mquinson [Sat, 22 Oct 2005 09:46:13 +0000 (09:46 +0000)]
And now, use the proper dox tag for parameters
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@1787
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Sat, 22 Oct 2005 09:13:05 +0000 (09:13 +0000)]
Fix broken links
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@1786
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Sat, 22 Oct 2005 09:11:48 +0000 (09:11 +0000)]
Fix broken links in documentation
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@1785
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Sat, 22 Oct 2005 09:07:51 +0000 (09:07 +0000)]
Sanitize the documentation of heap: group defined in module-xbt.doc, its content declared in header file and each elements documented in the C file
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@1784
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Sat, 22 Oct 2005 09:05:33 +0000 (09:05 +0000)]
reindent
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@1783
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Sat, 22 Oct 2005 08:48:53 +0000 (08:48 +0000)]
Search the doc files where they live
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@1782
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
alegrand [Fri, 21 Oct 2005 12:53:51 +0000 (12:53 +0000)]
On the 3.0.2 tracks...
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@1781
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
alegrand [Fri, 21 Oct 2005 12:43:29 +0000 (12:43 +0000)]
Getting ready for 3.0.1
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@1779
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Thu, 20 Oct 2005 22:14:04 +0000 (22:14 +0000)]
let's be optimistic and assume that things are ok when we cross-compile (it does when the target is win)
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@1778
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Thu, 20 Oct 2005 20:57:28 +0000 (20:57 +0000)]
fd_set is not portable on IRIX, but needs some extra headers
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@1777
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Thu, 20 Oct 2005 20:56:55 +0000 (20:56 +0000)]
Lost access to this machine
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@1776
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Thu, 20 Oct 2005 20:56:18 +0000 (20:56 +0000)]
I really dislike gcc on AIX
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@1775
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Thu, 20 Oct 2005 19:36:23 +0000 (19:36 +0000)]
Missing header on IRIX
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@1774
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Thu, 20 Oct 2005 19:07:13 +0000 (19:07 +0000)]
Yeah xlC, calm down, I'll write ansi C without fancy stupid things
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@1773
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Thu, 20 Oct 2005 17:42:35 +0000 (17:42 +0000)]
I told you that the buf plugin is dead
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@1772
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Thu, 20 Oct 2005 17:40:37 +0000 (17:40 +0000)]
amok_remoterr is dead; fix a logic issue: error was never rethrown because of poorly computed bound values
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@1771
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Thu, 20 Oct 2005 17:39:30 +0000 (17:39 +0000)]
Avoid double free
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@1770
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Thu, 20 Oct 2005 17:38:12 +0000 (17:38 +0000)]
Let it fit my testing infrastructure a bit better
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@1769
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Thu, 20 Oct 2005 17:14:45 +0000 (17:14 +0000)]
Eradicate the last bits of the buffered transport which used to live on top of either TCP or SG and was merged with TCP once SG used datadesc_cpy
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@1768
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Thu, 20 Oct 2005 13:22:28 +0000 (13:22 +0000)]
Messages which the payload datatype descriptor is NULL contain 0 bytes as payload (strange, isnt it?)
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@1767
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Thu, 20 Oct 2005 12:58:33 +0000 (12:58 +0000)]
Let me run valgrind on that shitty code without changing this script
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@1766
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Thu, 20 Oct 2005 11:00:32 +0000 (11:00 +0000)]
payload_size is used to memcpy the payload. Do not count the size of the pointed data (like the actual data of a matrix) in it (only count the size of the matrix_t in our example)
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@1765
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Thu, 20 Oct 2005 10:13:00 +0000 (10:13 +0000)]
I keep forgetting the correct name of this macro
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@1764
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Thu, 20 Oct 2005 10:12:04 +0000 (10:12 +0000)]
This file is needed to do make distcheck
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@1763
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Thu, 20 Oct 2005 09:49:44 +0000 (09:49 +0000)]
Let it compile properly
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@1762
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
alegrand [Thu, 20 Oct 2005 09:18:14 +0000 (09:18 +0000)]
bug fix (found by Cyril Banino).
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@1761
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
alegrand [Wed, 19 Oct 2005 14:28:53 +0000 (14:28 +0000)]
make g++ happy
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@1760
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
alegrand [Wed, 19 Oct 2005 14:27:09 +0000 (14:27 +0000)]
make g++ happy. Hope it still has the same meaning though...
