The patch titled
Subject: mm-gup_benchmark-handle-gup-failures-fix
has been removed from the -mm tree. Its filename was
mm-gup_benchmark-handle-gup-failures-fix.patch
This patch was dropped because it was folded into mm-gup_benchmark-handle-gup-failures.patch
------------------------------------------------------
From: Andrew Morton <akpm(a)linux-foundation.org>
Subject: mm-gup_benchmark-handle-gup-failures-fix
Cc: Huang Ying <ying.huang(a)intel.com>
Cc: Jonathan Corbet <corbet(a)lwn.net>
Cc: Kirill A. Shutemov <kirill.shutemov(a)linux.intel.com>
Cc: Linus Torvalds <torvalds(a)linux-foundation.org>
Cc: "Michael S. Tsirkin" <mst(a)redhat.com>
Cc: Peter Zijlstra <peterz(a)infradead.org>
Cc: <stable(a)vger.kernel.org>
Cc: Thomas Gleixner <tglx(a)linutronix.de>
Cc: Thorsten Leemhuis <regressions(a)leemhuis.info>
Signed-off-by: Andrew Morton <akpm(a)linux-foundation.org>
---
mm/gup_benchmark.c | 5 +++--
1 file changed, 3 insertions(+), 2 deletions(-)
diff -puN mm/gup_benchmark.c~mm-gup_benchmark-handle-gup-failures-fix mm/gup_benchmark.c
--- a/mm/gup_benchmark.c~mm-gup_benchmark-handle-gup-failures-fix
+++ a/mm/gup_benchmark.c
@@ -41,8 +41,9 @@ static int __gup_benchmark_ioctl(unsigne
}
nr = get_user_pages_fast(addr, nr, gup->flags & 1, pages + i);
- if (nr > 0)
- i += nr;
+ if (nr <= 0)
+ break;
+ i += nr;
}
end_time = ktime_get();
_
Patches currently in -mm which might be from akpm(a)linux-foundation.org are
i-need-old-gcc.patch
mm-gup_benchmark-handle-gup-failures.patch
mm-pagemap-fix-swap-offset-value-for-pmd-migration-entry-fix.patch
writeback-safer-lock-nesting-fix.patch
arm-arch-arm-include-asm-pageh-needs-personalityh.patch
ocfs2-without-quota-support-try-to-avoid-calling-quota-recovery-checkpatch-fixes.patch
mm.patch
list_lru-prefetch-neighboring-list-entries-before-acquiring-lock-fix.patch
mm-oom-cgroup-aware-oom-killer-fix.patch
mm-oom-docs-describe-the-cgroup-aware-oom-killer-fix-2-fix.patch
linux-next-rejects.patch
fs-fsnotify-account-fsnotify-metadata-to-kmemcg-fix.patch
kernel-forkc-export-kernel_thread-to-modules.patch
slab-leaks3-default-y.patch
The patch titled
Subject: autofs: mount point create should honour passed in mode
has been added to the -mm tree. Its filename is
autofs-mount-point-create-should-honour-passed-in-mode.patch
This patch should soon appear at
http://ozlabs.org/~akpm/mmots/broken-out/autofs-mount-point-create-should-h…
and later at
http://ozlabs.org/~akpm/mmotm/broken-out/autofs-mount-point-create-should-h…
Before you just go and hit "reply", please:
a) Consider who else should be cc'ed
b) Prefer to cc a suitable mailing list as well
c) Ideally: find the original patch on the mailing list and do a
reply-to-all to that, adding suitable additional cc's
*** Remember to use Documentation/process/submit-checklist.rst when testing your code ***
The -mm tree is included into linux-next and is updated
there every 3-4 working days
------------------------------------------------------
From: Ian Kent <raven(a)themaw.net>
Subject: autofs: mount point create should honour passed in mode
The autofs file system mkdir inode operation blindly sets the created
directory mode to S_IFDIR | 0555, ingoring the passed in mode, which can
cause selinux dac_override denials.
