof speculative preallocation. All other
+ * scenarios are indicative of a problem or misuse (such as mixing
+ * direct and mapped I/O).
+ *
+ * The file may be unmapped by the time we get here so we cannot
+ * reliably fail the I/O based on mapping. Instead, fail the I/O if this
+ * is a read or a write within eof. Otherwise, carry on but warn as a
+ * precuation if the file happens to be mapped.
+ */
+ if (direct && imap.br_startblock == DELAYSTARTBLOCK) {
+ if (!create || offset < i_size_read(VFS_I(ip))) {
+ WARN_ON_ONCE(1);
+ error = -EIO;
+ goto out_unlock;
+ }
+ WARN_ON_ONCE(mapping_mapped(VFS_I(ip)->i_mapping));
+ }
+
/* for DAX, we convert unwritten extents directly */
if (create &&
(!nimaps ||
@@ -1525,7 +1545,6 @@ __xfs_get_blocks(
set_buffer_new(bh_result);
if (imap.br_startblock == DELAYSTARTBLOCK) {
- BUG_ON(direct);
if (create) {
set_buffer_uptodate(bh_result);
set_buffer_mapped(bh_result);
Back to linux.kernel | Previous | Next — Previous in thread | Next in thread | Find similar | Unroll thread
[PATCH 4.4 00/91] 4.4.80-stable review Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2017-08-05 02:00 +0200
[PATCH 4.4 23/91] nfc: fdp: fix NULL pointer dereference Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2017-08-05 02:00 +0200
[PATCH 4.4 38/91] libnvdimm, btt: fix btt_rw_page not returning errors Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2017-08-05 02:00 +0200
[PATCH 4.4 06/91] md/raid5: add thread_group worker async_tx_issue_pending_all Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2017-08-05 02:00 +0200
[PATCH 4.4 04/91] powerpc/pseries: Fix of_node_put() underflow during reconfig remove Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2017-08-05 02:00 +0200
[PATCH 4.4 26/91] isdn/i4l: fix buffer overflow Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2017-08-05 02:00 +0200
[PATCH 4.4 29/91] mailbox: always wait in mbox_send_message for blocking Tx mode Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2017-08-05 02:00 +0200
[PATCH 4.4 05/91] crypto: authencesn - Fix digest_null crash Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2017-08-05 02:00 +0200
[PATCH 4.4 28/91] wil6210: fix deadlock when using fw_no_recovery option Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2017-08-05 02:00 +0200
[PATCH 4.4 54/91] x86/mce/AMD: Make the init code more robust Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2017-08-05 02:00 +0200
[PATCH 4.4 62/91] irqchip/keystone: Fix "scheduling while atomic" on rt Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2017-08-05 02:00 +0200
[PATCH 4.4 53/91] tpm: Replace device number bitmap with IDR Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2017-08-05 02:00 +0200
[PATCH 4.4 47/91] pstore: Use dynamic spinlock initializer Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2017-08-05 02:00 +0200
[PATCH 4.4 17/91] drm: rcar-du: Simplify and fix probe error handling Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2017-08-05 02:00 +0200
Re: [PATCH 4.4 17/91] drm: rcar-du: Simplify and fix probe error handling Ben Hutchings <ben.hutchings@codethink.co.uk> - 2017-08-07 17:20 +0200
Re: [PATCH 4.4 17/91] drm: rcar-du: Simplify and fix probe error handling Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2017-08-08 18:30 +0200
[PATCH 4.4 34/91] kaweth: fix oops upon failed memory allocation Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2017-08-05 02:00 +0200
[PATCH 4.4 07/91] drm/vmwgfx: Fix gcc-7.1.1 warning Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2017-08-05 02:00 +0200
[PATCH 4.4 25/91] isdn: Fix a sleep-in-atomic bug Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2017-08-05 02:00 +0200
[PATCH 4.4 39/91] ipmi/watchdog: fix watchdog timeout set on reboot Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2017-08-05 02:00 +0200
[PATCH 4.4 36/91] PM / Domains: defer dev_pm_domain_set() until genpd->attach_dev succeeds if present Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2017-08-05 02:00 +0200
[PATCH 4.4 03/91] net: reduce skb_warn_bad_offload() noise Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2017-08-05 02:00 +0200
[PATCH 4.4 32/91] mpt3sas: Dont overreach ioc->reply_post[] during initialization Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2017-08-05 02:00 +0200
[PATCH 4.4 08/91] drm/nouveau/bar/gf100: fix access to upper half of BAR2 Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2017-08-05 02:00 +0200
[PATCH 4.4 27/91] ath10k: fix null deref on wmi-tlv when trying spectral scan Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2017-08-05 02:00 +0200
[PATCH 4.4 09/91] KVM: PPC: Book3S HV: Context-switch EBB registers properly Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2017-08-05 02:00 +0200
[PATCH 4.4 60/91] drm/msm: Verify that MSM_SUBMIT_BO_FLAGS are set Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2017-08-05 02:00 +0200
[PATCH 4.4 22/91] xfs: dont BUG() on mixed direct and mapped I/O Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2017-08-05 02:00 +0200
[PATCH 4.4 15/91] drm/rcar: Nuke preclose hook Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2017-08-05 02:10 +0200
[PATCH 4.4 18/91] perf intel-pt: Fix ip compression Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2017-08-05 02:10 +0200
[PATCH 4.4 13/91] Revert "powerpc/numa: Fix percpu allocations to be NUMA aware" Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2017-08-05 02:10 +0200
[PATCH 4.4 11/91] KVM: PPC: Book3S HV: Reload HTM registers explicitly Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2017-08-05 02:10 +0200
[PATCH 4.4 14/91] Staging: comedi: comedi_fops: Avoid orphaned proc entry Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2017-08-05 02:10 +0200
[PATCH 4.4 16/91] drm: rcar-du: Perform initialization/cleanup at probe/remove time Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2017-08-05 02:10 +0200
[PATCH 4.4 19/91] perf intel-pt: Fix last_ip usage Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2017-08-05 02:10 +0200
[PATCH 4.4 10/91] KVM: PPC: Book3S HV: Restore critical SPRs to host values on guest exit Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2017-08-05 02:10 +0200
Re: [PATCH 4.4 00/91] 4.4.80-stable review Shuah Khan <shuahkh@osg.samsung.com> - 2017-08-05 04:00 +0200
Re: [PATCH 4.4 00/91] 4.4.80-stable review Guenter Roeck <linux@roeck-us.net> - 2017-08-05 08:20 +0200
csiph-web