Re: [PATCH] mm/highmem: Align-down to page the address for kunmap_flush_on_unmap()

[Matthew Wilcox]

On Fri, Jan 27, 2023 at 06:58:55PM +0100, Fabio M. De Francesco wrote:
> On giovedì 26 gennaio 2023 21:07:53 CET Matthew Wilcox wrote:
> > On Thu, Jan 26, 2023 at 03:33:46PM +0100, Fabio M. De Francesco wrote:
> > > If ARCH_HAS_FLUSH_ON_KUNMAP is defined (PA-RISC case), __kunmap_local()
> > > calls kunmap_flush_on_unmap(). The latter currently flushes the wrong
> > > address (as confirmed by Matthew Wilcox and Helge Deller). Al Viro
> > > proposed to call kunmap_flush_on_unmap() on an aligned-down to page
> > > address in order to fix this issue. Consensus has been reached on this
> > > solution.
> > > 
> > > Therefore, if ARCH_HAS_FLUSH_ON_KUNMAP is defined, call
> > > kunmap_flush_on_unmap() on an aligned-down to page address computed with
> > > the PTR_ALIGN_DOWN() macro.
> > 
> > You missed a spot.  Sent the version I've had in my tree for a few days.
> 
> I'm sorry for missing the other call site.
> Furthermore, your patch has a better commit message. 
> 
> However, I'm also sorry because I would have expected a different kind of 
> support from you. My patch could have been improved with v2.
> 
> Anyway, I assume that you missed that I had asked you and Al if anybody were 
> interested in doing these changes and probably you missed also that Ira 
> encouraged me to send a patch, whereas you did not show any interest...
> https://lore.kernel.org/lkml/3146373.5fSG56mABF@suse/

Sorry; I get too much email and I missed the last six emails in that
thread (just went back and read them).  According to git, I committed
this patch to my tree on Jan 20th, so if I'd read your email, I would
have told you to not bother.