But the function also checks if the caller is the daemon (as no-one else
should be able to do anything here) so there's no point in not honouring
the passed in mode, allowing the daemon to set appropriate mode when
required.
Link: http://lkml.kernel.org/r/152361593601.8051.14014139124905996173.stgit@pluto…
Signed-off-by: Ian Kent <raven(a)themaw.net>
Cc: <stable(a)vger.kernel.org>
Signed-off-by: Andrew Morton <akpm(a)linux-foundation.org>
---
fs/autofs4/root.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff -puN fs/autofs4/root.c~autofs-mount-point-create-should-honour-passed-in-mode fs/autofs4/root.c
--- a/fs/autofs4/root.c~autofs-mount-point-create-should-honour-passed-in-mode
+++ a/fs/autofs4/root.c
@@ -749,7 +749,7 @@ static int autofs4_dir_mkdir(struct inod
autofs4_del_active(dentry);
- inode = autofs4_get_inode(dir->i_sb, S_IFDIR | 0555);
+ inode = autofs4_get_inode(dir->i_sb, S_IFDIR | mode);
if (!inode)
return -ENOMEM;
d_add(dentry, inode);
_
Patches currently in -mm which might be from raven(a)themaw.net are
autofs-mount-point-create-should-honour-passed-in-mode.patch
From: Josef Bacik <jbacik(a)fb.com>
This fixes a use after free bug, we need to do the del_gendisk after we
cleanup the queue on the device.
Fixes: c6a4759ea0c9 ("nbd: add device refcounting")
cc: stable(a)vger.kernel.org
Signed-off-by: Josef Bacik <jbacik(a)fb.com>
---
drivers/block/nbd.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/block/nbd.c b/drivers/block/nbd.c
index 86258b00a1d4..e33da3e6aa20 100644
--- a/drivers/block/nbd.c
+++ b/drivers/block/nbd.c
@@ -174,8 +174,8 @@ static void nbd_dev_remove(struct nbd_device *nbd)
{
struct gendisk *disk = nbd->disk;
if (disk) {
- del_gendisk(disk);
blk_cleanup_queue(disk->queue);
+ del_gendisk(disk);
blk_mq_free_tag_set(&nbd->tag_set);
disk->private_data = NULL;
put_disk(disk);
--
2.14.3
The normal request completion can be done before or during handling
BLK_EH_RESET_TIMER, and this race may cause the request to never
be completed since driver's .timeout() may always return
BLK_EH_RESET_TIMER.
This issue can't be fixed completely by driver, since the normal
completion can be done between returning .timeout() and handing
BLK_EH_RESET_TIMER.
This patch fixes this race by introducing rq state of MQ_RQ_COMPLETE_IN_RESET,
and reading/writing rq's state by holding queue lock, which can be
per-request actually, but just not necessary to introduce one lock for
so unusual event.
Cc: Bart Van Assche <bart.vanassche(a)wdc.com>
Cc: Tejun Heo <tj(a)kernel.org>
Cc: Christoph Hellwig <hch(a)lst.de>
Cc: Ming Lei <ming.lei(a)redhat.com>
Cc: Sagi Grimberg <sagi(a)grimberg.me>
Cc: Israel Rukshin <israelr(a)mellanox.com>,
Cc: Max Gurtovoy <maxg(a)mellanox.com>
Cc: stable(a)vger.kernel.org
Signed-off-by: Ming Lei <ming.lei(a)redhat.com>
---
This is another way to fix this long-time issue, and turns out this
solution is much simpler.
block/blk-mq.c | 44 +++++++++++++++++++++++++++++++++++++++-----
block/blk-mq.h | 1 +
2 files changed, 40 insertions(+), 5 deletions(-)
diff --git a/block/blk-mq.c b/block/blk-mq.c
index 0dc9e341c2a7..12e8850e3905 100644
--- a/block/blk-mq.c
+++ b/block/blk-mq.c
@@ -630,10 +630,27 @@ void blk_mq_complete_request(struct request *rq)
* However, that would complicate paths which want to synchronize
* against us. Let stay in sync with the issue path so that
* hctx_lock() covers both issue and completion paths.
