2.6.8-rc2-mm1

2.6.8-rc2-mm1

[Andrew Morton]

ftp://ftp.kernel.org/pub/linux/kernel/people/akpm/patches/2.6/2.6.8-rc2/2.6.8-rc2-mm1/

- Added the bk-power tree to the -mm lineup.  This is Pat Mochel's big
  pmdisk/swsusp consolidation.  Please copy mochel@digitalimplant.org on any
  swsusp problems.

- Added Sam's bk-kbuild tree to the -mm lineup

- The bk-ieee1394 tree is uncontactable so the patch here is old.

- The bk-arm, bk-serial and bk-pcmcia trees have had their server turned
  off, so they're not here.

- The bk-usb repository is currently empty for some reason

- Added a big i2o rewrite.  This accidentally reverts lots of post-2.6.6
  fixes and needs work.

- The lingering bug in the IDE barrier code should be fixed

- Re-introduced the signal-race-fixes patch.  This breaks all architectures
  except x86, x86_64, s390 and ia64.

- If people have patches in here which are important for a 2.6.8 release,
  please let me know.

- This kernel is boot-tested and somewhat stress-tested on x86 and x86_64
  and ia64.  But given the amount of new stuff in here and its general state,
  things may be a bit flakey.




Changes since 2.6.8-rc1-mm1:


 linus.patch
 bk-acpi.patch
 bk-agpgart.patch
 bk-alsa.patch
 bk-cifs.patch
 bk-dma-declare-coherent-memory.patch
 bk-cpufreq.patch
 bk-drm.patch
 bk-ieee1394.patch
 bk-input.patch
 bk-kbuild.patch
 bk-libata.patch
 bk-mpc52xx.patch
 bk-netdev.patch
 bk-ntfs.patch
 bk-pci.patch
 bk-pnp.patch
 bk-power.patch
 bk-scsi.patch

 External trees

-kbuild-improve-kernel-build-with-separated-output.patch
-w9968cf-build-fix.patch
-ppc32-pmac_zilog-initialize-port-spinlock-on-all-init-paths.patch
-altix-serial-driver-2.patch
-ide_tf_pio_out_fixes.patch
-ide_tf_pio_out_prehandler.patch
-ide_tf_pio_out_error.patch
-ide_task_in_intr.patch
-ide_pre_task_out_intr.patch
-ide_pre_task_mulout_intr.patch
-ide_tf_no_partial.patch
-ide_non_tf_pio.patch
-ide_no_flagged_pio.patch
-i2c-i2c-devci2c_dev_init-cleanup.patch
-fix-warnings-drivers-net-sk98lin-skaddrc.patch
-bio_copy_user-cleanups.patch
-fix-airo-oops-on-removal.patch
-serious-performance-regression-due-to-nx-patch.patch
-fix-oops-in-device_platform_unregister.patch
-fix-3c59xc-uses-of-plain-integer-as-null-pointer.patch
-small-style-fixups-for-the-new-automount-code.patch
-ifndef-guard-percpu_counterh-and-blockgroup_lockh.patch
-floppyc-remove-superfluous-variable-initialization.patch
-unknown-symbol-in-sound-oss-kahluako-needs-unknown-symbol-udelay.patch
-remove-struct_cpy.patch
-autoselect-fatfs.patch
-fix-saa7146-compilation-on-268-rc1.patch
-fix-return-codes-after-i2c_add_driver-in-tea6415c.patch
-remove-outdated-stallion-contact-information.patch
-fix-ia64-early_printk-build-problem.patch
-fix-inode-state-corruption-268-rc1-bk1.patch

 Merged

+fixes-for-rcu_offline_cpu-rcu_move_batch-268-rc2.patch

 CPU hotplug fix

+unix98-pty-indices-leak.patch

 PTY leak fix

+sched-initialize-sched-domain-table.patch

 CPu scheduler fix

+compat_clock_getres-shouldnt-return-efault-if-res-==-null.patch

 posix timers fix

+bio-page-refcounting-fix.patch

 BIO layer page refcounting fix

+longhaul-fix.patch
+bk-netdev-axnet_cs-fix.patch
+bk-netdev-hp-plus-fix.patch
+bk-power-x86_64-fix.patch
+bk-acpi-x86_cpu_to_apicid-fix.patch

 Various fixes for bk trees.

+nmi-trigger-switch-support-for-debuggingupdated.patch
+nmi-trigger-switch-support-for-debuggingupdated-fix.patch

 x86 support for front-panel NMI switches

+make-i386-die-more-resilient-against-recursive-errors.patch

 Handle oops-inside-oops more cleanly on x86

+ppc64-page-align-emergency-stack-2.patch
+ppc64-remove-multiple-irq-optimisation.patch
+ppc64-cpu-hotplug-fix.patch
+ppc64-fix-ras-irq-handlers.patch
+ppc64-whitespace-cleanup-in-promc.patch
+ppc64-hvcs-driver.patch
+ppc64-smt-bugfix.patch

 PPC64 updates

+ppc32-fix-ppc44x-early-uart-setup.patch
+ppc32-export-some-dma-api-symbols.patch
+ppc32-fix-comment-in-arch-ppc-platforms-pmac_pcic.patch

 PPC32 updates

-x86-64-support-for-singlestep-into-32-bit-system-calls.patch

 I don't remember why I dropped this.

+barrier-flushing-fix.patch

 disk barrier fix

+perfctr-x86-init-bug.patch
+perfctr-k8-fix-for-internal-benchmarking-code.patch

 perfctr x86 fixes

+perfctr-inheritance-1-3-driver-updates.patch
+perfctr-inheritance-illegal-sleep-bug.patch
+perfctr-inheritance-2-3-kernel-updates.patch
+perfctr-inheritance-3-3-documentation-updates.patch

 Inherit perfctr settings across fork/exec

-kernelthread-idle-fix.patch
+kernelthread-idle-fix-2.patch

 updated CPU scheduler fix

+sched-consolidate-sched-domains.patch
+sched-consolidate-domains-fix.patch
+sched-consolidate-domains-fix-2.patch

 CPU scheduler code consolidation

+sched-domain-node-span-4.patch
+sched-domain-node-span-4-warning-fix.patch

 CPU scheduler NUMA fixes

+driver-model-and-sysfs-support-for-pcmcia-update.patch

 Additional pcmcia/sysfs integration work

+control-pktcdvd-with-an-auxiliary-character-device.patch
+control-pktcdvd-with-an-auxiliary-character-device-fix.patch

 CDRW/DVDRW packet writing updates

+rename-uml-console-device.patch

 UML build fix

-detect-too-early-schedule-attempts.patch

 Dropped - was a pain.

+per_cpu-per_cpu-cpu_gdt_table-fix.patch

 Fix per_cpu-per_cpu-cpu_gdt_table.patch

+making-i-dhash_entries-cmdline-work-as-it-use-to-fix.patch

 Fix making-i-dhash_entries-cmdline-work-as-it-use-to.patch

-x86-64-singlestep-through-sigreturn-system-call.patch
+x86-64-singlestep-through-sigreturn-system-call-2.patch

 Updated

+oom-show_free_areas.patch

 Print the memory zone state when oom-killings happen

+send_IPI_mask_bitmask-build-fix.patch

 Compile fix

+e1000-build-fix.patch

 e1000 compile fix

+pty_write-latency-fix.patch

 scheduling latency fix

+v850-define-find_first_bit.patch

 v850 build fix

+enable-all-events-for-initramfs.patch

 initramfs initialisation ordering fix

+radeonfb-64-bit-fix.patch

 fix radeonfb on x86_64

+arch-i386-kernel-smpc-gcc341-inlining-fix.patch
+net-sunrpc-xprtc-gcc341-inlining-fix.patch

 gcc-3.4.1 fixes

+fix-menuconfig-partial-inability-to-show-help-texts.patch

 menuconfig fixlet

+was-removal-of-sync-in-panic.patch

 Don't call sync() when panicing

+move-cache_reap-out-of-timer-context.patch
+move-cache_reap-out-of-timer-context-fix.patch

 interrupt latency fixes in slab

+switch-sgc-to-standard-jiffies-converters.patch

 sg.c jiffies cleanup

+sign-fix-in-swapfilec.patch

 signedness fixlet

+a-trivial-patch-for-removing-unnecessary-comment-in-mm-filemapc.patch

 Comment fix

+gettimeofday-nanoseconds-patch-makes-it-possible-for-the-posix-timer.patch

 posix timer resolution improvements

+fix-for-buffer-limit-for-long-in-sysctlc.patch

 sysctl buffer overflow fix

+quiet-down-per-zone-memory-stats.patch

 Less printk's on boot

+create-nodemask_t.patch

 Add nodemask_t abstraction

+ipmi_msghandler-module-load-failure-fix.patch

 impi module loading fix

+fat-kill-nls-default.patch

 Rework FATFS NLS handling

+add-ixdp2x01-board-support-to-cs89x0-driver.patch

 Additional board support for cs89x0.c

+remove-scripts-mkconfigs.patch

 Remove dead file

+267-msi-x-update.patch

 Big MSI driver update

+remove-dead-prototypes.patch

 Remove unused function prototypes

+s390-use-include-asm-generic-dma-mapping-brokenh.patch

 s390 build fix

+fix-readahead-breakage-for-sequential-after-random-reads.patch

 Fix readahead corner case

+cdrom-get_last_written-fix.patch

 cdrom driver fix

+update-mailing-list-for-osst.patch

 MAINTAINERS update

+fix-aic-for-db4.patch

 Teach the aic build process about db4

+m68k-68060-errata-i14.patch
+m68k-ifpsp060.patch
+m68k-sparse-missing-void.patch
+m68k-sparse-if-vs-ifdef.patch
+m68k-sparse-void-return.patch
+m68k-sparse-extern.patch
+m68k-sparse-inline.patch
+dsp56k-sparse-const.patch
+m68k-sparse-floating-point.patch
+dnfb-sparse-struct-init.patch
+amifb-sparse-=.patch
+m68k-hardirqh.patch
+dmasound-paths.patch
+m68k-bitops.patch
+m68k-checksum-include.patch
+m68k-pgalloc-fixup.patch
+m68k-maintainership.patch

 m68k and related updates

+depends-on-pci-multi-tech-synclink-applicom-serial.patch
+pci-warnings-moxa-serial.patch
+pci-warnings-specialix-serial.patch
+depends-on-pci-via686a-i2c.patch
+depends-on-pci-dma-api-ieee1394-core-and-sbp-2.patch
+depends-on-pci-fritzpci-pciv2-pnp-and-hysdn.patch
+pci-warnings-hisax-isdn.patch
+depends-on-pci-guillemot-maxi-radio-fm-2000.patch
+depends-on-pci-technisat-skystar2-pci.patch
+depends-on-pci-dma-api-cisco-aironet-34x-35x-4500-4800.patch
+depends-on-pci-toshiba-and-via-fir.patch
+depends-on-pci-matrox-1-wire.patch
+dallas-1-wire-delayh.patch

 Various !CONFIG_PCI fixes

+linux-mmzoneh-const.patch

 Add consts to mmzone.h (will probably drop this - it can cause tons of
 warnings)

+intel8x0c-to-include-ck804-audio-support.patch

 Additional sound driver device support

+cirrusfb-update-for-amiga-zorro.patch

 cirrusfb update

+get_random_bytes-returns-the-same-on-every-boot.patch

 make get_random_bytes() more random

+page_cache_readahead-unused-variable.patch

 cleanup

+remove-faulty-__inits-from-drivers-video-fbmemc-fwd.patch

 __init section fixes

+locking-optimization-for-cache_reap.patch

 slab latency fix

+export-all-functions-in-lib-stringc.patch

 Add missing exports

+hlist_for_each_safe-cleanup.patch

 list.h cleanup

+b44-add-47xx-support.patch

 Add device support to b44.c

+fbmon-edd-blacklist.patch

 blacklisting infrastructure for dodgy monitors

+signal-race-fix.patch
+signal-race-fix-ia64.patch
+signal-race-fix-s390.patch
+signal-race-fix-x86_64.patch

 Fixes for SMP signal race

+process-aggregates.patch
+process-aggregates-warning-fix.patch

 Process aggregating infrastructure

+d_unhash-consolidation.patch

 Code consolidation

+front-buttons-wouldnt-mute-ess-maestro.patch

 maestro driver fixlet

+ipv6-routec-gcc-341-fix-inline.patch

 gcc-3.4.1 fix

+config-file-for-laptop-mode.patch
+add-documentation-about-proc-sys-vm-laptop_mode-to-various-docs.patch
+automatically-disable-laptop-mode-when-battery-almost-runs-out.patch

 laptop_mode documentation updates

+ppc32-snd-powermac-requires-i2c.patch

 dependency fix

+ext2_readdir-retval-fix.patch

 Return the correct thing on error

+ncpfs-setattr-retval-fix.patch

 Check a return value

+allow-x86_64-to-reenable-interrupts-on-contention.patch

 Recuced interrupt latency on x86_64

+recommend-noapic-when-timer-via-ioapic-fails.patch

 Add helpful messages when the ioapic timer is discover to be busted.

