Follow @Openwall on Twitter for new release announcements and other news
[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Date: Thu, 21 May 2015 20:39:32 +0200
From: Aurelien Jarno <aurelien@...el32.net>
To: Richard Henderson <rth@...ddle.net>
Cc: Rich Felker <dalias@...c.org>, musl@...ts.openwall.com,
	qemu-devel@...gnu.org
Subject: Re: [Qemu-devel] Broken SuperH atomics in qemu app-level emulation

On 2015-05-20 12:16, Richard Henderson wrote:
> On 05/17/2015 08:05 PM, Rich Felker wrote:
> > At the very least qemu-sh4 should provide a correct value of AT_HWCAP
> > so we get working atomics with -cpu SH7785. And since the GUSA atomic
> > model can't work with app-level emulation, I really think qemu-sh4
> > should either default to -cpu SH7785 or always expose the
> > synco/mov.li/mov.co opcodes (and hwcap) regardless of the -cpu
> > setting.
> 
> I agree with AT_HWCAP and defaulting to SH7785.  Aurelien?

I agree, but for linux-user only. In general we should default to the
most capable CPU when possible. For the softmmu emulation, QEMU doesn't
emulate it fully, but only the instruction set, so we should keep the
current default CPU.

-- 
Aurelien Jarno                          GPG: 4096R/1DDD8C9B
aurelien@...el32.net                 http://www.aurel32.net

Powered by blists - more mailing lists

Confused about mailing lists and their use? Read about mailing lists on Wikipedia and check out these guidelines on proper formatting of your messages.