Re: [Valgrind-developers] [PATCH 1/2] x86: optimize XCHG to MOV for same-register forms

Christian Ludloff <[email protected]>
Newsgroups gmane.comp.gnu.binutils
Message-ID <CAKSQd8U6E=xOBoCF7EgGBNyc_Hiq+8vKL2FTWrxd=cx6w+pW3Q@mail.gmail.com>
> https://sourceware.org/bugzilla/show_bug.cgi?id=34343

There is an additional case that breaks because of the optimization.

Transmeta processors provided translation control via XCHG eSP,eSP.

67h,87h,E4h forced BOT on Crusoe, i.e. begin-of-translation.
66h,87h,E4h forced EOT on Crusoe, i.e. end-of-translation.

67h,87h,E4h forced BOT+EOT on Efficeon, i.e. break-translation-here.
66h,87h,E4h no longer forced EOT on Efficeon.

Granted, Transmeta processors are historical by now.

Still... this gas optimization is... perhaps better left to higher level tools.

--
C.
lmpx.com only provides a reader for public news (NNTP) servers. It is not affiliated with the servers or forums shown here and is not responsible for the content of articles, which is written by their respective authors.