Re: [GIT PULL] Kmap conversions for 5.12

From: David Sterba
Date: Fri Feb 26 2021 - 09:26:42 EST


On Thu, Feb 25, 2021 at 08:32:34AM -0800, Ira Weiny wrote:
> On Thu, Feb 25, 2021 at 02:12:52PM +0100, David Sterba wrote:
> > On Wed, Feb 24, 2021 at 09:59:12AM -0800, Ira Weiny wrote:
> > > On Wed, Feb 24, 2021 at 01:30:49PM +0100, David Sterba wrote:
> > > > On Tue, Feb 23, 2021 at 11:25:06AM -0800, Ira Weiny wrote:
> > > > > On Tue, Feb 23, 2021 at 09:13:42AM -0800, Linus Torvalds wrote:
> > > > > > On Tue, Feb 23, 2021 at 7:03 AM David Sterba <dsterba@xxxxxxxx> wrote:
> > > > [...]
> > > >
> > > > > Sorry. I will change it.
> > > >
> > > > Let me know how you want to proceed with the patchset/pull request.
> > >
> > > To be clear I'd like to just drop the 2 patches which use zero_user() for this
> > > merge window.
> > >
> > > I've already submitted some additional btrfs changes for 5.13[1]. I can rework
> > > these zero_user() patches and submit them through Andrew for 5.13 as separate
> > > set. That is what I meant by 'I will change it'.
> > >
> > > > I
> > > > can play the messenger again but now it seems a round of review is
> > > > needed and with some testing it'll be possible in some -rc. At that
> > > > point you may take the patches via the mm tree, unless Linus is ok with
> > > > a late pull.
> > >
> > > I'm ok with delaying the memzero_page() change to 5.13. There are a lot of
> > > kmap changes to come. But I'm trying to do them as smaller series just for
> > > this reason. I don't want valid changes to be denied due to my messing up just
> > > a few patches... :-( Hopefully you and Linus can forgive me on this one.
> > >
> > > Is ok to just drop them and merge the rest of this series in 5.12?
> >
> > Ok, no problem. Please let me know exactly which patches to drop, I'll
> > respin the branch. Thanks.
>
> Drop These 2:
>
> [PATCH V2 5/8] iov_iter: Remove memzero_page() in favor of zero_user()
> https://lore.kernel.org/lkml/20210210062221.3023586-6-ira.weiny@xxxxxxxxx/
>
> [PATCH V2 8/8] btrfs: convert to zero_user()
> https://lore.kernel.org/lkml/20210210062221.3023586-9-ira.weiny@xxxxxxxxx/
>
>
> Keep:
>
> [PATCH V2 1/8] mm/highmem: Lift memcpy_[to|from]_page to core
> [PATCH V2 2/8] mm/highmem: Convert memcpy_[to|from]_page() to kmap_local_page()
> [PATCH V2 3/8] mm/highmem: Introduce memcpy_page(), memmove_page(), and memset_page()
> [PATCH V2 4/8] mm/highmem: Add VM_BUG_ON() to mem*_page() calls
> ...
> [PATCH V2 6/8] btrfs: use memcpy_[to|from]_page() and kmap_local_page()
> [PATCH V2 7/8] btrfs: use copy_highpage() instead of 2 kmaps()
> ...
>
> I would resend but I'd rather keep the exact commits you had in your testing
> rather than potentially messing up the rebase this late.

Got it, thanks. It's easier for me to delete the patches once I have
them in the branch, that's been updated and now pushed to kernel org
again (https://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux.git/log/?h=kmap-conversion-for-5.12)

I'll add it to testing branches and let it test over the weekend,
sending the pull request next week.