| |||||||||||||
|
| |||||||||||||
The 2.6.31 kernel is out
Linus has released the 2.6.31 kernel. A
few of the major features in 2.6.31 include performance counter support, the
"fsnotify" notification infrastructure, kernel mode setting for ATI Radeon
chipsets, the kmemleak tool,
char drivers in user space
support, USB 3 support, and much more. As always, see the KernelNewbies 2.6.31
page for a much more exhaustive list.
(Log in to post comments)
The 2.6.31 kernel is out Posted Sep 10, 2009 9:46 UTC (Thu) by neilbrown (subscriber, #359) [Link]
Is this a first? Linus releasing a kernel just *Before* a weekly edition of lwn, rather than just *After* ???
The 2.6.31 kernel is out Posted Sep 10, 2009 12:57 UTC (Thu) by corbet (editor, #1) [Link] An hour later and it would have gone the other way. I think he was trying for the usual post-publication release, but forgot that we're not in the same time zone...:)
The 2.6.31 kernel is out Posted Sep 10, 2009 10:54 UTC (Thu) by ejmarkow (guest, #56170) [Link]
Now have it compiled and installed on my box. Very nice Kernel release! It just keeps getting better and better. Congrats.
The 2.6.31 kernel is out Posted Sep 10, 2009 12:02 UTC (Thu) by nowster (subscriber, #67) [Link]
Unfortunately the one line fix for cdc_acm didn't make it.
Patch: http://bugzilla.kernel.org/attachment.cgi?id=23046
Bug report: http://bugzilla.kernel.org/show_bug.cgi?id=14103
The 2.6.31 kernel is out Posted Sep 10, 2009 15:21 UTC (Thu) by ledow (guest, #11753) [Link]
Not every patch can make it in, not every patch has even been checked, not every kernel release can be held up while patches are double-checked (Hell, that'd mean there'd never *be* another kernel release). Even a one-line patch can break things seriously for a lot of people, and if it's not in and checked by the deadline it doesn't get it.
I never understand why people get so ratty that a patch that fixes *their* problem doesn't make it in - you're in possession of the patch by definition (and thus anyone else with the same hardware will notice the same problem and find the same patch) and it'll get in eventually if it's that important.
The 2.6.31 kernel is out Posted Sep 10, 2009 15:47 UTC (Thu) by larryr (guest, #4030) [Link]
I think it is pretty reasonable and typical to hold releases for P1 regressions with known, trivial fixes.
The 2.6.31 kernel is out Posted Sep 10, 2009 16:49 UTC (Thu) by mingo (subscriber, #31122) [Link] This bug was reported 9 days ago - that's borderline even for trivial fixes and generally they dont make it upstream unless affecting a significant number of users. Note how late it was apparently reported despite it being triggerable in -rc1 already [which was released on June 24th] - suggesting that it either was not critical enough for people to report sooner, or was not hit by enough people (or by the right people) to be reported in time. (or somehow it was reported but did not get on the regression list. Cc: Rafael to make sure it gets on the regression list.) Bugs that get reported in time and have trivial fixes do get fixed in the release. Also note that if the fix is trivial it will go into v2.6.31.1 - which will be released at around the same time a delayed v2.6.31 would have been released. So there's no delay for you in terms of getting a fixed stable kernel.
The 2.6.31 kernel is out Posted Sep 11, 2009 13:58 UTC (Fri) by ariveira (guest, #57833) [Link]
> Bugs that get reported in time and have trivial fixes do get fixed in the release.
Well I have reported this two releases ago and I'm still waiting XD
http://bugzilla.kernel.org/show_bug.cgi?id=13362
Maybe is just that my hard is foobared but it is annoying to carry the revert...
The 2.6.31 kernel is out Posted Sep 11, 2009 19:15 UTC (Fri) by mb (subscriber, #50428) [Link]
> > Bugs that get reported in time and have trivial fixes do get fixed in the release.
> Well I have reported this two releases ago and I'm still waiting XD > http://bugzilla.kernel.org/show_bug.cgi?id=13362
I think you missed the "and have trivial fixes" part...
The 2.6.31 kernel is out Posted Sep 11, 2009 19:37 UTC (Fri) by ariveira (guest, #57833) [Link]
A one line revert[1] is a trivial fix from my no developer point of view.
But I may be very well mistaken as I'm not a developer.
[1] And i have read multiple times that a regression with a bisected culprit
Sooner is better Posted Sep 10, 2009 19:32 UTC (Thu) by ncm (subscriber, #165) [Link] If they'd taken time to fix that, the announcement would have gone out after this issue of LWN went to press. I know which I prefer.
|
|
||||||||||||