- 02 Jun, 2005 1 commit
-
-
Moe Jette authored
-
- 01 Jun, 2005 7 commits
-
-
Danny Auble authored
update for race condition for bglbridge api. Added info about multi diminsion X dim into Administrators info for bluegene.html
-
Moe Jette authored
-
Moe Jette authored
-
Moe Jette authored
-
Moe Jette authored
-
Danny Auble authored
update for sfree to only run on SN of a BGL system. checking for the db2 files and doing a DLopen on them, creating a warning error if they aren't correct.
-
Moe Jette authored
-
- 27 May, 2005 3 commits
-
-
Christopher J. Morrone authored
but don't stop the rest of the build
-
Moe Jette authored
could cause job submit or update to fail.
-
Christopher J. Morrone authored
-
- 26 May, 2005 3 commits
-
-
Danny Auble authored
-
Moe Jette authored
-
Christopher J. Morrone authored
-
- 25 May, 2005 5 commits
-
-
Moe Jette authored
libdb2.so file present. Otherwise tell the user to run on the BGL SN.
-
Danny Auble authored
-
Christopher J. Morrone authored
-
Christopher J. Morrone authored
-
Christopher J. Morrone authored
the directory containing the proctrackext.exp file (AIX only).
-
- 24 May, 2005 11 commits
-
-
Moe Jette authored
-
Moe Jette authored
select_bluegene.so files (not build on AIX).
-
Moe Jette authored
type if SYSTEM_DIMENSIONS==0 (not ASCI-C compliant)
-
Moe Jette authored
-
Moe Jette authored
-
Christopher J. Morrone authored
-
Danny Auble authored
-
Danny Auble authored
-
Christopher J. Morrone authored
-
Danny Auble authored
-
Danny Auble authored
-
- 23 May, 2005 1 commit
-
-
Moe Jette authored
-
- 20 May, 2005 9 commits
-
-
Danny Auble authored
-
Christopher J. Morrone authored
-
Danny Auble authored
-
Danny Auble authored
-
Danny Auble authored
-
Christopher J. Morrone authored
"srun output overflow ("Need to rewind" in srun/_do_output_line)" When srun's stdout is consuming data slowly, srun can receive notice that the job has terminated before the output stream has been fully written. The IO thread will receives a SIGHUP to kick it out of its blocking poll. However in the slow stdout situation the SIGHUP can interrupt the fflush. When the fflush is interrupted, it appears to clear the stream buffer even though the data wasn't written out to the file descriptor, and we see data loss on stdout. To avoid this situation, this change makes signals to the IO thread go over a pipe rather than sending a signal. Also, some extra return code checking is done in io.c:_do_output_line().
-
Christopher J. Morrone authored
under AIX. This work-around bypasses the eio handling of /dev/null and instead directly makes pin[0] a file handle for /dev/null. The only problem with this fix is that it is not obvious elsewhere in the code that pin is not a pipe when using "srun --batch" and stdin is /dev/null.
-
Moe Jette authored
terminating).
-
Moe Jette authored
systems (different mechanisms to prevent line-feed).
-