+move-pit-code-to-timer_pit.patch
+move-pit-code-to-timer_pit-warning-fix.patch

 x86 timer code cleanup

+i2o-build_99.patch
+i2o-build_99-gcc295-fixes.patch

 i2o driver rewrite

+s390-core-changes.patch
+s390-zfcp-host-adapter.patch
+s390-network-driver-changes.patch

 S/390 updates

+dvb-major-number.patch

 Use the right major in DVB

+selinux-fix-clearing-of-new-personality-bit-on-security-transitions.patch

 SELinux fix

+activate-smbus-device-on-hp-d300l.patch

 SMBus quirk for some HP machines

+apic-output-reduction.patch

 Less printk's at boot.

+lost-error-code-in-rescan_partitions.patch

 Error code handling fix

+trivial-doc-patch-for-partitions.patch

 Comment fixes

+rename-config_pci_use_vector-to-config_pci_msi.patch

 Rename things in the MSI driver

+fix-ide-probe-double-detection.patch

 IDE probing fix

+fix-smm-failures-on-e750x-systems.patch

 SMM fixes for certain chipsets

+fix-bogus-ioctl-return-in-mtrr.patch

 mtrr return value fix

+serial-cs-and-unusable-port-size-ranges.patch

 Fix handling of broken GSM GPRS PCMCIA cards.

+remove-boot98.patch

 Remove PC9800 vestiges

+writepages-drops-bh-on-not-uptodate-page.patch

 Fix rare filesytem corruption.

+critical-x86-64-patches-for-268rc2.patch

 x86_64 critical updates





All 371 patches:


linus.patch

fixes-for-rcu_offline_cpu-rcu_move_batch-268-rc2.patch
  fixes for rcu_offline_cpu, rcu_move_batch

unix98-pty-indices-leak.patch
  Fix UNIX98 pty indices leak

sched-initialize-sched-domain-table.patch
  sched: initialize sched domain table

compat_clock_getres-shouldnt-return-efault-if-res-==-null.patch
  compat_clock_getres shouldn't return -EFAULT if res == NULL

bio-page-refcounting-fix.patch
  BIO page refcounting fix

sysfs-leaves-mount.patch
  sysfs backing store: add sysfs_dirent

sysfs-leaves-dir.patch
  sysfs backing store: add sysfs_dirent

sysfs-leaves-file.patch
  sysfs backing store: sysfs_create() changes

sysfs-leaves-bin.patch
  sysfs backing store: bin attribute changes

sysfs-leaves-symlink.patch
  sysfs backing store: sysfs_create_link changes

sysfs-leaves-misc.patch
  sysfs backing store: attribute groups and misc routines

bk-acpi.patch

bk-agpgart.patch

bk-alsa.patch

bk-cifs.patch

bk-dma-declare-coherent-memory.patch

bk-cpufreq.patch

bk-drm.patch

bk-ieee1394.patch

bk-input.patch

bk-kbuild.patch

bk-libata.patch

bk-mpc52xx.patch

bk-netdev.patch

bk-ntfs.patch

bk-pci.patch

bk-pnp.patch

bk-power.patch

bk-scsi.patch

mm.patch
  add -mmN to EXTRAVERSION

longhaul-fix.patch
  longhaul build fix

bk-netdev-axnet_cs-fix.patch
  bk-netdev-axnet_cs-fix

bk-netdev-hp-plus-fix.patch
  bk-netdev-hp-plus-fix

bk-power-x86_64-fix.patch
  bk-power x86_64 fixes

bk-acpi-x86_cpu_to_apicid-fix.patch
  bk-acpi-x86_cpu_to_apicid-fix

nmi-trigger-switch-support-for-debuggingupdated.patch
  NMI trigger switch support for debugging(updated)

nmi-trigger-switch-support-for-debuggingupdated-fix.patch
  nmi-trigger-switch-support-for-debuggingupdated-fix

make-i386-die-more-resilient-against-recursive-errors.patch
  Make i386 die() more resilient against recursive errors

kgdb-ga.patch
  kgdb stub for ia32 (George Anzinger's one)
  kgdbL warning fix
  kgdb buffer overflow fix
  kgdbL warning fix
  kgdb: CONFIG_DEBUG_INFO fix
  x86_64 fixes
  correct kgdb.txt Documentation link (against  2.6.1-rc1-mm2)
  kgdb: fix for recent gcc
  kgdb warning fixes
  THREAD_SIZE fixes for kgdb
  Fix stack overflow test for non-8k stacks
  kgdb-ga.patch fix for i386 single-step into sysenter
  fix TRAP_BAD_SYSCALL_EXITS on i386
  add TRAP_BAD_SYSCALL_EXITS config for i386

kgdboe-netpoll.patch
  kgdb-over-ethernet via netpoll
  kgdboe: fix configuration of MAC address

kgdb-x86_64-support.patch
  kgdb-x86_64-support.patch for 2.6.2-rc1-mm3
  kgdb-x86_64-warning-fixes

kgdb-ia64-support.patch
  IA64 kgdb support
  ia64 kgdb repair and cleanup
  ia64 kgdb fix

kgdb-ia64-fixes.patch
  kgdb: ia64 fixes

make-tree_lock-an-rwlock.patch
  make mapping->tree_lock an rwlock

radix_tree_tag_set-atomic.patch
  Make radix_tree_tag_set/clear atomic wrt the tag

radix_tree_tag_set-only-needs-read_lock.patch
  radix_tree_tag_set only needs read_lock()

must-fix.patch
  must fix lists update
  must fix list update
  mustfix update

must-fix-update-5.patch
  must-fix update

mustfix-lists.patch
  mustfix lists

ppc64-page-align-emergency-stack-2.patch
  page align emergency stack

ppc64-remove-multiple-irq-optimisation.patch
  ppc64: remove multiple IRQ optimisation

ppc64-cpu-hotplug-fix.patch
  ppc64: cpu hotplug fix

ppc64-fix-ras-irq-handlers.patch
  ppc64: Fix RAS irq handlers

ppc64-whitespace-cleanup-in-promc.patch
  ppc64: whitespace cleanup in prom.c

ppc64-hvcs-driver.patch
  ppc64: HVCS driver

ppc64-smt-bugfix.patch
  ppc64 SMT bugfix

ppc32-fix-ppc44x-early-uart-setup.patch
  ppc32: Fix PPC44x early uart setup

ppc32-export-some-dma-api-symbols.patch
  ppc32: export some DMA API symbols

ppc32-fix-comment-in-arch-ppc-platforms-pmac_pcic.patch
  ppc32: fix comment in arch/ppc/platforms/pmac_pci.c

ppc64-reloc_hide.patch

invalidate_inodes-speedup.patch
  invalidate_inodes speedup
  more invalidate_inodes speedup fixes

get_user_pages-handle-VM_IO.patch
  fix get_user_pages() against mappings of /dev/mem

fa311-mac-address-fix.patch
  wrong mac address with netgear FA311 ethernet card

pid_max-fix.patch
  Bug when setting pid_max > 32k

jbd-remove-livelock-avoidance.patch
  JBD: remove livelock avoidance code in journal_dirty_data()

journal_add_journal_head-debug.patch
  journal_add_journal_head-debug

list_del-debug.patch
  list_del debug check

oops-dump-preceding-code.patch
  i386 oops output: dump preceding code

lockmeter.patch
  lockmeter
  ia64 CONFIG_LOCKMETER fix

unplug-can-sleep.patch
  unplug functions can sleep

firestream-warnings.patch
  firestream warnings

ext3_rsv_cleanup.patch
  ext3 block reservation patch set -- ext3 preallocation cleanup

ext3_rsv_base.patch
  ext3 block reservation patch set -- ext3 block reservation
  ext3 reservations: fix performance regression
  ext3 block reservation patch set -- mount and ioctl feature
  ext3 block reservation patch set -- dynamically increase reservation window
  ext3 reservation ifdef cleanup patch
  ext3 reservation max window size check patch
  ext3 reservation file ioctl fix

ext3-reservation-default-on.patch
  ext3 reservation: default to on

ext3-lazy-discard-reservation-window-patch.patch
  ext3 lazy discard reservation window patch
  ext3 discard reservation in last iput fix patch
  Fix lazy reservation discard
  ext3 reservations: bad_inode fix
  ext3 reservation discard race fix

hugetlb_shm_group-sysctl-gid-0-fix.patch
  hugetlb_shm_group sysctl-gid-0-fix

really-ptrace-single-step-2.patch
  ptrace single-stepping fix

ipr-ppc64-depends.patch
  Make ipr.c require ppc

disk-barrier-core.patch
  disk barriers: core
  disk-barrier-core-tweaks

disk-barrier-ide.patch
  disk barriers: IDE
  disk-barrier-ide-symbol-expoprt
  disk-barrier ide warning fix

barrier-update.patch
  barrier update

barrier-flushing-fix.patch
  barrier flushing fix

disk-barrier-scsi.patch
  disk barriers: scsi

disk-barrier-dm.patch
  disk barriers: devicemapper

disk-barrier-md.patch
  disk barriers: MD

reiserfs-v3-barrier-support.patch
  reiserfs v3 barrier support
  reiserfs-v3-barrier-support-tweak

sync_dirty_buffer-retval.patch
  make sync_dirty_buffer() return something useful

ext3-barrier-support.patch
  ext3 barrier support

jbd-barrier-fallback-on-failure.patch
  jbd: barrier fallback on failure

ide-print-failed-opcode.patch
  ide: print failed opcode on IO errors
  From: Jens Axboe <axboe@suse.de>
  Subject: Re: ide errors in 7-rc1-mm1 and later

add-bh_eopnotsupp-for-testing.patch
  add BH_Eopnotsupp for testing async barrier failures

handle-async-barrier-failures.patch
  Handle async barrier failures

enable-suspend-resuming-of-e1000.patch
  Enable suspend/resuming of e1000

tty_io-hangup-locking.patch
  tty_io.c hangup locking

perfctr-core.patch
  From: Mikael Pettersson <mikpe@csd.uu.se>
  Subject: [PATCH][1/6] perfctr-2.7.3 for 2.6.7-rc1-mm1: core
  CONFIG_PERFCTR=n build fix
  From: Mikael Pettersson <mikpe@csd.uu.se>
  Subject: [PATCH][6/6] perfctr-2.7.3 for 2.6.7-rc1-mm1: misc

perfctr-i386.patch
  From: Mikael Pettersson <mikpe@csd.uu.se>
  Subject: [PATCH][2/6] perfctr-2.7.3 for 2.6.7-rc1-mm1: i386
  perfctr #if/#ifdef cleanup
  perfctr Dothan support
  perfctr x86_tests build fix

perfctr-x86-init-bug.patch
  perfctr x86 init bug

perfctr-k8-fix-for-internal-benchmarking-code.patch
  perfctr: K8 fix for internal benchmarking code

perfctr-x86_64.patch
  From: Mikael Pettersson <mikpe@csd.uu.se>
  Subject: [PATCH][3/6] perfctr-2.7.3 for 2.6.7-rc1-mm1: x86_64

perfctr-ppc.patch
  From: Mikael Pettersson <mikpe@csd.uu.se>
  Subject: [PATCH][4/6] perfctr-2.7.3 for 2.6.7-rc1-mm1: PowerPC
  perfctr ppc32 update
  perfctr update 4/6: PPC32 cleanups

