Creating nightly hwloc snapshot SVN tarball was a success.
Snapshot: hwloc 1.0a1r1823
Start time: Thu Mar 11 21:01:05 EST 2010
End time: Thu Mar 11 21:03:04 EST 2010
Your friendly daemon,
Cyrador
On Mar 11, 2010, at 3:57 PM, Samuel Thibault wrote:
> > What *is* hwloc's policy about thread safety, anyway?
>
> It would be odd to not be threadsafe, considering the target
> applications :) and I believe hwloc is already.
Hmm. I'm not so sure. Consider:
1. thread A calls hwloc_topology_ini
The hwloc_cpuset_next() will calculate the next cpu behind a given one.
Use hwloc_cpuset_first() and hwloc_cpuset_next() in hwloc_cpuset_foreach()
to sparsely iterate over a cpuset.
Regards,
Bert
---
doc/Makefile.am|1 +
include/hwloc/cpuset.h |9
It was not clear whether hwloc_cpuset_foreach_end() needs a terminating ';'
or not. And hwloc itself had both cases.
This converts hwloc_cpuset_foreach_{begin,end}() to use the do-while-(0)
idiom to enforce a terminating semi-colon.
Note: it is still possible to break and continue this loop.
Reg
Jeff Squyres, le Thu 11 Mar 2010 15:23:35 -0800, a écrit :
> On Mar 11, 2010, at 3:14 PM, Bert Wesarg wrote:
> > I do not know the policy from this library regarding thread safety, so
> > I decided to be on the safe side here.
>
> What *is* hwloc's policy about thread safety, anyway?
It would be
On Fri, Mar 12, 2010 at 00:18, Bert Wesarg wrote:
> On Fri, Mar 12, 2010 at 00:14, Bert Wesarg wrote:
>> On Fri, Mar 12, 2010 at 00:03, Brice Goglin wrote:
>>>
>>> Did you actually test this ? The way I am reading the manpage is that
>>> you need to open with setmntent and close with endmntent.
On Mar 11, 2010, at 3:14 PM, Bert Wesarg wrote:
> I do not know the policy from this library regarding thread safety, so
> I decided to be on the safe side here.
What *is* hwloc's policy about thread safety, anyway?
--
Jeff Squyres
jsquy...@cisco.com
For corporate legal information go to:
http:
On Fri, Mar 12, 2010 at 00:14, Bert Wesarg wrote:
> On Fri, Mar 12, 2010 at 00:03, Brice Goglin wrote:
>>
>> Did you actually test this ? The way I am reading the manpage is that
>> you need to open with setmntent and close with endmntent.
>
> I have read the man page, but only about getmntent, a
On Fri, Mar 12, 2010 at 00:03, Brice Goglin wrote:
>
> Did you actually test this ? The way I am reading the manpage is that
> you need to open with setmntent and close with endmntent.
I have read the man page, but only about getmntent, and than read the
source code for getmntent from glibc.
So y
Brice Goglin, le Fri 12 Mar 2010 00:03:37 +0100, a écrit :
> Also, isn't getmntent (without _r) enough here?
I guess it will have troubles if it is called concurrently in different
threads: the value returned by getmntent is typically not dynamically
allocated.
Samuel
Bert Wesarg wrote:
> This does multiple things at once:
>
> I) it uses getmntent_r(3) to parse lines from /proc/mounts
>
> II) while doing this, it uses the correct un-escape rules for this
> file format.
>
> The current code converts "\ " to " ", while linux uses a "\040" to
> " "
This does multiple things at once:
I) it uses getmntent_r(3) to parse lines from /proc/mounts
II) while doing this, it uses the correct un-escape rules for this
file format.
The current code converts "\ " to " ", while linux uses a "\040" to
" " escaping rule.
III) it accurate
Applied thanks!
Brice
Bert Wesarg wrote:
> When reading all tids from a process in
> topology-linux.c::hwloc_linux_get_proc_tids(), it used a
> exponential realloc algorithm to increase the storage size for the tids.
>
> Now it uses the number of links (.st_nlinks) from a stat() call to the
>
13 matches
Mail list logo