| 01:06.58 | *** join/#debianppc p8m (~p8m@162.244.177.51) |
| 01:41.07 | *** join/#debianppc neantpastel (~neantpast@dsl-10-148-173.b2b2c.ca) |
| 12:07.49 | *** join/#debianppc mauricfo (~mauricfo@177.44.49.168) |
| 12:25.13 | *** join/#debianppc klebers (Kleber@nat/ibm/x-wpzlpuoocwdvoqgx) |
| 14:07.00 | *** join/#debianppc tdaitx_ (~tdaitx@179.181.68.113) |
| 14:48.34 | *** join/#debianppc camm` (~user@nat-tvwna-outside-visitornet2-d-8.princeton.org) |
| 14:49.36 | camm` | anyone here work on gcc ppc64el? |
| 15:59.12 | *** join/#debianppc bikerr (~bike@230.41.92.62.static.cust.telenor.com) |
| 16:31.39 | *** join/#debianppc thierry-ibm (tfauck@nat/ibm/x-jckvydlergydczhn) |
| 16:37.38 | hartb | camm`, I'm sure we can find someone; what's up? |
| 16:56.25 | camm` | hartb: Finally getting around to filing my gcc bugs, one simple at -O0, one deep in floating point at default optimization, and wondered if someone was available to field questions. |
| 17:05.55 | hartb | camm`, looking for someone who can help. If want to ask the question here I can relay it if nothing else. But I have to head out for a bit; should be back in about 1.5 hours |
| 17:14.50 | *** join/#debianppc willschm (willschm@nat/ibm/x-mzeegjpoxhnpjqxh) |
| 17:14.56 | camm` | hartb: no worries -- let me get my data together and I'll ping you when ready. |
| 17:45.51 | *** join/#debianppc pankid (~pan@173-162-51-174-michigan.hfc.comcastbusiness.net) |
| 18:18.27 | *** join/#debianppc bikerr (~bike@230.41.92.62.static.cust.telenor.com) |
| 18:50.11 | bergner | camm`: if you open ppc64 LE gcc bugs, make sure to add myslef (bergner@gcc.gnu.org) and wschmidt@gcc.gnu.org to the bug's CC list |
| 18:50.39 | bergner | camm`: if you have just questions, you can start with me and I can forward on to others on my team if needed |
| 19:05.50 | hartb | bergner, thank you! |
| 19:15.41 | camm` | bergner: thanks so much. The first issue is that at -O0, register variables can be stored in e.g. $f31 and clobbered in function calls, i.e. they are not pushed/restored across the call. Is this a known issue, or should the details be posted? |
| 19:15.57 | camm` | (can't see it in the bug list) |
| 19:24.08 | bergner | f31 is a non-volatile register, so should be saved/restored. If you have a test case, then do open a bug |
| 19:24.28 | bergner | ...in https://gcc.gnu.org/bugzilla/ |
| 22:48.22 | *** join/#debianppc klebers (~Kleber@179.159.212.222) |
| 22:53.45 | *** join/#debianppc cascardo_ (~cascardo@2001:470:1f07:13f5:6a17:29ff:fe00:4f38) |
| 23:10.00 | *** join/#debianppc hartb (~hartb@32.97.110.56) |
| 23:25.02 | *** join/#debianppc tdaitx__ (~tdaitx@179.181.68.113) |