ock before
calling usb_unlink_urb.
CC:
Signed-off-by: Ming Lei
---
drivers/hid/usbhid/hid-core.c | 16 ++--
1 file changed, 10 insertions(+), 6 deletions(-)
diff --git a/drivers/hid/usbhid/hid-core.c b/drivers/hid/usbhid/hid-core.c
index aa1c503..b5d07da 100644
--- a/drivers/hid/usbhid/
On Fri, Apr 20, 2012 at 12:11 AM, Oliver Neukum wrote:
> Am Donnerstag, 19. April 2012, 15:51:04 schrieb Ming Lei:
>> The URB complete handler may be called by usb_unlink_urb directly,
>> so deadlock will be triggered in __usbhid_submit_report since
>> usbhid->lock is to
static inline void clear_cpu_bit(int nr, unsigned long __percpu *addr)
+{
+ unsigned long *fl = __this_cpu_ptr(addr);
+ clear_bit(nr, fl);
+}
+
+static inline int test_cpu_bit(int nr, unsigned long __percpu *addr)
+{
+ unsigned long *fl = __this_cpu_ptr(addr);
+ return test_bit(nr, fl);
+}
+
#endif
thanks,
--
Ming Lei
--
To unsubscribe from this list: send the line "unsubscribe stable" in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
On Fri, Apr 20, 2012 at 6:45 PM, Oliver Neukum wrote:
> Am Freitag, 20. April 2012, 12:17:51 schrieb Ming Lei:
>> On Fri, Apr 20, 2012 at 3:57 PM, Oliver Neukum wrote:
>> >
>> > You are racing with hid_irq_out(). It calls hid_submit_out()
>> > under lock. So
erstand you mean URB complete handler should be guaranteed
to be run only in IRQ context and can't be called by usb_unlink_urb,
don't I?
If yes, many hc drivers are still involved to be modified.
thanks,
--
Ming Lei
--
To unsubscribe from this list: send the line "unsubscribe stable&qu
On Sat, Apr 21, 2012 at 5:59 AM, Dmitry Torokhov
wrote:
> On Friday, April 20, 2012 06:17:51 PM Ming Lei wrote:
> Why don't you do something like this:
>
> urb_to_unlink = usbhid->urbout;
> usbhid->urbout = NULL;
This may trigger oops in hid_submit_out ca
nk_urb does?
just implement usb_unlink_urb as something like
tasklet_schedule(unlink_tasklet).
Then we can have a uniform lock requirement and no changes are involved
on host controller drivers.
Thanks,
--
Ming Lei
--
To unsubscribe from this list: send the line "unsubscribe stable" in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
the lock -- the race has already been lost.
>
> Does that solve your problem?
Not sure if that does work.
If the URB completes asynchronously after unlinking, its status is still
-ECONNRESET, so extra race may be caused without holding the lock
because complete handler will access some global d
On Sun, Apr 22, 2012 at 1:31 AM, Alan Stern wrote:
> On Sat, 21 Apr 2012, Ming Lei wrote:
>> How about always scheduling a tasklet to run what usb_unlink_urb does?
>> just implement usb_unlink_urb as something like
>> tasklet_schedule(unlink_tasklet).
>>
>>
On Sun, Apr 22, 2012 at 8:50 PM, Alan Stern wrote:
> On Sun, 22 Apr 2012, Ming Lei wrote:
>
>> > Although the kerneldoc doesn't actually say so, it should be safe to
>> > assume that usb_unlink_urb calls the completion routine directly _only_
>> > in cases
On Mon, Apr 23, 2012 at 11:42 PM, Alan Stern wrote:
> On Sun, 22 Apr 2012, Ming Lei wrote:
>> If the lock which is to be acquired in the URB complete handler is dropped
>> before calling usb_unlink_urb, one new submitted URB in complete handler
>> may be unlinked, as mentio
On Tue, Apr 24, 2012 at 10:22 PM, Oliver Neukum wrote:
> Am Dienstag, 24. April 2012, 06:19:00 schrieb Ming Lei:
>
>> @@ -486,11 +494,15 @@ static void hid_ctrl(struct urb *urb)
>>
>> if (usbhid->ctrlhead != usbhid->ctrltail && !hid_submit_ctrl(hi
On Tue, Apr 24, 2012 at 11:20 PM, Alan Stern wrote:
> On Tue, 24 Apr 2012, Ming Lei wrote:
>
>> > Instead of changing return codes or adding locks, how about
>> > implementing a small state machine for each URB?
