x86 updates for v2.6.30, final bits
Linus,
Please pull the latest x86-stage-3-for-linus git tree from:
git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip.git x86-stage-3-for-linus
This concludes the x86 merge window.
The MM PAT bits touching mm/memory.c and include/linux/mm.h were reviewed and acked by Nick, the DMAR bits are part of a longer x2apic fix series from Suresh. (Jesse Cc:-ed)
[ The sched.c bit is an odd one out: due to previous cpumask changes
it developed a dependency/conflict on the cpumask tree which depended on the x86 tree. Should i have started a separate branch for it? I didnt want to merge the x86 tree into the scheduler tree. We had really excessive dependencies and cross-merges in this cycle around the x86 tree and i very much hope this is an exception - there's been too many infrastructure projects going on in parallel. Nothing of this scale will happen in the .31 cycle - with the TLB, APIC, percpu unification in this cycle we've modernized and unified the whole x86 infrastructure. Future work will interact a lot less. ]
out-of-topic modifications in x86-stage-3-for-linus:
----------------------------------------------------drivers/pci/dmar.c # fa4b57c: x86, dmar: use atomic allocations
# 1531a6a: x86, dmar: start with sane state # eba67e5: x86, dmar: routines for disabling # 9d783ba: x86, x2apic: enable fault handlin # 0ac2491: x86, dmar: move page fault handlidrivers/pci/intel-iommu.c # 1531a6a: x86, dmar: start with sane state
# 9d783ba: x86, x2apic: enable fault handlin # 0ac2491: x86, dmar: move page fault handlidrivers/pci/intr_remapping.c # fa4b57c: x86, dmar: use atomic allocations
# 2e93456: x86, intr-remapping: fix free_irt
# 1531a6a: x86, dmar: start with sane state
# eba67e5: x86, dmar: routines for disabling
# 9d783ba: x86, x2apic: enable fault handlin
# 4c5502b: x86, x2apic: fix lock ordering duinclude/linux/dmar.h # 29b61be: x86, x2apic: cleanup ifdef CONFIG
# 1531a6a: x86, dmar: start with sane state # 9d783ba: x86, x2apic: enable fault handlininclude/linux/intel-iommu.h # eba67e5: x86, dmar: routines for disabling
# 9d783ba: x86, x2apic: enable fault handlininclude/linux/mm.h # 895791d: VM, x86, PAT: add a new vm flag t
# 4bb9c5c: VM, x86, PAT: Change is_linear_pf
kernel/sched.c # 7a46c59: cpuacct: reduce one NULL check in
mm/memory.c # 895791d: VM, x86, PAT: add a new vm flag t
# 4bb9c5c: VM, x86, PAT: Change is_linear_pfKnown risks:
- We had some boot regression trouble with the e820 changes, on systems with odd/small memory layouts:
x86: fix e820_update_range()
x86: make e820_update_range() handle small range update
- The very-early-brk changes had several regressions along the way:
x86: add brk allocation for very, very early allocations
- The IO-APIC/APIC cleanups were relatively trouble-free, but that code is always a risk area.
- No known regressions at the moment.
