Not a lot to say, the ChangeLog says it all (and I include the
-rc2 log too, since I forgot to actually ever post it). Mostly
random smaller stuff all over. The big merges were all in rc1 and
do not seem to have caused any huge headaches.
Linus
—
Summary of changes from v2.6.1-rc2 to v2.6.1-rc3
Adam Belay:
- Fix PnP BIOS call
Andi Kleen:
- Fix interrupt routing problem on x86-64
- [COMPAT]: Handle SO_TIMESTAMP cmsgs
Andrew Morton:
- MSI build fixes
- fix sysfs oops
- JFS fix for NFS on little-endian systems
- dvb: firmware fixes
- ia32 sched_clock() deadlock fix
- /proc/ppc64 and /proc/iSeries fixes from Linas Vepstas
- ppc64: Add missing section definition
Bartlomiej Zolnierkiewicz:
- cmd640.c: fix PCI type1 access
- ide-tape.c: stop abusing rq->flags
- remove dead and broken DISK_RECOVERY_TIME support
- fix oopses on rmmod in some OSS drivers
Ciaran McCreesh:
- [SPARC64]: Fix CONFIG_DRM_FFB=y build
Dave Jones:
- [AGPGART] printk level changes for amd64
- [AGPGART] Add support for Radeon IGP345M to ATI GART
driver
Dave Kleikamp:
- don’t clear i_sb
David Dillow:
- [SUNZILOG]: Register the correct number of ports, ignore
keyb/mouse lines
David S. Miller:
- [NET]: In dev_kfree_skb_any() use dev_kfree_skb_irq() if
irqs_disabled() - [NET]: Un-deprecate skb_linearize(), we can re-deprecate in
2.7.x
Hideaki Yoshifuji:
- [IPV6]: Kill obsolete functions (ip6_frag_xmit() and
ip6_build_xmit())
Ingo Molnar:
- [NET]: Do type checking in {udp,inet6,raw6,inet}_sk()
James Morris:
- [NETFILTER]: Add SELINUX priority values for ipv4/ipv6,
approved by Harald Welte
Jean Tourrilhes:
- [IRDA]: Fix locking in the ircomm-shutdown path
Keith M. Wesolowski:
- [SPARC32]: Fix BUG on swapout on srmmu systems
Linus Torvalds:
- Fix ttpci bogus use of floating point by casting the constant
expression properly. - Fix silly mremap test
- Don’t relocate non-allocated regions in modules
Martin Devera:
- [NET]: Make sure that class selected by priority is a leaf in
HTB scheduler
Matthew Wilcox:
- [SPARC64]: Use drivers/block/Kconfig
- [SPARC32]: Use drivers/block/Kconfig
Mitsuru Kanda:
- Fix my PGP fingerprint in the CREDITS file
Roman Zippel:
- generate an error if writing of kernel config failed
- fix gconf segfault problem
- gconf compile warning fixes
- gconf startup fixes
- qconf fix
Stephen Hemminger:
- [AF_PACKET]: Convert to seq_file
- [DECNET]: Better way to prevent decnet module unload
- [NET]: Fix multiple eth0 mixed PCI/ISA init
Tom ‘spot’ Callaway:
- [SPARC]: Add missing MODULE_LICENSE tags to various Sparc
driver - [SPARC64]: Fix sun_uflash MTD driver build
- [SPARC]: Add placeholder asm-{sparc,sparc64}/setup.h so MTD
builds - [SPARC64]: Export sys_close for solaris emulation
Ville Nuorvala:
- [IPV6]: Autoconfig link-local address on ip6-ip6 tunnel
device
Summary of changes from v2.6.1-rc1 to v2.6.1-rc2
Amit Gurdasani:
- [SERIAL] EISA ID for PnP modem
Andi Kleen:
- X86-64 merge
- Fix memset on x86-64
Bart De Schuymer:
- [BRIDGE NETFILTER]: Fix leaks and crashes in SKB handling
- [BRIDGE NETFILTER]: IPV6 needs the skb->nf_bridge leak fix
as well - [BRIDGE]: Fix build with vlan disabled, spurious ifdef
Ciaran McCreesh:
- [SPARC64]: Fix broken _PAGE_SZHUGE defines for 512K and
64K
Dave Jones:
- [CPUFREQ] Add support for 1GHz Centrino speedstep From: Youichi
Aso <aso@granite.phys.s.u-tokyo.ac.jp> - [AGPGART] Handle multiple AMD64 AGP bridges correctly on
UP - [AGPGART] Fix return check on request_mem_region() Do things
the way every other user of this function does. - [CPUFREQ] Fix powernow-k8 policy usage
- [CPUFREQ] Abort if there is a failure in aquiring “ownership”
of the SMI speedstep interface - [AGPGART] Fix two nasty bugs in the K8 AGP support
- [CPUFREQ] Use different attack with the Powernow-K7 bad bios