>> >
>> > Initially the state is A
uct usbhid_device {
unsigned long last_out;
/* record of last output for timeouts */
spinlock_t lock; /* fifo
spinlock */
+ spinlock_t unlink_lock;
ks it is enough to
avoid the race, isn't it?
> We'd be back at the original proposal.
Introducing a new flag to describe 'unlinked' is still OK, but
borrowing urb->unlinked is better, the bad is that it is private.
Thanks,
--
Ming Lei
--
To unsubscribe from this list: send the line "unsubscribe stable" in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
transfer requests en masse
> * @anchor: anchor the requests are bound to
> *
> diff --git a/include/linux/usb.h b/include/linux/usb.h
> index 73b68d1..23df8ae 100644
> --- a/include/linux/usb.h
> +++ b/include/linux/usb.h
> @@ -1379,6 +1379,8 @@ extern int usb_unlink_urb(struc
On Wed, Apr 25, 2012 at 6:52 PM, Oliver Neukum wrote:
> Am Mittwoch, 25. April 2012, 11:14:19 schrieb Ming Lei:
>> You need to add check of "usbhid->outtail != usbhid->outhead" above.
>
> Done. Could you test?
Basically my USB keyboard can work well as before wit
a bit big for kmalloc
- given that many devices have the partition count limit, so only
allocate disk_max_parts() partitions instead of 256 partitions
Cc: stable@vger.kernel.org
Signed-off-by: Ming Lei
---
block/partition-generic.c |2 +-
block/partitions/check.c | 35
arts;
>
> This is relatively unusual style.
> Could you break out this struct instead?
IMO, looks it isn't necessary since no one uses the type of the embedded
structure directly, and the style isn't introduced by this patch, which only
follows the previous one.
Thanks,
--
M
On Fri, Feb 1, 2013 at 4:33 PM, Yasuaki Ishimatsu
wrote:
> At least, you should use release_partitions() instead of kfree() here.
Good catch, thank you for pointing it out, and I will post v1 later
with the update.
Thanks,
--
Ming Lei
--
To unsubscribe from this list: send the l
a bit big for kmalloc
- given that many devices have the partition count limit, so only
allocate disk_max_parts() partitions instead of 256 partitions
Cc: stable@vger.kernel.org
Signed-off-by: Ming Lei
---
v1:
- add one missing free_partitions
- rename release_partitions as
On Sat, Feb 2, 2013 at 6:43 AM, Andrew Morton wrote:
> On Fri, 1 Feb 2013 20:23:12 +0800
I just return from holiday, sorry for the delay, and thanks for the review.