Thanks,
Ingo
------------------>
Akinobu Mita (3):
x86: unify kmap_atomic_pfn() and iomap_atomic_prot_pfn()
x86: debug check for kmap_atomic_pfn and iomap_atomic_prot_pfn() x86, mm: remove unnecessary include file from iomap_32.c
Américo Wang (1):
x86: ptrace, bts: fix an unreachable statement
Andi Kleen (19):
x86, mce: disable machine checks on suspend
x86, mce: don't disable machine checks during code patching x86, mce: always use separate work queue to run trigger
x86, mce: switch machine check polling to per CPU timer
x86, mce: don't set up mce sysdev devices with mce=off
x86, mce: disable machine checks on offlined CPUs
x86, mce: enable machine checks in 64-bit defconfig
x86, mce: implement dynamic machine check banks support
x86, mce: factor out duplicated struct mce setup into one function x86, mce: separate correct machine check poller and fatal exception handler x86, mce, cmci: export MAX_NR_BANKS
x86, mce, cmci: factor out threshold interrupt handler
x86, mce, cmci: avoid potential reentry of threshold interrupt x86, mce: replace machine check events logged interval with ratelimit x86, mce, cmci: use polled banks bitmap in machine check poller x86, mce, cmci: define MSR names and fields for new CMCI registers x86, mce, cmci: add CMCI support
x86, mce, cmci: disable CMCI on rebooting
x86, mce, cmci: recheck CMCI banks after APIC has been enabled on CPU #0
Andreas Herrmann (1):
x86: mtrr: don't modify RdDram/WrDram bits of fixed MTRRs
Andrew Morton (1):
x86: use smp_call_function_single() in arch/x86/kernel/cpu/mcheck/mce_amd_64.c
Bartlomiej Zolnierkiewicz (1):
x86: fix IO APIC resource allocation error message
Cyrill Gorcunov (2):
x86: linkage.h - guard assembler specifics by __ASSEMBLY__
x86: shrink __ALIGN and __ALIGN_STR definitions
Dimitri Sivanich (4):
x86: UV, SGI RTC: add generic system vector
x86: UV, SGI RTC: loop through installed UV blades
x86: UV, SGI RTC: add UV RTC clocksource/clockevents
x86: UV, SGI RTC: fix uv_time.c for UP
Dmitri Vorobiev (1):
x86: Fix a couple of sparse warnings in arch/x86/kernel/apic/io_apic.c
H. Peter Anvin (11):
x86, mce: use %ll instead of %L for 64-bit numbers x86, mce: remove invalid __cpuinit/__cpuexit annotations x86, mce, cmci: remove incorrect __cpuinit/__cpuexit annotations
x86: remove zImage support
x86: remove additional vestiges of the zImage/bzImage split
x86: use targets in the boot Makefile instead of CLEAN_FILES
x86: move brk initialization out of #ifdef CONFIG_BLK_DEV_INITRD
x86-32: tighten the bound on additional memory to map
x86-32: move _end to a dummy section
x86, setup: move 32-bit code to .text32
x86, setup: fix the setting of 480-line VGA modes
Hidetoshi Seto (1):
x86, mce: remove incorrect __cpuinit for intel_init_cmci()
Hiroshi Shimamoto (1):
x86: signal: check signal stack overflow properly
Huang Ying (4):
x86, mce: fix a race condition in mce_read()
x86, kexec: fix kexec x86 coding style
x86, kexec: x86_64: add identity map for pages at image->start x86, kexec: x86_64: add kexec jump support for x86_64
Ingo Molnar (7):
x86, mce: fix build failure in arch/x86/kernel/cpu/mcheck/threshold.c
x86: move init_memory_mapping() to common mm/init.c, build fix on 32-bit PAE
x86: UV, SGI RTC: add generic system vector, build fix on UP
x86: memcpy, clean up
x86: unify kmap_atomic_pfn() and iomap_atomic_prot_pfn(), fix
x86: cpu/common.c more cleanups
