Hello,
Here goes -pre4, which contains networking update, IA64 update,
PPC update, USB update, bunch of knfsd fixes, amongst others.
And finally merge most important part of -aa VM. Those changes
are fixing some OOM deadlocks, give us better per-zone balancing
and better reclaiming. The OOM killer has been removed.
I’ve been using it on my 256MB desktop: performance feels much
better, but it needs extensive testing, so please help.
Summary of changes from v2.4.23-pre3 to v2.4.23-pre4
<adsharma:unix-os.sc.intel.com>:
- ia64: IA-32 compatibility patch: FP denormal handling
- ia64: Correct NR_CPUS/cpu_online test order in CMC/CPE
polling
- ia64: Remove partial semtimedop32 stuff from upstream
- ia64: Merge to newer ACPI CA
- ia64: sys_ia32.c needs linux/quotacompat.h
- ia64: tlb.c whitespace cleanup to follow 2.5
- ia64: make cpu_relax() a barrier to be consistent with 2.5
- ia64: kernel/acpi.c: Whitespace changes to follow 2.5
- ia64: MCA: pass GP physical address to SAL
- ia64: minor bugfixes and whitespace cleanup to follow 2.5
- ia64: MCA: Find correct offset of OEM data (from Keith
Owens) - ia64: sal.h: Backport spelling and other trivial changes from
2.5 - ia64: Comment changes to fix “correctable” usage
- ia64: Fix check for binutils that supports “hint”
instructions - ia64: Update configs for upstream changes
- ia64: Use ARRAY_SIZE(), fix formatting, remove static
initializers to zero - ia64 unwind: (unw_access_ar): initialize struct pt_regs *pt
before using it to get AR_CSD & AR_SSD - ia64: Update defconfig to new generic config
- ia64: initialize bootmem early for acpi_table_init()
- ia64: Use $(CC), not $(AS), when checking for “hint @pause”
support in binutils - ia64: Clarify ACPI available_cpus handling
- ia64: TRIVIAL: Remove extraneous ‘`’
- ia64: minstate.h: whitespace changes to reduce diffs with
2.5 - ia64: Fix minstate comments
- ia64: fix SAVE_RESET so OS INIT handler works again
- ia64: Remove AIC7XXX driver from ski defconfig
- 2.4 HCDP early printk support
<chas:nrl.navy.mil>:
- [ATM]: In atm_getaddr() do not copy_to_user() with locks
held
<daniel:deadlock.et.tudelft.nl>:
- Implement LCD display support in atyfb driver
<eric:lammerts.org>:
- fix current->user->processes leak in
reparent_to_init()
<erikj:subway.americas.sgi.com>:
- ia64: 9/3/2003 SGI update
<erlend-a:us.his.no>:
- [CRYPTO]: Add alg. type to /proc/crypto output
<joris:struyve.be>:
- unusual_devs.h entry
<karlis:mt.lv>:
- [BRIDGE]: kfree –> kfree_skb
<marcelo:logos.cnet>:
- Mehmet Ceyran/Alan Cox: Longer i810_audio.c retries
- aa VM merge: Per-zone watermark changes, add
lower_zone_reserve_ratio - aa VM merge: page reclaiming logic changes: Kills oom
killer - aa VM merge: Page accounting helpers changes
- aa VM merge: tunables
- aa VM merge: Kill PF_MEMDIE
- aa VM merge: Fixup page reclaiming changes patch
- Changed EXTRAVERSION to -pre4
- Cset exclude: [email protected]|ChangeSet|20030912113656|10550
<matthewc:cse.unsw.edu.au>:
- smpboot.c, acpi.c
Alan Cox:
- Fix ymfpci oops
Alex Williamson:
- ia64: Use PAL_HALT_LIGHT in cpu_idle
- ia64: New CMC/CPE polling
- ia64: Update to CMC/CPE polling
- ia64: Rename SAL_CALLSAFE to
SALCALL_REENTRANT
Arjan van de Ven:
- LSB compliance fix in mprotect
Arun Sharma:
- ia64: translate F_GETLK64/F_SETLK64 to F_GETLK/F_SETLK
- ia64: fix memory leak in sys32_execve path