> Ming Lei wrote:
>
>> Currently, sizeof(struct parsed_partitions) may be 64KB in 32bit arch,
>> so it is
Hi Greg,
The attachment patch is backported from the commit below:
>From 7643721471117d5f62ca36f328d3dc8d84af4402 Mon Sep 17 00:00:00 2001
From: Ming Lei
Date: Fri, 22 Feb 2013 03:05:03 +
Subject: [PATCH] usbnet: smsc95xx: fix suspend failure
and this patch can be applied cleanly agai
commit:
From 7643721471117d5f62ca36f328d3dc8d84af4402 Mon Sep 17 00:00:00 2001
Subject: [PATCH] usbnet: smsc95xx: fix suspend failure
Cc:
Cc: Steve Glendinning
Signed-off-by: Ming Lei
---
drivers/net/usb/smsc95xx.c |6 ++
1 file changed, 6 insertions(+)
diff --git a/drivers/n
) Atr=01(Isoc) MxPS= 33 Ivl=1ms
E: Ad=03(O) Atr=01(Isoc) MxPS= 33 Ivl=1ms
I: If#= 1 Alt= 5 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E: Ad=83(I) Atr=01(Isoc) MxPS= 49 Ivl=1ms
E: Ad=03(O) Atr=01(Isoc) MxPS= 49 Ivl=1ms
Cc:
Cc: Gustavo Padovan
Signed-off-by: Ming Lei
---
drivers
) Atr=01(Isoc) MxPS= 33 Ivl=1ms
E: Ad=03(O) Atr=01(Isoc) MxPS= 33 Ivl=1ms
I: If#= 1 Alt= 5 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E: Ad=83(I) Atr=01(Isoc) MxPS= 49 Ivl=1ms
E: Ad=03(O) Atr=01(Isoc) MxPS= 49 Ivl=1ms
Cc:
Cc: Gustavo Padovan
Signed-off-by: Ming Lei
---
v1
and sent out.
Thanks,
--
Ming Lei
--
To unsubscribe from this list: send the line "unsubscribe stable" in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
) MxPS= 33 Ivl=1ms
I: If#= 1 Alt= 5 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E: Ad=83(I) Atr=01(Isoc) MxPS= 49 Ivl=1ms
E: Ad=03(O) Atr=01(Isoc) MxPS= 49 Ivl=1ms
Cc:
Cc: Gustavo Padovan
Signed-off-by: Ming Lei
---
drivers/bluetooth/ath3k.c |2 ++
drivers/bluetooth/btusb.c
e problem since
the lock is always held in readdir path.
Reported-by: Dave Jones
Tested-by: Sasha Levin
Cc:
Signed-off-by: Ming Lei
---
fs/sysfs/dir.c | 13 -
1 file changed, 12 insertions(+), 1 deletion(-)
diff --git a/fs/sysfs/dir.c b/fs/sysfs/dir.c
index 2fbdff6..c9e1660 100644
gered later.
This patch returns immeadiately under the situation for fixing the bug,
and it is reasonable to return from readdir() when filldir() fails.
Reported-by: Dave Jones
Tested-by: Sasha Levin
Cc:
Signed-off-by: Ming Lei
---
fs/sysfs/dir.c |4
1 file changed, 4 insertions(+
On Thu, Mar 21, 2013 at 10:41 AM, Li Zefan wrote:
>
> In fact the same race exists between readdir() and read()/write()...
Fortunately, no read()/write() are implemented on sysfs directory, :-)
Thanks,
--
Ming Lei
--
To unsubscribe from this list: send the line "unsubscribe stable&q
On Thu, Mar 21, 2013 at 11:28 AM, Li Zefan wrote:
> On 2013/3/21 11:17, Ming Lei wrote:
>> On Thu, Mar 21, 2013 at 10:41 AM, Li Zefan wrote:
>>>
>>> In fact the same race exists between readdir() and read()/write()...
>>
>> Fortunately, no read()/wri
On Fri, Mar 22, 2013 at 1:48 PM, Li Zefan wrote:
> On 2013/3/21 12:48, Ming Lei wrote:
>
> Yes, it can...As I said, it's irrelevant, because it's vfs that changes
> file->f_pos.
>
> SYSCALL_DEFINE3(read, unsigned int, fd, char __user *, buf, size_t, count)
>
sb&m=136368139627876&w=2
Cc: stable
Signed-off-by: Ming Lei
---
Cc stable since the previous patchset is marked as stable.
drivers/usb/serial/usb-serial.c |1 +
1 file changed, 1 insertion(+)
diff --git a/drivers/usb/serial/usb-serial.c b/drivers/usb/serial/usb-serial.c
index 2e70e
Also I appreciate it if you may share your test case...
