On Wed, Nov 21, 2018 at 05:10:25PM +0100, Christoph Hellwig wrote:
> On Wed, Nov 21, 2018 at 11:31:36PM +0800, Ming Lei wrote:
> > > But while looking over this I wonder why we even need the max_seg_len
> > > here. The only thing __bvec_iter_advance does it to move bi_bvec_done
> > > and bi_idx
On Wed, Nov 21, 2018 at 05:08:11PM +0100, Christoph Hellwig wrote:
> On Wed, Nov 21, 2018 at 11:06:11PM +0800, Ming Lei wrote:
> > bvec_iter_* is used for single-page bvec in current linus tree, and there
> > are
> > lots of users now:
> >
> > [linux]$ git grep -n "bvec_iter_*" ./ | wc
> >
Actually..
I think we can kill this code entirely. If we look at what the
clustering setting is really about it is to avoid ever merging a
segement that spans a page boundary. And we should be able to do
that with something like this before your series:
---
>From
On Wed, Nov 21, 2018 at 05:10:25PM +0100, Christoph Hellwig wrote:
> No - I think we can always use the code without any segment in
> bvec_iter_advance. Because bvec_iter_advance only operates on the
> iteractor, the generation of an actual single-page or multi-page
> bvec is left to the caller
On Wed, Nov 21, 2018 at 11:48:13PM +0800, Ming Lei wrote:
> I guess the correct check should be:
>
> end_addr = vec_addr + bv->bv_offset + bv->bv_len;
> if (same_page &&
> (end_addr & PAGE_MASK) != (page_addr & PAGE_MASK))
>
On Wed, Nov 21, 2018 at 11:06:11PM +0800, Ming Lei wrote:
> bvec_iter_* is used for single-page bvec in current linus tree, and there are
> lots of users now:
>
> [linux]$ git grep -n "bvec_iter_*" ./ | wc
> 191 995 13242
>
> If we have to switch it first, it can be a big change, just
On Wed, Nov 21, 2018 at 03:55:02PM +0100, Christoph Hellwig wrote:
> On Wed, Nov 21, 2018 at 11:23:23AM +0800, Ming Lei wrote:
> > if (bio->bi_vcnt > 0) {
> > - struct bio_vec *bv = >bi_io_vec[bio->bi_vcnt - 1];
> > + struct bio_vec bv;
> > + struct bio_vec *seg =
On Wed, Nov 21, 2018 at 03:33:55PM +0100, Christoph Hellwig wrote:
> > + non-cluster.o
>
> Do we really need a new source file for these few functions?
>
> > default:
> > + if (!blk_queue_cluster(q)) {
> > + blk_queue_non_cluster_bio(q, bio);
> >
On Wed, Nov 21, 2018 at 02:32:44PM +0100, Christoph Hellwig wrote:
> > +#define bio_iter_mp_iovec(bio, iter) \
> > + segment_iter_bvec((bio)->bi_io_vec, (iter))
>
> Besides the mp naming we'd like to get rid off there also is just
> a single user of this macro,
On Wed, Nov 21, 2018 at 02:19:28PM +0100, Christoph Hellwig wrote:
> On Wed, Nov 21, 2018 at 11:23:10AM +0800, Ming Lei wrote:
> > This patch introduces helpers of 'segment_iter_*' for multipage
> > bvec support.
> >
> > The introduced helpers treate one bvec as real multi-page segment,
> > which
On Wed, Nov 21, 2018 at 11:23:23AM +0800, Ming Lei wrote:
> if (bio->bi_vcnt > 0) {
> - struct bio_vec *bv = >bi_io_vec[bio->bi_vcnt - 1];
> + struct bio_vec bv;
> + struct bio_vec *seg = >bi_io_vec[bio->bi_vcnt - 1];
>
> - if (page ==
> + non-cluster.o
Do we really need a new source file for these few functions?
> default:
> + if (!blk_queue_cluster(q)) {
> + blk_queue_non_cluster_bio(q, bio);
> + return;
I'd name this
On Wed, Nov 21, 2018 at 09:45:25AM +0200, Nikolay Borisov wrote:
> > + bio_for_each_segment_all()
> > + bio_first_bvec_all()
> > + bio_first_page_all()
> > + bio_last_bvec_all()
> > +
> > +* The following helpers iterate over single-page bvecs. The passed 'struct
> > +bio_vec' will contain
On Wed, Nov 21, 2018 at 11:23:20AM +0800, Ming Lei wrote:
> This patch introduces one extra iterator variable to
> bio_for_each_segment_all(),
> then we can allow bio_for_each_segment_all() to iterate over multi-page bvec.
