On Mon, Apr 11, 2022 at 11:55:14AM +0900, Damien Le Moal wrote:
> On 4/11/22 11:19, Damien Le Moal wrote:
> > On 4/11/22 10:04, Ming Lei wrote:
> >> On Mon, Apr 11, 2022 at 09:50:57AM +0900, Damien Le Moal wrote:
> >>> On 4/11/22 09:36, Ming Lei wrote:
> On Mon, Apr 11, 2022 at 09:18:56AM +090
Dne 11. 04. 22 v 0:03 Demi Marie Obenour napsal(a):
For quite a while, I have wanted to write a tool to manage thin volumes
that is not based on LVM. The main thing holding me back is that the
current dm-thin interface is extremely error-prone. The only per-thin
metadata stored by the kernel is
On Mon, Apr 11, 2022 at 06:38:38PM +0900, Damien Le Moal wrote:
> Commit 0fbb4d93b38b ("dm: add dm_submit_bio_remap interface") changed
> the alloc_io() function to delay the initialization of struct dm_io
> orig_bio field, leaving this field as NULL until the first call to
> __split_and_process_bi
On Mon, Apr 11, 2022 at 04:42:59PM +0900, Damien Le Moal wrote:
> On 4/11/22 16:34, Ming Lei wrote:
> > On Mon, Apr 11, 2022 at 11:55:14AM +0900, Damien Le Moal wrote:
> >> On 4/11/22 11:19, Damien Le Moal wrote:
> >>> On 4/11/22 10:04, Ming Lei wrote:
> On Mon, Apr 11, 2022 at 09:50:57AM +090
On Sat, Apr 09 2022 at 1:15P -0400,
Christoph Hellwig wrote:
> On Fri, Apr 08, 2022 at 11:42:51AM -0400, Mike Snitzer wrote:
> > I think we can achieve the goal of efficient cloning/remapping for
> > both usecases simply by splitting out the bio_set_dev() and leaving it
> > to the caller to pick
On Mon, Apr 11 2022 at 12:58P -0400,
Mike Snitzer wrote:
> On Sat, Apr 09 2022 at 1:15P -0400,
> Christoph Hellwig wrote:
>
> > On Fri, Apr 08, 2022 at 11:42:51AM -0400, Mike Snitzer wrote:
> > > I think we can achieve the goal of efficient cloning/remapping for
> > > both usecases simply by s
When bio_{alloc,init}_clone are passed a bdev, bio_init() will call
bio_associate_blkg() so the __bio_clone() work to initialize blkcg
isn't needed.
Signed-off-by: Mike Snitzer
---
block/bio.c | 3 ---
1 file changed, 3 deletions(-)
diff --git a/block/bio.c b/block/bio.c
index 7892f1108ca6..698
On 04/06, Christoph Hellwig wrote:
> Secure erase is a very different operation from discard in that it is
> a data integrity operation vs hint. Fully split the limits and helper
> infrastructure to make the separation more clear.
>
> Signed-off-by: Christoph Hellwig
> ---
> block/blk-core.c
Dne 11. 04. 22 v 19:22 Demi Marie Obenour napsal(a):
On Mon, Apr 11, 2022 at 10:16:02AM +0200, Zdenek Kabelac wrote:
Dne 11. 04. 22 v 0:03 Demi Marie Obenour napsal(a):
Your proposal actually breaks this sequence and would move things to the
state of 'guess at which states we are now'. (and IM
On Tue, Apr 5, 2022 at 12:48 PM Jane Chu wrote:
>
> There is no _set_memory_prot internal helper, while coming across
> the code, might as well fix the comment.
Looks good,
Reviewed-by: Dan Williams
>
> Reviewed-by: Christoph Hellwig
> Signed-off-by: Jane Chu
> ---
> arch/x86/mm/pat/set_mem
I notice that none of the folks from "X86 MM" are on the cc, added.
On Tue, Apr 5, 2022 at 12:49 PM Jane Chu wrote:
>
> Relocate the twin mce functions to arch/x86/mm/pat/set_memory.c
> file where they belong.
>
> Signed-off-by: Jane Chu
> ---
> arch/x86/include/asm/set_memory.h | 52 --
On Tue, Apr 5, 2022 at 12:48 PM Jane Chu wrote:
>
> The set_memory_uc() approach doesn't work well in all cases.
> For example, when "The VMM unmapped the bad page from guest
> physical space and passed the machine check to the guest."
> "The guest gets virtual #MC on an access to that page.
> Wh
On Wed, Apr 6, 2022 at 10:31 PM Christoph Hellwig wrote:
>
> On Wed, Apr 06, 2022 at 05:32:31PM +, Jane Chu wrote:
> > Yes, I believe Dan was motivated by avoiding the dm dance as a result of
> > adding .recovery_write to dax_operations.
> >
> > I understand your point about .recovery_write is
Khazhismel Kumykov writes:
> mixing sched_clock() and ktime_get_ns() will give bad results, don't do
> it
>
> Fixes: 2613eab11996 ("dm mpath: add Historical Service Time Path Selector")
> Signed-off-by: Khazhismel Kumykov
Looks good.