Thanks,
--
Ming Lei
--
To unsubscribe from this list: send the line "unsubscribe stable" in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Hi Zefan,
On Tue, Mar 26, 2013 at 4:45 PM, Ming Lei wrote:
> On Tue, Mar 26, 2013 at 3:30 PM, Li Zefan wrote:
>>> Considered that vfs_read()/vfs_write on sysfs dir is almost doing nothing,
>>> the
>>> above problem may only exist in theory.
>>
>> The r
On Tue, Mar 26, 2013 at 10:03 PM, Ming Lei wrote:
>
> If you mean the test code on link[1], I can't reproduce the
> warning with the two sysfs fix patches in 4 hours's test.
>
> [1], https://patchwork.kernel.org/patch/2160771/
You are right, looks it is not a problem
st code on the
link:
https://patchwork.kernel.org/patch/2160771/
This patch fixes the use after free under this situation.
Cc:
Reported-by: Li Zefan
Signed-off-by: Ming Lei
---
fs/sysfs/dir.c | 15 +++
1 file changed, 11 insertions(+), 4 deletions(-)
diff --git a/fs/sys
ot;fix" should be discussed further, because in theory
we can't avoid all GFP_KERNEL allocation in both device
reset path and runtime resume path, which may call other
kernel component APIs with GFP_KERNEL allocation.
So hope to figure out one generic enough solution.
Thanks,
--
Ming Lei
--
To
4e7f 100644
> --- a/drivers/base/firmware_class.c
> +++ b/drivers/base/firmware_class.c
> @@ -38,7 +38,7 @@ MODULE_LICENSE("GPL");
>
> /* Builtin firmware support */
>
> -#ifdef CONFIG_FW_LOADER
> +#ifdef CONFIG_FIRMWARE_IN_KERNEL
This might not be correct whe
will break
this case.
>
> Perhaps the compile problem is solved in newer kernels (by always
> generating an empty builtin firmware list?) but the #ifdef is still
> incorrect.
Looks the problem hasn't been reported before.
Thanks,
--
Ming Lei
--
To unsubscribe from this list: send t
parent lock if it has to fix the
races.
Cc: Alan Stern
Cc: stable@vger.kernel.org
Signed-off-by: Ming Lei
---
drivers/base/core.c |8
1 file changed, 8 insertions(+)
diff --git a/drivers/base/core.c b/drivers/base/core.c
index 346be8b..cbc8bd2 100644
--- a/drivers/base/core.c
> down to which you prefer: a hang during shutdown, or a crash. :-)
Considered that the device or driver can be logged, either hang or
crash will be fixed later by someone, :-)
Thanks,
--
Ming Lei
--
To unsubscribe from this list: send the line "unsubscribe stable" in
the body of
nd its parent lock(if it has) to
fix the races.
Cc: Alan Stern
Cc: stable@vger.kernel.org
Signed-off-by: Ming Lei
---
v2:
- take Alan's suggestion to use device_trylock to avoid
hanging during shutdown by buggy device or driver
- hold parent reference counter
d
);
>
> Even if the parent can't be locked, you should still try to lock the
> device.
I have considered doing it, but which may consume another 1sec.
Also if the parent lock has been held, it is very possibly that the
device can't be probed or removed at the same time, so just lo
On Tue, Jun 12, 2012 at 12:02 AM, Alan Stern wrote:
> On Mon, 11 Jun 2012, Ming Lei wrote:
>>
>> I have considered doing it, but which may consume another 1sec.
>>
>> Also if the parent lock has been held, it is very possibly that the
>> device can't be probe
Without clearing OPEN flag in failure path, runtime or system resume
may submit interrupt/rx URB and start tx queue mistakenly on a
interface in DOWN state.
