mbox series

[GIT,PULL] percpu changes for v6.2-rc1

Message ID Y5dhRrNFzNX1Z64/@fedora (mailing list archive)
State New
Headers show
Series [GIT,PULL] percpu changes for v6.2-rc1 | expand

Pull-request

git://git.kernel.org/pub/scm/linux/kernel/git/dennis/percpu.git for-6.2

Message

Dennis Zhou Dec. 12, 2022, 5:13 p.m. UTC
Hi Linus,

Baoquan was nice enough to run some clean ups for percpu. There will be
a minor merge conflict with Vlastimil's slab PR [1]. This is due to
adjusting the PERCPU_DYNAMIC_EARLY_SIZE up for proper early percpu
reservation size on arm64 with 64k pages.

[1] https://lore.kernel.org/all/20221122163634.3b21cf71@canb.auug.org.au/

Some new eventual work is potential BPF accounting of percpu.

Thanks,
Dennis

The following changes since commit f0c4d9fc9cc9462659728d168387191387e903cc:

  Linux 6.1-rc4 (2022-11-06 15:07:11 -0800)

are available in the Git repository at:

  git://git.kernel.org/pub/scm/linux/kernel/git/dennis/percpu.git for-6.2

for you to fetch changes up to d667c94962c1c81ef587ac91dc5c01a1cfe339c7:

  mm/percpu: remove unused PERCPU_DYNAMIC_EARLY_SLOTS (2022-11-07 22:59:28 -0800)

----------------------------------------------------------------
Baoquan He (7):
      mm/percpu: remove unused pcpu_map_extend_chunks
      mm/percpu: use list_first_entry_or_null in pcpu_reclaim_populated()
      mm/percpu: Update the code comment when creating new chunk
      mm/percpu: add comment to state the empty populated pages accounting
      mm/percpu: replace the goto with break
      mm/percpu.c: remove the lcm code since block size is fixed at page size
      mm/percpu: remove unused PERCPU_DYNAMIC_EARLY_SLOTS

 include/linux/percpu.h |  7 +++----
 mm/percpu.c            | 44 ++++++++++++++++++--------------------------
 2 files changed, 21 insertions(+), 30 deletions(-)

Comments

Linus Torvalds Dec. 13, 2022, 12:11 a.m. UTC | #1
On Mon, Dec 12, 2022 at 9:13 AM Dennis Zhou <dennis@kernel.org> wrote:
>
> Baoquan was nice enough to run some clean ups for percpu. There will be
> a minor merge conflict with Vlastimil's slab PR [1]. This is due to
> adjusting the PERCPU_DYNAMIC_EARLY_SIZE up for proper early percpu
> reservation size on arm64 with 64k pages.

Well, the merge was trivial, but...

>   git://git.kernel.org/pub/scm/linux/kernel/git/dennis/percpu.git for-6.2

Argh, this is a plain branch. I was really hoping that we had all
moved on to using signed tags for pulls.

I've pulled this, because I still don't absolutely _require_ signed
tags for kernel.org accounts, but can you please try to start using
them in the future?

The plain unsigned branches (and unsigned tags) are getting rare
enough that I'm slowly starting to consider making them a hard
requirement. Not this release, probably not the next one, but...

                Linus
pr-tracker-bot@kernel.org Dec. 13, 2022, 12:36 a.m. UTC | #2
The pull request you sent on Mon, 12 Dec 2022 09:13:42 -0800:

> git://git.kernel.org/pub/scm/linux/kernel/git/dennis/percpu.git for-6.2

has been merged into torvalds/linux.git:
https://git.kernel.org/torvalds/c/ca1443c7e75a28c6fde5c67cb1904b624cf43c36

Thank you!
Dennis Zhou Dec. 13, 2022, 3:41 p.m. UTC | #3
On Mon, Dec 12, 2022 at 04:11:28PM -0800, Linus Torvalds wrote:
> On Mon, Dec 12, 2022 at 9:13 AM Dennis Zhou <dennis@kernel.org> wrote:
> >
> > Baoquan was nice enough to run some clean ups for percpu. There will be
> > a minor merge conflict with Vlastimil's slab PR [1]. This is due to
> > adjusting the PERCPU_DYNAMIC_EARLY_SIZE up for proper early percpu
> > reservation size on arm64 with 64k pages.
> 
> Well, the merge was trivial, but...
> 
> >   git://git.kernel.org/pub/scm/linux/kernel/git/dennis/percpu.git for-6.2
> 
> Argh, this is a plain branch. I was really hoping that we had all
> moved on to using signed tags for pulls.
> 
> I've pulled this, because I still don't absolutely _require_ signed
> tags for kernel.org accounts, but can you please try to start using
> them in the future?
> 
> The plain unsigned branches (and unsigned tags) are getting rare
> enough that I'm slowly starting to consider making them a hard
> requirement. Not this release, probably not the next one, but...
> 

Got it. I'll send you signed tags from now on. Thanks for pulling this.

Thanks,
Dennis

>                 Linus