perfctr-ppc32-buglet-fix.patch
  perfctr ppc32 buglet fix

perfctr-virtualised-counters.patch
  From: Mikael Pettersson <mikpe@csd.uu.se>
  Subject: [PATCH][5/6] perfctr-2.7.3 for 2.6.7-rc1-mm1: virtualised counters
  perfctr update 6/6: misc minor cleanups
  perfctr update 3/6: __user annotations
  perfctr-cpus_complement-fix
  perfctr cpumask cleanup

perfctr-ifdef-cleanup.patch
  perfctr ifdef cleanup

perfctr-update-2-6-kconfig-related-updates.patch
  perfctr update 2/6: Kconfig-related updates

perfctr-update-5-6-reduce-stack-usage.patch
  perfctr update 5/6: reduce stack usage

perfctr-low-level-documentation.patch
  perfctr low-level documentation

perfctr-documentation-update.patch
  perfctr documentation update

perfctr-inheritance-1-3-driver-updates.patch
  perfctr inheritance 1/3: driver updates

perfctr-inheritance-illegal-sleep-bug.patch
  perfctr inheritance illegal sleep bug

perfctr-inheritance-2-3-kernel-updates.patch
  perfctr inheritance 2/3: kernel updates

perfctr-inheritance-3-3-documentation-updates.patch
  perfctr inheritance 3/3: documentation updates

ext3-online-resize-patch.patch
  ext3: online resizing

ext3-online-resize-warning-fix.patch
  ext3-online-resize-warning-fix

sched-clean-init-idle.patch
  sched: cleanup init_idle()

sched-clean-fork.patch
  sched: cleanup, improve sched <=> fork APIs

sched-clean-fork-rename-wake_up_new_process-wake_up_new_task.patch
  sched: rename wake_up_new_process -> wake_up_new_task

kernelthread-idle-fix-2.patch
  kernel thread idle fix

sched-misc-cleanups-2.patch
  sched: misc cleanups #2

sched-unlikely-rt_task.patch
  sched: make rt_task unlikely

sched-misc.patch
  sched: sched misc changes

sched-misc-fix-rt.patch
  sched: fix RT scheduling & interactivity estimator

sched-no-balance-clone.patch
  sched: disable balance on clone

sched-remove-balance-clone.patch
  sched: remove balance on clone

sched-fork-hotplug-cleanuppatch.patch
  sched: fork hotplug hanling cleanup

sched-consolidate-sched-domains.patch
  sched: consolidate sched domains

sched-consolidate-domains-fix.patch
  sched: fix for sched-consolidate-domains

sched-consolidate-domains-fix-2.patch
  another sched consolidate domains fix

sched-domain-node-span-4.patch
  sched: limit cpuspan of node scheduler domains

sched-domain-node-span-4-warning-fix.patch
  sched-domain-node-span-4-warning-fix

memory-backed-inodes-fix.patch
  memory-backed inodes fix

ext3_bread-cleanup.patch
  ext3_bread() cleanup

flexible-mmap-2.6.7-mm3-A8.patch
  i386 virtual memory layout rework

flexible-mmap-bug-fix.patch
  flexible-mmap BUG fix

flexible-mmap-updatepatch-267-mm5.patch
  flexible-mmap update

driver-model-and-sysfs-support-for-pcmcia-1-3.patch
  driver model and sysfs support for PCMCIA (1/3)

driver-model-and-sysfs-support-for-pcmcia-update.patch
  driver model and sysfs support for PCMCIA update

update-drivers-net-pcmcia-2-3.patch
  update drivers/net/pcmcia (2/3)

update-drivers-net-wireless-3-3.patch
  update drivers/net/wireless (3/3)

posix-locking-fix-to-posix_same_owner.patch
  posix locking: Minimal fix to posix_same_owner()

posix-locking-fix-to-locking-code.patch
  posix locking: more locking code fixes

posix-locking-fix-up-nfs4statec.patch
  posix locking: Fix up nfs4state.c

posix-locking-fix-up-lockd.patch
  posix locking: Fix up lockd to make use of the new interface

posix-locking-fl_owner_t-to-pid-mapping.patch
  posix locking: mapping between fl_owner_t and client-side "pid"

dvdrw-support-for-267-bk13.patch
  DVD+RW support for 2.6.7-bk13

cdrw-packet-writing-support-for-267-bk13.patch
  CDRW packet writing support

dvd-rw-packet-writing-update.patch
  Packet writing support for DVD-RW and DVD+RW discs.

fix-race-in-pktcdvd-kernel-thread-handling.patch
  Fix race in pktcdvd kernel thread handling

fix-open-close-races-in-pktcdvd.patch
  Fix open/close races in pktcdvd

packet-writing-review-fixups.patch
  packet writing: review fixups

remove-pkt_dev-from-struct-pktcdvd_device.patch
  Remove pkt_dev from struct pktcdvd_device

packet-writing-docco.patch
  packet writing documentation

convert-packet-writing-to-seq_file.patch
  packet writing: convert to seq_file

control-pktcdvd-with-an-auxiliary-character-device.patch
  Control pktcdvd with an auxiliary character device

control-pktcdvd-with-an-auxiliary-character-device-fix.patch
  control-pktcdvd-with-an-auxiliary-character-device-fix

r8169_napi-help-text-2.patch
  R8169_NAPI help text

no-sysgood-for-ptrace-singlestep.patch
  Don't use SYSGOOD for ptrace singlestep

err2-6-hashbin_remove_this-locking-fix.patch
  err2-6: hashbin_remove_this() locking fix

dm-use-idr.patch
  devicemapper: use an IDR tree for tracking minors

ipc-1-3-add-refcount-to-ipc_rcu_alloc.patch
  ipc: Add refcount to ipc_rcu_alloc

ipc-2-3-remove-sem_revalidate.patch
  ipc: remove sem_revalidate

ipc-3-3-enforce-semvmx-limit-for-undo.patch
  ipc: enforce SEMVMX limit for undo

cleanup-of-ipc-msgc.patch
  cleanup of ipc/msg.c

sk98lin-procfs-fix.patch
  sk98lin procfs fix

cpufreq-driver-for-nforce2-kernel-267.patch
  cpufreq driver for nForce2

allow-modular-ide-pnp.patch
  allow modular ide-pnp

uml-base-patch.patch
  uml: Uml base patch

rename-uml-console-device.patch
  uml: rename console_device

uml-readds-just-for-now-ghashh-for-uml.patch
  uml: Readds (just for now) ghash.h for UML

uml-avoid-that-gcc-breaks-uml-with-unit-at-a-time-compilation-mode.patch
  uml: Avoid that gcc breaks UML with "unit at a time" compilation mode.

uml-fixes-an-host-fd-leak-caused-by-hostfs.patch
  uml: Fixes an host fd leak caused by hostfs.

uml-adds-legacy_pty-config-option.patch
  uml: Adds LEGACY_PTY config option

uml-makes-make-help-arch=um-work.patch
  uml: Makes "make help ARCH=um" work.

uml-fixes-fixdepc-to-support-arch-um-include-uml-configh.patch
  uml: Fixes "fixdep.c" to support arch/um/include/uml-config.h.

uml-kill-useless-warnings.patch
  uml: Kill useless warnings

uml-avoids-compile-failure-when-host-misses-tkill.patch
  uml: Avoids compile failure when host misses tkill().

uml-reduces-code-in-_user-files-by-moving-it-in-_kern-files-if-already-possible.patch
  uml: Reduces code in *_user files, by moving it in _kern files if already possible.

uml-fixes-raw-and-uses-it-in-check_one_sigio-also-fixes-a-silly-panic-eintr-returned-by-call.patch
  uml: Fixes raw() and uses it in check_one_sigio; also fixes a silly panic (EINTR returned by call).

uml-folds-hostaudio_userc-into-hostaudio_kernc.patch
  uml: Folds hostaudio_user.c into hostaudio_kern.c.

uml-use-ptrace_scemu-the-so-called-sysemu-to-reduce-syscall-cost.patch
  uml: Use PTRACE_SCEMU (the so-called SYSEMU) to reduce syscall cost.

uml-adds-the-nosysemu-command-line-parameter-to-disable-sysemu.patch
  uml: Adds the "nosysemu" command line parameter to disable SYSEMU

uml-adds-proc-sysemu-to-toggle-sysemu-usage.patch
  uml: Adds /proc/sysemu to toggle SYSEMU usage.

uml-fix-for-sysemu-patches.patch
  uml: Fix for sysemu patches

uml-handles-correctly-errno-==-eintr-in-lots-of-places.patch
  uml: Handles correctly errno == EINTR in lots of places.

uml-adds-some-exports.patch
  uml: Adds some exports

uml-avoids-a-panic-for-a-legal-situation.patch
  uml: Avoids a panic for a legal situation

uml-removes-dead-code-in-trap_kernc.patch
  uml: Removes dead code in trap_kern.c

uml-make-malloc-call-vmalloc-if-needed-needed-for-hostfs-on-26-host.patch
  uml: Make malloc() call vmalloc if needed. Needed for hostfs on 2.6 host.

uml-little-kmalloc.patch
  uml: little-kmalloc

uml-fix-os_process_pc-and-os_process_parent-for-corner-cases.patch
  uml: Fix os_process_pc and os_process_parent for corner cases.

fix-warnings-in-net-irda.patch
  sparse: fix warnings in net/irda/*

i810_audio-mmio-support.patch
  i810_audio MMIO support

i810_audio-mmio-support-2.patch
  i810_audio MMIO support #2

i810_audio-fix-the-error-path-of-resource-management.patch
  i810_audio: Fix the error path of resource management

fix-drivers-isdn-hisax-avm_pcic-build-warning-when.patch
  Fix drivers/isdn/hisax/avm_pci.c build warning when !CONFIG_ISAPNP

idr-stale-comment.patch
  idr.c: remove stale comment

idr-comments-updates.patch
  idr comments updates

schedule-profiling.patch
  schedule() profiling
  From: Arjan van de Ven <arjanv@redhat.com>
  Subject: Re: schedule profileing

add-a-few-might_sleep-checks.patch
  Add a few might_sleep() checks

add-a-few-might_sleep-checks-fix.patch
  add-a-few-might_sleep-checks fix

release_task-may-sleep.patch
  permit sleeping in release_task()

ia64-ptrace-fix-fix.patch
  Make get_user_pages() work again for ia64 gate area

possible-buglet-in-drivers-input-joystick-tmdcc.patch
  Possible buglet in drivers/input/joystick/tmdc.c

crc16-renaming-in-via-velocity-ethernet-driver.patch
  CRC16 renaming in VIA Velocity ethernet driver

per_cpu-per_cpu-cpu_gdt_table.patch
  percpu: cpu_gdt_table

per_cpu-per_cpu-cpu_gdt_table-fix.patch
  per_cpu-per_cpu-cpu_gdt_table-fix

per_cpu-per_cpu-init_tss.patch
  percpu: init_tss

per_cpu-per_cpu-cpu_tlbstate.patch
  percpu: cpu_tlbstate

gcc35-advansys.c.patch
  gcc-3.5 fixes

gcc35-alps_tdlb7.c.patch
  gcc-3.5 fixes

gcc35-always-inline.patch
  gcc-3.5 fixes

gcc35-arlan.h.patch
  gcc-3.5 fixes

gcc35-auerswald.c.patch
  gcc-3.5 fixes

gcc35-dabusb.c.patch
  gcc-3.5 fixes

gcc35-ds.c.patch
  gcc-3.5 fixes

gcc35-fixmap.h.patch
  gcc-3.5 fixes

gcc35-fore200e.c.patch
  gcc-3.5 fixes

gcc35-index.html.patch
  gcc-3.5 fixes

gcc35-ip6_fib.c.patch
  gcc-3.5 fixes

gcc35-iphase.h.patch
  gcc-3.5 fixes

gcc35-irttp.h.patch
  gcc-3.5 fixes

gcc35-mtrr.h.patch
  gcc-3.5 fixes

gcc35-netrom.h.patch
  gcc-3.5 fixes

gcc35-pppoe.c.patch
  gcc-3.5 fixes

gcc35-sonypi.patch
  gcc-3.5 fixes

gcc35-sp887x.c.patch
  gcc-3.5 fixes

gcc35-tda1004x.c.patch
  gcc-3.5 fixes

gcc35-transport.h.patch
  gcc-3.5 fixes

gcc35-ufs_fs.h.patch
  gcc-3.5 fixes

gcc35-usblp.c.patch
  gcc-3.5 fixes

gcc35-videodev.c.patch
  gcc-3.5 fixes

gcc35-wavefront_fx.c.patch
  gcc-3.5 fixes

gcc35-xfrm6_state.c.patch
  gcc-3.5 fixes

fix-rivafbs-nv_arch_-cleanup-debug-backlight-control-on-ppc.patch
  Fix rivafb's NV_ARCH_, cleanup DEBUG, backlight control on ppc

fix-rivafbs-nv_arch_-cleanup-debug-backlight-control-on-ppc-fix.patch
  fix-rivafbs-nv_arch_-cleanup-debug-backlight-control-on-ppc fix

dev-zero-vs-hugetlb-mappings.patch
  /dev/zero vs hugetlb mappings.