Cc: stable@vger.kernel.org
Signed-off-by: Ming Lei
---
drivers/net/usb/usbnet.c |4 +++-
1 file changed, 3 insertions(+), 1 deletion
This patch decreases dev->suspend_count in the -EBUSY failure path
of usbnet_suspend. Without the change, the later runtime suspend
will do nothing except for increasing dev->suspend_count.
Cc: stable@vger.kernel.org
Signed-off-by: Ming Lei
---
drivers/net/usb/usbnet.c |1 +
1 file c
patch, usbnet can't recieve any packets from peer
in runtime suspend state if runtime PM is enabled and
autosuspend_delay is set as zero.
Cc: stable@vger.kernel.org
Signed-off-by: Ming Lei
---
drivers/net/usb/usbnet.c | 42 ++
1 file change
On Wed, Jun 13, 2012 at 12:20 PM, Ming Lei wrote:
> If usbnet is resumed by remote wakeup, generally there are
> some packets comming to be handled, so allocate and submit
> rx URBs in usbnet_resume to avoid delays introduced by tasklet.
> Otherwise, usbnet may have been runtime susp
On Sat, Jun 16, 2012 at 6:03 AM, Greg Kroah-Hartman
wrote:
> On Mon, Jun 11, 2012 at 01:13:20PM +0800, Ming Lei wrote:
>> Firstly, .shutdown callback may touch a uninitialized hardware
>> if dev->driver is set and .probe is not completed.
>>
>> Secondly, device_shu
On Mon, Jun 18, 2012 at 7:22 AM, David Miller wrote:
> From: Ming Lei
> Date: Fri, 15 Jun 2012 10:22:16 +0800
>
>> David, sorry, the 'GFP_ATOMIC' above should be 'flags', so could
>> you take the fixed version from attachment? Or could you do it by
>
can be
exploited in reality.
>
>> >> Signed-off-by: Ming Lei
>> >> ---
>> >> v2:
>> >> - take Alan's suggestion to use device_trylock to avoid
>> >> hanging during shutdown by buggy device or driver
>> >>
On Thu, Jun 21, 2012 at 6:37 AM, Greg Kroah-Hartman
wrote:
> On Tue, Jun 19, 2012 at 10:00:36AM +0800, Ming Lei wrote:
>> so I marked it as -stable because I have explained how the race can be
>> exploited in reality.
>
> Ok, but as this has been there since when, 2
n, do you have objections on the new one?
Thanks,
--
Ming Lei
--
To unsubscribe from this list: send the line "unsubscribe stable" in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
On Fri, Jun 22, 2012 at 12:12 AM, Alan Stern wrote:
> On Fri, 22 Jun 2012, Ming Lei wrote:
>
>> On Thu, Jun 21, 2012 at 9:49 PM, Greg Kroah-Hartman
>> wrote:
>>
>> > We also have not made the rule that a shutdown will complete in a
>> > specific amount
: stable@vger.kernel.org
Signed-off-by: Ming Lei
---
drivers/usb/core/driver.c |4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)
diff --git a/drivers/usb/core/driver.c b/drivers/usb/core/driver.c
index ddd820d..c228c9c 100644
--- a/drivers/usb/core/driver.c
+++ b/drivers/usb/core
On Thu, Sep 20, 2012 at 3:29 PM, Oliver Neukum wrote:
> On Thursday 20 September 2012 13:28:51 Ming Lei wrote:
>> If one usb interface driver returns failure from its
>> suspend(auto) callback inside usb_suspend_both, its
>> resume() callback should be called to recover
ist' according to 'VARIETIES OF MEMORY BARRIER' part
of Documentation/memory-barriers.txt.
Reported-and-Tested-by: Russell King
Cc:
Signed-off-by: Ming Lei
---
drivers/base/bus.c |2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/base/bus.c b/drivers
described in the commit log, and avoid the problem.
> the driver model code. All in all, what with the error path issue, and
> now the blatently wrong description, I'm not gaining much confidence in
> Ming Lei.
