[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: qemu disassembler status
From: |
Richard Henderson |
Subject: |
Re: qemu disassembler status |
Date: |
Sun, 13 Sep 2020 17:26:58 -0700 |
User-agent: |
Mozilla/5.0 (X11; Linux x86_64; rv:68.0) Gecko/20100101 Thunderbird/68.10.0 |
On 9/12/20 12:03 AM, Thomas Huth wrote:
> Is capstone good enough already to replace libvixl?
Yes as far as the version that we currently have imported, but no if we were to
update vixl. In particular, capstone has no support for SVE or other v8.2+
additions.
> And what about the other old disassemblers that we have in disasm/ ?
> Could some of them be replaced by capstone, too?
We could definitely lose the old x86 disassembler. Which has no support for
AVX, which we use in TCG. Possibly some of the others too, but not many.
> Or shall we try to pursue the idea of adding a GPLv3 helper program that
> could link against recent versions of libopcode?
It might be worthwhile, yes.
> And what about new disassembler files like the Loongson 2F disassembler
> that has been proposed two months ago? Shall we enforce that people try
> to add such stuff to capstone first?
I don't know what the level of difficulty is there.
I certainly do not thing a complete new standalone disassembler is worthwhile.
Exactly what to suggest instead is a problem though.
r~