Hello and welcome!
-Arnaldo Carvalho de Melo has perf tools fixes :
Anton Blanchard (6): perf symbols: Fix ppc64 SEGV in dso__load_sym with debuginfo files perf symbols: /proc/kallsyms does not sort module symbols perf symbols: Preserve symbol scope when parsing /proc/kallsyms perf symbols: Add some heuristics for choosing the best duplicate symbol perf symbols: Synthesize anonymous mmap events perf sort: Fix symbol sort output by separating unresolved samples by type Arnaldo Carvalho de Melo (1): perf top: Fix userspace sample addr map offset Darren Hart (1): perf tools: Add support for disabling -Werror via WERROR=0 David Ahern (2): perf record: Create events initially disabled and enable after init perf tool: Fix endianness handling of u32 data in samples Masami Hiramatsu (1): perf probe: Fix regression of variable finder Stephane Eranian (1): perf symbols: Fix issue with binaries using 16-bytes buildids (v2)
-H. Peter Anvin has updates on the kernel.org situation :
Hi All, As you know, we've been working very hard on building a new kernel.org infrastructure from the ground up. This new infrastructure will no longer have shell access to the git repositories; instead we will be running git using the gitolite web glue. Gitolite uses ssh keys to push into it, so we will start sending out new ssh credentials to the active developers who had kernel.org accounts before - please contact me or John so we can establish secure communications (ideally through gpg keys). One disadvantage of moving away from shell access is that gitolite has very limited on-server scripting - so for those of you using git hooks on kernel.org, you'll need to investigate how scripting works in gitolite before this will be functional. We are planning to get the first trees up and running early next week with broader access happening in early October. We will be bringing up additional developer services later, as resources permit and we can do so securely. We understand that the kernel.org outage has inconvenienced each of you as well as the larger Linux community. And, adjusting to the new tools will take some additional time and hassle. We apologize for that. But I want to assure you that getting the kernel.org development machines back up is the highest priority of kernel.org.
-Guenter Roeck has hwmon updates for 3.1 :
Chris Peterson (1): hwmon: (w83791d) rename prototype parameter from 'register' to 'reg' Guenter Roeck (1): hwmon: (coretemp) Don't use threshold registers for tempX_max Jan Beulich (3): hwmon: (coretemp) don't use kernel assigned CPU number as platform device ID hwmon: (coretemp) constify static data hwmon: (coretemp) remove struct platform_data * parameter from create_core_data() Jean Delvare (2): hwmon: (coretemp) Drop duplicate function get_pkg_tjmax hwmon: (coretemp) Let the user force TjMax Roland Stigge (1): hwmon: (ds620) Fix handling of negative temperatures
-John W. Linville has wireless fixes, mostly consisting of
Johannes Berg (1): iwlagn: fix dangling scan request John W. Linville (1): Merge branch 'master' of git://git.infradead.org/users/linville/wireless into for-davem Jouni Malinen (1): cfg80211: Fix validation of AKM suites Larry Finger (1): rtlwifi: rtl8192cu: Fix unitialized struct Mohammed Shafi Shajakhan (1): ath9k: Fix a dma warning/memory leak Rajkumar Manoharan (1): ath9k_hw: Fix Rx DMA stuck for AR9003 chips Stanislaw Gruszka (2): iwlegacy: fix command queue timeout iwlegacy: do not use interruptible waits
-Linus announces -rc8 :
