Hello, folks and welcome once again to Kernel Weekly News!
-We start this week with Guan Xuetao’s pull request which adds support for a new hardware
architecture, the unicore32. Here’s the sparse description :
UniCore32 ISA is defined and developed by Micro-Processor R&D Center of Peking University, and over the years, the CPUs and SoCs using UniCore32 ISA have been successfully applied in a variety of products in China.
-Takashi Iwai proposes sound fixes for -rc5, Thomas Gleixner has RTC updates (a quite large update, it
seems), John W. Linville updates the wireless tree (for .38), there are oprofile updates from Robert Richter,
aimed @ 2.6.38, Chris Mason has btrfs updates and Dan Williams has quite a lot of fixes for -rc5 regarding
dmaengine.
-Ingo Molnar has x86 updates in a git pull request that will probably be used in 2.6.39, while he also has
perf fixes and a timer and a locking fix; Dave Airlie announces drm fixes with the following :
Slightly bigger than I'd prefer, but we got a lot of endian fixes for radeon spread out all over the place, Alex has hopefully discovered the correct answer for the PLL nightmares, also a race condition on GPU reset that was causing an oops and full machine hang. Usually smattering of Intel fixes, RC6 power saving is off again due to hard hangs on some machines, and a resume regression fix. Hopefully along with Len's ACPI pull we are rid of most of the regressions now.
-Linus Torvalds, as almost every week, announces another release candidate, this time
2.6.38-rc5 :
Another week, another -rc. Nothing much stands out here - we've fixed some regressions (including, hopefully, a BUG_ON() due to the new RCU filename lookup that Ubuntu people saw), and things look fairly normal. Most of the changes are pretty spread out and small, with drivers/gpu (radeon and i915) somewhat standing out from the pack. And to a lesser degree drivers/rtc (converting some missed drivers to the alarm_irq_enable methods) along with some ext4 updates (mainly the "serialize unaligned async DIO" thing) standing out as being larger than average. There's also some device tree documentation movement that looks huge if you look at traditional patches, but it's just moving files out of the powerpc directory (since the whole device tree thing is generic these days). But most of it is pretty small. The appended ShortLog gives some flavor of it,
The shortlog isn’t that short, so you are invited to use the web to see it.
-James Bottomley proposes a git pull for the SCSI subsystem, saying
This is slightly bigger than I would like. It's mostly small fixes, especially the module removal oops related ones (slub poisoning oops on module removal and configfs shutdown oops) in the new target stuff. The big diff is the removal of a /proc interface in target (the mib code) that shouldn't have made it in in the first place. We need to do this in -rc so we don't get tied into a deprecated interface (it will be reintroduced at a /sys based interface in one of the merge windows). The patch is available here: master.kernel.org:/pub/scm/linux/kernel/git/jejb/scsi-rc-fixes-2.6.git
-Dave Miller has a pull req for sparc, Greg Kroah-Hartman announced 2.6.32.29,
2.6.36.4 and 2.6.37.1, Arnaldo Carvalho de Melo has perf improvements and
fixes and Dave Airlie has a git pull request for drm.
That’s it, I’m out till next week, same time as always, have a good weekend!
