[v2,0/3] fs/ecryptfs: Replace kmap{,_atomic}() with kmap_local_page()

Message ID 20230426172223.8896-1-fmdefrancesco@gmail.com
Headers
Series fs/ecryptfs: Replace kmap{,_atomic}() with kmap_local_page() |

Message

Fabio M. De Francesco April 26, 2023, 5:22 p.m. UTC
  kmap() and kmap_atomic() have been deprecated in favor of
kmap_local_page().

Therefore, replace kmap() and kmap_atomic() with kmap_local_page().

Tested in a QEMU/KVM x86_32 VM, 6GB RAM, booting a kernel with
HIGHMEM64GB enabled.

v1->v2: Patches 1/3 and 2/3 were not grouped in a series. Patch 3/3 is
new. Therefore, currently one series groups all conversions needed if
fs/ecryptfs. The subject of 2/3 changed to differentiate between it and
3/3. Furthermore, the commit messages of 1/3 and 2/3 have been largely
reworked.

Fabio M. De Francesco (3):
  fs/ecryptfs: Replace kmap() with kmap_local_page()
  fs/ecryptfs: Use kmap_local_page() in ecryptfs_write()
  fs/ecryptfs: Use kmap_local_page() in copy_up_encrypted_with_header()

 fs/ecryptfs/crypto.c     |  8 ++++----
 fs/ecryptfs/mmap.c       |  4 ++--
 fs/ecryptfs/read_write.c | 12 ++++++------
 3 files changed, 12 insertions(+), 12 deletions(-)
  

Comments

Fabio M. De Francesco June 28, 2023, 1:50 p.m. UTC | #1
On mercoledì 26 aprile 2023 19:22:20 CEST Fabio M. De Francesco wrote:
> kmap() and kmap_atomic() have been deprecated in favor of
> kmap_local_page().
> 
> Therefore, replace kmap() and kmap_atomic() with kmap_local_page().

After two months from submission, I haven't received any comments on this 
short series yet, except for a "Reviewed by" tag from Ira on patch 1/3 only.

I would appreciate any comments/reviews/acks and would especially like to know 
if anything is preventing these patches from being applied.

Thank you all in advance,

Fabio

> Tested in a QEMU/KVM x86_32 VM, 6GB RAM, booting a kernel with
> HIGHMEM64GB enabled.
> 
> v1->v2: Patches 1/3 and 2/3 were not grouped in a series. Patch 3/3 is
> new. Therefore, currently one series groups all conversions needed if
> fs/ecryptfs. The subject of 2/3 changed to differentiate between it and
> 3/3. Furthermore, the commit messages of 1/3 and 2/3 have been largely
> reworked.
> 
> Fabio M. De Francesco (3):
>   fs/ecryptfs: Replace kmap() with kmap_local_page()
>   fs/ecryptfs: Use kmap_local_page() in ecryptfs_write()
>   fs/ecryptfs: Use kmap_local_page() in copy_up_encrypted_with_header()
> 
>  fs/ecryptfs/crypto.c     |  8 ++++----
>  fs/ecryptfs/mmap.c       |  4 ++--
>  fs/ecryptfs/read_write.c | 12 ++++++------
>  3 files changed, 12 insertions(+), 12 deletions(-)
> 
> --
> 2.40.0
  
Eric Biggers June 30, 2023, 2:31 a.m. UTC | #2
On Wed, Jun 28, 2023 at 03:50:30PM +0200, Fabio M. De Francesco wrote:
> On mercoledì 26 aprile 2023 19:22:20 CEST Fabio M. De Francesco wrote:
> > kmap() and kmap_atomic() have been deprecated in favor of
> > kmap_local_page().
> > 
> > Therefore, replace kmap() and kmap_atomic() with kmap_local_page().
> 
> After two months from submission, I haven't received any comments on this 
> short series yet, except for a "Reviewed by" tag from Ira on patch 1/3 only.
> 
> I would appreciate any comments/reviews/acks and would especially like to know 
> if anything is preventing these patches from being applied.
> 

eCryptfs is in "Odd Fixes" status.  See the thread
https://lore.kernel.org/ecryptfs/ZB4nYykRg6UwZ0cj@sequoia/T/#u

I would suggest that if Tyler is not responding, that Christian or Al take these
patches through the VFS tree instead.

FWIW, I took a quick look at these three patches, and all look correct.  I'm not
sure I want to give a formal R-b, as I don't want people to start bothering me
about eCryptfs stuff because they saw my name on it :-)

- Eric
  
Christian Brauner June 30, 2023, 7:38 a.m. UTC | #3
On Thu, Jun 29, 2023 at 07:31:07PM -0700, Eric Biggers wrote:
> On Wed, Jun 28, 2023 at 03:50:30PM +0200, Fabio M. De Francesco wrote:
> > On mercoledì 26 aprile 2023 19:22:20 CEST Fabio M. De Francesco wrote:
> > > kmap() and kmap_atomic() have been deprecated in favor of
> > > kmap_local_page().
> > > 
> > > Therefore, replace kmap() and kmap_atomic() with kmap_local_page().
> > 
> > After two months from submission, I haven't received any comments on this 
> > short series yet, except for a "Reviewed by" tag from Ira on patch 1/3 only.
> > 
> > I would appreciate any comments/reviews/acks and would especially like to know 
> > if anything is preventing these patches from being applied.
> > 
> 
> eCryptfs is in "Odd Fixes" status.  See the thread
> https://lore.kernel.org/ecryptfs/ZB4nYykRg6UwZ0cj@sequoia/T/#u
> 
> I would suggest that if Tyler is not responding, that Christian or Al take these
> patches through the VFS tree instead.
> 
> FWIW, I took a quick look at these three patches, and all look correct.  I'm not
> sure I want to give a formal R-b, as I don't want people to start bothering me
> about eCryptfs stuff because they saw my name on it :-)

Ah well, you saw right through that. :)
That usually means you've been doing kernel development for way too long...
  