hugetlbfs-vm_pgoff-bugs.patch
  hugetlbfs vm_pgoff bugs

hugetlbfs-private-mappings.patch
  hugetlbfs private mappings

net-kconfig-crc16-fix.patch
  net/Kconfig crc16 warning fix

preset-loops_per_jiffy-for-faster-booting.patch
  preset loops_per_jiffy for faster booting

define-inline-as-__attribute__always_inline-also-for-gcc-=-34.patch
  #define inline as __attribute__((always_inline)) also for gcc >= 3.4

gcc-34-and-broken-inlining.patch
  clean up __always_inline__ usage

handle-undefined-symbols.patch
  Fail if vmlinux contains undefined symbols

split-generic_file_aio_write-into-buffered-and-direct-i-o-parts.patch
  split generic_file_aio_write into buffered and direct I/O parts

unknown-symbol-in-drivers-scsi-pcmcia-fdomain_csko.patch
  fdomain_cs needs ISA

radeonfb-cleanup-and-little-fixes.patch
  radeonfb: cleanup and little fixes

making-i-dhash_entries-cmdline-work-as-it-use-to.patch
  Make i/dhash_entries cmdline work as it use to.

making-i-dhash_entries-cmdline-work-as-it-use-to-fix.patch
  making-i-dhash_entries-cmdline-work-as-it-use-to-fix

fix-double-reset-in-aic7xxx-driver.patch
  Fix double reset in aic7xxx driver

rivafb-i2c-fixes.patch
  Rivafb I2C fixes

jbd-recovery-latency-fix.patch
  jbd recovery latency fix

truncate_inode_pages-latency-fix.patch
  truncate_inode_pages-latency-fix

journal_clean_checkpoint_list-latency-fix.patch
  journal_clean_checkpoint_list latency fix

kjournald-smp-latency-fix.patch
  kjournald-smp-latency-fix

unmap_vmas-smp-latency-fix.patch
  unmap_vmas-smp-latency-fix

__cleanup_transaction-latency-fix.patch
  __cleanup_transaction-latency-fix

prune_dcache-latency-fix.patch
  prune_dcache-latency-fix

filemap_sync-latency-fix.patch
  filemap_sync-latency-fix

slab-latency-fix.patch
  slab-latency-fix

get_user_pages-latency-fix.patch
  get_user_pages-latency-fix

oom-show_free_areas.patch
  oom-killer: call show_free_areas

send_IPI_mask_bitmask-build-fix.patch
  send_IPI_mask_bitmask() build fix

e1000-build-fix.patch
  e1000 build fix

pty_write-latency-fix.patch
  pty_write-latency-fix

v850-define-find_first_bit.patch
  v850: Define find_first_bit

enable-all-events-for-initramfs.patch
  Enable all events for initramfs

radeonfb-64-bit-fix.patch
  radeonfb x86_64 fix

arch-i386-kernel-smpc-gcc341-inlining-fix.patch
  arch/i386/kernel/smp.c gcc341 inlining fix

net-sunrpc-xprtc-gcc341-inlining-fix.patch
  net/sunrpc/xprt.c gcc341 inlining fix

fix-menuconfig-partial-inability-to-show-help-texts.patch
  Fix menuconfig partial inability to show help texts.

was-removal-of-sync-in-panic.patch
  remove sync() from panic

move-cache_reap-out-of-timer-context.patch
  Move cache_reap out of timer context

move-cache_reap-out-of-timer-context-fix.patch
  move-cache_reap-out-of-timer-context-fix

switch-sgc-to-standard-jiffies-converters.patch
  switch sg.c to standard jiffies converters

sign-fix-in-swapfilec.patch
  sign fix in swapfile.c

a-trivial-patch-for-removing-unnecessary-comment-in-mm-filemapc.patch
  Remove dead comment in mm/filemap.c

gettimeofday-nanoseconds-patch-makes-it-possible-for-the-posix-timer.patch
  gettimeofday nanoseconds patch

fix-for-buffer-limit-for-long-in-sysctlc.patch
  fix for buffer limit for long in sysctl.c

quiet-down-per-zone-memory-stats.patch
  quieten down per-zone memory stats

x86-64-singlestep-through-sigreturn-system-call-2.patch
  Fix x86-64 singlestep through sigreturn system call

create-nodemask_t.patch
  Create nodemask_t

ipmi_msghandler-module-load-failure-fix.patch
  ipmi_msghandler module load failure fix

fat-kill-nls-default.patch
  FAT: kill nls default

add-ixdp2x01-board-support-to-cs89x0-driver.patch
  Add IXDP2x01 board support to CS89x0 driver

remove-scripts-mkconfigs.patch
  remove scripts/mkconfigs

267-msi-x-update.patch
  MSI-X Update

remove-dead-prototypes.patch
  remove dead prototypes

s390-use-include-asm-generic-dma-mapping-brokenh.patch
  s390: Use include/asm-generic/dma-mapping-broken.h

fix-readahead-breakage-for-sequential-after-random-reads.patch
  fix readahead breakage for sequential after random reads

cdrom-get_last_written-fix.patch
  Subject: cdrom.c get_last_written fixup

update-mailing-list-for-osst.patch
  MAINTAINERS: update mailing list for osst

fix-aic-for-db4.patch
  fix aic driver build for db4

m68k-68060-errata-i14.patch
  M68k 68060 errata I14

m68k-ifpsp060.patch
  M68k ifpsp060

m68k-sparse-missing-void.patch
  m68k sparse missing void

m68k-sparse-if-vs-ifdef.patch
  m68k sparse #if vs. #ifdef

m68k-sparse-void-return.patch
  m68k sparse void return

m68k-sparse-extern.patch
  m68k sparse extern

m68k-sparse-inline.patch
  m68k sparse inline

dsp56k-sparse-const.patch
  dsp56k sparse const

m68k-sparse-floating-point.patch
  m68k sparse floating point

dnfb-sparse-struct-init.patch
  dnfb sparse struct init

amifb-sparse-=.patch
  amifb sparse &=

m68k-hardirqh.patch
  m68k hardirq.h

dmasound-paths.patch
  dmasound paths

m68k-bitops.patch
  M68k bitops

m68k-checksum-include.patch
  M68k checksum include

m68k-pgalloc-fixup.patch
  M68k pgalloc fixup

m68k-maintainership.patch
  M68k Maintainership

depends-on-pci-multi-tech-synclink-applicom-serial.patch
  depends on PCI: Multi-Tech, SyncLink, Applicom serial

pci-warnings-moxa-serial.patch
  !PCI warnings: Moxa serial

pci-warnings-specialix-serial.patch
  !PCI warnings: Specialix serial

depends-on-pci-via686a-i2c.patch
  depends on PCI: VIA686A i2c

depends-on-pci-dma-api-ieee1394-core-and-sbp-2.patch
  depends on PCI DMA API: IEEE1394 core and SBP-2

depends-on-pci-fritzpci-pciv2-pnp-and-hysdn.patch
  depends on PCI: Fritz!PCI/PCIv2/PnP and HYSDN

pci-warnings-hisax-isdn.patch
  !PCI warnings: Hisax ISDN

depends-on-pci-guillemot-maxi-radio-fm-2000.patch
  depends on PCI: Guillemot MAXI Radio FM 2000

depends-on-pci-technisat-skystar2-pci.patch
  depends on PCI: Technisat Skystar2 PCI

depends-on-pci-dma-api-cisco-aironet-34x-35x-4500-4800.patch
  depends on PCI DMA API: Cisco/Aironet 34X/35X/4500/4800

depends-on-pci-toshiba-and-via-fir.patch
  depends on PCI: Toshiba and VIA FIR

depends-on-pci-matrox-1-wire.patch
  depends on PCI: Matrox 1-wire

dallas-1-wire-delayh.patch
  Dallas 1-wire delay.h

linux-mmzoneh-const.patch
  <linux/mm{,zone}.h> const

intel8x0c-to-include-ck804-audio-support.patch
  intel8x0.c to include CK804 audio support

cirrusfb-update-for-amiga-zorro.patch
  cirrusfb: update for amiga (zorro)

get_random_bytes-returns-the-same-on-every-boot.patch
  get_random_bytes() returns the same on every boot

page_cache_readahead-unused-variable.patch
  page_cache_readahead unused variable

remove-faulty-__inits-from-drivers-video-fbmemc-fwd.patch
  remove faulty __init's from drivers/video/fbmem.c

locking-optimization-for-cache_reap.patch
  slab: locking optimization for cache_reap

export-all-functions-in-lib-stringc.patch
  Export all functions in lib/string.c

hlist_for_each_safe-cleanup.patch
  hlist_for_each_safe cleanup

b44-add-47xx-support.patch
  b44: add 47xx support

fbmon-edd-blacklist.patch
  fbcom: EDD-based blacklisting

signal-race-fix.patch
  signal handling race fix

signal-race-fix-ia64.patch
  signal-race-fix: ia64

signal-race-fix-s390.patch
  signal-race fixes for s390

signal-race-fix-x86_64.patch
  signal-race-fixes: x86-64 support

process-aggregates.patch
  Process Aggregates (PAGG)

process-aggregates-warning-fix.patch
  process-aggregates warning fix

d_unhash-consolidation.patch
  d_unhash consolidation

front-buttons-wouldnt-mute-ess-maestro.patch
  front buttons wouldn't mute ESS Maestro

ipv6-routec-gcc-341-fix-inline.patch
  ipv6/route.c gcc-341 fix inline

config-file-for-laptop-mode.patch
  Config file for laptop mode.

add-documentation-about-proc-sys-vm-laptop_mode-to-various-docs.patch
  Add documentation about /proc/sys/vm/laptop_mode to various docs.

automatically-disable-laptop-mode-when-battery-almost-runs-out.patch
  Automatically disable laptop mode when battery almost runs out.

ppc32-snd-powermac-requires-i2c.patch
  ppc32: snd-powermac requires i2c

ext2_readdir-retval-fix.patch
  ext2_readdir() return value fix

ncpfs-setattr-retval-fix.patch
  ncpfs: setattr return value fix

allow-x86_64-to-reenable-interrupts-on-contention.patch
  Allow x86_64 to reenable interrupts on contention

recommend-noapic-when-timer-via-ioapic-fails.patch
  Recommend 'noapic' when timer via IOAPIC fails

move-pit-code-to-timer_pit.patch
  x86: move PIT code to timer_pit

move-pit-code-to-timer_pit-warning-fix.patch
  move-pit-code-to-timer_pit-warning-fix

i2o-build_99.patch
  i20 rewrite

i2o-build_99-gcc295-fixes.patch
  i2o-build_99-gcc295-fixes

s390-core-changes.patch
  s390: core changes

s390-zfcp-host-adapter.patch
  s390: zfcp host adapter.

s390-network-driver-changes.patch
  s390: network driver changes

dvb-major-number.patch
  From: Alan Cox <alan@redhat.com>
  Subject: DVB major number

selinux-fix-clearing-of-new-personality-bit-on-security-transitions.patch
  selinux: fix clearing of new personality bit on security transitions

activate-smbus-device-on-hp-d300l.patch
  activate SMBus device on hp d300l

apic-output-reduction.patch
  Subject: Re: Fw: [Fwd: Re: [Kernel-janitors] [PATCH] IO-APIC debug message 	reducti]

lost-error-code-in-rescan_partitions.patch
  lost error code in rescan_partitions

trivial-doc-patch-for-partitions.patch
  trivial doc patch for partitions

rename-config_pci_use_vector-to-config_pci_msi.patch
  rename CONFIG_PCI_USE_VECTOR to CONFIG_PCI_MSI

fix-ide-probe-double-detection.patch
  Fix ide probe double detection

fix-smm-failures-on-e750x-systems.patch
  fix SMM failures on E750x systems

fix-bogus-ioctl-return-in-mtrr.patch
  Subject: PATCH: fix bogus ioctl return in mtrr

serial-cs-and-unusable-port-size-ranges.patch
  serial-cs and unusable port size ranges

remove-boot98.patch
  remove boot98

writepages-drops-bh-on-not-uptodate-page.patch
  writepages drops bh on not uptodate page

critical-x86-64-patches-for-268rc2.patch
  Critical x86-64 patches for 2.6.8rc2

Re: 2.6.8-rc2-mm1

[Jesse Barnes]

[-- Attachment #1: Type: text/plain, Size: 375 bytes --]

On Wednesday, July 28, 2004 2:04 am, Andrew Morton wrote:
> ftp://ftp.kernel.org/pub/linux/kernel/people/akpm/patches/2.6/2.6.8-rc2/2.6
>.8-rc2-mm1/

sd->groups isn't initialized this early (I know because I ran into this when I 
first tried to merge Suresh's patch too).  You've already got the real fix in 
your tree, you just need this little bit torn out.