x86: mpparse: clean up code by introducing a few helper functions, fix
Jan Beulich (14):
x86: remove leftover unwind annotations
x86-64: move save_paranoid into .kprobes.text
x86-64: remove unnecessary spill/reload of rbx from memcpy
x86: fix HYPERVISOR_update_descriptor()
x86: move a few device initialization objects into .devinit.rodata
x86: move various CPU initialization objects into .cpuinit.rodata
x86: smarten /proc/interrupts output
x86, 32-bit: also limit NODES_HIGH_SHIFT here
x86, 32-bit: also use cpuinfo_x86's x86_{phys,virt}_bits members
x86: move save_mr() into .meminit.text
x86: properly __init-annotate recent early_printk additions
x86: clean up output resulting from update_mptable option
x86: fix code paths used by update_mptable
x86: create a non-zero sized bm_pte only when needed
Jaswinder Singh Rajput (22):
x86: cpu architecture debug code
x86: cpu architecture debug code, build fix, cleanup
x86: cpu_debug add write support for MSRs
x86: cpu/common.c cleanups
x86: entry_32.S fix compile warnings - fix work mask bit width
x86: cpu_debug add support for various AMD CPUs
x86: cpu/mttr/cleanup.c fix compilation warning
x86: kprobes.c fix compilation warning
x86: mpparse: clean up code by introducing a few helper functions
x86: mpparse.c introduce smp_dump_mptable helper function
x86: mpparse.c introduce check_physptr helper function
x86: topology.c cleanup
x86: kdebugfs.c cleanup
x86: i8253 cleanup
x86: rtc.c cleanup
x86: io_delay.c cleanup
x86: pci-nommu.c cleanup
x86: cpu/cpu.h cleanup
x86: irq.c use same path for show_interrupts
x86: irq.c keep CONFIG_X86_LOCAL_APIC interrupts together
x86: apic/io_apic.c define msi_ir_chip and ir_ioapic_chip all the time
x86: e820 fix various signedness issues in setup.c and e820.c
Jeremy Fitzhardinge (18):
x86: clean up old gcc warnings
x86-32: use specific __vmalloc_start_set flag in __virt_addr_valid
x86: pre-initialize boot_cpu_data.x86_phys_bits to avoid system_state tests
x86-32: make sure virt_addr_valid() returns false for fixmap addresses
x86: make section delimiter symbols part of their section
x86: add brk allocation for very, very early allocations
x86-32: use brk segment for allocating initial kernel pagetable
x86: use brk allocation for DMI
x86: allow extend_brk users to reserve brk space
x86, paravirt: prevent gcc from generating the wrong addressing mode
x86-32: make sure we map enough to fit linear map pagetables
x86-32: remove ALLOCATOR_SLOP from head_32.S
x86/brk: make the brk reservation symbols inaccessible from C x86/brk: put the brk reservations in their own section
x86/32: no need to use set_pte_present in set_pte_vaddr
x86: with the last user gone, remove set_pte_present
Revert "x86: create a non-zero sized bm_pte only when needed" x86/dmi: fix dmi_alloc() section mismatches
KOSAKI Motohiro (1):
x86, mce: use round_jiffies() instead round_jiffies_relative()
Li Zefan (1):
cpuacct: reduce one NULL check in fast-path
Pallipadi, Venkatesh (3):
VM, x86, PAT: Change is_linear_pfn_mapping to not use vm_pgoff VM, x86, PAT: add a new vm flag to track full pfnmap at mmap x86, PAT, PCI: Change vma prot in pci_mmap to reflect inherited prot
Pekka Enberg (22):
x86: move devmem_is_allowed() to common mm/init.c
x86: set_highmem_pages_init() cleanup, #2
x86: move free_initrd_mem() to common mm/init.c
x86: init_memory_mapping() trivial cleanups
x86: add gbpages support to 32-bit init_memory_mapping()