>
> The below is actually what's happening, according to
On Fri, Sep 28, 2012 at 11:31 AM, anish singh
wrote:
> Hello Ming,
> Though I am not an expert in this driver core area but
> I have been following this fix.So have some queries below:
>
> On Fri, Sep 28, 2012 at 6:22 AM, Ming Lei wrote:
>> Inside bus_add_driver(), one devic
tach because driver_attach didn't
see the device in the bus.
So the 'devb' will be missed to be probed in the bus, won't it?
Thanks,
--
Ming Lei
--
To unsubscribe from this list: send the line "unsubscribe stable" in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
On Fri, Sep 28, 2012 at 10:13 PM, Russell King - ARM Linux
wrote:
> On Fri, Sep 28, 2012 at 10:07:22PM +0800, Ming Lei wrote:
>> On Fri, Sep 28, 2012 at 9:55 PM, Russell King - ARM Linux
>> wrote:
>> >> I do not mention threads case in one CPU because the context in
ssell's nice explanation.
Sorry, how do you know I didn't look at Russell's explanation?
Thanks,
--
Ming Lei
--
To unsubscribe from this list: send the line "unsubscribe stable" in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
On Fri, Sep 28, 2012 at 11:05 PM, Russell King - ARM Linux
wrote:
> On Fri, Sep 28, 2012 at 10:44:13PM +0800, Ming Lei wrote:
>> On Fri, Sep 28, 2012 at 10:31 PM, anish kumar
>> wrote:
>>
>> > Ming,
>> > Somehow I feel here that the problem reported by Russ
On Wed, Nov 21, 2012 at 10:01 PM, Solomon Peachy wrote:
> On Wed, Nov 21, 2012 at 09:35:28AM +0800, Ming Lei wrote:
>> Solomon, I can't duplicate the build failure with your .config on
>> 3.7-rc5-next.
>
> Okay, so it's since been fixed.
>
>> > * Th
_copy_worker);
>
> @@ -689,6 +697,7 @@ fail:
> mutex_unlock(&epdata->lock);
>
> if (unlikely(value)) {
> + kfree(priv->iv);
> kfree(priv);
> put_ep(epdata);
> } else
> --
>
ted-by: Dongsu Park
Signed-off-by: Ming Lei
---
block/blk-mq.c | 17 +++--
1 file changed, 11 insertions(+), 6 deletions(-)
diff --git a/block/blk-mq.c b/block/blk-mq.c
index 6804028..9da05b3 100644
--- a/block/blk-mq.c
+++ b/block/blk-mq.c
@@ -1580,15 +1580,20 @@ stat
has been disabled, otherwise
still use-after-free on tags can be triggered.
Cc:
Reported-by: Dongsu Park
Signed-off-by: Ming Lei
---
block/blk-mq.c | 13 ++---
1 file changed, 10 insertions(+), 3 deletions(-)
diff --git a/block/blk-mq.c b/block/blk-mq.c
index 67f01a0..6804028 100644
--
has been unmapped, otherwise
still use-after-free on tags can be triggered.
Cc:
Reported-by: Dongsu Park
Tested-by: Dongsu Park
Signed-off-by: Ming Lei
---
block/blk-mq.c | 16 +---
1 file changed, 13 insertions(+), 3 deletions(-)
diff --git a/block/blk-mq.c b/block/blk-mq.c
in
Tested-by: Dongsu Park
Signed-off-by: Ming Lei
---
block/blk-mq.c | 34 +-
1 file changed, 13 insertions(+), 21 deletions(-)
diff --git a/block/blk-mq.c b/block/blk-mq.c
index 1fccb98..76f460e 100644
--- a/block/blk-mq.c
+++ b/block/blk-mq.c
@@ -1574,22 +1574,6
@vger.kernel.org (v4.0)
Reported-by: Justin M. Forbes
Tested-by: Justin M. Forbes
Signed-off-by: Ming Lei
---
drivers/block/loop.c | 19 +--
drivers/block/loop.h | 2 ++
2 files changed, 19 insertions(+), 2 deletions(-)
diff --git a/drivers/block/loop.c b/drivers/block/loop.c
index
On Wed, Apr 29, 2015 at 12:36 AM, Jeff Moyer wrote:
> Ming Lei writes:
>
>> If there are too many pending per work I/O, too many
>> high priority work thread can be generated so that
>> system performance can be effected.
