--- Comment #17 from jvdelisle at gcc dot gnu dot org 2010-05-13 20:34
---
I believe this is fixed now.
--
jvdelisle at gcc dot gnu dot org changed:
What|Removed |Added
--- Comment #14 from nightstrike at gmail dot com 2010-04-29 11:53 ---
I'm running the testsuite now
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=43844
--- Comment #15 from dannysmith at users dot sourceforge dot net
2010-04-30 03:02 ---
(In reply to comment #13)
Kai, what about the GetTempPath? Cf. the rough draft in attachment 20460
[edit] ?
Compare choose_tmpdir() in libiberty/make_temp_file.c
Danny
--
--- Comment #16 from kargl at gcc dot gnu dot org 2010-04-30 03:23 ---
(In reply to comment #15)
(In reply to comment #13)
Kai, what about the GetTempPath? Cf. the rough draft in attachment 20460
[edit] [edit] ?
Compare choose_tmpdir() in libiberty/make_temp_file.c
Can't use
--- Comment #13 from burnus at gcc dot gnu dot org 2010-04-26 06:11 ---
Kai, what about the GetTempPath? Cf. the rough draft in attachment 20460 ?
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=43844
--- Comment #10 from ktietz at gcc dot gnu dot org 2010-04-24 12:02 ---
So I investigated this issue about mktemp in more detail and found finally the
first-scope and second-scope bug here.
Old logic was opening files and as long as mktemp failed on a second call,
things were working
--- Comment #11 from jvdelisle at gcc dot gnu dot org 2010-04-24 12:17
---
Yes, OK to commit to trunk. Thanks!
--
jvdelisle at gcc dot gnu dot org changed:
What|Removed |Added
--- Comment #12 from ktietz at gcc dot gnu dot org 2010-04-24 12:25 ---
(In reply to comment #11)
Yes, OK to commit to trunk. Thanks!
Ok, applied to trunk at revision 158686.
Kai
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=43844
--- Comment #9 from jvdelisle at gcc dot gnu dot org 2010-04-23 18:19
---
I discussed with Kai on IRC today and have approved Kai to commit the patch.
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=43844
--- Comment #3 from burnus at gcc dot gnu dot org 2010-04-22 06:57 ---
I also do not like:
if (tempdir == NULL)
tempdir = DEFAULT_TEMPDIR; /* This is /tmp. */
My feeling is that, e.g., on Windows this won't work; I do not know whether
before the environment variables
--- Comment #4 from ktietz at gcc dot gnu dot org 2010-04-22 07:42 ---
(In reply to comment #3)
My feeling is that, e.g., on Windows this won't work; I do not know whether
before the environment variables GFORTRAN_TMPDIR, TMP and TEMP are used
or not, but /tmp does exist on many
--- Comment #5 from burnus at gcc dot gnu dot org 2010-04-22 08:03 ---
(In reply to comment #3)
or not, but /tmp does exist on many Windows systems.
I wanted to write: does not.
(In reply to comment #4)
I would use here the platform API GetTempPath. This seems to be the best
--- Comment #6 from burnus at gcc dot gnu dot org 2010-04-22 09:21 ---
Created an attachment (id=20460)
-- (http://gcc.gnu.org/bugzilla/attachment.cgi?id=20460action=view)
Draft patch for GetTempPath; fix for mktemp
(In reply to comment #2)
The if and do .. while block may execute
--- Comment #7 from burnus at gcc dot gnu dot org 2010-04-22 09:25 ---
Side note:
template = get_mem (strlen (tempdir) + 20);
Can one change the 20 into 19 or 18 + 1 ? The string /gfortrantmpXX
has 18 characters and '\0' has one.
--
--- Comment #8 from jvdelisle at gcc dot gnu dot org 2010-04-22 13:52
---
I will fix all this on trunk tonight and if we get good test results I will
port it back.
Thanks Kai. My eye was telling me something was not right there. Thanks Tobias
too.
I have a mingw build mostly
--- Comment #1 from jvdelisle at gcc dot gnu dot org 2010-04-22 00:41
---
This does not look right to me:
static int
tempfile (st_parameter_open *opp)
{
const char *tempdir;
char *template;
int fd;
tempdir = getenv (GFORTRAN_TMPDIR);
if (tempdir == NULL)
tempdir =
--- Comment #2 from jvdelisle at gcc dot gnu dot org 2010-04-22 00:44
---
The if and do .. while block may execute mktemp more than once. ??
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=43844
17 matches
Mail list logo