x86: find_early_table_space() unification
x86: move pgd_base out of init_memory_mapping()
x86: ifdef 32-bit specific setup in init_memory_mapping()
x86: remove unnecessary save_mr() sanity check
x86: rename after_init_bootmem to after_bootmem in mm/init_32.c
x86: unify kernel_physical_mapping_init() call in init_memory_mapping()
x86: add table start and end sanity checks to 32-bit init_memory_mapping()
x86: ifdef 32-bit and 64-bit setup in init_memory_mapping()
x86: ifdef 32-bit and 64-bit pfn setup in init_memory_mapping()
x86: ifdef 32-bit and 64-bit NR_RANGE_MR for save_mr() unification
x86: add stub init_gbpages() for 32-bit init_memory_mapping()
x86: move init_memory_mapping() to common mm/init.c
x86: fix up some bad global variable names in mm/init.c
x86: unify kernel_physical_mapping_init() function signatures
x86: move function and variable declarations to asm/init.h
x86: fix uninitialized variable in init_memory_mapping()
x86: rename do_not_nx to disable_nx in mm/init_64.c
Ravikiran G Thirumalai (1):
Revert "x86: don't compile vsmp_64 for 32bit"
Rusty Russell (2):
x86, uv: fix cpumask iterator in uv_bau_init()
x86: Correct behaviour of irq affinity
Sebastian Andrzej Siewior (1):
x86/centaur: merge 32 & 64 bit version
Stoyan Gaydarov (1):
x86: BUG to BUG_ON changes
Stuart Bennett (1):
x86: mmiotrace: quieten spurious warning message
Suresh Siddha (14):
x86, x2apic: fix lock ordering during IRQ migration
x86, dmar: move page fault handling code to dmar.c
x86, x2apic: enable fault handling for intr-remapping
x86, dmar: routines for disabling queued invalidation and intr remapping x86, dmar: start with sane state while enabling dma and interrupt-remapping x86, intr-remapping: fix free_irte() to clear all the IRTE entries x86, x2apic: use virtual wire A mode in disable_IO_APIC() with interrupt-remapping x86, x2apic: fix clear_local_APIC() in the presence of x2apic
x86, x2apic: cleanup the IO-APIC level migration with interrupt-remapping x86, x2apic: cleanup ifdef CONFIG_INTR_REMAP in io_apic code
x86, ioapic: Fix non atomic allocation with interrupts disabled
x86: fix broken irq migration logic while cleaning up multiple vectors x86, dmar: use atomic allocations for QI and Intr-remapping init
x86: add x2apic_wrmsr_fence() to x2apic flush tlb paths
Thomas Gleixner (3):
x86: convert obsolete irq_desc_t typedef to struct irq_desc
x86: disable __do_IRQ support
x86: reduce preemption off section in exit thread
Wang Chen (2):
x86: don't define __this_fixmap_does_not_exist()
x86: early_ioremap_init(), use __fix_to_virt(), because we are sure it's safe
Yinghai Lu (23):
x86: fix bootmem cross node for 32bit numa
x86: make 32-bit init_memory_mapping range change more like 64-bit
x86: ioremap mptable
x86: reserve exact size of mptable
x86: fix bootmem cross node for 32bit numa, cleanup
x86/doc: mini-howto for using earlyprintk=dbgp
x86: make "memtest" like "memtest=17"
x86: remove smp_apply_quirks()/smp_checks()
x86: fix warning about nodeid
x86: more MTRR debug printouts
x86: print out mtrr_range_state when user specify size
x86: separate mtrr cleanup/mtrr_e820 trim to separate file
x86: fix e820_update_range()
x86: make e820_update_range() handle small range update
x86: fix get_mtrr() warning about smp_processor_id() with CONFIG_PREEMPT=y
x86: print the continous part of fixed mtrrs together
x86-32: compute initial mapping size more accurately