Reviewed-by: Gabriel Krisman Bertazi
> ---
> drivers/md/
On Tue, Apr 5, 2022 at 12:48 PM Jane Chu wrote:
>
> Introduce DAX_RECOVERY flag to dax_direct_access(). The flag is
> not set by default in dax_direct_access() such that the helper
> does not translate a pmem range to kernel virtual address if the
> range contains uncorrectable errors. When the f
On Tue, Apr 12, 2022 at 08:33:04AM +0900, Damien Le Moal wrote:
> On 4/11/22 23:18, Ming Lei wrote:
> This fixes the issue:
>
> diff --git a/drivers/md/dm.c b/drivers/md/dm.c
> index 3c5fad7c4ee6..3dd6735450c5 100644
> --- a/drivers/md/dm.c
> +++ b/drivers/md/dm.c
> >>
On Tue, Apr 12, 2022 at 09:28:46AM +0900, Damien Le Moal wrote:
> On 4/12/22 09:09, Ming Lei wrote:
> > On Tue, Apr 12, 2022 at 08:33:04AM +0900, Damien Le Moal wrote:
> >> On 4/11/22 23:18, Ming Lei wrote:
> >> This fixes the issue:
> >>
> >> diff --git a/drivers/md/dm.c b/drivers/md/d
Some storage arrays can be accessed using multiple protocols at the same
time. I've have customers request the ability to set different values
for the path specific timeouts, like fast_io_fail_tmo, based on the
protocol used to access the path. In order to make this possible, this
patchset adds a
Instead of dev_loss, fast_io_fail, and eh_deadline belonging to the
multipath structure, have them belong to the path structure. This means
that they are selected per path, and that sysfs_set_scsi_tmo() doesn't
assume that all paths of a multipath device will have the same value.
Currently they wil
Instead of allocating a copy when the original string is going to be
freed right after the merge, just steal the pointer. Also, merge_mpe()
can't get called with NULL arguments.
Signed-off-by: Benjamin Marzinski
---
libmultipath/config.c | 16 +---
1 file changed, 5 insertions(+), 11
For the config values that exist in the proctol subsection of the device
configs, paths will now also check the pctable when checking a hwentry
for a value. Values in a matching pcentry will be used in preference to
values in that hwentry.
Signed-off-by: Benjamin Marzinski
---
libmultipath/props
Instead of always indenting two tabs, which is what is needed in for
the protocol subsection of the device subsection, indent a variable
amount. This will be needed in a future patch.
Signed-off-by: Benjamin Marzinski
---
libmultipath/print.c | 28 +---
1 file changed, 21
Some storage arrays can be accessed using multiple protocols at the same
time. In these cases, users may want to set path attributes
differently, depending on the protocol that the path is using. To allow
this, add a protocol subsection to the device subsection in
multipath.conf, which allows sele
Future patches will make use of this, so move it out of
snprint_path_protocol()
Signed-off-by: Benjamin Marzinski
---
libmultipath/print.c | 17 -
libmultipath/structs.c | 18 ++
libmultipath/structs.h | 1 +
3 files changed, 19 insertions(+), 17 deletions(-)
Signed-off-by: Benjamin Marzinski
---
multipath/multipath.conf.5 | 4
1 file changed, 4 insertions(+)
diff --git a/multipath/multipath.conf.5 b/multipath/multipath.conf.5
index 605b46e0..97695da4 100644
--- a/multipath/multipath.conf.5
+++ b/multipath/multipath.conf.5
@@ -1636,6 +1636,8 @@
Add a procotol subsection to the overrides section, just like the one in
the device subsection. This allows users to a protocol specific
parameters to all device configurations.
Signed-off-by: Benjamin Marzinski
---
libmultipath/dict.c| 76 --
libmulti
Signed-off-by: Benjamin Marzinski
---
multipath/multipath.conf.5 | 38 ++
1 file changed, 38 insertions(+)
diff --git a/multipath/multipath.conf.5 b/multipath/multipath.conf.5
index 97695da4..dab1d58f 100644
--- a/multipath/multipath.conf.5
+++ b/multipath/mul
On Tue, Apr 5, 2022 at 12:48 PM Jane Chu wrote:
>
> Refactor the pmem_clear_poison() in order to share common code
> later.
>
I would just add a note here about why, i.e. to factor out the common
shared code between the typical write path and the recovery write
path.
> Signed-off-by: Jane Chu
>
On Tue, Apr 5, 2022 at 12:48 PM Jane Chu wrote:
>
> Introduce DAX_RECOVERY flag to dax_direct_access(). The flag is
> not set by default in dax_direct_access() such that the helper
> does not translate a pmem range to kernel virtual address if the
> range contains uncorrectable errors. When the f
On Mon, Apr 11, 2022 at 09:57:36PM -0700, Dan Williams wrote:
> So how about change 'int flags' to 'enum dax_access_mode mode' where
> dax_access_mode is:
>
> /**
> * enum dax_access_mode - operational mode for dax_direct_access()
> * @DAX_ACCESS: nominal access, fail / trim access on encounteri
On Mon, Apr 11, 2022 at 01:33:58PM -0400, Mike Snitzer wrote:
> When bio_{alloc,init}_clone are passed a bdev, bio_init() will call
> bio_associate_blkg() so the __bio_clone() work to initialize blkcg
> isn't needed.
No, unfortunately it isn't as simple as that. There are bios that do
not use the
31 matches
Mail list logo
