| |||||||||||||
RCU and open file accountingRCU and open file accountingPosted Mar 9, 2006 15:45 UTC (Thu) by swiftone (guest, #17420)Parent article: RCU and open file accounting It is a fairly significant patch to merge at this late point in the cycle, I thought the 2.6 series is currently at an "ongoing" cycle, with no 2.7 planned. That would mean that the only way for significant changes to show up would be to do exactly this -- percolate through the -mm tree and get into the mainline kernel. If I have a flawed understanding of the current kernel design process, please let me know.
(Log in to post comments)
RCU and open file accounting Posted Mar 9, 2006 16:23 UTC (Thu) by copsewood (subscriber, #199) [Link] If 2.6.16 can't be destabilised in this way, the alternative would probably be for this patch to go into 2.6.17.
In the cycle towards 2.6.16 Posted Mar 9, 2006 16:36 UTC (Thu) by AnswerGuy (guest, #1256) [Link] ... I have to suspect that Jon was speaking to the upcoming 2.6.16 release.
There's been a concerted effort to keep the larger patches in the early builds so it would have been more appropriate for a 2.6.15.{1,2,3} ... and not for 2.6.16-rc6 ... for example.
Anyway, it could wait until 2.6.16.1 ... there aren't that many users with hundreds of CPUs and over 10GiB of RAM that need to worry about make -j8192 running out of steam.
JimD
In the cycle towards 2.6.16 Posted Mar 9, 2006 17:36 UTC (Thu) by madscientist (subscriber, #16861) [Link] I'm quite confident that there's no way a patch like this would ever be considered for a bugfix release like 2.6.16.1. It's far too big and intrusive, and, honestly, not severe enough to justify the risk.
It'll either squeek into 2.6.16, or be deferred until 2.6.17.
In the cycle towards 2.6.16 Posted Mar 9, 2006 19:05 UTC (Thu) by PaulMcKenney (subscriber, #9624) [Link] Looks to me like it actually did go into the 2.6.16 stream.
|
|||||||||||||