Christian Brauner June 30, 2023, 8:45 a.m. UTC | #4
On Wed, 26 Apr 2023 19:22:20 +0200, Fabio M. De Francesco wrote:
> kmap() and kmap_atomic() have been deprecated in favor of
> kmap_local_page().
> 
> Therefore, replace kmap() and kmap_atomic() with kmap_local_page().
> 
> Tested in a QEMU/KVM x86_32 VM, 6GB RAM, booting a kernel with
> HIGHMEM64GB enabled.
> 
> [...]

Picking this up. Please tell me if this should be routed somewhere else.
vfs.misc will be rebased once v6.5-rc1 is released.

---

Applied to the vfs.misc branch of the vfs/vfs.git tree.
Patches in the vfs.misc branch should appear in linux-next soon.

Please report any outstanding bugs that were missed during review in a
new review to the original patch series allowing us to drop it.

It's encouraged to provide Acked-bys and Reviewed-bys even though the
patch has now been applied. If possible patch trailers will be updated.

Note that commit hashes shown below are subject to change due to rebase,
trailer updates or similar. If in doubt, please check the listed branch.

tree:   https://git.kernel.org/pub/scm/linux/kernel/git/vfs/vfs.git
branch: vfs.misc

[1/3] fs/ecryptfs: Replace kmap() with kmap_local_page()
      https://git.kernel.org/vfs/vfs/c/7a367455b6a5
[2/3] fs/ecryptfs: Use kmap_local_page() in ecryptfs_write()
      https://git.kernel.org/vfs/vfs/c/55f13011af9d
[3/3] fs/ecryptfs: Use kmap_local_page() in copy_up_encrypted_with_header()
      https://git.kernel.org/vfs/vfs/c/de9f5a15080f
  
Fabio M. De Francesco July 1, 2023, 10:46 a.m. UTC | #5
On venerdì 30 giugno 2023 10:45:17 CEST Christian Brauner wrote:
> On Wed, 26 Apr 2023 19:22:20 +0200, Fabio M. De Francesco wrote:
> > kmap() and kmap_atomic() have been deprecated in favor of
> > kmap_local_page().
> > 
> > Therefore, replace kmap() and kmap_atomic() with kmap_local_page().
> > 
> > Tested in a QEMU/KVM x86_32 VM, 6GB RAM, booting a kernel with
> > HIGHMEM64GB enabled.
> > 
> > [...]
> 
> Picking this up.

Well, you anticipated me before I could ask you to take these patches through 
the VFS tree, as suggested by Eric.

> Please tell me if this should be routed somewhere else.
> vfs.misc will be rebased once v6.5-rc1 is released.

Actually, I really don't care which route they take, what really matters to me 
is that they get upstream one way or another :-)

Thank you very much,

Fabio

> ---
> 
> Applied to the vfs.misc branch of the vfs/vfs.git tree.
> Patches in the vfs.misc branch should appear in linux-next soon.
> 
> Please report any outstanding bugs that were missed during review in a
> new review to the original patch series allowing us to drop it.
> 
> It's encouraged to provide Acked-bys and Reviewed-bys even though the
> patch has now been applied. If possible patch trailers will be updated.
> 
> Note that commit hashes shown below are subject to change due to rebase,
> trailer updates or similar. If in doubt, please check the listed branch.
> 
> tree:   https://git.kernel.org/pub/scm/linux/kernel/git/vfs/vfs.git
> branch: vfs.misc
> 
> [1/3] fs/ecryptfs: Replace kmap() with kmap_local_page()
>       https://git.kernel.org/vfs/vfs/c/7a367455b6a5
> [2/3] fs/ecryptfs: Use kmap_local_page() in ecryptfs_write()
>       https://git.kernel.org/vfs/vfs/c/55f13011af9d
> [3/3] fs/ecryptfs: Use kmap_local_page() in copy_up_encrypted_with_header()
>       https://git.kernel.org/vfs/vfs/c/de9f5a15080f
  
Christian Brauner Aug. 18, 2023, 1:18 p.m. UTC | #6
On Thu, Aug 17, 2023 at 07:13:56PM +0200, Fabio M. De Francesco wrote:
> On venerdì 30 giugno 2023 10:45:17 CEST Christian Brauner wrote:
> > On Wed, 26 Apr 2023 19:22:20 +0200, Fabio M. De Francesco wrote:
> > > kmap() and kmap_atomic() have been deprecated in favor of
> > > kmap_local_page().
> > > 
> > > Therefore, replace kmap() and kmap_atomic() with kmap_local_page().
> > > 
> > > Tested in a QEMU/KVM x86_32 VM, 6GB RAM, booting a kernel with
> > > HIGHMEM64GB enabled.
> > > 
> > > [...]
> > 
> > Picking this up. Please tell me if this should be routed somewhere else.
> > vfs.misc will be rebased once v6.5-rc1 is released.
> 
> Christian,
> 
> v6.5-rc1 has been released since a while, but I can't yet see this series. Are 
> there problems with these patches that stop their merge?

Nothing stops them. I just planned to send all of this for v6.6. as I
didn't see a need to sent it earlier.

This message made neomutt crash like crazy btw. So I had to get create
to be able to reply to so hopefully that message gets through...