>
> Given it is just one mechannical & simple change on all
>
On Wed, Nov 21, 2018 at 11:23:19AM +0800, Ming Lei wrote:
> bch_bio_alloc_pages() is always called on one new bio, so it is safe
> to access the bvec table directly. Given it is the only kind of this
> case, open code the bvec table access since bio_for_each_segment_all()
> will be changed to
> diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
> index 1ad6eafc43f2..a281b6737b61 100644
> --- a/include/linux/blkdev.h
> +++ b/include/linux/blkdev.h
> @@ -805,6 +805,10 @@ struct req_iterator {
> __rq_for_each_bio(_iter.bio, _rq) \
>
On Wed, 2018-11-21 at 12:20 +0100, Hannes Reinecke wrote:
> On 11/21/18 11:50 AM, Martin Wilck wrote:
> >
> > The concern that I have is with the design of the polling dmevent
> > API,
> > in particular the fact that the default action is EVENT_REMOVE. If,
> > for
> > whatever reason, one map is
> +#define bio_iter_mp_iovec(bio, iter) \
> + segment_iter_bvec((bio)->bi_io_vec, (iter))
Besides the mp naming we'd like to get rid off there also is just
a single user of this macro, please just expand it there.
> +#define segment_iter_bvec(bvec, iter)
On Wed, Nov 21, 2018 at 11:23:10AM +0800, Ming Lei wrote:
> This patch introduces helpers of 'segment_iter_*' for multipage
> bvec support.
>
> The introduced helpers treate one bvec as real multi-page segment,
> which may include more than one pages.
Unless I'm missing something these bvec vs
On 11/21/18 11:50 AM, Martin Wilck wrote:
Hi Ben,
I recently encountered a strange and disturbing phenomenon with the
latest upstream code.
The system in question had two iSCSI connections on separate networks.
On one of them, I tested iSCSI disconnect and re-connects, and broken
storage
Hi Ben,
I recently encountered a strange and disturbing phenomenon with the
latest upstream code.
The system in question had two iSCSI connections on separate networks.
On one of them, I tested iSCSI disconnect and re-connects, and broken
storage configurations (different LUNs diffenent sizes
It's a big difference if a map is flushed from DM (changing kernel
state) or just removed from internal multipathd tables. Convey
this information in log messages.
Signed-off-by: Martin Wilck
---
libmultipath/structs_vec.c | 6 +++---
libmultipath/structs_vec.h | 3 ++-
multipathd/main.c
Add unit tests for the bitmask management functions the previous
patch added in util.h.
Signed-off-by: Martin Wilck
---
tests/util.c | 98
1 file changed, 98 insertions(+)
diff --git a/tests/util.c b/tests/util.c
index 839effd2..e6d4b9ab
On Tue, Nov 20, 2018 at 09:35:07PM -0800, Sagi Grimberg wrote:
>
> > > Wait, I see that the bvec is still a single array per bio. When you said
> > > a table I thought you meant a 2-dimentional array...
> >
> > I mean a new 1-d table A has to be created for multiple bios in one rq,
> > and build
In coalesce_paths() and ev_add_path(), we check for size mismatch.
We should do it here, too.
Signed-off-by: Martin Wilck
---
libmultipath/structs_vec.c | 6 ++
1 file changed, 6 insertions(+)
diff --git a/libmultipath/structs_vec.c b/libmultipath/structs_vec.c
index 561ff4ac..a0b69ce5
Reduce the verbosity of pointless messages from pathinfo(), in
particular those that are printed on every checker invocation.
Signed-off-by: Martin Wilck
---
libmultipath/discovery.c | 10 ++
1 file changed, 6 insertions(+), 4 deletions(-)
diff --git a/libmultipath/discovery.c
It's normal that SCSI devices don't support this page. Only RDAC
devices need it.
Signed-off-by: Martin Wilck
---
libmultipath/discovery.c | 4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)
diff --git a/libmultipath/discovery.c b/libmultipath/discovery.c
index 1c87277f..7f983a63 100644
This was important during development of the refcounting code, but
now it isn't any more.
Signed-off-by: Martin Wilck
---
libmultipath/prio.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/libmultipath/prio.c b/libmultipath/prio.c
index 17acfd05..0590218d 100644
---
The remove_map_and_stop_waiter() call which follows immediately
will do the same thing.