+ *
+ * Cover complete vs BLK_EH_RESET_TIMER race in slow path with
+ * helding queue lock.
*/
hctx_lock(hctx, &srcu_idx);
if (blk_mq_rq_aborted_gstate(rq) != rq->gstate)
__blk_mq_complete_request(rq);
+ else {
+ unsigned long flags;
+ bool need_complete = false;
+
+ spin_lock_irqsave(q->queue_lock, flags);
+ if (!blk_mq_rq_aborted_gstate(rq))
+ need_complete = true;
+ else
+ blk_mq_rq_update_state(rq, MQ_RQ_COMPLETE_IN_RESET);
+ spin_unlock_irqrestore(q->queue_lock, flags);
+
+ if (need_complete)
+ __blk_mq_complete_request(rq);
+ }
hctx_unlock(hctx, srcu_idx);
}
EXPORT_SYMBOL(blk_mq_complete_request);
@@ -814,24 +831,41 @@ static void blk_mq_rq_timed_out(struct request *req, bool reserved)
{
const struct blk_mq_ops *ops = req->q->mq_ops;
enum blk_eh_timer_return ret = BLK_EH_RESET_TIMER;
+ unsigned long flags;
req->rq_flags |= RQF_MQ_TIMEOUT_EXPIRED;
if (ops->timeout)
ret = ops->timeout(req, reserved);
+again:
switch (ret) {
case BLK_EH_HANDLED:
__blk_mq_complete_request(req);
break;
case BLK_EH_RESET_TIMER:
/*
- * As nothing prevents from completion happening while
- * ->aborted_gstate is set, this may lead to ignored
- * completions and further spurious timeouts.
+ * The normal completion may happen during handling the
+ * timeout, or even after returning from .timeout(), so
+ * once the request has been completed, we can't reset
+ * timer any more since this request may be handled as
+ * BLK_EH_RESET_TIMER in next timeout handling too, and
+ * it has to be completed in this situation.
+ *
+ * Holding the queue lock to cover read/write rq's
+ * aborted_gstate and normal state, so the race can be
+ * avoided completely.
*/
- blk_mq_rq_update_aborted_gstate(req, 0);
- blk_add_timer(req);
+ spin_lock_irqsave(req->q->queue_lock, flags);
+ if (blk_mq_rq_state(req) != MQ_RQ_COMPLETE_IN_RESET) {
+ blk_mq_rq_update_aborted_gstate(req, 0);
+ blk_add_timer(req);
+ } else {
+ blk_mq_rq_update_state(req, MQ_RQ_IN_FLIGHT);
+ ret = BLK_EH_HANDLED;
+ goto again;
+ }
+ spin_unlock_irqrestore(req->q->queue_lock, flags);
break;
case BLK_EH_NOT_HANDLED:
break;
diff --git a/block/blk-mq.h b/block/blk-mq.h
index 88c558f71819..6dc242fc785a 100644
--- a/block/blk-mq.h
+++ b/block/blk-mq.h
@@ -35,6 +35,7 @@ enum mq_rq_state {
MQ_RQ_IDLE = 0,
MQ_RQ_IN_FLIGHT = 1,
MQ_RQ_COMPLETE = 2,
+ MQ_RQ_COMPLETE_IN_RESET = 3,
MQ_RQ_STATE_BITS = 2,
MQ_RQ_STATE_MASK = (1 << MQ_RQ_STATE_BITS) - 1,
--
2.9.5
Hi,
Not sure if you received my email from last week.
We offer following image editing services:
images cutting out, clipping path, masking
jewelry photos retouching
beauty photos retouching
also wedding photos etc
If you want to test our quality of work.
You may send us one photo with instruction and we will work on it.
Hope to hear from you soon.
Regards,
Ross
The Studio Manager