>>
>> This patch limits the max pen
On Fri, May 1, 2015 at 12:59 AM, Jeff Moyer wrote:
> Ming Lei writes:
>
>> On Wed, Apr 29, 2015 at 12:36 AM, Jeff Moyer wrote:
>>> Ming Lei writes:
>>>
>>>> If there are too many pending per work I/O, too many
>>>> high priority work threa
On Fri, May 22, 2015 at 8:36 PM, Josh Boyer wrote:
> On Tue, May 5, 2015 at 7:49 AM, Ming Lei wrote:
>> If there are too many pending per work I/O, too many
>> high priority work thread can be generated so that
>> system performance can be effected.
>>
>>
}
>
> device_unlock(dev);
> - if (dev->parent)
> - device_unlock(dev->parent);
> + if (parent)
> + device_unlock(parent);
>
> put_device(dev);
> - put_de
toph Hellwig
Cc: stable@vger.kernel.org (v4.0)
Signed-off-by: Ming Lei
---
block/blk-mq.c | 8 ++--
1 file changed, 6 insertions(+), 2 deletions(-)
diff --git a/block/blk-mq.c b/block/blk-mq.c
index e68b71b..594eea0 100644
--- a/block/blk-mq.c
+++ b/block/blk-mq.c
@@ -1600,6 +1600,7 @@ static int
On Thu, Jun 4, 2015 at 9:14 PM, Ming Lei wrote:
> Now blk_cleanup_queue() can be called before calling
> del_gendisk()[1], inside which hctx->ctxs is touched
> from blk_mq_unregister_hctx(), but the variable has
> been freed by blk_cleanup_queue() at that time.
>
> So this p
ng the oops reported by Stefan.
[1], 6cd18e711dd8075 (block: destroy bdi before blockdev is
unregistered)
Reported-by: Stefan Seyfried
Cc: NeilBrown
Cc: Christoph Hellwig
Cc: stable@vger.kernel.org (v4.0)
Signed-off-by: Ming Lei
---
V1:
- fix commit log
block/blk-mq.c | 8 +
U Standard PC (i440FX + PIIX, 1996), BIOS
Bochs 01/01/2011^M
[ 359.055682] task: 8802161cc000 ti: 88021b4a8000 task.ti:
88021b4a8000^M
[ 359.055693] RIP: 0010:[] []
__kmalloc+0xe8/0x152^M
Cc:
Signed-off-by: Ming Lei
---
block/blk-mq-sysfs.c | 25 ++---
1 fi
439.790911] ---[ end trace d40af58949325661 ]---^M
Cc:
Signed-off-by: Ming Lei
---
block/blk-flush.c | 15 ++-
block/blk-mq-tag.c | 4 ++--
block/blk-mq-tag.h | 12
block/blk-mq.c | 16 +---
block/blk.h| 6 ++
5 files changed, 35 insertion
blem reported by nirinA raseliarison:
http://lkml.org/lkml/2013/6/14/188
Reported-and-tested-by: nirinA raseliarison
Cc: Guenter Roeck
Cc: Bjorn Helgaas
Cc: stable
Signed-off-by: Ming Lei
---
drivers/base/firmware_class.c | 27 ++-
1 file changed, 18 insertions(+),
On Tue, Jun 18, 2013 at 7:59 AM, Greg Kroah-Hartman
wrote:
> On Sat, Jun 15, 2013 at 04:36:38PM +0800, Ming Lei wrote:
>> fw_priv->buf is accessed in both request_firmware_load() and
>> writing to sysfs file of 'loading' context, but not protected
>> by '
On Tue, Jun 18, 2013 at 12:05 PM, Guenter Roeck wrote:
>>
> I may be missing something, but why would mainline not need it ?
