| === JanC_ is now known as JanC | ||
| doob | hi, I have a question regarding the state of perf in the xenial ubuntu kernel source | 06:38 |
|---|---|---|
| doob | While looking for the append option of the `perf record` command I noticed that this is missing in the current ubuntu kernel source of perf (I'm running 4.4.0-78-generic #99~14.04.2-Ubuntu SMP Thu Apr 27 18:49:46 UTC 2017 x86_64 x86_64 x86_64 GNU/Linux) | 06:39 |
| doob | This is surprising to me since the official sources seam to have this since version 2.6.31 https://github.com/torvalds/linux/commit/abaff32a03e26e5d6674cb2a26ad882efe7493a3 | 06:41 |
| doob | Is there anyone at the ubuntu kernel team especially responsible/taking care of maintaining `perf` to ask about why this is and whats the general state of perf in ubuntu? | 06:45 |
| apw | doob, hmmm, well noone speific looks after it. if it has an issue that needs investigating do file a bug against the kernel | 06:56 |
| apw | doob, as that is produced as part of the kernel build | 06:56 |
| doob | I was just wondering if it has just been forgotten somehow or if there is reason to patch away or not include this compared to the main kernel tree | 06:57 |
| doob | ok | 06:58 |
| apw | doob, we do not carry any patches for perf | 06:58 |
| doob | but do you compare/sync with the upstream kernel tree at some point regulary or does it mean you never ever get upstream code for perf as long as nobody files a bug report? | 06:59 |
| doob | I'm not sure if I got you right with "we do not carry any patches for perf" | 07:00 |
| apw | doob, perf is kernel version specific and carried without modification as far as i recall | 07:02 |
| doob | Ok. This is what I was always thinking how it is. This is why I'm supprised to not find the code in the ubuntu tree that is in the upstream tree since 2.6.31. | 07:03 |
| apw | doob, the source for 4.4 has the append option | 07:05 |
| doob | Am I looking at the wrong place? http://kernel.ubuntu.com/git/ubuntu/ubuntu-xenial.git/tree/tools/perf/builtin-record.c | 07:08 |
| apw | doob, oh no i am wrong, that is an append option in another command | 07:14 |
| apw | doob, the -A was removed a long time ago | 07:14 |
| apw | commit 563aecb2e63a16f86c7daabfdcfee23f3e7c5955 | 07:14 |
| apw | Author: Jiri Olsa <jolsa@redhat.com> | 07:14 |
| apw | Date: Wed Jun 5 13:35:06 2013 +0200 | 07:14 |
| apw | perf record: Remove -A/--append option | 07:14 |
| apw | It's no longer working and needed. | 07:14 |
| apw | Quite straightforward discussion/vote was in here: | 07:14 |
| apw | http://marc.info/?t=137028288300004&r=1&w=2 | 07:14 |
| apw | (the commit's words) | 07:14 |
| doob | Ok. Thx. In the meantime I also noted I did a mistake. I think I grabbed some very old commit message from GH and was misleaded by the branch information saying that it is in all the current branches. But of course the commit will stay there for ever even if the code has been removed. Sorry for the confusion. | 07:16 |
| apw | np | 07:18 |
| === klebers_ is now known as klebers | ||
| === klebers_ is now known as klebers | ||
| === dannf` is now known as dannf | ||
| === jdstrand_ is now known as jdstrand | ||
| === jdstrand is now known as jdstrand_ | ||
| === jdstrand_ is now known as jdstrand | ||
| === JanC_ is now known as JanC | ||
| dmj_s76 | sforshee: I was able to test the package for bug1686815 for the bluetooth issue. It fixes the bug. | 22:05 |
| jjohansen | sforshee: sorry for the dely on this, I finally tracked down the memory corruption, this passes the regression tests, except for long_path which is deliberately broken atm, and will have to be fixed in some follow on patches | 23:23 |
| jjohansen | http://paste.ubuntu.com/24583768/ | 23:23 |
| jjohansen | sforshee: I have included the fixes that are on 4.12, so it should be a clean rebase to the 4.12 kernel | 23:24 |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!