[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date: Mon, 10 Oct 2022 21:55:47 +0200
From: "Arnd Bergmann" <arnd@...db.de>
To: "Nick Desaulniers" <ndesaulniers@...gle.com>,
"Andrew Morton" <akpm@...ux-foundation.org>
Cc: "Alexander Viro" <viro@...iv.linux.org.uk>,
"Nathan Chancellor" <nathan@...nel.org>,
"Tom Rix" <trix@...hat.com>, "Xiaoming Ni" <nixiaoming@...wei.com>,
"Luis Chamberlain" <mcgrof@...nel.org>,
"Andi Kleen" <ak@...ux.intel.com>, linux-fsdevel@...r.kernel.org,
linux-kernel@...r.kernel.org, llvm@...ts.linux.dev,
"Christoph Hellwig" <hch@....de>,
"Eric Dumazet" <edumazet@...gle.com>,
"Darrick J. Wong" <darrick.wong@...cle.com>,
"Greg Kroah-Hartman" <gregkh@...uxfoundation.org>
Subject: Re: [PATCH] fs/select: mark do_select noinline_for_stack for 32b
On Fri, Oct 7, 2022, at 10:11 PM, Nick Desaulniers wrote:
> +#ifdef CONFIG_64BIT
> +#define noinline_for_stack_32b
> +#else
> +#define noinline_for_stack_32b noinline_for_stack
> +#endif
> +
> +noinline_for_stack_32b
I don't see much value in making it behave differently for 32 bit:
it doesn't reduce the total frame size on 32-bit machines but only
hides the warning. The bug you are working around also looks i386
specific (because of limited number of registers vs ubsan needing
a lot of them), so just make it a simple 'noinline_for_stack'.
Arnd
Powered by blists - more mailing lists