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

Note that libc-hacker is a closed list. You may look at the archives of this list, but subscription and posting are not open.

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] Fix *printf %La with IBM long double format

From: Steven Munroe <munroesj at us dot ibm dot com>
To: Jakub Jelinek <jakub at redhat dot com>
Cc: Ulrich Drepper <drepper at redhat dot com>, Glibc hackers <libc-hacker at sources dot redhat dot com>, Ryan Arnold <rsa at us dot ibm dot com>
Date: Wed, 06 Jun 2007 08:48:18 -0500
Subject: Re: [PATCH] Fix *printf %La with IBM long double format
References: <20070604130209.GM3081@sunsite.mff.cuni.cz>

Jakub Jelinek wrote:
> Hi!
>
> As the attached testcase shows, printf_fphex for ldbl-128ibm
> has a bunch of issues when the lower double is denormal or has exponent 1.
> E.g. the implicit mantissa bit needs to be also set for exponent 1,
> the difference between exponents might be smaller than 53 (when the
> smaller double is denormal and the bigger double has exponent less than
> 53, etc.
>
>   
I spotted this while debugging nextafterl and put it on the list for
Ryan to look at. But you got there first

Unfortunately I have been busy with POWER6 perf tuning issues...
> I've also included the unrelated workaround for test-misc.c failures
> - the tests from 2007-05-03 aren't really testable on ppc/ppc64.
>   

Thanks!

References:
- [PATCH] Fix *printf %La with IBM long double format
  - From: Jakub Jelinek

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