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@1759
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
alegrand [Wed, 19 Oct 2005 14:16:01 +0000 (14:16 +0000)]
make g++ happy
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@1758
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
alegrand [Tue, 18 Oct 2005 10:18:31 +0000 (10:18 +0000)]
final reference
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@1757
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
alegrand [Mon, 17 Oct 2005 15:11:54 +0000 (15:11 +0000)]
stupid example of msg_process_suspend and resume
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@1756
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
alegrand [Mon, 17 Oct 2005 15:11:13 +0000 (15:11 +0000)]
Return an error code when the simulation deadlocks. Avoid deadlock in this simulation.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@1755
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
alegrand [Mon, 17 Oct 2005 15:10:27 +0000 (15:10 +0000)]
Getting rid of old printf
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@1754
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
alegrand [Tue, 11 Oct 2005 14:44:21 +0000 (14:44 +0000)]
moved to %f instead of %g. It's usefull when values get large...
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@1753
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
alegrand [Tue, 11 Oct 2005 08:16:29 +0000 (08:16 +0000)]
Document last changes.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@1752
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
alegrand [Tue, 11 Oct 2005 08:16:13 +0000 (08:16 +0000)]
Add two convenient functions for Cyril Banino.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@1751
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Mon, 3 Oct 2005 19:50:09 +0000 (19:50 +0000)]
The buf plugin is dead on SG. Thanks to the verbosity of simulated exceptions for making this error so obvious
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@1750
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Mon, 3 Oct 2005 19:47:48 +0000 (19:47 +0000)]
Get the simulated exceptions as verbose as real ones
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@1749
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Mon, 3 Oct 2005 19:39:40 +0000 (19:39 +0000)]
Write yet another changelog entry for Arnaud (bugfix for Flavien); Get the simulated exceptions as verbose as real ones
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@1748
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
alegrand [Mon, 3 Oct 2005 15:44:06 +0000 (15:44 +0000)]
Add some debugging messages.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@1747
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
alegrand [Mon, 3 Oct 2005 15:43:30 +0000 (15:43 +0000)]
Make SURF complain when you ask to perform a communication between two non-connected cards.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@1746
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
alegrand [Mon, 3 Oct 2005 13:30:06 +0000 (13:30 +0000)]
removed a brutal exit. It doesn't make sense in MSG... the whole process
should not die whenever a context dies.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@1745
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
alegrand [Mon, 3 Oct 2005 11:54:26 +0000 (11:54 +0000)]
Add the ability to print the message location in the code.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@1744
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Sun, 2 Oct 2005 19:28:02 +0000 (19:28 +0000)]
Add a yummy matrix multiplication rpc thing
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@1743
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Sun, 2 Oct 2005 12:33:52 +0000 (12:33 +0000)]
Some more debugs
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@1742
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Sun, 2 Oct 2005 12:32:34 +0000 (12:32 +0000)]
I must have been drunk when I did the previous version. I'd like to convince you that I was only asleep, but ...
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@1741
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Sun, 2 Oct 2005 12:29:39 +0000 (12:29 +0000)]
more explicit logs
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@1740
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Fri, 30 Sep 2005 19:45:53 +0000 (19:45 +0000)]
put the list of all existing categories in a more visible location in the doc
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@1739
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Fri, 30 Sep 2005 12:50:52 +0000 (12:50 +0000)]
Bug fix: Do not use the same name 'count' for two different variables (one
being the total amount of copied data while the other is the array size)
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@1738
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Fri, 30 Sep 2005 11:50:25 +0000 (11:50 +0000)]
allow the size of multidimentional objects to be given thru annotations in parsing macro
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@1737
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Fri, 30 Sep 2005 07:05:29 +0000 (07:05 +0000)]
Damn. It's harder than expected. First finish this damn article and come back later
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@1736
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Sun, 25 Sep 2005 06:38:45 +0000 (06:38 +0000)]
Don't forget to run the ./msg_test_surfxml_bypassed test
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@1735
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Sun, 25 Sep 2005 06:37:42 +0000 (06:37 +0000)]
let it run
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@1734
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Fri, 23 Sep 2005 19:08:40 +0000 (19:08 +0000)]
Let's stick to ascii and avoid non-breaking spaces
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@1733
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
alegrand [Wed, 21 Sep 2005 16:23:22 +0000 (16:23 +0000)]
Let's document my modifications by myself !!! :)
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@1732
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Wed, 21 Sep 2005 14:32:29 +0000 (14:32 +0000)]
Make the error message both correct and explicit about what's going wrong (%s is the whole block which should be name:value, not only the option name since we failed to parse the name at this point)
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@1731
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
alegrand [Wed, 21 Sep 2005 14:10:54 +0000 (14:10 +0000)]
"Malformated" is not correct and malformation is only for fetal development. :)
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@1730
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
alegrand [Wed, 21 Sep 2005 14:05:32 +0000 (14:05 +0000)]
help compilation on grid5000 with the paranoid compilation mode...