problems - [AGPGART] Fix MAX_HAMMER_GARTS off by one
- [CPUFREQ] Fix rounding in longhaul
- [AGPGART] Mask memory after allocation We missed a few cases
where we need to do this. - [AGPGART] Merge missing chunk of NVIDIA nForce agpgart
driver - [AGPGART] Remove duplicate programming of AGP command
register - [CPUFREQ] on P4s no TSC adjustment is necessary From Dominik
Brodowski - [CPUFREQ] Disable debug output in speedstep-smi driver
- [CPUFREQ] speedstep hcakers cnat spel
- [CPUFREQ] Disable smi_detect_freqs() call on systems which do
not support it [BUG #1422] Dominik Brodowski. - [CPUFREQ] Support for 533 MHz FSB in speedstep driver
- [CPUFREQ] Detect CPU speed without relying on cpu_khz
- [CPUFREQ] remove unneeded #ifdefs in
include/linux/cpufreq.h - [CPUFREQ] use latency in nanoseconds sometimes nanoseconds are
used, sometimes microseconds, sometimes even something else. - [CPUFREQ] Do something about “cpufreq: change failed”
- [CPUFREQ] do not leak memory in powernow-k8 From Pavel
Machek - [CPUFREQ] Typo fix in drivers/cpufreq/Kconfig
- [CPUFREQ] Missing .owner entry in speedstep-smi driver
David Mosberger:
- ia64: Fix a ptrace-bug that caused “strace -f” to crash the
inferior process. The root-cause of the problem was that ptrace()
tried to copy the portion of the register backing store that landed
on the kernel stack back to users-space, but the resulting state
was inconsistent if the inferior process was in the middle of a
system-call (as would always be the case for strace). - ia64: Rearrange ia64_do_signal() such that it is possible for a
debugger to cancel system-call restart. - ia64: Allow system-call number to be changed during system-call
tracing (both for native and x86 system call tracing). This is
needed by recent versions of strace and UML likes to do that,
too. - ia64: Remove the old ia64_ni_syscall()/sys32_ni_syscall()
routines which are overly verbose and replace them with calls to
sys_ni_syscall(). - ia64: fix perfmon bug causing lost samples
- ia64: Merge patch by Arun Sharma: hook up lots of ia32
syscalls - ia64: Rename efi_get_time() in the simulator’s
firmware-emulator to avoid name-clash with declaration of routine
of the same name in <linux/efi.h>. - ia64: Update defconfig
David S. Miller:
- [TG3]: Do not drop existing GRC_MODE_HOST_STACKUP when writing
to GRC_MODE - [TG3]: Do not set RX_MODEKEEPVLAN_TAG
when ASF is enabled - [TG3]: Clear on-chip stats/status block after resetting
flow-through queues - [TG3]: Update version and release date
- [TG3]: Update to latest non-5705 TSO firmware
- [TG3]: Update version and reldate
- [MEDIA]: ttusb_dec.c needs linux/init.h
- [SPARC64]: Update defconfig
François Romieu:
- [TG3]: Fix bogus return value in tg3_init_one()
Hideaki Yoshifuji:
- [NET]: Fix comment type in skbuff.h
James Bottomley:
- MSI broke voyager build
Jeroen Vreeken:
- [AX25]: Missing spin_unlock() and recvmsg reported dst instead
of src
Linus Torvalds:
- Remove generated files from revision control
- Remove dead files
- Fix IDE “PIO WRITE wait for ready” test under extreme interrupt
load - Don’t allow mremap of zero-sized areas
Michael Hunold:
- DVB: Update documentation and credits
- DVB: Fix feed list handling bugs in demux
- DVB: Fixes for frontend drivers
- DVB: Add static firmware compilation again
- DVB: Revamp of the TTUSB-DEC driver
- DVB: Fix memory usage of ttpci driver
Rene Herman:
- [SERIAL] add PnP ID to 8250_pnp.c
Russell King:
- Fix “echo -n 3 > /sys/…/power/state”
- [SERIAL] Remove old RSA resource handlign
- [ARM] Fix more gcc3 build errors
- [ARM] Fix cachepolicy=<foo>
- [ARM] Kill dma-isa.c warning
- [ARM] Report more detail when unable to resolve module
relocations
Xose Vazquez Perez:
- [TG3]: Add new device IDs