Re: linux-next: Fixes tag needs some work in the block tree
From: Song Liu
Date: Tue Aug 15 2023 - 12:52:40 EST
On Wed, Aug 16, 2023 at 12:31 AM Jens Axboe <axboe@xxxxxxxxx> wrote:
>
> On 8/15/23 10:26 AM, Song Liu wrote:
> > On Wed, Aug 16, 2023 at 12:20?AM Jens Axboe <axboe@xxxxxxxxx> wrote:
> >>
> >> On 8/15/23 10:17 AM, Song Liu wrote:
> >>> On Wed, Aug 16, 2023 at 12:07?AM Jens Axboe <axboe@xxxxxxxxx> wrote:
> >>>>
> >>>> On 8/15/23 9:58 AM, Song Liu wrote:
> >>>>> On Tue, Aug 15, 2023 at 11:08?PM Jens Axboe <axboe@xxxxxxxxx> wrote:
> >>>>>>
> >>>>>> On 8/14/23 9:21 PM, Yu Kuai wrote:
> >>>>>>> Hi,
> >>>>>>>
> >>>>>>> ? 2023/08/15 7:19, Stephen Rothwell ??:
> >>>>>>>> Hi all,
> >>>>>>>>
> >>>>>>>> In commit
> >>>>>>>>
> >>>>>>>> 71070e7bd25d ("md/raid10: fix a 'conf->barrier' leakage in raid10_takeover()")
> >>>>>>>>
> >>>>>>>> Fixes tag
> >>>>>>>>
> >>>>>>>> Fixes: 4d27e927344a ("md: don't quiesce in mddev_suspend()")
> >>>>>>>>
> >>>>>>>> has these problem(s):
> >>>>>>>>
> >>>>>>>> - Target SHA1 does not exist
> >>>>>>>
> >>>>>>> Sorry about this, the commit was picked from previous md-next branch,
> >>>>>>> and I didn't know that it will change...
> >>>>>>
> >>>>>> It's most likely not your fault, I'm guessing Song rebased that branch
> >>>>>> and this is why the sha changed. Song?
> >>>>>
> >>>>> Yeah, it is my fault. I am really sorry for this issue.
> >>>>>
> >>>>>>
> >>>>>>> Is there any way that I can fix this ?
> >>>>>>
> >>>>>> Only way to fix this is if I rewind my branch and Song updates his
> >>>>>> branch to not have this issue, and then sends a new pull request.
> >>>>>
> >>>>> Yes please. Let's fix this.
> >>>>
> >>>> Please send a new one, I've dropped the old one.
> >>>
> >>> Is this the for-6.6/block branch? I haven't seen the change yet. The top
> >>> commit on the branch (as I see) is
> >>>
> >>> 306d74055754 ("ublk: Switch to memdup_user_nul() helper")
> >>
> >> Where are you looking? It's at:
> >
> > I am looking at
> >
> > https://git.kernel.org/pub/scm/linux/kernel/git/axboe/linux-block.git
> >
> > It appears to be some lag from git server?
>
> That's just a backup, I always pull and push from git.kernel.dk. The
> korg one will lag about an hour, depending on when things are pushed.
> I've never used the korg one as the canonical location.
Got it. I will update my git config.
Here is the updated PR:
The following changes since commit 66a6a5d0ec852eaced589da066376e69397cd71e:
ublk: Switch to memdup_user_nul() helper (2023-08-15 10:06:54 -0600)
are available in the Git repository at:
https://git.kernel.org/pub/scm/linux/kernel/git/song/md.git
tags/md-next-20230814-resend
for you to fetch changes up to 0d0bd28c500173bfca78aa840f8f36d261ef1765:
md/raid5-cache: fix null-ptr-deref for r5l_flush_stripe_to_raid()
(2023-08-15 09:40:27 -0700)
----------------------------------------------------------------
Li Lingfeng (1):
md: Hold mddev->reconfig_mutex when trying to get mddev->sync_thread
WANG Xuerui (5):
raid6: remove the <linux/export.h> include from recov.c
raid6: guard the tables.c include of <linux/export.h> with __KERNEL__
raid6: test: cosmetic cleanups for the test Makefile
raid6: test: make sure all intermediate and artifact files are .gitignored
raid6: test: only check for Altivec if building on powerpc hosts
Yu Kuai (3):
md/raid5-cache: fix a deadlock in r5l_exit_log()
md/raid10: fix a 'conf->barrier' leakage in raid10_takeover()
md/raid5-cache: fix null-ptr-deref for r5l_flush_stripe_to_raid()
Zhang Shurong (1):
md: raid1: fix potential OOB in raid1_remove_disk()
drivers/md/md-cluster.c | 8 ++++----
drivers/md/md.c | 9 +++++----
drivers/md/md.h | 2 +-
drivers/md/raid1.c | 8 ++++++--
drivers/md/raid10.c | 3 +--
drivers/md/raid5-cache.c | 14 ++++++++------
drivers/md/raid5.c | 2 +-
lib/raid6/mktables.c | 2 ++
lib/raid6/recov.c | 1 -
lib/raid6/test/.gitignore | 3 +++
lib/raid6/test/Makefile | 50
++++++++++++++++++++++++++------------------------
11 files changed, 57 insertions(+), 45 deletions(-)
create mode 100644 lib/raid6/test/.gitignore
Thanks,
Song