lists.openwall.net   lists  /  announce  owl-users  owl-dev  john-users  john-dev  passwdqc-users  yescrypt  popa3d-users  /  oss-security  kernel-hardening  musl  sabotage  tlsify  passwords  /  crypt-dev  xvendor  /  Bugtraq  Full-Disclosure  linux-kernel  linux-netdev  linux-ext4  linux-hardening  linux-cve-announce  PHC 
Open Source and information security mailing list archives
 
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date:   Wed, 7 Jul 2021 10:36:38 -0400
From:   John David Anglin <dave.anglin@...l.net>
To:     Arnd Bergmann <arnd@...db.de>,
        Abd-Alrhman Masalkhi <abd.masalkhi@...il.com>
Cc:     Christoph Hellwig <hch@...radead.org>,
        Jens Axboe <axboe@...nel.dk>, bernie@...eler.com,
        Parisc List <linux-parisc@...r.kernel.org>,
        linux-block <linux-block@...r.kernel.org>,
        Linux Kernel Mailing List <linux-kernel@...r.kernel.org>,
        Dan Carpenter <dan.carpenter@...cle.com>
Subject: Re: div_u64/do_div stack size usage, was Re: [v3] block: Removed a
 warning while compiling with a cross compiler for parisc

On 2021-07-07 4:17 a.m., Arnd Bergmann wrote:
> I have also tried it with the other gcc versions and shown that it
> happens with every older compiler as well, but it does not happen
> with gcc-10 or gcc-11, which bring the frame size down to 164 or
> 172 bytes respectively. gcc-10 also fixes similar warnings for
> net/ipv4/tcp_input.c, net/sunrpc/stats.c and lib/xxhash.c that
> fly under the radar with the default warning level.
>
> My first thought was this was a result of -finline-functions getting
> enabled by default in gcc-10, but enabling it on gcc-9 did not
> help here. It's likely that the gcc behavior was fixed in the process
> of enabling the more aggressive inliner by default though.
A number of improvements were made to the calculation of RTX costs in gcc-10 and gcc-11.
These dramatically affect inlining and the compilation time for xxhash.c on parisc.  The problem
is pretty much fixed for the 32-bit target but more work is needed for the 64-bit target.

See:
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=87256

Dave

-- 
John David Anglin  dave.anglin@...l.net


Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