Skip to contentSkip to conversation

md-fixes 20201212

GIT PULL In review Song Liu 3 participants 5 messages Dec 12, 2020

Subsystems (2)
lore
Share & tools
lore

Song Liu

author
#0 lore raw

Hi Jens,

Please consider pulling the following changes on top of tag block-5.10-2020-12-05. This is to fix raid10 data corruption [1] in 5.10-rc7.

Tests run on this change:

1. md raid10: tested discard on raid10 device works properly (zero mismatch_cnt).
2. dm raid10: tested discard_granularity and discard_max_bytes was set properly.

Thanks,
Song

[1] https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1907262/

The following changes since commit 7e7986f9d3ba69a7375a41080a1f8c8012cb0923:

block: use gcd() to fix chunk_sectors limit stacking (2020-12-01 11:02:55 -0700)

are available in the Git repository at:

ssh://[email protected]/pub/scm/linux/kernel/git/song/md.git md-fixes

for you to fetch changes up to 0d5c7b890229f8a9bb4b588b34ffe70c62691143:

Revert "md: add md_submit_discard_bio() for submitting discard bio" (2020-12-12 00:51:41 -0800)

----------------------------------------------------------------
Song Liu (7):
Revert "dm raid: remove unnecessary discard limits for raid10" Revert "dm raid: fix discard limits for raid1 and raid10" Revert "md/raid10: improve discard request for far layout" Revert "md/raid10: improve raid10 discard request"
Revert "md/raid10: pull codes that wait for blocked dev into one function" Revert "md/raid10: extend r10bio devs to raid disks"
Revert "md: add md_submit_discard_bio() for submitting discard bio"

 drivers/md/dm-raid.c |   9 +++++
 drivers/md/md.c      |  20 ----------
 drivers/md/md.h      |   2 -
 drivers/md/raid0.c   |  14 ++++++-
 drivers/md/raid10.c  | 423 +++++++++++++++++++++++++++++------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
 drivers/md/raid10.h  |   1 -
 6 files changed, 78 insertions(+), 391 deletions(-)
#1 lore raw
Quoted 39 lines · Song Liu

Hi Jens,

Please consider pulling the following changes on top of tag block-5.10-2020-12-05. This is to fix raid10 data corruption [1] in 5.10-rc7.

Tests run on this change:

1. md raid10: tested discard on raid10 device works properly (zero mismatch_cnt).
2. dm raid10: tested discard_granularity and discard_max_bytes was set properly.

Thanks,
Song

[1] https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1907262/

The following changes since commit 7e7986f9d3ba69a7375a41080a1f8c8012cb0923:

block: use gcd() to fix chunk_sectors limit stacking (2020-12-01 11:02:55 -0700)

are available in the Git repository at:

ssh://[email protected]/pub/scm/linux/kernel/git/song/md.git md-fixes

for you to fetch changes up to 0d5c7b890229f8a9bb4b588b34ffe70c62691143:

Revert "md: add md_submit_discard_bio() for submitting discard bio" (2020-12-12 00:51:41 -0800)

----------------------------------------------------------------
Song Liu (7):
Revert "dm raid: remove unnecessary discard limits for raid10" Revert "dm raid: fix discard limits for raid1 and raid10" Revert "md/raid10: improve discard request for far layout" Revert "md/raid10: improve raid10 discard request"
Revert "md/raid10: pull codes that wait for blocked dev into one function" Revert "md/raid10: extend r10bio devs to raid disks"
Revert "md: add md_submit_discard_bio() for submitting discard bio"

Quoted 8 lines
 drivers/md/dm-raid.c |   9 +++++
 drivers/md/md.c      |  20 ----------
 drivers/md/md.h      |   2 -
 drivers/md/raid0.c   |  14 ++++++-
 drivers/md/raid10.c  | 423 +++++++++++++++++++++++++++++------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
 drivers/md/raid10.h  |   1 -
 6 files changed, 78 insertions(+), 391 deletions(-)

Jens already pulled 95% of these changes no? why are you sending a pull that ignores that fact?

And as I stated here:
https://www.redhat.com/archives/dm-devel/2020-December/msg00253.html

I don't agree with reverting commit e0910c8e4f87bb9 ("dm raid: fix discard limits for raid1 and raid10"). Espeiclaly not like you've done, given it was marked for stable any follow-on fix/revert needs to be as well.

Simply changing 'struct mddev' chunk_sectors members from int to 'unsigned int' is the better way forward I think.

Mike

--
dm-devel mailing list
[email protected]
https://www.redhat.com/mailman/listinfo/dm-devel

#2 lore raw
Quoted 57 lines · Mike Snitzer
Quoted 39 lines · Song Liu

Hi Jens,

Please consider pulling the following changes on top of tag block-5.10-2020-12-05. This is to fix raid10 data corruption [1] in 5.10-rc7.