Another week, another -rc at git://github.com/torvalds/linux and this time it really is quite noticeably smaller. The diffstat is pretty tiny, with some coretemp and clock_ops patches standing out, along with a small perf-tool update and everything else being pretty much one- or few-liners. And not that many of those either. If you are an automount users (either the autofs variety or the nfs one), please give this a try. There's been some changes in the automount logic: they are small and fairly obvious, and I doubt anybody will notice anything at all, but I'd still ask automount people to try it out. Other than that, it really is getting quite calm. We're _still_ discussing some USB unplug issues in the block layer, but there are patches floating around. So if somebody can actually trigger problems reliably, please contact Jens and Cc me. Other than that - please just test it out for any surprises.. Linus --- Alex Deucher (1): drm/radeon/kms: fix DDIA enable on some rs690 systems Anton Blanchard (6): perf symbols: Fix ppc64 SEGV in dso__load_sym with debuginfo files perf symbols: /proc/kallsyms does not sort module symbols perf symbols: Preserve symbol scope when parsing /proc/kallsyms perf symbols: Add some heuristics for choosing the best duplicate symbol perf symbols: Synthesize anonymous mmap events perf sort: Fix symbol sort output by separating unresolved samples by type Arnaldo Carvalho de Melo (2): perf top: Fix userspace sample addr map offset perf python: Add missing perf_event__parse_sample 'swapped' parm Avi Kivity (1): KVM: x86 emulator: fix Src2CL decode Axel Lin (2): ASoC: bf5xx-ad73311: Fix prototype for bf5xx_probe TPM: tpm_nsc: Fix a double free of pdev in cleanup_nsc Ben Hutchings (2): ALSA: fm801: Fix double free in case of error in tuner detection ALSA: fm801: Gracefully handle failure of tuner auto-detect Chris Peterson (1): hwmon: (w83791d) rename prototype parameter from 'register' to 'reg' Christian Borntraeger (2): [S390] kvm: fix address mode switching [S390] kvm: extension capability for new address space layout Darren Hart (1): perf tools: Add support for disabling -Werror via WERROR=0 Dave Airlie (1): Revert "drm/radeon/kms: fix typo in r100_blit_copy" David Ahern (2): perf record: Create events initially disabled and enable after init perf tool: Fix endianness handling of u32 data in samples David Henningsson (1): ALSA: HDA: No power nids on 92HD93 David Rientjes (1): thp: fix khugepaged defrag tunable documentation Fabio Estevam (1): spi/imx: Fix spi-imx when the hardware SPI chipselects are used Geert Uytterhoeven (2): zorro: Defer device_register() until all devices have been identified TPM: TCG_ATMEL should depend on HAS_IOPORT Guenter Roeck (1): hwmon: (coretemp) Don't use threshold registers for tempX_max Heiko Stuebner (1): ARM: S3C2443: Fix bit-reset in setrate of clk_armdiv Jan Beulich (3): hwmon: (coretemp) don't use kernel assigned CPU number as platform device ID hwmon: (coretemp) constify static data hwmon: (coretemp) remove struct platform_data * parameter from create_core_data() Jarkko Nikula (1): ASoC: omap-mcbsp: Do not attempt to change DAI sysclk if stream is active Jean Delvare (2): hwmon: (coretemp) Drop duplicate function get_pkg_tjmax hwmon: (coretemp) Let the user force TjMax Jeff Harris (1): spi: Fix WARN when removing spi-fsl-spi module Lars-Peter Clausen (1): ASoC: ssm2602: Re-enable oscillator after suspend Linus Torvalds (3): vfs pathname lookup: Add LOOKUP_AUTOMOUNT flag vfs: remove LOOKUP_NO_AUTOMOUNT flag Linux 3.1-rc8 Marek Szyprowski (1): ARM: S5P: fix incorrect loop iterator usage on gpio-interrupt Mark Brown (2): ASoC: Remove bitrotted wm8962_resume() ASoC: Ensure we generate a driver name Masami Hiramatsu (1): perf probe: Fix regression of variable finder Oleg Nesterov (1): ptrace: PTRACE_LISTEN forgets to unlock ->siglock Peter Huewe (2): TPM: Call tpm_transmit with correct size TPM: Zero buffer after copying to userspace Rafael J. Wysocki (1): PM / Clocks: Do not acquire a mutex under a spinlock Randy Dunlap (2): scsi: SCSI_ISCI needs to select SCSI_SAS_HOST_SMP, fixes build error scsi: fix qla2xxx printk format warning Roland Stigge (1): hwmon: (ds620) Fix handling of negative temperatures Russell King (3): ARM: nommu: fix warning with checksyscalls.sh ARM: fix vmlinux.lds.S discarding sections ARM: dma-mapping: free allocated page if unable to map Stephane Eranian (1): perf symbols: Fix issue with binaries using 16-bytes buildids (v2) Sylwester Nawrocki (2): ARM: S5PV210: Rename sclk_cam clocks for FIMC media driver