Thanks,
Jesse

[-- Attachment #2: sched-merge-fix.patch --]
[-- Type: text/plain, Size: 559 bytes --]

diff -Napur -X /home/jbarnes/dontdiff linux-2.6.8-rc2-mm1.orig/kernel/sched.c linux-2.6.8-rc2-mm1/kernel/sched.c
--- linux-2.6.8-rc2-mm1.orig/kernel/sched.c	2004-07-28 09:36:36.000000000 -0700
+++ linux-2.6.8-rc2-mm1/kernel/sched.c	2004-07-28 09:34:45.000000000 -0700
@@ -3770,8 +3770,6 @@ __init static void arch_init_sched_domai
 		cpumask_t nodemask = node_to_cpumask(cpu_to_node(i));
 
 #ifdef CONFIG_NUMA
-		if (i != first_cpu(sd->groups->cpumask))
-			continue;
 		sd = &per_cpu(node_domains, i);
 		group = cpu_to_node_group(i);
 		*sd = SD_NODE_INIT;

Re: 2.6.8-rc2-mm1 (compile stats)

[John Cherry]

Linux 2.6 (mm tree) Compile Statistics (gcc 3.2.2)

Kernel            bzImage   bzImage  bzImage  modules  bzImage  modules
                (defconfig) (allno) (allyes) (allyes) (allmod) (allmod)
--------------- ---------- -------- -------- -------- -------- --------
2.6.8-rc2-mm1     0w/0e     0w/0e    88w/9e    3w/0e   1w/0e     81w/0e
2.6.8-rc1-mm1     0w/0e     0w/0e    88w/9e    5w/0e   1w/0e     87w/0e
2.6.7-mm7         0w/0e     0w/0e    89w/9e    5w/0e   1w/0e     84w/0e
2.6.7-mm6         0w/0e     0w/0e    85w/9e    5w/0e   1w/0e     80w/0e
2.6.7-mm5         0w/0e     0w/0e    92w/0e    5w/0e   1w/0e     87w/0e
2.6.7-mm4         0w/0e     0w/0e    94w/0e    5w/0e   1w/0e     89w/0e
2.6.7-mm3         0w/0e     0w/0e    90w/6e    5w/0e   1w/0e     86w/0e
2.6.7-mm2         0w/0e     0w/0e   109w/0e    7w/0e   1w/0e    106w/0e
2.6.7-mm1         0w/0e     5w/0e   108w/0e    5w/0e   1w/0e    104w/0e
2.6.7-rc3-mm2     0w/0e     5w/0e   105w/10e   5w/0e   2w/0e    100w/2e
2.6.7-rc3-mm1     0w/0e     5w/0e   104w/10e   5w/0e   2w/0e    100w/2e
2.6.7-rc2-mm2     0w/0e     5w/0e   109w/10e   5w/0e   2w/0e    105w/2e
2.6.7-rc2-mm1     0w/0e    12w/0e   158w/13e   5w/0e   3w/0e    153w/4e
2.6.7-rc1-mm1     0w/0e     6w/0e   108w/0e    5w/0e   2w/0e    104w/0e
2.6.6-mm5         0w/0e     0w/0e   109w/5e    5w/0e   2w/0e    110w/0e
2.6.6-mm4         0w/0e     0w/0e   112w/9e    5w/0e   2w/5e    106w/1e
2.6.6-mm3         3w/9e     0w/0e   120w/26e   5w/0e   2w/0e    114w/10e
2.6.6-mm2         4w/11e    0w/0e   120w/24e   6w/0e   2w/0e    118w/9e
2.6.6-mm1         1w/0e     0w/0e   118w/25e   6w/0e   2w/0e    114w/10e
2.6.6-rc3-mm2     0w/0e     0w/0e   117w/ 0e   8w/0e   2w/0e    116w/0e
2.6.6-rc3-mm1     0w/0e     0w/0e   120w/10e   8w/0e   2w/0e    152w/2e
2.6.6-rc2-mm2     0w/0e     1w/5e   118w/ 0e   8w/0e   3w/0e    118w/0e
2.6.6-rc2-mm1     0w/0e     0w/0e   115w/ 0e   7w/0e   3w/0e    116w/0e
2.6.6-rc1-mm1     0w/0e     0w/7e   122w/ 0e   7w/0e   4w/0e    122w/0e
2.6.5-mm6         0w/0e     0w/0e   123w/ 0e   7w/0e   4w/0e    124w/0e
2.6.5-mm5         0w/0e     0w/0e   119w/ 0e   7w/0e   4w/0e    120w/0e
2.6.5-mm4         0w/0e     0w/0e   120w/ 0e   7w/0e   4w/0e    121w/0e
2.6.5-mm3         0w/0e     1w/0e   121w/12e   7w/0e   3w/0e    123w/0e
2.6.5-mm2         0w/0e     0w/0e   128w/12e   7w/0e   3w/0e    134w/0e
2.6.5-mm1         0w/0e     5w/0e   122w/ 0e   7w/0e   3w/0e    124w/0e
2.6.5-rc3-mm4     0w/0e     0w/0e   124w/ 0e   8w/0e   4w/0e    126w/0e
2.6.5-rc3-mm3     0w/0e     5w/0e   129w/14e   8w/0e   4w/0e    129w/6e
2.6.5-rc3-mm2     0w/0e     5w/0e   130w/14e   8w/0e   4w/0e    129w/6e
2.6.5-rc3-mm1     0w/0e     5w/0e   129w/ 0e   8w/0e   4w/0e    129w/0e
2.6.5-rc2-mm5     0w/0e     5w/0e   130w/ 0e   8w/0e   4w/0e    129w/0e
2.6.5-rc2-mm4     0w/0e     5w/0e   134w/ 0e   8w/0e   3w/0e    133w/0e
2.6.5-rc2-mm3     0w/0e     5w/0e   134w/ 0e   8w/0e   3w/0e    133w/0e
2.6.5-rc2-mm2     0w/0e     5w/0e   137w/ 0e   8w/0e   3w/0e    134w/0e
2.6.5-rc2-mm1     0w/0e     5w/0e   136w/ 0e   8w/0e   3w/0e    134w/0e
2.6.5-rc1-mm2     0w/0e     5w/0e   135w/ 5e   8w/0e   3w/0e    133w/0e
2.6.5-rc1-mm1     0w/0e     5w/0e   135w/ 5e   8w/0e   3w/0e    133w/0e
2.6.4-mm2         1w/2e     5w/2e   144w/10e   8w/0e   3w/2e    144w/0e
2.6.4-mm1         1w/0e     5w/0e   146w/ 5e   8w/0e   3w/0e    144w/0e
2.6.4-rc2-mm1     1w/0e     5w/0e   146w/12e  11w/0e   3w/0e    147w/2e
2.6.4-rc1-mm2     1w/0e     5w/0e   144w/ 0e  11w/0e   3w/0e    145w/0e
2.6.4-rc1-mm1     1w/0e     5w/0e   147w/ 5e  11w/0e   3w/0e    147w/0e
2.6.3-mm4         1w/0e     5w/0e   146w/ 0e   7w/0e   3w/0e    142w/0e
2.6.3-mm3         1w/2e     5w/2e   146w/15e   7w/0e   3w/2e    144w/5e
2.6.3-mm2         1w/8e     5w/0e   140w/ 0e   7w/0e   3w/0e    138w/0e
2.6.3-mm1         1w/0e     5w/0e   143w/ 5e   7w/0e   3w/0e    141w/0e
2.6.3-rc3-mm1     1w/0e     0w/0e   144w/13e   7w/0e   3w/0e    142w/3e
2.6.3-rc2-mm1     1w/0e     0w/265e 144w/ 5e   7w/0e   3w/0e    145w/0e
2.6.3-rc1-mm1     1w/0e     0w/265e 141w/ 5e   7w/0e   3w/0e    143w/0e
2.6.2-mm1         2w/0e     0w/264e 147w/ 5e   7w/0e   3w/0e    173w/0e
2.6.2-rc3-mm1     2w/0e     0w/265e 146w/ 5e   7w/0e   3w/0e    172w/0e
2.6.2-rc2-mm2     0w/0e     0w/264e 145w/ 5e   7w/0e   3w/0e    171w/0e
2.6.2-rc2-mm1     0w/0e     0w/264e 146w/ 5e   7w/0e   3w/0e    172w/0e
2.6.2-rc1-mm3     0w/0e     0w/265e 144w/ 8e   7w/0e   3w/0e    169w/0e
2.6.2-rc1-mm2     0w/0e     0w/264e 144w/ 5e  10w/0e   3w/0e    171w/0e
2.6.2-rc1-mm1     0w/0e     0w/264e 144w/ 5e  10w/0e   3w/0e    171w/0e
2.6.1-mm5         2w/5e     0w/264e 153w/11e  10w/0e   3w/0e    180w/0e
2.6.1-mm4         0w/821e   0w/264e 154w/ 5e   8w/1e   5w/0e    179w/0e
2.6.1-mm3         0w/0e     0w/0e   151w/ 5e  10w/0e   3w/0e    177w/0e
2.6.1-mm2         0w/0e     0w/0e   143w/ 5e  12w/0e   3w/0e    171w/0e
2.6.1-mm1         0w/0e     0w/0e   146w/ 9e  12w/0e   6w/0e    171w/0e
2.6.1-rc2-mm1     0w/0e     0w/0e   149w/ 0e  12w/0e   6w/0e    171w/4e
2.6.1-rc1-mm2     0w/0e     0w/0e   157w/15e  12w/0e   3w/0e    185w/4e
2.6.1-rc1-mm1     0w/0e     0w/0e   156w/10e  12w/0e   3w/0e    184w/2e
2.6.0-mm2         0w/0e     0w/0e   161w/ 0e  12w/0e   3w/0e    189w/0e
2.6.0-mm1         0w/0e     0w/0e   173w/ 0e  12w/0e   3w/0e    212w/0e

Web page with links to complete details:
   http://developer.osdl.org/cherry/compile/

John

Re: 2.6.8-rc2-mm1

[Peter Osterlund]

Andrew Morton <akpm@osdl.org> writes:

> ftp://ftp.kernel.org/pub/linux/kernel/people/akpm/patches/2.6/2.6.8-rc2/2.6.8-rc2-mm1/
[...]
> Changes since 2.6.8-rc1-mm1:
[...]
> +control-pktcdvd-with-an-auxiliary-character-device.patch
> +control-pktcdvd-with-an-auxiliary-character-device-fix.patch
> 
>  CDRW/DVDRW packet writing updates

The control-pktcdvd-with-an-auxiliary-character-device patch
introduced a door locking bug.

    pktsetup, mount, umount -> door remains locked.

