This is the mail archive of the libc-alpha@sourceware.org mailing list for the glibc project.

Index Nav:	[Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav:	[Date Prev] [Date Next]	[Thread Prev] [Thread Next]
Other format:	[Raw text]

Re: [PATCH] Disable use of FMA instructions in branred

From: Andreas Schwab <schwab at linux-m68k dot org>
To: David Miller <davem at davemloft dot net>
Cc: munroesj at us dot ibm dot com, munroesj at linux dot vnet dot ibm dot com, carlos at systemhalted dot org, libc-alpha at sourceware dot org
Date: Sat, 17 Mar 2012 09:24:56 +0100
Subject: Re: [PATCH] Disable use of FMA instructions in branred
References: <m2obrwshli.fsf@igel.home><CADZpyixoYuJd4pKaxy_N1uaRQ1z29VAsszbJ2cHL0FG1M0rYCQ@mail.gmail.com><1331914823.7288.9.camel@spokane1.rchland.ibm.com><20120316.205747.1324879371265298854.davem@davemloft.net>

David Miller <davem@davemloft.net> writes:

> I personally find this change highly suspect, especially because we
> specifically have a multi-arch variant on x86-64 to build this file
> with FMA4 enabled.

That has obviously never been properly tested.

> I suspect we actually have either GCC bug or an unreasonable
> assumption in the brandred.c code.

It is just the same bug as the use of libc_feholdexcept_setround_53bit.

Andreas.

-- 
Andreas Schwab, schwab@linux-m68k.org
GPG Key fingerprint = 58CA 54C7 6D53 942B 1756  01D3 44D5 214B 8276 4ED5
"And now for something completely different."

References:
- [PATCH] Disable use of FMA instructions in branred
  - From: Andreas Schwab
- Re: [PATCH] Disable use of FMA instructions in branred
  - From: Carlos O'Donell
- Re: [PATCH] Disable use of FMA instructions in branred
  - From: Steven Munroe
- Re: [PATCH] Disable use of FMA instructions in branred
  - From: David Miller

Index Nav:	[Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav:	[Date Prev] [Date Next]	[Thread Prev] [Thread Next]