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@1729
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Wed, 21 Sep 2005 11:21:45 +0000 (11:21 +0000)]
Let the hack to make solaris happy work even when bash is not installed, and when /bin/sh points to a *really* old shell not setting $SHELL correctly
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@1728
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
alegrand [Wed, 21 Sep 2005 10:43:35 +0000 (10:43 +0000)]
a few casts to help compilation with gcc-4.0
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@1727
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
alegrand [Tue, 20 Sep 2005 12:37:41 +0000 (12:37 +0000)]
solving the pthread problem. Henri, give it a try on your Mac. I'd really like to know whether it solved your problem or not...
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@1726
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Mon, 19 Sep 2005 20:45:34 +0000 (20:45 +0000)]
Damn *slow* m68k
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@1725
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Fri, 9 Sep 2005 10:51:10 +0000 (10:51 +0000)]
Kill an internal function marked 'KILLME' and surviving from the times where libdata didn't exist
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@1724
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Fri, 9 Sep 2005 10:40:37 +0000 (10:40 +0000)]
Document last changes (gras perf improvement) and some cosmetics
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@1723
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Fri, 9 Sep 2005 10:39:39 +0000 (10:39 +0000)]
That's done, now
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@1722
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Fri, 9 Sep 2005 10:38:38 +0000 (10:38 +0000)]
Do not free the datacontainer on dynar_reset. The corresponding perf gain exactly the point of this function, isn't it?
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@1721
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Fri, 9 Sep 2005 10:37:40 +0000 (10:37 +0000)]
Yet another dataset regeneration (some machines need a loooong time to recompile SG)
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@1720
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Fri, 9 Sep 2005 10:34:50 +0000 (10:34 +0000)]
Perf improvement: Change libdata to a set so that we can search for stuff by ID (and thus reduce the insane amount of dict lookups)
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@1719
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Fri, 9 Sep 2005 10:33:17 +0000 (10:33 +0000)]
Do not use IOV for now, it doesn't seem implemented here, leading to serious performance *loss*
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@1718
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Fri, 9 Sep 2005 10:31:08 +0000 (10:31 +0000)]
Perf improvement: Change libdata to a set so that we can search for stuff by ID (and thus reduce the insane amount of dict lookups); Kill some dead code
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@1717
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Fri, 9 Sep 2005 10:25:06 +0000 (10:25 +0000)]
Perf improvement: Change libdata to a set so that we can search for stuff by ID (and thus reduce the insane amount of dict lookups)
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@1716
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Fri, 9 Sep 2005 10:19:46 +0000 (10:19 +0000)]
Perf improvement: reduce the amount of cbps creation/destruction by making it static and using a (newly created) cbps_reset (based on dynar_reset)
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@1715
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Fri, 9 Sep 2005 09:40:25 +0000 (09:40 +0000)]
Move some stuff unrelated to dynar out of dynar.h (into misc.h until I find a better place)
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@1714
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Thu, 8 Sep 2005 22:49:05 +0000 (22:49 +0000)]
documentation fix
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@1713
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Thu, 8 Sep 2005 22:07:59 +0000 (22:07 +0000)]
Rename BEGIN_DECL since we changed the usual definition to a macro with arguments (since the usual version did fool doxygen)
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@1712
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Thu, 8 Sep 2005 21:43:59 +0000 (21:43 +0000)]
this one comes from ia64
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@1711
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Thu, 8 Sep 2005 21:32:47 +0000 (21:32 +0000)]
Regenerate this one on AIX
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@1710
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Thu, 8 Sep 2005 21:26:35 +0000 (21:26 +0000)]
This data is used within the TRY/CATCH block; mark it volatile to avoid issues on non-x86
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@1709
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Thu, 8 Sep 2005 21:11:28 +0000 (21:11 +0000)]
Change an assert to an exception raising point (to ease debugging)
