Re: [PATCH v14 1/9] xen: introduce generic non-atomic test_*bit()

[Oleksii <oleksii.kurochko@xxxxxxxxx>]

On Wed, 2024-07-03 at 09:24 +0200, Jan Beulich wrote:
> On 02.07.2024 13:01, Oleksii Kurochko wrote:
> > The following generic functions were introduced:
> > * test_bit
> > * generic__test_and_set_bit
> > * generic__test_and_clear_bit
> > * generic__test_and_change_bit
> > 
> > These functions and macros can be useful for architectures
> > that don't have corresponding arch-specific instructions.
> > 
> > Also, the patch introduces the following generics which are
> > used by the functions mentioned above:
> > * BITOP_BITS_PER_WORD
> > * BITOP_MASK
> > * BITOP_WORD
> > * BITOP_TYPE
> > 
> > The following approach was chosen for generic*() and arch*() bit
> > operation functions:
> > If the bit operation function that is going to be generic starts
> > with the prefix "__", then the corresponding generic/arch function
> > will also contain the "__" prefix. For example:
> >  * test_bit() will be defined using arch_test_bit() and
> >    generic_test_bit().
> >  * __test_and_set_bit() will be defined using
> >    arch__test_and_set_bit() and generic__test_and_set_bit().
> > 
> > Signed-off-by: Oleksii Kurochko <oleksii.kurochko@xxxxxxxxx>
> > Reviewed-by: Jan Beulich <jbeulich@xxxxxxxx>
> > ---
> > Changes in V14:
> >  - Nothing changed. Only Rebase.
> > ---
> > Changes in V13:
> >  - add Reviewed-by: Jan Beulich <jbeulich@xxxxxxxx>
> > ---
> > Changes in V12:
> >  - revert change of moving the definition of BITS_PER_BYTE from
> > <arch>/bitops.h to xen/bitops.h.
> >    ( a separate patch will be provided to put BITS_PER_BYTE to
> > proper place )
> 
> Oleksii - seeing that this wasn't actually done (as noticed by
> Michal), my
> intention would be to adjust the patch while committing. Please let
> me know
> shortly if there is anything I'm overlooking, possibly implying the
> intended
> adjustment shouldn't be done (and further suggesting that the
> revision log
> then is wrong and/or incomplete). As indicated, I'll need another
> reply of
> yours here (and for subsequent patches; maybe simply the entire
> series)
> anyway, as a release-ack is still missing.
The changes log is correct.

If we are going to drop BITS_PER_BYTE in xen/bitops.h, then it will
need to be added back to <arm>/bitops.h as it was done in "Changes in
V12". (This change was lost somewhere due to an incorrect rebase by
me.).

Regarding Release-Acked-By, I was okay to have this patch series during
Soft Code Freeze but now we are in Hard Code Freeze state where I am
expecting to see only bug/security fixes.

Unfortunately, it would be better based on current state of release to
wait 4.20 staging branch.

Based on that let me know if you want me to sent a new patch series
version with BITS_PER_BYTE adjustments or it still could be fix during
the commit when 4.20 staging will be available?

~ Oleksii

> 
> Andrew - this is going to be the last time I ask that you please come
> forward
> with any concrete objections to this or any other patch in the
> series, if
> indeed there were any. Without you doing so, I'm going to commit this
> series
> (or the parts thereof that are properly ready) as soon as a release
> ack has
> arrived, but - to allow a little bit of time - no earlier than
> tomorrow
> morning. That'll be on the basis that so far possible objections of
> yours are
> purely hearsay; you never voiced any on the list, and even on Matrix
> it was
> only along the lines of "there is something".
> 
> Jan