x86: put initial_pg_tables into .bss
x86: fix 64k corruption-check
x86: print out more info in e820_update_range()
x86: MTRR workaround for system with stange var MTRRs
x86: fix set_extra_move_desc calling
x86: use default_cpu_mask_to_apicid for 64bit
[email protected] (3):
x86, CPA: Add a flag parameter to cpa set_clr()
x86, PAT: Add support for struct page pointer array in cpa set_clr x86, CPA: Add set_pages_arrayuc and set_pages_array_wb
Documentation/x86/earlyprintk.txt | 101 +++ arch/x86/Kconfig | 15 +- arch/x86/Kconfig.cpu | 17 +- arch/x86/Makefile | 29 +- arch/x86/boot/Makefile | 53 +- arch/x86/boot/header.S | 29 +- arch/x86/boot/pm.c | 44 -- arch/x86/boot/pmjump.S | 1 + arch/x86/boot/setup.ld | 3 +- arch/x86/boot/tools/build.c | 9 +- arch/x86/boot/video-vga.c | 22 +- arch/x86/include/asm/apic.h | 37 +- arch/x86/include/asm/apicdef.h | 1 + arch/x86/include/asm/boot.h | 4 - arch/x86/include/asm/cacheflush.h | 3 + arch/x86/include/asm/cpu_debug.h | 226 ++++++ arch/x86/include/asm/desc.h | 3 +- arch/x86/include/asm/dmi.h | 19 +- arch/x86/include/asm/e820.h | 2 +- arch/x86/include/asm/entry_arch.h | 2 + arch/x86/include/asm/hardirq.h | 1 + arch/x86/include/asm/highmem.h | 1 + arch/x86/include/asm/hw_irq.h | 1 + arch/x86/include/asm/init.h | 18 + arch/x86/include/asm/io_apic.h | 5 +- arch/x86/include/asm/irq.h | 1 + arch/x86/include/asm/irq_remapping.h | 2 - arch/x86/include/asm/irq_vectors.h | 5 + arch/x86/include/asm/kexec.h | 13 +- arch/x86/include/asm/linkage.h | 19 +- arch/x86/include/asm/mce.h | 35 +- arch/x86/include/asm/msidef.h | 1 + arch/x86/include/asm/msr-index.h | 5 + arch/x86/include/asm/page_32_types.h | 5 + arch/x86/include/asm/page_types.h | 6 - arch/x86/include/asm/paravirt.h | 19 +- arch/x86/include/asm/pat.h | 5 + arch/x86/include/asm/pgtable-2level.h | 7 - arch/x86/include/asm/pgtable-3level.h | 17 - arch/x86/include/asm/pgtable.h | 2 - arch/x86/include/asm/pgtable_32.h | 3 - arch/x86/include/asm/pgtable_32_types.h | 5 + arch/x86/include/asm/pgtable_types.h | 1 + arch/x86/include/asm/processor.h | 5 +- arch/x86/include/asm/sections.h | 7 + arch/x86/include/asm/setup.h | 39 +- arch/x86/include/asm/uv/uv_hub.h | 4 + arch/x86/include/asm/xen/hypercall.h | 2 + arch/x86/kernel/Makefile | 4 +- arch/x86/kernel/alternative.c | 17 +- arch/x86/kernel/apic/apic.c | 35 +- arch/x86/kernel/apic/apic_flat_64.c | 18 +- arch/x86/kernel/apic/io_apic.c | 292 ++++---- arch/x86/kernel/apic/probe_64.c | 7 + arch/x86/kernel/apic/x2apic_cluster.c | 6 + arch/x86/kernel/apic/x2apic_phys.c | 6 + arch/x86/kernel/check.c | 8 +- arch/x86/kernel/cpu/Makefile | 5 +- arch/x86/kernel/cpu/addon_cpuid_features.c | 2 +- arch/x86/kernel/cpu/amd.c | 54 ++- arch/x86/kernel/cpu/centaur.c | 36 +- arch/x86/kernel/cpu/centaur_64.c | 37 - arch/x86/kernel/cpu/common.c | 396 ++++++----- arch/x86/kernel/cpu/cpu.h | 25 +- arch/x86/kernel/cpu/cpu_debug.c | 901 +++++++++++++++++++++++ arch/x86/kernel/cpu/cyrix.c | 16 +- arch/x86/kernel/cpu/intel.c | 32 +- arch/x86/kernel/cpu/intel_cacheinfo.c | 8 +- arch/x86/kernel/cpu/mcheck/Makefile | 1 + arch/x86/kernel/cpu/mcheck/mce_32.c | 14 - arch/x86/kernel/cpu/mcheck/mce_64.c | 530 ++++++++++---- arch/x86/kernel/cpu/mcheck/mce_amd_64.c | 62 +- arch/x86/kernel/cpu/mcheck/mce_intel_64.c | 207 ++++++- arch/x86/kernel/cpu/mcheck/threshold.c | 29 + arch/x86/kernel/cpu/mtrr/Makefile | 2 +- arch/x86/kernel/cpu/mtrr/cleanup.c | 1101 ++++++++++++++++++++++++++++ arch/x86/kernel/cpu/mtrr/generic.c | 202 ++++-- arch/x86/kernel/cpu/mtrr/main.c | 1069 +-------------------------- arch/x86/kernel/cpu/mtrr/mtrr.h | 4 + arch/x86/kernel/cpu/transmeta.c | 2 +- arch/x86/kernel/cpu/umc.c | 2 +- arch/x86/kernel/e820.c | 142 +++-- arch/x86/kernel/early_printk.c | 20 +- arch/x86/kernel/entry_32.S | 18 +- arch/x86/kernel/entry_64.S | 6 +- arch/x86/kernel/head32.c | 5 +- arch/x86/kernel/head64.c | 2 +- arch/x86/kernel/head_32.S | 76 +- arch/x86/kernel/i8253.c | 68 +- arch/x86/kernel/io_delay.c | 27 +- arch/x86/kernel/irq.c | 88 ++- arch/x86/kernel/irqinit_32.c | 3 + arch/x86/kernel/irqinit_64.c | 3 + arch/x86/kernel/kdebugfs.c | 82 +-- arch/x86/kernel/kprobes.c | 2 +- arch/x86/kernel/kvm.c | 7 - arch/x86/kernel/machine_kexec_32.c | 17 +- arch/x86/kernel/machine_kexec_64.c | 99 +++- arch/x86/kernel/mmconf-fam10h_64.c | 2 +- arch/x86/kernel/mpparse.c | 384 +++++----- arch/x86/kernel/paravirt.c | 1 - arch/x86/kernel/pci-nommu.c | 20 +- arch/x86/kernel/process.c | 5 +- arch/x86/kernel/ptrace.c | 3 +- arch/x86/kernel/quirks.c | 3 +- arch/x86/kernel/relocate_kernel_32.S | 24 +- arch/x86/kernel/relocate_kernel_64.S | 189 ++++- arch/x86/kernel/rtc.c | 20 +- arch/x86/kernel/setup.c | 58 +- arch/x86/kernel/signal.c | 48 +- arch/x86/kernel/smpboot.c | 78 -- arch/x86/kernel/tlb_uv.c | 3 +- arch/x86/kernel/topology.c | 14 +- arch/x86/kernel/uv_time.c | 393 ++++++++++ arch/x86/kernel/visws_quirks.c | 2 +- arch/x86/kernel/vmi_32.c | 6 - arch/x86/kernel/vmlinux_32.lds.S | 21 +- arch/x86/kernel/vmlinux_64.lds.S | 101 ++- arch/x86/kernel/vsmp_64.c | 12 +- arch/x86/lguest/boot.c | 8 - arch/x86/lib/memcpy_64.S | 143 ++-- arch/x86/mm/highmem_32.c | 24 +- arch/x86/mm/init.c | 344 +++++++++ arch/x86/mm/init_32.c | 256 ++----- arch/x86/mm/init_64.c | 280 +------- arch/x86/mm/iomap_32.c | 26 +- arch/x86/mm/ioremap.c | 52 +- arch/x86/mm/kmmio.c | 2 +- arch/x86/mm/memtest.c | 3 + arch/x86/mm/numa_32.c | 5 +- arch/x86/mm/pageattr.c | 147 +++- arch/x86/mm/pat.c | 5 +- arch/x86/mm/pgtable_32.c | 2 +- arch/x86/mm/tlb.c | 5 - arch/x86/pci/common.c | 4 +- arch/x86/pci/fixup.c | 4 +- arch/x86/pci/i386.c | 3 + arch/x86/xen/mmu.c | 7 +- drivers/pci/dmar.c | 296 ++++++++- drivers/pci/intel-iommu.c | 220 +----- drivers/pci/intr_remapping.c | 110 ++- include/linux/dmar.h | 52 ++- include/linux/intel-iommu.h | 5 +- include/linux/mm.h | 3 +- kernel/sched.c | 5 +- mm/memory.c | 6 +- 146 files changed, 6468 insertions(+), 3585 deletions(-) create mode 100644 Documentation/x86/earlyprintk.txt create mode 100755 arch/x86/include/asm/cpu_debug.h create mode 100644 arch/x86/include/asm/init.h delete mode 100644 arch/x86/kernel/cpu/centaur_64.c create mode 100755 arch/x86/kernel/cpu/cpu_debug.c create mode 100644 arch/x86/kernel/cpu/mcheck/threshold.c create mode 100644 arch/x86/kernel/cpu/mtrr/cleanup.c create mode 100644 arch/x86/kernel/uv_time.c
Quoted 8 lines · Ingo Molnar
[ The
sched.cbit is an odd one out: due to previous cpumask changes
it developed a dependency/conflict on the cpumask tree which depended on the x86 tree. Should i have started a separate branch for it? I didnt want to merge the x86 tree into the scheduler tree. We had really excessive dependencies and cross-merges in this cycle around the x86 tree and i very much hope this is an exception
The problem is that you've been hoping for this "exception" for the last three kernel releases.
