What will be in the x86-64/x86 2.6.21 merge


I will post the existing patches in batches for closer review.

Can be already all viewed at ftp://ftp.firstfloor.org/pub/ak/x86_64/quilt/patches/

Highlights:

- Using %fs instead of %gs for PDA register (Jeremy F.)
- Removed the old static rom probing on x86-64. We really trust e820 now.
This is slightly experimental still. We'll see how it does.
- uncached copy user
Not a clear win, but also not a loss. Follow i386.
- Fix 32bit EFI with regparms.
- Fix boot slowdown as VT guest (Zach Amsden)
- VMI for paravirtualized VMware
* first paravirt ops client
* still some changes missing which need more work
- Various changes in APIC routing setup (Ingo Molnar)
- Various changes in mmconfig handling (Olivier Galibert, OGAWA Hirofumi, me)
* Share code between i386/x86-64
* Be more aggressive at ignoring bogus MCFG tables
* Now supports white lists for some chipsets (currently only Intel 945/915)
- Some patches for the upcomming AMD Family10 CPUs.
- More init section reference fixes from Vivek
- Some preparation patches for Perfmon
- New NUMA hash function for x86-64 and related changes (Amul Shah)
- Support a trigger on machine check events on x86-64
- NMI watchdog fixes for Core2 from Venkatesh
- Fix a HPET timer calibration issue on systems with long SMM events at boot (Jack Steiner)
- Fix compat a.out signals on x86-64
- Lots of small stuff

Not included yet. Might or might not make .21:

- Solution for Nvidia IOMMU corruptions. We could default to iommu=soft for
nvidia, but I was still hoping for a workaround from the hardware vendors.
- vDSO support (still trouble with newer toolkits)
- Xen paravirt ops support from Jeremy/Chris
- Dynamic command line from Bernhard Walle
- Fast getcpu from Dean (requires vDSO)
- Rewritten RAID XOR functions
- Fixes for empty nodes from mm
- Fake node improvements for x86-64
- New dynamic IRQ 0 probing to work around all chipset issues
- lguest
* still seems heavily in development. Not sure it will be ready in time.

Not likely to make .21:
- PAT support (needs some more work and lot more testing)
- Early firewire support for firescope at early boot

Full list of patches:

Note I will do at least a second review pass before sending them to Linus,
so there might be changes. No guarantee all patches will make it.

defconfig-update
i386-defconfig-update
Defconfig updates

copy-user-nocache
Support cache bypassing copy_from_user for file system workloads
Wasn't a clear win, but also not a clear loss.
#io-apic-reuse
#vdso
Not yet

#last-branch
Private Debugging

make-the-numa-hash-function-nodemap-allocation
convert-i386-pda-code-to-use-%fs
kernel-mode-faults-pollute-current-thead
Fix for UML

revert-i386-fix-the-verify_quirk_intel_irqbalance
revert-x86_64-mm-add-genapic_force
revert-x86_64-mm-fix-the-irqbalance-quirk-for-e7320-e7520-e7525
optimize-fix-apic-mode-setup
always-use-physical-delivery-mode-on-8-cpus
remove-clustered-apic-mode
default-to-physical-mode-on-hotplug-cpu-kernels
genapic changes from Ingo

x86_64-make-the-numa-hash-function-nodemap-allocation-fix-fix
fix-a-typo-in-an-irq-handler-name

mmconfig-share
only-call-unreachable_devices-when-type-1-is-available.
detect-and-support-the-e7520-and-the-945g-gz-p-pl
reserve-resources-but-only-when-were-sure-about-them.
mmconfig-ioremap
mmconfig-reject-broken-table
mmconfig changes

get-rid-of-arch_have_xtime_lock

a-memcpy-that-tries-to-reduce-cache-pressure
use-memcpy_uncached_read-in-rdma-interrupt-handler-to-reduce-packet-loss
Infiniband uncached copy support

improved-iommu-documentation
do-not-always-end-the-stack-trace-with-ulong_max
e820-include
Misc fixes

page-allocation-hooks-for-vmi-backend
paravirt-cpu-hypercall-batching-mode
iopl-handling-for-paravirt-guests
smp-boot-hook-for-paravirt
vmi-backend-for-paravirt-ops
vmi-timer
profile-pc-badness
kprobe-rpl-fix
vmi-timer-race
paravirt-debug-defaults-off
VMI

