>
> Hi Martin:
>Here I add condlog(2, "start funcname"), pthread_cleanup_push(func_print,
> NULL)
> in every pthread_create func. When these two core happened, "exit tur_thread"
> are less than "start tur_thread". So the trouble may be in tur_thread.
>
If I made no mistake, these coredumps
On 3/1/21 2:44 PM, bchatel...@cines.fr wrote:
I try to use multipath with SAS disk, transported by Fiber Channel, on
Dell Compellent.
My volume is detected on 2x R440 PowerEdge hosted by Centos 8 and
orchestrated by oVirt.
[...]
# multipath -ll
36000d31003d5c210 dm-3 COMPELNT,Com
> Note that unlike all other threads, TUR threads are _detached_ threads.
> multipathd tries to cancel them, but it has no way to verify that they
> actually stopped. It may be just a normal observation that you can't
> see the messages when a TUR thread terminates, in particular if the
> program
Hi Lixiaokeng,
thanks for your continued efforts!
On Mon, 2021-03-01 at 22:53 +0800, lixiaokeng wrote:
>
> Hi Martin:
> Here I add condlog(2, "start funcname"),
> pthread_cleanup_push(func_print, NULL)
> in every pthread_create func. When these two core happened, "exit
> tur_thread"
> are le
On Tue, 2021-03-02 at 16:41 +0800, lixiaokeng wrote:
>
> >
> > Hi Martin:
> > Here I add condlog(2, "start funcname"),
> > pthread_cleanup_push(func_print, NULL)
> > in every pthread_create func. When these two core happened, "exit
> > tur_thread"
> > are less than "start tur_thread". So the t
It's work good.
I have add this line to device in multipath.conf :
path_grouping_policy "group_by_prio"
I have something like this :
#multipath -ll
36000d31003d5c210 dm-3 COMPELNT,Compellent Vol
size=1.5T features='1 queue_if_no_path' hwhandler='1 alua' wp=rw
|-+- policy='se
On 2021/3/2 19:07, Martin Wilck wrote:
> On Tue, 2021-03-02 at 16:41 +0800, lixiaokeng wrote:
>>
>>>
>>> Hi Martin:
>>> Here I add condlog(2, "start funcname"),
>>> pthread_cleanup_push(func_print, NULL)
>>> in every pthread_create func. When these two core happened, "exit
>>> tur_thread"
>>>
On Fri, 26 Feb 2021, JeffleXu wrote:
>
>
> On 2/20/21 3:38 AM, Mikulas Patocka wrote:
> >
> >
> > On Mon, 8 Feb 2021, Jeffle Xu wrote:
> >
> >> Offer one fastpath of bio-based polling when bio submitted to dm device
> >> is not split.
> >>
> >> In this case, there will be only one bio subm
Introduce a function that submits bio to a request-based device driver.
The function doesn't offload requests to current->bio_list. It is expected
to be called from device mapper, where current->bio_list is already set
up.
Signed-off-by: Mikulas Patocka
---
block/blk-core.c | 16 +++
Hi
This is reworked patchset for device mapper I/O polling.
The polling happens inside device mapper and the polling cookie is not
returned to the upper-layer code (because returning it would be unsafe).
Mikulas
--
dm-devel mailing list
dm-devel@redhat.com
https://listman.redhat.com/mailman/lis
Use submit_bio_noacct_mq_direct and pass the returned cookie through the
device mapper stack. The cookie and queue is stored in the structure
clone_info.
Signed-off-by: Mikulas Patocka
---
drivers/md/dm.c | 42 +++---
1 file changed, 23 insertions(+), 19 de
Support I/O polling if submit_bio_noacct_mq_direct returned non-empty
cookie.
Signed-off-by: Mikulas Patocka
---
drivers/md/dm.c |5 +
1 file changed, 5 insertions(+)
Index: linux-2.6/drivers/md/dm.c
===
--- linux-2.6.orig
Don't clear REQ_HIPRI for bio-based drivers. Device mapper will need to
see this flag in order to support polling.
Signed-off-by: Mikulas Patocka
---
block/blk-core.c |2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
Index: linux-2.6/block/blk-core.c
On Tue, 2021-03-02 at 20:44 +0800, lixiaokeng wrote:
>
> > Note that unlike all other threads, TUR threads are _detached_
> > threads.
> > multipathd tries to cancel them, but it has no way to verify that
> > they
> > actually stopped. It may be just a normal observation that you
> > can't
> > see
Hi lixiaokeng,
On Tue, 2021-03-02 at 16:29 +0100, Martin Wilck wrote:
> On Tue, 2021-03-02 at 20:44 +0800, lixiaokeng wrote:
> >
> >
>
> The stacks you have shown indicate that the instruction pointers were
> broken. That would suggest something similar as dicussed in the ML
> thread leading to
On 3/2/21 4:41 PM, bchatel...@cines.fr wrote:
It's work good.
I have add this line to device in multipath.conf :
path_grouping_policy "group_by_prio
That's not enough to run properly in active/passive mode with ALUA.
Mainly, because the default value of failback is "manual".
This is a min
16 matches
Mail list logo
