The bug affects not only %f but also %g, and it can lead to a SIGSEGV.
Reproduce with the bash 3.2.1 built-in:
$ LC_ALL=en_US.UTF-8 bash -c "printf \"%'.5gn\" 999.996"
1000
$ LC_ALL=en_US.UTF-8 bash -c "printf \"%'.4gn\" .996"
Segmentation fault
$ LC_ALL=en_US.UTF-8 bash -c "printf \"
Jim reminded me that I should have credited Bastien for
the bug report and initial patch for the openat-die
issue. Sorry about that. I installed this:
>From 422ef85558e07719d5aa28a1bab67d789ecb3e70 Mon Sep 17 00:00:00 2001
From: Paul Eggert
Date: Wed, 12 Jan 2011 09:51:47 -0800
Subject: [PATCH]
Thanks for the confirmation. I opened a glibc bug on this.
http://sourceware.org/bugzilla/show_bug.cgi?id=12394
Here is some sample code to demonstrate the problem (attached to the bug
report):
#include
#include
int
main(void)
{
double x;
setlocale(LC_NUMERIC, "");
while (scanf("%l
Eric Blake wrote:
> [adding bug-gnulib]
>
> On 01/11/2011 08:40 PM, Andrew J. Schorr wrote:
>> On Tue, Jan 11, 2011 at 10:31:26PM -0500, Andrew J. Schorr wrote:
>>> Using gawk 3.1.8 on Fedora 13 64-bit linux, I see this:
>>>
>>> bash-4.1$ (echo 100; echo 99.9998071828) | gawk -v "sq='"
>>>
[adding bug-gnulib]
On 01/11/2011 08:40 PM, Andrew J. Schorr wrote:
> On Tue, Jan 11, 2011 at 10:31:26PM -0500, Andrew J. Schorr wrote:
>> Using gawk 3.1.8 on Fedora 13 64-bit linux, I see this:
>>
>> bash-4.1$ (echo 100; echo 99.9998071828) | gawk -v "sq='" '{printf
>> "%" sq ".2f\n", $1
On 01/12/2011 03:25 AM, Bastien ROUCARIES wrote:
>> And, rather than checking != ENOTSUP, it might be safer to check ==
>> ENOMEM, so that you are minimizing the impact of your change. The whole
>> point of patch 3 appears to be avoiding the risk of the fchdir()
>> fallback on the rare systems whe
Le mardi 11 janvier 2011 19:36:36, Eric Blake a écrit :
> On 01/11/2011 10:50 AM, Bastien ROUCARIES wrote:
> > I have also corrected a bug openat not testing NULL path. I return EFAULT
> > like my Linux here.
>
> I disagree with this change.
[...]
> :), and we are not in a position to check for
Rob Vermaas wrote:
> while building latest rcs revision we encountered a problem with
> gnulib. The build fails with
>
> building save-cwd.o
> gcc -std=gnu99 -std=gnu99 -DHAVE_CONFIG_H -I. -g -O2 -c -o
> save-cwd.o save-cwd.c
> save-cwd.c:33:21: fatal error: xgetcwd.h: No such file or directo
Hi,
while building latest rcs revision we encountered a problem with
gnulib. The build fails with
building save-cwd.o
gcc -std=gnu99 -std=gnu99 -DHAVE_CONFIG_H -I. -g -O2 -c -o
save-cwd.o save-cwd.c
save-cwd.c:33:21: fatal error: xgetcwd.h: No such file or directory
compilation terminated