The problem is that pktsetup opens the cdrom device in non-blocking
mode, which doesn't lock the door. mount then opens the cdrom device
again in blocking mode, which does lock the door. umount closes the
blocking mode open, but the door remains locked, because
cdrom.c:cdrom_release() only unlocks the door on the last release, it
doesn't care that the only remaining open is non-blocking.

Signed-off-by: Peter Osterlund <petero2@telia.com>
---

 linux-petero/drivers/block/pktcdvd.c |    2 ++
 1 files changed, 2 insertions(+)

diff -puN drivers/block/pktcdvd.c~packet-door-unlock drivers/block/pktcdvd.c
--- linux/drivers/block/pktcdvd.c~packet-door-unlock	2004-07-26 17:36:21.425126472 +0200
+++ linux-petero/drivers/block/pktcdvd.c	2004-07-26 17:36:21.442123888 +0200
@@ -1981,6 +1981,8 @@ static void pkt_release_dev(struct pktcd
 	if (flush && pkt_flush_cache(pd))
 		DPRINTK("pktcdvd: %s not flushing cache\n", pd->name);
 
+	pkt_lock_door(pd, 0);
+
 	q = bdev_get_queue(pd->bdev);
 	pkt_set_speed(pd, 0xffff, 0xffff);
 	spin_lock_irq(q->queue_lock);
_

-- 
Peter Osterlund - petero2@telia.com
http://w1.894.telia.com/~u89404340

Re: 2.6.8-rc2-mm1

[Johannes Stezenbach]

On Wed, Jul 28, 2004 at 02:04:44AM -0700, Andrew Morton wrote:
> 
> - If people have patches in here which are important for a 2.6.8 release,
>   please let me know.
...
> +dvb-major-number.patch
> 
>  Use the right major in DVB

I would like to see this patch go into 2.6.8. We already changed
the major number in linuxtv.org CVS and announced it on our
website, so this might help keep the time of confusion for
DVB users short.

The patch below should go along with it. It fixes some breakage
in dvb_usercopy() introduced by Al Viro's sparse cleanups in -rc2.
(A similar patch might have been mailed already by Michael Hunold.)

Signed-off-by: Johannes Stezenbach <js@convergence.de>

--- linux-2.6.8-rc2/drivers/media/dvb/dvb-core/dvb_functions.c.orig	2004-07-29 00:19:50.000000000 +0200
+++ linux-2.6.8-rc2/drivers/media/dvb/dvb-core/dvb_functions.c	2004-07-29 00:20:05.000000000 +0200
@@ -36,7 +36,7 @@ int dvb_usercopy(struct inode *inode, st
         /*  Copy arguments into temp kernel buffer  */
         switch (_IOC_DIR(cmd)) {
         case _IOC_NONE:
-                parg = NULL;
+                parg = (void *) arg;
                 break;
         case _IOC_READ: /* some v4l ioctls are marked wrong ... */
         case _IOC_WRITE:

Johannes

Re: 2.6.8-rc2-mm1

[viro]

On Thu, Jul 29, 2004 at 12:24:55AM +0200, Johannes Stezenbach wrote:
> Signed-off-by: Johannes Stezenbach <js@convergence.de>
> 
> --- linux-2.6.8-rc2/drivers/media/dvb/dvb-core/dvb_functions.c.orig	2004-07-29 00:19:50.000000000 +0200
> +++ linux-2.6.8-rc2/drivers/media/dvb/dvb-core/dvb_functions.c	2004-07-29 00:20:05.000000000 +0200
> @@ -36,7 +36,7 @@ int dvb_usercopy(struct inode *inode, st
>          /*  Copy arguments into temp kernel buffer  */
>          switch (_IOC_DIR(cmd)) {
>          case _IOC_NONE:
> -                parg = NULL;
> +                parg = (void *) arg;

Mind explaining why it is the right thing to do?  You are creating a kernel
pointer out of value passed to you by userland and feed it to a function
that expects a kernel pointer.  Which is an invitation for trouble - if
it ends up dereferenced, we are screwed and won't notice that.

Re: 2.6.8-rc2-mm1

[Adam Kropelin]

On Wed, Jul 28, 2004 at 02:04:44AM -0700, Andrew Morton wrote:
> 
> ftp://ftp.kernel.org/pub/linux/kernel/people/akpm/patches/2.6/2.6.8-rc2/2.6.8-rc2-mm1/

<snip>

> - If people have patches in here which are important for a 2.6.8 release,
>   please let me know.

There's a trivial yet fairly important fix to hiddev.h in bk-input that
it would be nice to get merged before 2.6.8. Distros have been
shipping the current in-tree (broken) version with their kernel headers
packages so a number of userspace apps cannot build. I've broken out the
patch below.

There's also a hiddev oops-on-removal fix that ought to be merged fairly
soon, but I can understand if Vojtech wants that tested a bit longer
first.

--Adam

diff -Nru a/include/linux/hiddev.h b/include/linux/hiddev.h
--- a/include/linux/hiddev.h	2004-07-27 18:37:32 -07:00
+++ b/include/linux/hiddev.h	2004-07-27 18:37:32 -07:00
@@ -128,10 +128,11 @@
 
 /* hiddev_usage_ref_multi is used for sending multiple bytes to a control.
  * It really manifests itself as setting the value of consecutive usages */
+#define HID_MAX_MULTI_USAGES 1024
 struct hiddev_usage_ref_multi {
 	struct hiddev_usage_ref uref;
 	__u32 num_values;
-	__s32 values[HID_MAX_USAGES];
+	__s32 values[HID_MAX_MULTI_USAGES];
 };
 
 /* FIELD_INDEX_NONE is returned in read() data from the kernel when flags
@@ -211,6 +212,11 @@
 /*
  * In-kernel definitions.
  */
+
+struct hid_device;
+struct hid_usage;
+struct hid_field;
+struct hid_report;
 
 #ifdef CONFIG_USB_HIDDEV
 int hiddev_connect(struct hid_device *);

Re: 2.6.8-rc2-mm1

[Paul Jackson]

Andrew,

The patch:

  gcc35-index.html.patch

in the broken-out patch set is _very_ trivial.
So much so that quilt complains:

  Only garbage was found in the patch input.

I trust I won't be missing much if I delete this
one from the series file ;).

-- 
                          I won't rest till it's the best ...
                          Programmer, Linux Scalability
                          Paul Jackson <pj@sgi.com> 1.650.933.1373

Re: 2.6.8-rc2-mm1

[Andrew Morton]

Adam Kropelin <akropel1@rochester.rr.com> wrote:
>
> > - If people have patches in here which are important for a 2.6.8 release,
> >   please let me know.
> 
> There's a trivial yet fairly important fix to hiddev.h in bk-input that
> it would be nice to get merged before 2.6.8. Distros have been
> shipping the current in-tree (broken) version with their kernel headers
> packages so a number of userspace apps cannot build. I've broken out the
> patch below.
> 
> There's also a hiddev oops-on-removal fix that ought to be merged fairly
> soon, but I can understand if Vojtech wants that tested a bit longer
> first.

OK, thanks.  Vojtech, can you please sort this matter out?

Re: 2.6.8-rc2-mm1

[Johannes Stezenbach]

On Wed, Jul 28, 2004 at 11:44:23PM +0100, viro@parcelfarce.linux.theplanet.co.uk wrote:
> On Thu, Jul 29, 2004 at 12:24:55AM +0200, Johannes Stezenbach wrote:
> > Signed-off-by: Johannes Stezenbach <js@convergence.de>
> > 
> > --- linux-2.6.8-rc2/drivers/media/dvb/dvb-core/dvb_functions.c.orig	2004-07-29 00:19:50.000000000 +0200
> > +++ linux-2.6.8-rc2/drivers/media/dvb/dvb-core/dvb_functions.c	2004-07-29 00:20:05.000000000 +0200
> > @@ -36,7 +36,7 @@ int dvb_usercopy(struct inode *inode, st
> >          /*  Copy arguments into temp kernel buffer  */
> >          switch (_IOC_DIR(cmd)) {
> >          case _IOC_NONE:
> > -                parg = NULL;
> > +                parg = (void *) arg;
> 
> Mind explaining why it is the right thing to do?  You are creating a kernel
> pointer out of value passed to you by userland and feed it to a function
> that expects a kernel pointer.  Which is an invitation for trouble - if
> it ends up dereferenced, we are screwed and won't notice that.

This is a hack introduced by someone years ago. The "pointer" is
actually an integer argument, e.g. in include/linux/dvb/audio.h:

#define AUDIO_SET_MUTE             _IO('o', 6)

actually takes an integer argument (!0 mute, 0 unmute), so one can write

	if (ioctl(fd, AUDIO_SET_MUTE, 1) == -1)
		perror("mute");

It is unusual (maybe even wrong?), but we cannot change it without
losing binary API compatibility. However, I see that sparse might
flag this as a possible bug :-(

Johannes

Re: 2.6.8-rc2-mm1

[Andrew Morton]

Johannes Stezenbach <js@convergence.de> wrote:
>
> This is a hack introduced by someone years ago. The "pointer" is
> actually an integer argument, e.g. in include/linux/dvb/audio.h:
> 
> #define AUDIO_SET_MUTE             _IO('o', 6)
> 
> actually takes an integer argument (!0 mute, 0 unmute), so one can write
> 
> 	if (ioctl(fd, AUDIO_SET_MUTE, 1) == -1)
> 		perror("mute");

Is it a boolean argument?

If so, we could change the code to do

	parg = (void *)(arg ? 1 : 0);

so if someone dereferences it they'll get a nice oops.

Re: 2.6.8-rc2-mm1

[Johannes Stezenbach]

Andrew Morton wrote:
> Johannes Stezenbach <js@convergence.de> wrote:
> >
> > This is a hack introduced by someone years ago. The "pointer" is
> > actually an integer argument, e.g. in include/linux/dvb/audio.h:
> > 
> > #define AUDIO_SET_MUTE             _IO('o', 6)
> > 
> > actually takes an integer argument (!0 mute, 0 unmute), so one can write
> > 
> > 	if (ioctl(fd, AUDIO_SET_MUTE, 1) == -1)
> > 		perror("mute");
> 
> Is it a boolean argument?
> 
> If so, we could change the code to do
> 
> 	parg = (void *)(arg ? 1 : 0);
> 
> so if someone dereferences it they'll get a nice oops.

Unfortunately there are a few more ioctls which use enums,
e.g. AUDIO_CHANNEL_SELECT has an argument of type
audio_channel_select_t etc. (yes, I know, those typedefs
should go). Or even DMX_SET_BUFFER_SIZE...

Johannes

Re: 2.6.8-rc2-mm1

[Andrew Morton]

Johannes Stezenbach <js@convergence.de> wrote:
>
> Andrew Morton wrote:
> > Johannes Stezenbach <js@convergence.de> wrote:
> > >
> > > This is a hack introduced by someone years ago. The "pointer" is
> > > actually an integer argument, e.g. in include/linux/dvb/audio.h:
> > > 
> > > #define AUDIO_SET_MUTE             _IO('o', 6)
> > > 
> > > actually takes an integer argument (!0 mute, 0 unmute), so one can write
> > > 
> > > 	if (ioctl(fd, AUDIO_SET_MUTE, 1) == -1)
> > > 		perror("mute");
> > 
> > Is it a boolean argument?
> > 
> > If so, we could change the code to do
> > 
> > 	parg = (void *)(arg ? 1 : 0);
> > 
> > so if someone dereferences it they'll get a nice oops.
> 
> Unfortunately there are a few more ioctls which use enums,
> e.g. AUDIO_CHANNEL_SELECT has an argument of type
> audio_channel_select_t etc. (yes, I know, those typedefs
> should go). Or even DMX_SET_BUFFER_SIZE...

OK.  Well how's about we add a fifth argument to that callback function, of
type `int'?  Stick this integer in there and leave the fourth argument NULL
for this class of ioctl?

Re: 2.6.8-rc2-mm1

[Adrian Bunk]

The following issue comes from Linus' tree:

JFFS2_COMPRESSION_OPTIONS is asked even if JFFS2_FS support isn't 
selected.