The details differ, but you do seem to mix things up too much. I'm not at all happy. I think quality control is slipping, because there's this absolutely _humongous_ amount of crap that gets in through you. You seem to have a hard time saying "no".
And yes, you boot-test things pretty well, but I really wish you had more focus. This "everything under the sun" thing is very annoying, and I think you are too damn eager to merge the random new feature of the day.
So instead of "hoping", how about you look at making sure it really _does_ become an exception. And that really fundamentally means that it can't happen every release.
How about trying one release to just say "no" if you start seeing all these kinds of things. We don't allow non-x86 architectures to just tie things together this way. The fact that you have the same tree seems to just encourage badness by making it "easier" to just mix things up.
Linus
Quoted 27 lines · Linus Torvalds
Quoted 8 lines · Ingo Molnar
[ The
sched.cbit is an odd one out: due to previous cpumask changes
it developed a dependency/conflict on the cpumask tree which depended on the x86 tree. Should i have started a separate branch for it? I didnt want to merge the x86 tree into the scheduler tree. We had really excessive dependencies and cross-merges in this cycle around the x86 tree and i very much hope this is an exceptionThe problem is that you've been hoping for this "exception" for the last three kernel releases.
The details differ, but you do seem to mix things up too much. I'm not at all happy. I think quality control is slipping, because there's this absolutely _humongous_ amount of crap that gets in through you. You seem to have a hard time saying "no".
And yes, you boot-test things pretty well, but I really wish you had more focus. This "everything under the sun" thing is very annoying, and I think you are too damn eager to merge the random new feature of the day.
So instead of "hoping", how about you look at making sure it really _does_ become an exception. And that really fundamentally means that it can't happen every release.
Quoted 5 lines
How about trying one release to just say "no" if you start seeing all these kinds of things. We don't allow non-x86 architectures to just tie things together this way. The fact that you have the same tree seems to just encourage badness by making it "easier" to just mix things up.
Yes, will do that - sorry about this. We'll also take more care and will try to cook up more scripting to make sure such crappy-looking circular criss-cross merges and compound trees dont happen again.
The remaining trees we have for this merge window are all single-purpose. We sent five of them in the last couple of hours and they have the kind of structure that we intend to use in the future too. If you still see _any_ problems with them, please let me know so i can fix the workflow ASAP.
Ingo
Quoted 8 lines · Ingo Molnar
Linus,
Please pull the latest x86-stage-3-for-linus git tree from:
git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip.git x86-stage-3-for-linusThis concludes the x86 merge window.
I'm planning on getting Xen changes into this merge window. I was expecting you to at least take the branches which have an impact on the arch/x86 side, but I guess there was a miscommunication there.
However, since all your pending x86 changes are now sent, do you have any objections me sending out pull requests for the Xen changes I've previously posted? If not, I'll start tomorrow.
Thanks,
J