mbox series

[GIT,PULL] hotfixes for 6.1-rc2

Message ID 20221020213327.2c284af5925b6a8ed915fe9e@linux-foundation.org
State New
Headers
Series [GIT,PULL] hotfixes for 6.1-rc2 |

Pull-request

git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm tags/mm-hotfixes-stable-2022-10-20

Message

Andrew Morton Oct. 21, 2022, 4:33 a.m. UTC 
  Linus, please merge this batch of hotfixes.

The following changes since commit 9abf2313adc1ca1b6180c508c25f22f9395cc780:

  Linux 6.1-rc1 (2022-10-16 15:36:24 -0700)

are available in the Git repository at:

  git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm tags/mm-hotfixes-stable-2022-10-20

for you to fetch changes up to 97061d441110528dc02972818f2f1dad485107f9:

  nouveau: fix migrate_to_ram() for faulting page (2022-10-20 21:27:24 -0700)

----------------------------------------------------------------
17 hotfixes, mainly for MM.  5 are cc:stable and the remainder address
post-6.0 issues.

----------------------------------------------------------------
Alexey Romanov (1):
      zsmalloc: zs_destroy_pool: add size_class NULL check

Alistair Popple (1):
      nouveau: fix migrate_to_ram() for faulting page

Andrew Morton (3):
      Merge branch 'master' into mm-hotfixes-stable
      Merge branch 'master' into mm-hotfixes-stable
      mm/mmap.c: __vma_adjust(): suppress uninitialized var warning

Colin Ian King (1):
      init: Kconfig: fix spelling mistake "satify" -> "satisfy"

Dan Carpenter (1):
      mailmap: update Dan Carpenter's email address

Hugh Dickins (1):
      mm: /proc/pid/smaps_rollup: fix maple tree search

Joseph Qi (2):
      ocfs2: fix BUG when iput after ocfs2_mknod fails
      ocfs2: clear dinode links count in case of error

Liam Howlett (2):
      mm/mempolicy: fix mbind_range() arguments to vma_merge()
      mm/mmap: fix MAP_FIXED address return on VMA merge

Liam R. Howlett (1):
      mm/page_alloc: reduce potential fragmentation in make_alloc_exact()

Martin Liska (1):
      gcov: support GCC 12.1 and newer compilers

Mel Gorman (1):
      mm/huge_memory: do not clobber swp_entry_t during THP split

Mike Kravetz (2):
      mm/mmap: undo ->mmap() when mas_preallocate() fails
      hugetlb: fix memory leak associated with vma_lock structure

Qais Yousef (1):
      mailmap: update email for Qais Yousef

Rik van Riel (1):
      mm,hugetlb: take hugetlb_lock before decrementing h->resv_huge_pages

 .mailmap                               |  4 +++-
 drivers/gpu/drm/nouveau/nouveau_dmem.c |  1 +
 fs/ocfs2/namei.c                       | 23 ++++++++++-----------
 fs/proc/task_mmu.c                     |  2 +-
 init/Kconfig                           |  2 +-
 kernel/gcov/gcc_4_7.c                  | 18 +++++++++++++++--
 mm/huge_memory.c                       | 11 +++++++++-
 mm/hugetlb.c                           | 37 +++++++++++++++++++++++++---------
 mm/mempolicy.c                         | 17 ++++++++++------
 mm/mmap.c                              | 20 +++++++++---------
 mm/page_alloc.c                        | 20 ++++++++++--------
 mm/zsmalloc.c                          |  3 +++
 12 files changed, 107 insertions(+), 51 deletions(-)

Comments

Linus Torvalds Oct. 21, 2022, 7:40 p.m. UTC | #1 
On Thu, Oct 20, 2022 at 9:33 PM Andrew Morton <akpm@linux-foundation.org> wrote:
>
> Andrew Morton (3):
>       Merge branch 'master' into mm-hotfixes-stable
>       Merge branch 'master' into mm-hotfixes-stable

I've pulled this, but both of these merges look entirely invalid.

First off, one of them is most definitely pointless - there's no
activity between the first and the second merge, so that first merge
shouldn't have been done in the first place.

But the second merge is bad by definition too, in that it doesn't have
any explanation for it.

I've said this before - and I hope that some day I just won't have to
say it again - if you cannot explain a merge, you shouldn't be *doing*
a merge.

It really is that simple. If you have no clear explanation for why you
need to do that merge, then you simply shouldn't have done it.

For example, that first merge clearly CANNOT have a valid reason for
existing. There's simply no valid explanation for doing it,
considering that there was no work on top of it that could possibly be
a cause tor that merge.

So that whole "if you cannot explain why you do a merge, don't do the
merge" rule should have been a hard "NO" on it, and should have made
it obvious that it shouldn't have happened.

Please don't do pointless merges.

            Linus
pr-tracker-bot@kernel.org Oct. 21, 2022, 7:43 p.m. UTC | #2 
The pull request you sent on Thu, 20 Oct 2022 21:33:27 -0700:

> git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm tags/mm-hotfixes-stable-2022-10-20

has been merged into torvalds/linux.git:
https://git.kernel.org/torvalds/c/440b7895c990a63869a9d55e5c2502dd501a124e

Thank you!