> Or do you mean "mainline plus 3.9" ?
Yes, mainline need it of course, sorry for not mentioning that explicitly.
Thanks,
--
Ming Lei
--
To unsu
From: Ming Lei
This patch is the arm version of below patch, which is already in
-mm tree:
http://marc.info/?t=13825144456&r=1&w=2
Without this patch, the below syscall may return failure when
'/proc/sys/vm/mmap_min_addr' is set 32768:
mmap(0, size, PROT
ter on ARM when CONFIG_PAGE_OFFSET
is defined as the original problem is only on ARM.
Cc:
Cc: Rusty Russell
Singed-off-by: Ming Lei
---
scripts/link-vmlinux.sh |4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)
diff --git a/scripts/link-vmlinux.sh b/scripts/link-vmlinux.sh
index 32b10f
: Jonathan Austin
Signed-off-by: Ming Lei
---
arch/arm/Kconfig |1 +
1 file changed, 1 insertion(+)
diff --git a/arch/arm/Kconfig b/arch/arm/Kconfig
index 799ef94..db708c4 100644
--- a/arch/arm/Kconfig
+++ b/arch/arm/Kconfig
@@ -1560,6 +1560,7 @@ endchoice
config PAGE_OFFSET
orted patch for 3.14.
Thanks,
--
Ming Lei
From d1152f1cc1b1784691746bcfc18729ac4b7eb1b8 Mon Sep 17 00:00:00 2001
From: Ming Lei
Date: Sun, 8 Jun 2014 10:33:34 +0800
Subject: [PATCH] [PATCH] virtio_blk: fix race between start and stop queue
When there isn't enough vring descriptor for addin
c of mq_flush_run() will never be run
- fs sync hangs forever
This patch introduces one exclusive work_struct inside
request_queue for queuing flush request only to fix the
problem.
Cc: #3.14
Signed-off-by: Ming Lei
---
Another simple fix is to disable ipi for flush request, but lo
On Mon, May 19, 2014 at 11:18 PM, Christoph Hellwig wrote:
> On Mon, May 19, 2014 at 11:05:50PM +0800, Ming Lei wrote:
>> Another simple fix is to disable ipi for flush request, but looks
>> this one should be better.
>
> I think the first thing is to bite the bullet and
On Mon, May 19, 2014 at 11:18 PM, Christoph Hellwig wrote:
> On Mon, May 19, 2014 at 11:05:50PM +0800, Ming Lei wrote:
>> Another simple fix is to disable ipi for flush request, but looks
>> this one should be better.
>
> I think the first thing is to bite the bullet and
On Tue, May 20, 2014 at 11:23 PM, Christoph Hellwig wrote:
> On Tue, May 20, 2014 at 11:20:25AM +0800, Ming Lei wrote:
>> - the conflict on the two structures just happens with flush
>> requests because rq->requeue_work is only used to queue
>> flush requests
>
>
On Wed, May 21, 2014 at 1:36 PM, Christoph Hellwig wrote:
> On Wed, May 21, 2014 at 01:16:14PM +0800, Ming Lei wrote:
>> I am wondering if virtio-blk is trivial block driver, :-)
>
> It's about as simple as it gets.
>
>> > The scsi-mq work that I plant to sub
d it.
Looks writing over ext4(especially sync writing) can survive
with Christoph's patch now, thanks Christoph.
Reported-and-tested-by: Ming Lei
Thanks,
--
Ming Lei
--
To unsubscribe from this list: send the line "unsubscribe stable" in
the body of a message to majord..
94 matches
Mail list logo