Signed-off-by: Martin Wilck
---
multipathd/main.c | 1 -
1 file changed, 1 deletion(-)
diff --git a/multipathd/main.c b/multipathd/main.c
index 79c8ed1e..0c248046 100644
--- a/multipathd/main.c
+++
When there are paths with the same WWID but different sizes, and
coalesce_paths() walks the pathvec, it checks paths _after_
the current one for size mismatch and sets ACT_REJECT. However,
these paths will be reached in the main loop later, and this time
the already handled paths will not be
These messages, printed for every checker invocation, were helpful
in the past but they don't make much sense under normal operations.
The same information can be derived from journal time stamps on
recent Linux distributions.
Signed-off-by: Martin Wilck
---
multipathd/main.c | 2 +-
1 file
This will make log level 4 actually usable.
Signed-off-by: Martin Wilck
---
libmultipath/util.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/libmultipath/util.c b/libmultipath/util.c
index 66c47611..b5d2e706 100644
--- a/libmultipath/util.c
+++ b/libmultipath/util.c
@@
With verbosity level 3, thousands of "path state = running" messages
are logged, which are pretty much irrelevant, as the checker state
takes precedence. Also the "get_state" message is totally useless.
With this patch, the path state is reported exactly once per path
and check.
Signed-off-by:
Unlike the other filter functions, filter_property() is called
on every pathinfo() call, and prints a message with -v3 every
time. Avoid that by using a lower log priority for the call
from pathinfo().
Signed-off-by: Martin Wilck
---
libmultipath/blacklist.c | 54
The thread startup/stop messages aren't very interesting.
Decrease their verbosity level to v4.
Signed-off-by: Martin Wilck
---
libmultipath/checkers/tur.c | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/libmultipath/checkers/tur.c b/libmultipath/checkers/tur.c
index
These messages are printed from update_multipath_strings, and
thus from check_path(), which is too much for -v3.
Signed-off-by: Martin Wilck
---
libmultipath/dmparser.c | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/libmultipath/dmparser.c b/libmultipath/dmparser.c
Since 1e79548, we don't fallback from /sys/dev/block to
/proc/partitions anyway. The strncmp() at "skip_proc"
is doomed to fail. So we might as well return immediately.
Also, decrease the log level; whether this failure is
noteworthy is rather up to the callers to decide.
Signed-off-by: Martin
No need to log at -v3 if nothing to do.
Signed-off-by: Martin Wilck
---
libmpathpersist/mpath_persist.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/libmpathpersist/mpath_persist.c b/libmpathpersist/mpath_persist.c
index 2ffe56ea..0b9c7149 100644
---
checker.name doesn't exist any more.
Signed-off-by: Martin Wilck
---
tests/hwtable.c | 84 -
1 file changed, 42 insertions(+), 42 deletions(-)
diff --git a/tests/hwtable.c b/tests/hwtable.c
index 1cd788ac..789481ff 100644
--- a/tests/hwtable.c
This is an important event which we want to see with -v3.
Signed-off-by: Martin Wilck
---
libmultipath/structs_vec.c | 6 +++---
multipathd/main.c | 1 +
2 files changed, 4 insertions(+), 3 deletions(-)
diff --git a/libmultipath/structs_vec.c b/libmultipath/structs_vec.c
index
If internal libmultipath APIs change (such as, recently, the
checker API), test programs will fail because they'll link
with the standard system prioritizer / checker / foreign APIs.
Make sure we always link with our own shared libraries.
Signed-off-by: Martin Wilck
---
tests/Makefile | 7
Hi Christophe,
most of the patches in this series reduce log levels of frequently
printed messages at verbosity level 3. My goal was to limit the
output of multipathd to one line per path per checker invocation,
which is sufficient to track multipathd's view of path health in
the logs.
The
From: Huaisheng Ye
Reconstruct origin_dax_direct_access and expand functions
origin_dax_copy_to/from_iter for DAX operations of dm-snapshot.
Here is the call trace of origin_dax_copy_to_iter,
origin_dax_copy_to_iter will callin dm-linear (-real) for further
dax_operation.
[518597.924019] Call
From: Huaisheng Ye
Reconstruct origin_dax_direct_access and expand functions
origin_dax_copy_to/from_iter for DAX operations of dm-snapshot.
Here is the call trace of origin_dax_copy_to_iter,
origin_dax_copy_to_iter will callin dm-linear (-real) for further
dax_operation.
