| --- Log opened Sat Jan 10 00:00:49 2015 | ||
| olofk | Trying to understand these gcc/libc interdependencies. Do I need to build a stage 2 gcc for every libc? | 12:31 |
|---|---|---|
| olofk | Something like this: | 12:31 |
| olofk | 1. binutils | 12:31 |
| olofk | 2. gcc stage 1 | 12:31 |
| olofk | for c in libc: | 12:31 |
| olofk | libc && gcc stage 2 | 12:31 |
| olofk | Worst attempt at pseudo code ever | 12:32 |
| olofk | Or is it: | 12:36 |
| olofk | 1. binutils | 12:36 |
| olofk | 2. for c in libc: | 12:36 |
| olofk | gcc stage 1 && libc && gcc stage 2 | 12:36 |
| poke53282 | binutils, linux-headers, gcc stage 1, libc, gcc stage 2 | 13:01 |
| poke53282 | stage 2 is least needed by uclibc, musl and glibc. | 13:03 |
| ysangkok | poke53282: when to use throw, and when to use message.Debug? some fatal errors use message.Debug, i don't understand that | 15:29 |
| poke53282 | message.Debug is used for everything. | 15:38 |
| poke53282 | And if his problem is critical also abort is executed. | 15:38 |
| poke53282 | One plan is have a new function message.Error, which sends the abort signal itself. | 15:39 |
| poke53282 | So, everytime, the abort function is called, change message.Debug to message.Error | 15:39 |
| ysangkok | ok | 15:45 |
| poke53282 | This are just a few lines of code. | 15:51 |
| poke53282 | I can implement it | 15:51 |
| ysangkok | poke53282: it would be useful if there was a way to invoke console.warn too... for example, if the file sizes don't match | 15:57 |
| mor1kx | [mor1kx] bandvig pushed 1 new commit to withfpu: https://github.com/openrisc/mor1kx/commit/ada2df5dabd711b43f8878f511001cca4b039c09 | 16:01 |
| mor1kx | mor1kx/withfpu ada2df5 Andrey Bacherov: Size optimization: share post-operation (add/sub, mul/div, i2f) align shifter . | 16:01 |
| poke53282 | ysangkok: You mean message.Warn | 16:02 |
| ysangkok | ah soo | 16:02 |
| ysangkok | all right :) | 16:02 |
| stekern | poke53282: the stdatomic-compare-exchange-1 fail seems to be related to ltoi | 16:03 |
| stekern | -i | 16:03 |
| stekern | it doesn't fail without -flto but does with | 16:04 |
| poke53282 | try with and without inlining | 16:04 |
| stekern | was that directed to me? | 16:07 |
| poke53282 | Yes | 16:08 |
| poke53282 | ysangkok: Added a commit. | 16:09 |
| stekern | ok, I don't understand | 16:09 |
| poke53282 | You are free to use those functions. | 16:09 |
| ysangkok | great :D | 16:09 |
| poke53282 | stekern: well I think, inlining and -flto ist partly related. | 16:11 |
| stekern | maybe, but there's nothing related to inlining in that test | 16:11 |
| poke53282 | Ahh, Ok# | 16:11 |
| stekern | http://pastie.org/9824059 | 16:12 |
| stekern | that's the part of the testcase that fails | 16:12 |
| poke53282 | Did you try it on your hardware? | 16:13 |
| stekern | and that atomic_compare_xxx call boils down to the atomic stuff blueCmd added | 16:13 |
| stekern | I'm running in or1ksim | 16:13 |
| stekern | I'm diffing the disasm of the failing and passing now | 16:14 |
| stekern | can't see any difference in the actual atomic related code that's generated though | 16:16 |
| stekern | so might be a general lto problem for us | 16:16 |
| poke53282 | diff with objdump -d ? | 16:22 |
| stekern | yes | 16:22 |
| ysangkok | poke53282: since the debug messages in the worker are relayed to the master using Send, it is hard to get backtraces for worker problems | 18:07 |
| ysangkok | poke53282: using console.warn, Chrome would show the correct line in the console in the right side... but this would require no indirection | 18:08 |
| ysangkok | poke53282: currently i just set a breakpoint on the worker debug function | 18:08 |
| poke53282 | Yes, the lines are lost. | 22:37 |
| poke53282 | But I had never problems with such things. The error message string is more than enough. | 22:38 |
| ysangkok | poke53282: new PR https://github.com/s-macke/jor1k/pull/51 | 23:21 |
| poke53282 | great, how did you change the base.xml files? | 23:31 |
| poke53282 | Ahh, manually I guess | 23:32 |
| poke53282 | Another hack gone. | 23:41 |
| --- Log closed Sun Jan 11 00:00:50 2015 | ||
Generated by irclog2html.py 2.15.2 by Marius Gedminas - find it at mg.pov.lt!