Chas Williams:
- [ATM]: If clip isn’t a module don’t __MODDECUSE_COUNT()
- [ATM]: #define’ing pci_pool_create() breaks
CONFIG_MODVERSION - [ATM]: Backport lane/mpoa module locking cleanup from
2.6.x
David Mosberger:
- ia64: handle_fpu_swa() scaling fix
- ia64: Backtraces of all processes on INIT, warning cleanup
Greg Kroah-Hartman:
- USB: fix data toggle problem for pl2303 driver
- USB: update usb-serial.h with spelling fixes and get and set
functions - USB: backport some pl2303 B0 fixes
- USB: fix oops when yanking a usb-serial device from the system
with the port still opened - USB: fix copy_from_user call in acm.c
- USB: fix copy_from_user call in aiptek.c
- USB: fix copy_to_user call in uhci-debug.h
- USB: fix copy_to_user call in mdc800 driver
- USB: remove duplicated copy_from_user call in stv680
driver - USB: fix copy_to_user calls in vicam driver
Harald Welte:
- [NE
Full ChangelogTFILTER]: NAT range calculation fix
Jack Steiner:
- ia64: discontig/NUMA support
- ia64: Add ia64_imva() and a few more ia64_tpa() uses
- ia64: add support for non-identity mapped kernels
- ia64: remove some SN1 remnants, add a bit more SN2 support
Jean Tourrilhes:
- wireless extension update: 802.11a/802.11g fixes
Jens Axboe:
- Add NEC iStorage to SCSI blacklist
Keith M. Wesolowski:
- [SPARC32]: Ignore btfixups in .text.exit
Keith Owens:
- ia64: Clean up several warnings (no functional change)
- ia64: Correct typo in UNW_DPRINT() call
- ia64: Fix more UNW_DPRINT() typos
- ia64: Delete some generated ia64 files that were being left by
make mrproper
Marc-Christian Petersen:
- Fixup ‘make xconfig’ problem caused by fetchop Config.in
change
Martin Hicks:
- ia64: max user stack size of main thread configurable via
RLIMIT_STACK
Matthew Wilcox:
- ia64: return PCI domain for pci_controller_num()
Neil Brown:
- knfsd: Lock client list while detaching locks
- knfsd: Set d_op when creating a parent directory during nfsd
fh->dentry conversion - knfsd: lockd fails to purge blocked NLM_LOCKs
- Fix typo in umem.c
- knfsd: Make sure nfs/tcp socket only gets closed
- knfsd: Change name of a #define in nfsd to match 2.6
- knfsd: Make sure nfsd replies from the address the request was
sent to
Oleg Drokin:
- [2.4] Rocketport driver compile fix
Paul Fulghum:
- synclink update
- synclinkmp update
- synclink_cs update
- n_hdlc update
- synclink drivers fixup
Paul Mackerras:
- PPC32: Handle single-stepped emulated instructions
correctly - PPC32: Fix for highmem on machines with 64-bit PTEs (e.g.
PPC440) - PPC32: Simplify VMALLOC_START, make it just a variable
- PPC32: Fix a typo in the PPC 440GP support
- PPC32: Fix a bug where TLB entries didn’t get execute
permission on 40x
Ralf Bächle:
- avoid glibc conflict
Seth Rohit:
- ia64: use “hint @pause” in cpu_relax() and spinlock
contention - ia64: patch to use >256MB purges
- ia64: Restructure pt_regs and optimize syscall path
- ia64: Correct .unwabi for PT_REGS_SAVES (should be “3,
‘i'”)
Stephen Hemminger:
- [BRIDGE]: Clear hw checksum flags when bridging
Stéphane Eranian:
- ia64: Fix perfmon usage of rum/srsm and sum/ssm
Tom Rini:
- PPC32: Add Magic SysRq support to MPC8260 platforms
- PPC32: Minor bootwrapper fixups
Tony Luck:
- ia64: cleaning up the INIT code (Backported from 2.5 by Bjorn
Helgaas) - ia64: Trim granules correctly in efi_memmap_walk()