[518597.924019] Call
From: Huaisheng Ye
If use DAX way to mount the origin device of dm_snapshot, when
merging back snapshot to origin, system call table_load during the
construction of snapshot-merge will try to detect new dm_table's type
equals to the existed md's type or not.
The existed type equals to
From: Huaisheng Ye
Changes
v1->v2:
Add NULL funtions for origin_dax_direct_access and
origin_dax_copy_from/to_iter in order to avoid building
error when CONFIG_DAX_DRIVER has NOT been enabled.
[v1]: https://lkml.org/lkml/2018/11/20/759
This series patches are used to
On 21.11.18 г. 5:23 ч., Ming Lei wrote:
> Now multi-page bvec is supported, some helpers may return page by
> page, meantime some may return segment by segment, this patch
> documents the usage.
>
> Signed-off-by: Ming Lei
> ---
> Documentation/block/biovecs.txt | 24
On Thu, Nov 15, 2018 at 04:53:06PM +0800, Ming Lei wrote:
> QUEUE_FLAG_NO_SG_MERGE has been killed, so kill BLK_MQ_F_SG_MERGE too.
>
> Cc: Dave Chinner
> Cc: Kent Overstreet
> Cc: Mike Snitzer
> Cc: dm-devel@redhat.com
> Cc: Alexander Viro
> Cc: linux-fsde...@vger.kernel.org
> Cc: Shaohua Li
On Fri, Nov 16, 2018 at 02:59:22PM +0100, Christoph Hellwig wrote:
> On Thu, Nov 15, 2018 at 06:18:11PM -0800, Omar Sandoval wrote:
> > This commit message wasn't very clear. Is it the case that
> > QUEUE_FLAG_NO_SG_MERGE is no longer set by any drivers?
>
> I think he wants to say that not doing
On Fri, Nov 16, 2018 at 10:19:56AM +0100, Christoph Hellwig wrote:
> On Thu, Nov 15, 2018 at 02:18:47PM -0800, Omar Sandoval wrote:
> > My only reason to prefer unsigned int is consistency. unsigned int is
> > much more common in the kernel:
> >
> > $ ag --cc -s 'unsigned\s+int' | wc -l
> >
From: Huaisheng Ye
Sometimes dm_swap_table couldn't be used in time.
For example, during the origin construction of
dm-snapshot, the sys_ioctl table_load will try to
detect the lower device origin-real whether support
direct access or not. But origin-real md's struct
dm-table pointer has not
From: Huaisheng Ye
Sometimes dm_swap_table couldn't be used in time.
For example, during the origin construction of
dm-snapshot, the sys_ioctl table_load will try to
detect the lower device origin-real whether support
direct access or not. But origin-real md's struct
dm-table pointer has not
From: Huaisheng Ye
This series patches are used to realize the dax_operations for dm-snapshot
with persistent memory device.
Here are the steps about how to verify the function.
1. Configure the persistent memory to fs-dax mode and create namespace with
ndctl;
2. find them in /dev;
# ndctl
Dear Steffen,
I stumbled upon this old thread[1] about the possibility of getting
pcrypt work for dm-crypt.
About around the same time dm-crypt got parallel support[2] but as far
as I understand the patch only enabled parallelism if there are
multiple simultaneous requests, but still goes single
From: Huaisheng Ye
If use DAX way to mount the origin device of dm_snapshot, when
merging back snapshot to origin, system call table_load during the
construction of snapshot-merge will try to detect new dm_table's type
equals to the existed md's type or not.
The existed type equals to
Ming Lei writes:
> On Thu, Nov 15, 2018 at 05:59:36PM -0800, Omar Sandoval wrote:
>> On Thu, Nov 15, 2018 at 04:53:02PM +0800, Ming Lei wrote:
>> > Now multi-page bvec can cover CONFIG_THP_SWAP, so we don't need to
>> > increase BIO_MAX_PAGES for it.
>>
>> You mentioned to it in the cover
On Tue, 2018-11-20 at 22:24 -0600, Benjamin Marzinski wrote:
> Commit 1f962693 didn't deal with all of cases where a path that was
> part
> of a multipath device could be removed. verify_paths() removes any
> path
> that no longer exists in sysfs. mpp->hwe needs to be updated here as
> well,
On Tue, Nov 20, 2018 at 09:35:07PM -0800, Sagi Grimberg wrote:
>> Given it is over TCP, I guess it should be doable for you to preallocate one
>> 256-bvec table in one page for each request, then sets the max segment size
>> as
>> (unsigned int)-1, and max segment number as 256, the preallocated
58 matches
Mail list logo
