Re: [PATCH v2 0/2] Convert regulator drivers to use devm_kmemdup_array()

[Andy Shevchenko]

On Thu, Mar 06, 2025 at 02:37:57PM +0200, Andy Shevchenko wrote:
> On Thu, Mar 06, 2025 at 01:32:19PM +0200, Raag Jadav wrote:
> > On Thu, Mar 06, 2025 at 01:04:32PM +0200, Andy Shevchenko wrote:
> > > On Thu, Mar 06, 2025 at 12:14:59PM +0200, Raag Jadav wrote:
> > > > On Thu, Mar 06, 2025 at 11:19:31AM +0200, Andy Shevchenko wrote:
> > > > > On Thu, Mar 06, 2025 at 04:15:55AM +0200, Raag Jadav wrote:
> > > > > > On Fri, Feb 28, 2025 at 09:22:47PM +0000, Mark Brown wrote:
> > > > > > > On Fri, 28 Feb 2025 12:50:55 +0530, Raag Jadav wrote:
> > > > > > > > This series converts regulator drivers to use the newly introduced[1]
> > > > > > > > devm_kmemdup_array() helper. This depends on changes available on
> > > > > > > > immutable tag[2].
> > > > > > > > 
> > > > > > > > [1] https://lore.kernel.org/r/20250212062513.2254767-1-raag.jadav@xxxxxxxxx
> > > > > > > > [2] https://lore.kernel.org/r/Z7xGpz3Q4Zj6YHx7@xxxxxxxxxxxxxxxxxx
> > > > > > > > 
> > > > > > > > [...]
> > > > > > > 
> > > > > > > Applied to
> > > > > > > 
> > > > > > >    https://git.kernel.org/pub/scm/linux/kernel/git/broonie/regulator.git for-next
> > > > > > 
> > > > > > Thank you.
> > > > > > 
> > > > > > Unless there's a nuance intended in the merge series title that I'm unable
> > > > > > to understand, it probably seems incomplete.
> > > > > 
> > > > > I believe it's an issue somewhere in the scripts. The long Subject line is
> > > > > split in the mailbox and that's probably is not supported by the machinery.
> > > > > 
> > > > > You are not the only one who reports this issue.
> > > > 
> > > > While I'm not well educated on the machinery, I'm also seeing devm_kmemdup_array()
> > > > introduction commit reordered in -next and thinking perhaps it can cause issues
> > > > with bisect, especially after final merge into Linus' tree?
> > > 
> > > I;m not sure what you exactly pointing out here. Mark seems applied only
> > > necessary part of the immutable tag, which have the same effect. The Git merges
> > > only once the stuff as long as it has the same hash.
> > 
> > $ git describe
> > next-20250306
> > 
> > $ git log --oneline --grep raag
> 
> --author="Raag ..."

...

> > 1f4c7f3b3afa Merge patch series "Split devres APIs to device/devres.h and introduce devm_kmemdup_array()"
> > a103b833ac38 devres: Introduce devm_kmemdup_array()
> > b8c38ccb2ca5 input: ipaq-micro-keys: use devm_kmemdup_array()
> > cdcc09a495a4 input: sparse-keymap: use devm_kmemdup_array()
> > a0d78eec8839 iio: adc: xilinx-xadc-core: use devm_kmemdup_array()
> > 86068aca7548 pinctrl: pxa2xx: use devm_kmemdup_array()
> > 91bfcc7a2fdb pinctrl: tangier: use devm_kmemdup_array()
> > d795fb90d6c6 pinctrl: cherryview: use devm_kmemdup_array()
> > f192c8447f4e pinctrl: baytrail: copy communities using devm_kmemdup_array()
> > 753764aa8eb5 pinctrl: intel: copy communities using devm_kmemdup_array()
> > a21cad931276 driver core: Split devres APIs to device/devres.h
> > 18311a766c58 err.h: move IOMEM_ERR_PTR() to err.h
> > 
> > I'm expecting commit a103b833ac38 to be before its users, or perhaps I'm
> > doing something wrong here?
> 
> $ git tag --contains a103b833ac38
> ib-devres-iio-input-pinctrl-v6.15
> next-20250225
> next-20250226
> next-20250227
> next-20250228
> next-20250303
> next-20250304
> next-20250305
> next-20250306

Ah, and this:

$ git cat-file commit 753764aa8eb5
tree 7d99d7bb009118f7c0d8cbac9aa2ae321fc74785
parent a103b833ac3806b816bc993cba77d0b17cf801f1
author Raag Jadav <raag.jadav@xxxxxxxxx> 1739341506 +0530
committer Andy Shevchenko <andriy.shevchenko@xxxxxxxxxxxxxxx> 1740391732 +0200

pinctrl: intel: copy communities using devm_kmemdup_array()
...

Most likely it uses the one which was merged earlier by Stephen, who merged
the ASoC before merging my immutable tag.

-- 
With Best Regards,
Andy Shevchenko