Re: fs/ntfs/aops.c:378:12: warning: stack frame size (2216) exceeds limit (1024) in 'ntfs_read_folio'

[Nathan Chancellor <nathan@kernel.org>]

On Mon, Aug 15, 2022 at 04:37:09PM +0200, Arnd Bergmann wrote:
> On Mon, Aug 15, 2022 at 3:48 PM Arnd Bergmann <arnd@kernel.org> wrote:
> 
> > I have no problems with a patch removing support for 256KB pages if that
> > helps, as Hexagon is the only architecture to support this and there are close
> > to zero Linux users anway. This would leave only three warnings for 64KB

Right, I had brought up at least adjusting the dependencies of 256KB
pages so that it could not be selected with CONFIG_COMPILE_TEST to
reduce the number of warnings that would appear in randconfigs.

https://lore.kernel.org/YoAlvnyjEbYV4T1L@dev-arch.thelio-3990X/

I suspect removing it outright would be fine too.

> > pages in allmodconfig:
> >
> > fs/mpage.c:131:20: error: stack frame size (1128) exceeds limit (1024)
> > in 'do_mpage_readpage' [-Werror,-Wframe-larger-than]
> > fs/mpage.c:447:12: error: stack frame size (1264) exceeds limit (1024)
> > in '__mpage_writepage' [-Werror,-Wframe-larger-than]
> > fs/ext4/readpage.c:223:5: error: stack frame size (1208) exceeds limit
> > (1024) in 'ext4_mpage_readpages' [-Werror,-Wframe-larger-than]
> 
> I looked into these a bit more and found that these are arrays of sector_t,
> which could be either 32-bit or 64-bit wide before 72deb455b5ec
> ("block: remove CONFIG_LBDAF"), but is now always 64-bit, so having
> an array of 128 of these (65536/512) adds a 1KB to the stack and will
> cause a warning. It's only slightly over the limit, and there are very few
> 32-bit systems that allow 64KB pages to trigger that warning.
> 
> I see now that ppc440 also supports 256KB pages and has the same
> problem as hexagon, but also has been broken since the start of the
> git history in this regard:
> 
> fs/mpage.c:638:1: error: the frame size of 4280 bytes is larger than
> 2048 bytes [-Werror=frame-larger-than=]
> 
> I don't know if anyone strongly cares about 256KB pages on
> ppc44x any more, but given this, I'm fairly sure that they are
> not using block based file systems. So we could just make
> CONFIG_BLOCK depend on PAGE_SIZE_LESS_THAN_256KB
> globally instead of dropping 256KB pages everywhere.

That doesn't sound like an unreasonable solution.

Cheers,
Nathan