The patch below adds a dependency on JFFS2_FS to 
JFFS2_COMPRESSION_OPTIONS.

I've also added a dependency on EXPERIMENTAL which seemed to be logical 
after reading the description of this option (but even if you disagree 
with this, please add the dependency on JFFS2_FS).


Signed-off-by: Adrian Bunk <bunk@fs.tum.de>

--- linux-2.6.8-pre2-mm1/fs/Kconfig.old	2004-07-29 16:24:35.000000000 +0200
+++ linux-2.6.8-pre2-mm1/fs/Kconfig	2004-07-29 16:24:55.000000000 +0200
@@ -1088,18 +1088,19 @@
 	  including a link to the mailing list where details of the remaining
 	  work to be completed for NAND flash support can be found, see the 
 	  JFFS2 web site at <http://sources.redhat.com/jffs2>.
 
 	  Say 'N' unless you have NAND flash and you are willing to test and
 	  develop JFFS2 support for it.
 
 config JFFS2_COMPRESSION_OPTIONS
 	bool "Advanced compression options for JFFS2"
+	depends on JFFS2_FS && EXPERIMENTAL
 	default n
 	help
 	  Enabling this option allows you to explicitly choose which
 	  compression modules, if any, are enabled in JFFS2. Removing
 	  compressors and mean you cannot read existing file systems,
 	  and enabling experimental compressors can mean that you
 	  write a file system which cannot be read by a standard kernel.
 
 	  If unsure, you should _definitely_ say 'N'.

2.6.8-rc2-mm1: NTFS compile error with gcc 2.95

[Adrian Bunk]

On Wed, Jul 28, 2004 at 02:04:44AM -0700, Andrew Morton wrote:
>...
> Changes since 2.6.8-rc1-mm1:
> 
>...
>  bk-ntfs.patch
>...

This causes the following compile error when using gcc 2.95:

<--  snip  -->

...
  LD      .tmp_vmlinux1
fs/built-in.o(.text+0x14425f): In function `ntfs_find_vcn':
: undefined reference to `__cmpdi2'
fs/built-in.o(.text+0x144272): In function `ntfs_find_vcn':
: undefined reference to `__cmpdi2'
make: *** [.tmp_vmlinux1] Error 1

<--  snip  -->


cu
Adrian

-- 

       "Is there not promise of rain?" Ling Tan asked suddenly out
        of the darkness. There had been need of rain for many days.
       "Only a promise," Lao Er said.
                                       Pearl S. Buck - Dragon Seed

Re: 2.6.8-rc2-mm1: NTFS compile error with gcc 2.95

[Jan-Benedict Glaw]

[-- Attachment #1: Type: text/plain, Size: 1695 bytes --]

On Thu, 2004-07-29 16:41:49 +0200, Adrian Bunk <bunk@fs.tum.de>
wrote in message <20040729144149.GC2349@fs.tum.de>:
> On Wed, Jul 28, 2004 at 02:04:44AM -0700, Andrew Morton wrote:
> >...
> > Changes since 2.6.8-rc1-mm1:
> >...
> >  bk-ntfs.patch
> >...
> 
> This causes the following compile error when using gcc 2.95:
> 
> <--  snip  -->
> 
> ...
>   LD      .tmp_vmlinux1
> fs/built-in.o(.text+0x14425f): In function `ntfs_find_vcn':
> : undefined reference to `__cmpdi2'
> fs/built-in.o(.text+0x144272): In function `ntfs_find_vcn':
> : undefined reference to `__cmpdi2'
> make: *** [.tmp_vmlinux1] Error 1
> 
> <--  snip  -->

GCC wanted to make a compare on a 8byte integer (so probably long long),
but decided there isn't an appropriate insn on that hardware platform.
So instead of emitting assembler, it generated a function call that
would have resulted in a call to libgcc. However, the Linux kernel asks
to *not* link that lib (eg. think about a gcc compiled for i686 (so is
libgcc) while compiling for i386).

With the that constraint in mind, there isn't really a "nice" solution.
Maybe Linux should provide a function of that name (and several others,
too). OTOH, the H8/300 port explicitely links in libgcc, but that's only
an option if even the oldest processor supports *all* known CPU
instructions.

MfG, JBG

-- 
Jan-Benedict Glaw       jbglaw@lug-owl.de    . +49-172-7608481             _ O _
"Eine Freie Meinung in  einem Freien Kopf    | Gegen Zensur | Gegen Krieg  _ _ O
 fuer einen Freien Staat voll Freier Bürger" | im Internet! |   im Irak!   O O O
ret = do_actions((curr | FREE_SPEECH) & ~(NEW_COPYRIGHT_LAW | DRM | TCPA));

[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

Re: 2.6.8-rc2-mm1

[Michael Hunold]

Hi,

On 07/29/04 01:24, Johannes Stezenbach wrote:
> On Wed, Jul 28, 2004 at 11:44:23PM +0100, viro@parcelfarce.linux.theplanet.co.uk wrote:
> 
>>On Thu, Jul 29, 2004 at 12:24:55AM +0200, Johannes Stezenbach wrote:
>>
>>>Signed-off-by: Johannes Stezenbach <js@convergence.de>
>>>
>>>--- linux-2.6.8-rc2/drivers/media/dvb/dvb-core/dvb_functions.c.orig	2004-07-29 00:19:50.000000000 +0200
>>>+++ linux-2.6.8-rc2/drivers/media/dvb/dvb-core/dvb_functions.c	2004-07-29 00:20:05.000000000 +0200
>>>@@ -36,7 +36,7 @@ int dvb_usercopy(struct inode *inode, st
>>>         /*  Copy arguments into temp kernel buffer  */
>>>         switch (_IOC_DIR(cmd)) {
>>>         case _IOC_NONE:
>>>-                parg = NULL;
>>>+                parg = (void *) arg;
>>
>>Mind explaining why it is the right thing to do?  You are creating a kernel
>>pointer out of value passed to you by userland and feed it to a function
>>that expects a kernel pointer.  Which is an invitation for trouble - if
>>it ends up dereferenced, we are screwed and won't notice that.
> 
> 
> This is a hack introduced by someone years ago. The "pointer" is
> actually an integer argument, e.g. in include/linux/dvb/audio.h:
> 
> #define AUDIO_SET_MUTE             _IO('o', 6)
> 
> actually takes an integer argument (!0 mute, 0 unmute), so one can write
> 
> 	if (ioctl(fd, AUDIO_SET_MUTE, 1) == -1)
> 		perror("mute");
> 
> It is unusual (maybe even wrong?), but we cannot change it without
> losing binary API compatibility. However, I see that sparse might
> flag this as a possible bug :-(

Is this convenient trick considered harmful?
Or is it a creative way of using ioctls?

We're currently using this stuff in the overhauled DVB v4 API, too. So 
before we finally establish the DVB v4 API, I'd like to know if this is 
a no-no.

Comments?

CU
Michael.

Re: 2.6.8-rc2-mm1

[David Woodhouse]

On Thu, 2004-07-29 at 16:36 +0200, Adrian Bunk wrote:
> The following issue comes from Linus' tree:
> 
> JFFS2_COMPRESSION_OPTIONS is asked even if JFFS2_FS support isn't 
> selected.
> 
> The patch below adds a dependency on JFFS2_FS to 
> JFFS2_COMPRESSION_OPTIONS.

That was already at bk://linux-mtd.bkbits.net/mtd-2.6 for Linus to pull.

> I've also added a dependency on EXPERIMENTAL which seemed to be logical 
> after reading the description of this option (but even if you disagree 
> with this, please add the dependency on JFFS2_FS).

Not correct.

-- 
dwmw2

[patch] Re: 2.6.8-rc2-mm1: NTFS compile error with gcc 2.95

[Adrian Bunk]

On Thu, Jul 29, 2004 at 05:54:11PM +0200, Jan-Benedict Glaw wrote:
> > 
> > This causes the following compile error when using gcc 2.95:
> > 
> > <--  snip  -->
> > 
> > ...
> >   LD      .tmp_vmlinux1
> > fs/built-in.o(.text+0x14425f): In function `ntfs_find_vcn':
> > : undefined reference to `__cmpdi2'
> > fs/built-in.o(.text+0x144272): In function `ntfs_find_vcn':
> > : undefined reference to `__cmpdi2'
> > make: *** [.tmp_vmlinux1] Error 1
> > 
> > <--  snip  -->
> 
> GCC wanted to make a compare on a 8byte integer (so probably long long),
> but decided there isn't an appropriate insn on that hardware platform.
> So instead of emitting assembler, it generated a function call that
> would have resulted in a call to libgcc. However, the Linux kernel asks
> to *not* link that lib (eg. think about a gcc compiled for i686 (so is
> libgcc) while compiling for i386).
>...

We had the same problem somewhere else recently.

This problem occurs only with gcc 2.95 and case statements.

> MfG, JBG


What about the following patch to fix NTFS compilation with gcc 2.95?


Signed-off-by: Adrian Bunk <bunk@fs.tum.de>

--- linux-2.6.8-rc2-mm1-full/fs/ntfs/attrib.c.old	2004-07-29 22:34:21.000000000 +0200
+++ linux-2.6.8-rc2-mm1-full/fs/ntfs/attrib.c	2004-07-29 22:36:15.000000000 +0200
@@ -1113,15 +1113,11 @@
 			}
 			rl++;
 		}
-		switch (rl->lcn) {
-		case (LCN)LCN_RL_NOT_MAPPED:
-			break;
-		case (LCN)LCN_ENOENT:
-			err = -ENOENT;
-			break;
-		default:
-			err = -EIO;
-			break;
+		if (rl->lcn != (LCN)LCN_RL_NOT_MAPPED) {
+			if (rl->lcn == (LCN)LCN_ENOENT)
+				err = -ENOENT;
+			else
+				err = -EIO;
 		}
 	}
 	if (!need_write)

Re: 2.6.8-rc2-mm1

[Andrew Morton]

Michael Hunold <hunold@convergence.de> wrote:
>
> > This is a hack introduced by someone years ago. The "pointer" is
> > actually an integer argument, e.g. in include/linux/dvb/audio.h:
> > 
> > #define AUDIO_SET_MUTE             _IO('o', 6)
> > 
> > actually takes an integer argument (!0 mute, 0 unmute), so one can write
> > 
> > 	if (ioctl(fd, AUDIO_SET_MUTE, 1) == -1)
> > 		perror("mute");
> > 
> > It is unusual (maybe even wrong?), but we cannot change it without
> > losing binary API compatibility. However, I see that sparse might
> > flag this as a possible bug :-(
> 
> Is this convenient trick considered harmful?
> Or is it a creative way of using ioctls?
> 
> We're currently using this stuff in the overhauled DVB v4 API, too. So 
> before we finally establish the DVB v4 API, I'd like to know if this is 
> a no-no.

It's a no-no.  Please define the interface to be as typesafe and as
__user-correct as possible.

2.6.8-rc2-mm1: DVB: "errno" undefined

[Adrian Bunk]

I'm getting the following errors when trying to compile 2.6.8-rc2-mm1 as 
modular as possible (using gcc 2.95):

<--  snip  -->

...
*** Warning: "errno" [drivers/media/dvb/frontends/tda1004x.ko] undefined!
*** Warning: "errno" [drivers/media/dvb/frontends/sp887x.ko] undefined!
*** Warning: "errno" [drivers/media/dvb/frontends/alps_tdlb7.ko] undefined!
...

<--  snip  -->

cu
Adrian

-- 

       "Is there not promise of rain?" Ling Tan asked suddenly out
        of the darkness. There had been need of rain for many days.
       "Only a promise," Lao Er said.
                                       Pearl S. Buck - Dragon Seed

Re: 2.6.8-rc2-mm1: DVB: "errno" undefined

[Kenneth Aafløy]

On Thursday 29 July 2004 23:27, you wrote:
> I'm getting the following errors when trying to compile 2.6.8-rc2-mm1 as
> modular as possible (using gcc 2.95):
[snip]
> *** Warning: "errno" [drivers/media/dvb/frontends/tda1004x.ko] undefined!
> *** Warning: "errno" [drivers/media/dvb/frontends/sp887x.ko] undefined!
> *** Warning: "errno" [drivers/media/dvb/frontends/alps_tdlb7.ko] undefined!
[snip]

This is still not fixed because we (linuxtv.org) have not submitted the 
changes necessary following this thread:
http://marc.theaimsgroup.com/?l=linux-kernel&m=108344912617115&w=2

This is about firmware loading in those modules, and we are working on 
converting those modules to i2c_kernel to take advantage of firmware_class.