move-startup_32-in-text.head-section
break-init-in-two-parts-to-avoid-modpost-warnings
section warning fixes from Vivek

mcheck-include
add-idle-notifier
Idle notifier for perfmon

improve-sched_clock-on-i686
romsignature-checksum-cleanup
fix-fake-numa-for-x86_64-machines-with-big-io-hole
remove-fastcall-references-in-x86_64-code
use-constant-instead-of-raw-number-in-x86_64-ioperm.c

handle-32-bit-perfmon-counter-writes-cleanly-in-x86_64-nmi_watchdog
handle-32-bit-perfmon-counter-writes-cleanly-in-i386-nmi_watchdog
handle-32-bit-perfmon-counter-writes-cleanly-in-oprofile
Core2 NMI watchdog fixes

config_physical_align-limited-to-4m
cleanup-doc-x86_64-files
list-x86_64-quilt-tree
simplify-notify_page_fault
Misc fixes

mce_amd-issues
mce-trigger
Machine check trigger support

remove-get_pmd
fix-tlbflush-comment
msr-on-cpu
kconfig-typos
msr-scf-single
mtrr-scf-single
fix-preprocessor-condition
mtrr-compat
MTRR support for 32bit compat clients
apm-seqfile
fix-size_or_mask-and-size_and_mask
48 bit MTRR support step 1.
ignore-long-smi-interrupts-in-clock-calibration-code-update-1
Fix HPET calibration problem when SMI interrupts
putreg-check
unexport-supported-pte
aout-no-vdso
Fix a.out
fs-gs-clear
VT boot performance fix
iommu-boundary
efi-regparms
Fix EFI on 32bit
remove-rom-reservation
Trust e820 completely. Muahaha!
define-dma-noncoherent-api-functions
robustify-bad_dma_address-handling
fix-laptop-bootup-hang-in-init_acpi
all-transmeta-cpus-have-constant-tscs
avoid-gcc-extension
support-classic-mediagxm
Support for older Cyrix CPUs
entrys-end-endproc-annotations
clean-up-sparsemem-memory_present-call
arch-i386-kernel-alternativec-should-include-asm-bugsh
remove-unused-kernel-config-option-x86_xadd
Misc cleanups
update-io-apic-dest-field-to-8-bit-for-xapic
Fix kexec on es7000 with 8 bit APIC IDs
avoid-warning-message-livelock
minor-patch-for-compilation-warning-in-x86_64-signal-code
dont-include-bugsh
add-option-to-show-more-code-in-oops-reports
32-bit-ptrace-mangles-sixth-system-call-argument
geode-configuration-fixes
survive-having-no-irq-mapping-for-a-vector
fix-gcc-check
Misc changes
paravirt-remove-fastcall
Small cleanup
fam10-cpuid
fam10-l3cache
fam10-nmi-watchdog
AMD family10 support

fix-microcode-warning
i386-fix-transmeta-warning
fails-to-detect-mediagx


-Andi
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/

Relevant Pages

  • 2.6.12-mm2
    ... I'm holding off on the 80-odd pcmcia patches until we get ... Fix damage from recent swsusp cleanups ... Various net fixes ... arch/cris updates ...
    (Linux-Kernel)
  • 2.6.7-mm4
    ... - Merged support for the 64-bit SuperH architecture ... Fix alignment of .init.setup section (fixes the kernel boot parameter woes) ... ext3 block reservation patch set -- ext3 preallocation cleanup ...
    (Linux-Kernel)
  • [git pull][resend] drm-intel fixes for 2.6.32-rc5
    ... Here's a set of fixes for the regressions introduced with the ... the new chipset support as new pre-release revs have come out. ... Fix render reclock availability detection. ... fix to setup display reference clock control on Ironlake ...
    (Linux-Kernel)
  • Re: Q824143 -- how to get patch?!?
    ... There are 2 sorts of patches. ... client or clients that are have an issue. ... They are available from product support if the engineer while working with ... the client identifies that this fix will fix the specific problem the client ...
    (microsoft.public.security)
  • 2.6.6-rc2-mm1
    ... - Input driver fixes, cleanups, features. ... Fix create_singlethread_workqueue.patch ... Implement SMT support within sched-domains for ppc64 ... sched: cpu load management cleanup ...
    (Linux-Kernel)