Tests run on this change:

1. md raid10: tested discard on raid10 device works properly (zero mismatch_cnt).
2. dm raid10: tested discard_granularity and discard_max_bytes was set properly.

Thanks,
Song

[1] https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1907262/

The following changes since commit 7e7986f9d3ba69a7375a41080a1f8c8012cb0923:

block: use gcd() to fix chunk_sectors limit stacking (2020-12-01 11:02:55 -0700)

are available in the Git repository at:

ssh://[email protected]/pub/scm/linux/kernel/git/song/md.git md-fixes

for you to fetch changes up to 0d5c7b890229f8a9bb4b588b34ffe70c62691143:

Revert "md: add md_submit_discard_bio() for submitting discard bio" (2020-12-12 00:51:41 -0800)

----------------------------------------------------------------
Song Liu (7):
Revert "dm raid: remove unnecessary discard limits for raid10" Revert "dm raid: fix discard limits for raid1 and raid10" Revert "md/raid10: improve discard request for far layout" Revert "md/raid10: improve raid10 discard request"
Revert "md/raid10: pull codes that wait for blocked dev into one function" Revert "md/raid10: extend r10bio devs to raid disks"
Revert "md: add md_submit_discard_bio() for submitting discard bio"

Quoted 8 lines
 drivers/md/dm-raid.c |   9 +++++
 drivers/md/md.c      |  20 ----------
 drivers/md/md.h      |   2 -
 drivers/md/raid0.c   |  14 ++++++-
 drivers/md/raid10.c  | 423 +++++++++++++++++++++++++++++------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
 drivers/md/raid10.h  |   1 -
 6 files changed, 78 insertions(+), 391 deletions(-)

Jens already pulled 95% of these changes no? why are you sending a pull that ignores that fact?

And as I stated here:
https://www.redhat.com/archives/dm-devel/2020-December/msg00253.html

Quoted 7 lines

I don't agree with reverting commit e0910c8e4f87bb9 ("dm raid: fix discard limits for raid1 and raid10"). Espeiclaly not like you've done, given it was marked for stable any follow-on fix/revert needs to be as well.

Simply changing 'struct mddev' chunk_sectors members from int to 'unsigned int' is the better way forward I think.

I agree, that'd be simpler. Mike, care to send a real patch for that against block-5.10?

Signature · 6 lines
Jens Axboe

--
dm-devel mailing list
[email protected]
https://www.redhat.com/mailman/listinfo/dm-devel
#3 lore raw

PATCH 1/1 md: change mddev 'chunk_sectors' from int to unsigned

Commit e2782f560c29 ("Revert "dm raid: remove unnecessary discard limits for raid10"") exposed compiler warnings introduced by commit e0910c8e4f87 ("dm raid: fix discard limits for raid1 and raid10"):

In file included from ./include/linux/kernel.h:14,

                 from ./include/asm-generic/bug.h:20,
                 from ./arch/x86/include/asm/bug.h:93,
                 from ./include/linux/bug.h:5,
                 from ./include/linux/mmdebug.h:5,
                 from ./include/linux/gfp.h:5,
                 from ./include/linux/slab.h:15,
                 from drivers/md/dm-raid.c:8:
drivers/md/dm-raid.c: In function ‘raid_io_hints’:
./include/linux/minmax.h:18:28: warning: comparison of distinct pointer types lacks a cast
  (!!(sizeof((typeof(x) *)1 == (typeof(y) *)1)))
                            ^~
./include/linux/minmax.h:32:4: note: in expansion of macro ‘__typecheck’
   (__typecheck(x, y) && __no_side_effects(x, y))
    ^~~~~~~~~~~
./include/linux/minmax.h:42:24: note: in expansion of macro ‘__safe_cmp’
  __builtin_choose_expr(__safe_cmp(x, y), \
                        ^~~~~~~~~~
./include/linux/minmax.h:51:19: note: in expansion of macro ‘__careful_cmp’
 #define min(x, y) __careful_cmp(x, y, <)
                   ^~~~~~~~~~~~~
./include/linux/minmax.h:84:39: note: in expansion of macro ‘min’
  __x == 0 ? __y : ((__y == 0) ? __x : min(__x, __y)); })
                                       ^~~
drivers/md/dm-raid.c:3739:33: note: in expansion of macro ‘min_not_zero’
   limits->max_discard_sectors = min_not_zero(rs->md.chunk_sectors,
                                 ^~~~~~~~~~~~

Fix this by changing the chunk_sectors member of 'struct mddev' from int to 'unsigned int' to match the type used for the 'chunk_sectors' member of 'struct queue_limits'. Various MD code still uses 'int' but none of it appears to ever make use of signed int; and storing positive signed int in unsigned is perfectly safe.