ARM: EXYNOS4: Rename sclk_cam clocks for FIMC driver Takashi Iwai (2): ALSA: hda/realtek - Avoid bogus HP-pin assignment ALSA: hda/realtek - Don't detect LO jack when identical with HP Thomas Pfaff (2): ALSA: usb-audio - clear chip->probing on error exit ALSA: usb-audio: Check for possible chip NULL pointer before clearing probing flag Trond Myklebust (1): VFS: Fix the remaining automounter semantics regressions Will Deacon (2): ARM: 7091/1: errata: D-cache line maintenance operation by MVA may not succeed ARM: 7099/1: futex: preserve oldval in SMP __futex_atomic_op Zhao Jin (1): KVM: MMU: fix incorrect return of spte
-David Miller announces networking updates :
1) Fix two SKB leaks in ipv6 multicast cost, from Ben Greear. 2) BNX2X needs to hit the registers differently for HW Attention and Wake-On-Lan, from Dmitry Kravkov. 3) Missing break statement in dcbnl handling of bnx2x, fix from Shmulik Ravid. 4) dst_alloc() return value not checked in IPSEC, fix from Madalin Bucur. 5) Fix calculation of do_bcast in batman-adv stack, from Antonio Quartulli. 6) Wireless: a) Fix DMA getting stuck on ath9k AR9003 chips, from Rajkumar Manoharan. b) Mirror iwlagn command queue timeout fix in iwlegacy, from Johannes Berg. c) Similarly, mirror iwlagn interruptible wait fix in iwlegacy, from Stanislaw Gruszka. d) NL80211_ATTR_AKM_SUITES netlink blobs were not validated using the correct variable, fix from Jouni Malinen. e) iwl_scan_request() leaves ->scan_request and ->scan_vif dangling on failure, causing crashes later, fix from Johannes Berg. f) tcb_desc needs to be initialized fully in rtlwifi USB, from Larry Finger. g) DMA mappings left dangling on module unload in ath9k driver, from Mohammed Shafi. Another bug found by the DMA debugging code. Please pull, thanks a lot!
-James Bottomley has scsi fixes for -rc8 :
James Bottomley (1): 3w-9xxx: fix iommu_iova leak Mark Salyzyn (1): libsas: fix failure to revalidate domain for anything but the first expander child. Neil Horman (1): cxgb3i: convert cdev->l2opt to use rcu to prevent NULL dereference Randy Dunlap (1): scsi: qla4xxx needs libiscsi.o Vasily Averin (1): aacraid: reset should disable MSI interrupt
-Mauro Carvalho Chehab announces few media updates, mainly uvc and omap3 :
Archit Taneja (1): [media] OMAP_VOUT: Fix build break caused by update_mode removal in DSS2 Dave Young (1): [media] v4l: Make sure we hold a reference to the v4l2_device before using it Hans Verkuil (1): [media] v4l: Fix use-after-free case in v4l2_device_release Joerg Roedel (1): [media] omap3isp: Fix build error in ispccdc.c Laurent Pinchart (1): [media] uvcvideo: Fix crash when linking entities Ming Lei (1): [media] uvcvideo: Set alternate setting 0 on resume if the bus has been reset
-Greg Kroah-Hartman announces 3.0.4 in a somehow unusual way :
As I'm getting more and more requests every day for the latest stable kernel patch, I figured it's best if I just email it out to the list, so that everyone can have a copy of it without having to dig for the past 4 patches to have something that works. So here's the 3.0.4 patch, which can be applied on top of the 3.0 release from Linus. You can get a copy of the 3.0 release from various places around the web, I'll not be able to mail that to the whole list. I'm also signing this message, with my "traditional" GPG key, which is in the web of trust, so as to try to provide a bit of validity to this message. If anyone has any questions about this, please let me know.
The patch is too long to paste here, so I suggest you go here
https://lkml.org/lkml/2011/9/28/430 to see it.
-Martin Schwidefsky of IBM has -rc7 patches for, you guessed it, s390 :
Carsten Otte (1): [S390] gmap: always up mmap_sem properly Martin Schwidefsky (1): [S390] Do not clobber personality flags on exec Peter Oberparleiter (1): [S390] cio: fix cio_tpi ignoring adapter interrupts
-Keith Packard has drm-intel fixes, Arnaldo Carvalho de Melo has perf improvements/fixes,
and these are the final news for this week. See you in October, everyone. :-)