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@1708
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Thu, 8 Sep 2005 20:54:22 +0000 (20:54 +0000)]
Write a changelog entry for Henry
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@1707
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Thu, 8 Sep 2005 20:52:44 +0000 (20:52 +0000)]
Officially start the work on next release
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@1706
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Thu, 8 Sep 2005 20:10:21 +0000 (20:10 +0000)]
Regenerate some of the datatest after last wire protocol changes
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@1705
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Thu, 8 Sep 2005 20:07:57 +0000 (20:07 +0000)]
Be prehistoric-compiler-friendly
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@1704
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Thu, 8 Sep 2005 19:37:43 +0000 (19:37 +0000)]
Deal with the fact that some prehistoric arches don't have vector IO
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@1703
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Thu, 8 Sep 2005 19:31:33 +0000 (19:31 +0000)]
Load the headers we need, don't rely on some random other ones to load this one
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@1702
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Thu, 8 Sep 2005 19:29:17 +0000 (19:29 +0000)]
Adapt to recent changes in transport layer
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@1701
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Thu, 8 Sep 2005 18:54:47 +0000 (18:54 +0000)]
Rework the Transport layer to simplify it and improve its performance
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@1700
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Thu, 8 Sep 2005 18:43:26 +0000 (18:43 +0000)]
Rework the hexadecimal displayer I use to debug the Transport layer, and give it a proper header file
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@1699
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Thu, 8 Sep 2005 14:41:36 +0000 (14:41 +0000)]
reduce verbosity
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@1698
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Thu, 8 Sep 2005 14:40:55 +0000 (14:40 +0000)]
Check whether the readv function is available or not
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@1697
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Thu, 8 Sep 2005 08:06:26 +0000 (08:06 +0000)]
Fixup the context build
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@1696
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Mon, 5 Sep 2005 22:55:04 +0000 (22:55 +0000)]
The buffered transport is now RL only (will allow a zero-copy version)
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@1695
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Mon, 5 Sep 2005 22:46:23 +0000 (22:46 +0000)]
Tweak gras_trp_*_recv() prototype. This is now (sock, char*data,int size, int bufsize). Size is the data I absolutely want to get, and buffsize is the size of my buffer in which I copy them. This allows the buffer transport to get more data than mandatory of tcp and file transports (thus minimizing the call to the syscall read() without having to separately exchange the buffer size)
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@1694
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Mon, 5 Sep 2005 22:39:19 +0000 (22:39 +0000)]
SG don't get buffered by default (since only measurement sockets use it now)
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@1693
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Mon, 5 Sep 2005 22:38:24 +0000 (22:38 +0000)]
Avoid a useless bprintf (and test whether NULL msg exceptions are ok)
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@1692
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Mon, 5 Sep 2005 22:37:51 +0000 (22:37 +0000)]
msg now has a RL and a SG specific files
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@1691
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Mon, 5 Sep 2005 22:37:14 +0000 (22:37 +0000)]
A new little function I did need
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@1690
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Mon, 5 Sep 2005 22:36:25 +0000 (22:36 +0000)]
make sure no variable seem to be used uninitialized to this stupid gcc which gets fooled by TRY/CATCH blocks
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@1689
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Mon, 5 Sep 2005 22:35:30 +0000 (22:35 +0000)]
Test gras_datadesc_copy(); make sure no variable seem to be used uninitialized to gcc; reorder the tests to test the easier first
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@1688
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
mquinson [Mon, 5 Sep 2005 22:32:30 +0000 (22:32 +0000)]
Make sure that the code still compiles with the freaking paranoid gcc warning options, now that they can be turned on again (mainly assigning dummy values to variables which get initialised within a TRY/CATCH block)
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk@1687
48e7efb5-ca39-0410-a469-
dd3cf9ba447f