Fixes e2782f560c29 ("Revert "dm raid: remove unnecessary discard limits for raid10"")
Fixes e0910c8e4f87 ("dm raid: fix discard limits for raid1 and raid10")
Cc stable@vger,kernel.org # e0910c8e4f87 was marked for stable@
---
 drivers/md/md.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/md/md.h +2 -2
index 2175a5ac4f7c..bb645bc3ba6d 100644
@@ -311,7 +311,7 @@ struct mddev {
 	int				external;	/* metadata is
 							 * managed externally */
 	char				metadata_type[17]; /* externally set*/
-	int				chunk_sectors;
+	unsigned int			chunk_sectors;
 	time64_t			ctime, utime;
 	int				level, layout;
 	char				clevel[16];
@@ -339,7 +339,7 @@ struct mddev {
 	 */
 	sector_t			reshape_position;
 	int				delta_disks, new_level, new_layout;
-	int				new_chunk_sectors;
+	unsigned int			new_chunk_sectors;
 	int				reshape_backwards;
 
 	struct md_thread		*thread;	/* management thread */
Signature · 6 lines
2.15.0

--
dm-devel mailing list
[email protected]
https://www.redhat.com/mailman/listinfo/dm-devel
author Reviewed-by on v1 1/1
#4 lore raw

Re: [dm-devel] [PATCH] md: change mddev 'chunk_sectors' from int to unsigned

Quoted 36 lines · Mike Snitzer

Commit e2782f560c29 ("Revert "dm raid: remove unnecessary discard limits for raid10"") exposed compiler warnings introduced by commit e0910c8e4f87 ("dm raid: fix discard limits for raid1 and raid10"):

In file included from ./include/linux/kernel.h:14,

                from ./include/asm-generic/bug.h:20,
                from ./arch/x86/include/asm/bug.h:93,
                from ./include/linux/bug.h:5,
                from ./include/linux/mmdebug.h:5,
                from ./include/linux/gfp.h:5,
                from ./include/linux/slab.h:15,
                from drivers/md/dm-raid.c:8:
drivers/md/dm-raid.c: In function ‘raid_io_hints’:
./include/linux/minmax.h:18:28: warning: comparison of distinct pointer types lacks a cast
 (!!(sizeof((typeof(x) *)1 == (typeof(y) *)1)))
                           ^~

./include/linux/minmax.h:32:4: note: in expansion of macro ‘__typecheck’
(__typecheck(x, y) && __no_side_effects(x, y))
^~~~~~~~~~~
./include/linux/minmax.h:42:24: note: in expansion of macro ‘__safe_cmp’

 __builtin_choose_expr(__safe_cmp(x, y), \
                       ^~~~~~~~~~
./include/linux/minmax.h:51:19: note: in expansion of macro ‘__careful_cmp’
#define min(x, y) __careful_cmp(x, y, <)
                  ^~~~~~~~~~~~~
./include/linux/minmax.h:84:39: note: in expansion of macro ‘min’
 __x == 0 ? __y : ((__y == 0) ? __x : min(__x, __y)); })
                                      ^~~
drivers/md/dm-raid.c:3739:33: note: in expansion of macro ‘min_not_zero’
  limits->max_discard_sectors = min_not_zero(rs->md.chunk_sectors,
                                ^~~~~~~~~~~~
Quoted 5 lines

Fix this by changing the chunk_sectors member of 'struct mddev' from int to 'unsigned int' to match the type used for the 'chunk_sectors' member of 'struct queue_limits'. Various MD code still uses 'int' but none of it appears to ever make use of signed int; and storing positive signed int in unsigned is perfectly safe.

Thanks for the quick fix and thorough analysis. I also checked MD code and didn't see any use of negative chunk_sectors, so this change is safe. I will convert the rest use of signed chunk_sectors in 5.11.

Quoted 6 lines
Fixes e2782f560c29 ("Revert "dm raid: remove unnecessary discard limits for raid10"")
Fixes e0910c8e4f87 ("dm raid: fix discard limits for raid1 and raid10")
Cc stable@vger,kernel.org # e0910c8e4f87 was marked for stable@
Quoted 29 lines

---
drivers/md/md.h | 4 ++--

1 file changed, 2 insertions(+), 2 deletions(-)
drivers/md/md.h +2 -2
index 2175a5ac4f7c..bb645bc3ba6d 100644
@@ -311,7 +311,7 @@ struct mddev {
 	int				external;	/* metadata is
 							 * managed externally */
 	char				metadata_type[17]; /* externally set*/
-	int				chunk_sectors;
+	unsigned int			chunk_sectors;
 	time64_t			ctime, utime;
 	int				level, layout;
 	char				clevel[16];
@@ -339,7 +339,7 @@ struct mddev {
 	 */
 	sector_t			reshape_position;
 	int				delta_disks, new_level, new_layout;
-	int				new_chunk_sectors;
+	unsigned int			new_chunk_sectors;
 	int				reshape_backwards;
 
 	struct md_thread		*thread;	/* management thread */
Signature · 1 line
2.15.0

--
dm-devel mailing list
[email protected]
https://www.redhat.com/mailman/listinfo/dm-devel