Could the offending modules be marked as broken or something untill 
linuxtv-dvb can test and submit the dvb frontend updates, which should be 
sometime soon, but probably not in time for 2.6.8?

Kenneth

Re: [patch] Re: 2.6.8-rc2-mm1: NTFS compile error with gcc 2.95

[Adrian Bunk]

On Thu, Jul 29, 2004 at 10:51:53PM +0100, Anton Altaparmakov wrote:
>...
> Thanks.  Looks good.  Have you verified that it fixes the gcc-2.95 
> compilation?  If so I will apply it to my tree and Andrew will get it when 
> he does his next pull of the ntfs-2.6-devel repository...

Sure, I verified the compilation with gcc 2.95 before sending this 
patch.

> Thanks,
> 
> 	Anton

Thanks
Adrian

-- 

       "Is there not promise of rain?" Ling Tan asked suddenly out
        of the darkness. There had been need of rain for many days.
       "Only a promise," Lao Er said.
                                       Pearl S. Buck - Dragon Seed

Re: 2.6.8-rc2-mm1: DVB: "errno" undefined

[Adrian Bunk]

On Fri, Jul 30, 2004 at 12:44:13AM +0200, Kenneth Aafløy wrote:
> On Thursday 29 July 2004 23:27, you wrote:
> > I'm getting the following errors when trying to compile 2.6.8-rc2-mm1 as
> > modular as possible (using gcc 2.95):
> [snip]
> > *** Warning: "errno" [drivers/media/dvb/frontends/tda1004x.ko] undefined!
> > *** Warning: "errno" [drivers/media/dvb/frontends/sp887x.ko] undefined!
> > *** Warning: "errno" [drivers/media/dvb/frontends/alps_tdlb7.ko] undefined!
> [snip]
> 
> This is still not fixed because we (linuxtv.org) have not submitted the 
> changes necessary following this thread:
> http://marc.theaimsgroup.com/?l=linux-kernel&m=108344912617115&w=2
> 
> This is about firmware loading in those modules, and we are working on 
> converting those modules to i2c_kernel to take advantage of firmware_class.
> 
> Could the offending modules be marked as broken or something untill 
> linuxtv-dvb can test and submit the dvb frontend updates, which should be 
> sometime soon, but probably not in time for 2.6.8?

The removal of errno from this three drivers is currently only in -mm.

So unless someone forwards them (they were sent by Andi Kleen as gcc 3.5 
build fixes, but he apparently didn't test a modular build) to Linus 
which hopefully won't happen before the affected modules are properly 
fixed, Linus' tree isn't affected.

> Kenneth

cu
Adrian

-- 

       "Is there not promise of rain?" Ling Tan asked suddenly out
        of the darkness. There had been need of rain for many days.
       "Only a promise," Lao Er said.
                                       Pearl S. Buck - Dragon Seed

Re: 2.6.8-rc2-mm1: DVB: "errno" undefined

[Arnd Bergmann]

[-- Attachment #1: Type: text/plain, Size: 6617 bytes --]

On Freitag, 30. Juli 2004 01:24, Adrian Bunk <bunk@fs.tum.de> wrote:

> The removal of errno from this three drivers is currently only in -mm.
> 
> So unless someone forwards them (they were sent by Andi Kleen as gcc 3.5 
> build fixes, but he apparently didn't test a modular build) to Linus 
> which hopefully won't happen before the affected modules are properly 
> fixed, Linus' tree isn't affected.

Actually, the problem has its origin in my removal of all in-kernel
syscalls (except execve, which is non-trivial) earlier this year.
This change was blindly reverted by the maintainer, while at the same
time the local errno variable was removed. See also
http://linux.bkbits.net:8080/linux-2.5/hist/drivers/media/dvb/frontends/tda1004x.c

This patch is the one that was already merged earlier. I'm now also
removing the definitions for the kernel syscalls on i386 to make it
harder to reintroduce them again. This was already done for ppc64,
the others should probably follow.

Signed-off-by: Arnd Bergmann <arnd@arndb.de>

 drivers/media/dvb/frontends/alps_tdlb7.c |   10 ++++------
 drivers/media/dvb/frontends/sp887x.c     |    9 ++++-----
 drivers/media/dvb/frontends/tda1004x.c   |   10 ++++------
 drivers/net/wireless/prism54/isl_38xx.c  |    2 --
 include/asm-i386/unistd.h                |    8 --------
 5 files changed, 12 insertions(+), 27 deletions(-)

===== drivers/media/dvb/frontends/alps_tdlb7.c 1.14 vs edited =====
--- 1.14/drivers/media/dvb/frontends/alps_tdlb7.c	Wed Jul 14 02:09:53 2004
+++ edited/drivers/media/dvb/frontends/alps_tdlb7.c	Fri Jul 30 16:00:28 2004
@@ -28,8 +28,6 @@
     
 */  
 
-
-#define __KERNEL_SYSCALLS__
 #include <linux/module.h>
 #include <linux/init.h>
 #include <linux/vmalloc.h>
@@ -148,13 +146,13 @@
 	loff_t filesize;
 	char *dp;
 
-	fd = open(fn, 0, 0);
+	fd = sys_open(fn, 0, 0);
 	if (fd == -1) {
                 printk("%s: unable to open '%s'.\n", __FUNCTION__, fn);
 		return -EIO;
 	}
 
-	filesize = lseek(fd, 0L, 2);
+	filesize = sys_lseek(fd, 0L, 2);
 	if (filesize <= 0 || filesize < SP8870_FIRMWARE_OFFSET + SP8870_FIRMWARE_SIZE) {
 	        printk("%s: firmware filesize to small '%s'\n", __FUNCTION__, fn);
 		sys_close(fd);
@@ -168,8 +166,8 @@
 		return -EIO;
 	}
 
-	lseek(fd, SP8870_FIRMWARE_OFFSET, 0);
-	if (read(fd, dp, SP8870_FIRMWARE_SIZE) != SP8870_FIRMWARE_SIZE) {
+	sys_lseek(fd, SP8870_FIRMWARE_OFFSET, 0);
+	if (sys_read(fd, dp, SP8870_FIRMWARE_SIZE) != SP8870_FIRMWARE_SIZE) {
 		printk("%s: failed to read '%s'.\n",__FUNCTION__, fn);
 		vfree(dp);
 		sys_close(fd);
===== drivers/media/dvb/frontends/sp887x.c 1.12 vs edited =====
--- 1.12/drivers/media/dvb/frontends/sp887x.c	Wed Jul 14 02:09:55 2004
+++ edited/drivers/media/dvb/frontends/sp887x.c	Fri Jul 30 16:00:28 2004
@@ -12,7 +12,6 @@
    next 0x4000 loaded. This may change in future versions.
  */
 
-#define __KERNEL_SYSCALLS__
 #include <linux/kernel.h>
 #include <linux/vmalloc.h>
 #include <linux/module.h>
@@ -209,13 +208,13 @@
 
 	// Load the firmware
 	set_fs(get_ds());
-	fd = open(sp887x_firmware, 0, 0);
+	fd = sys_open(sp887x_firmware, 0, 0);
 	if (fd < 0) {
 		printk(KERN_WARNING "%s: Unable to open firmware %s\n", __FUNCTION__,
 		       sp887x_firmware);
 		return -EIO;
 	}
-	filesize = lseek(fd, 0L, 2);
+	filesize = sys_lseek(fd, 0L, 2);
 	if (filesize <= 0) {
 		printk(KERN_WARNING "%s: Firmware %s is empty\n", __FUNCTION__,
 		       sp887x_firmware);
@@ -237,8 +236,8 @@
 	// read it!
 	// read the first 16384 bytes from the file
 	// ignore the first 10 bytes
-	lseek(fd, 10, 0);
-	if (read(fd, firmware, fw_size) != fw_size) {
+	sys_lseek(fd, 10, 0);
+	if (sys_read(fd, firmware, fw_size) != fw_size) {
 		printk(KERN_WARNING "%s: Failed to read firmware\n", __FUNCTION__);
 		vfree(firmware);
 		sys_close(fd);
===== drivers/media/dvb/frontends/tda1004x.c 1.14 vs edited =====
--- 1.14/drivers/media/dvb/frontends/tda1004x.c	Wed Jul 14 02:09:55 2004
+++ edited/drivers/media/dvb/frontends/tda1004x.c	Fri Jul 30 16:02:15 2004
@@ -32,7 +32,6 @@
  */
 
 
-#define __KERNEL_SYSCALLS__
 #include <linux/kernel.h>
 #include <linux/vmalloc.h>
 #include <linux/module.h>
@@ -40,7 +39,6 @@
 #include <linux/string.h>
 #include <linux/slab.h>
 #include <linux/fs.h>
-#include <linux/unistd.h>
 #include <linux/fcntl.h>
 #include <linux/errno.h>
 #include <linux/syscalls.h>
@@ -397,13 +395,13 @@
 
 	// Load the firmware
 	set_fs(get_ds());
-	fd = open(tda1004x_firmware, 0, 0);
+	fd = sys_open(tda1004x_firmware, 0, 0);
 	if (fd < 0) {
 		printk("%s: Unable to open firmware %s\n", __FUNCTION__,
 		       tda1004x_firmware);
 		return -EIO;
 	}
-	filesize = lseek(fd, 0L, 2);
+	filesize = sys_lseek(fd, 0L, 2);
 	if (filesize <= 0) {
 		printk("%s: Firmware %s is empty\n", __FUNCTION__,
 		       tda1004x_firmware);
@@ -434,8 +432,8 @@
 	}
 
 	// read it!
-	lseek(fd, fw_offset, 0);
-	if (read(fd, firmware, fw_size) != fw_size) {
+	sys_lseek(fd, fw_offset, 0);
+	if (sys_read(fd, firmware, fw_size) != fw_size) {
 		printk("%s: Failed to read firmware\n", __FUNCTION__);
 		vfree(firmware);
 		sys_close(fd);
===== drivers/net/wireless/prism54/isl_38xx.c 1.3 vs edited =====
--- 1.3/drivers/net/wireless/prism54/isl_38xx.c	Sat Jun  5 13:45:32 2004
+++ edited/drivers/net/wireless/prism54/isl_38xx.c	Fri Jul 30 16:06:25 2004
@@ -18,8 +18,6 @@
  *
  */
 
-#define __KERNEL_SYSCALLS__
-
 #include <linux/version.h>
 #include <linux/module.h>
 #include <linux/types.h>
===== include/asm-i386/unistd.h 1.39 vs edited =====
--- 1.39/include/asm-i386/unistd.h	Wed Jul 14 02:09:34 2004
+++ edited/include/asm-i386/unistd.h	Fri Jul 30 16:07:06 2004
@@ -431,15 +431,7 @@
  * won't be any messing with the stack from main(), but we define
  * some others too.
  */
-static inline _syscall0(pid_t,setsid)
-static inline _syscall3(int,write,int,fd,const char *,buf,off_t,count)
-static inline _syscall3(int,read,int,fd,char *,buf,off_t,count)
-static inline _syscall3(off_t,lseek,int,fd,off_t,offset,int,count)
-static inline _syscall1(int,dup,int,fd)
 static inline _syscall3(int,execve,const char *,file,char **,argv,char **,envp)
-static inline _syscall3(int,open,const char *,file,int,flag,int,mode)
-static inline _syscall1(int,close,int,fd)
-static inline _syscall3(pid_t,waitpid,pid_t,pid,int *,wait_stat,int,options)
 
 asmlinkage int sys_modify_ldt(int func, void __user *ptr, unsigned long bytecount);
 asmlinkage long sys_mmap2(unsigned long addr, unsigned long len,

[-- Attachment #2: signature --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

2.6.8-rc2-mm1 breaks PPPoE for me (was: 2.6.8-rc2-mm1)

[Matthias Andree]

On Wed, 28 Jul 2004, Andrew Morton wrote:

> ftp://ftp.kernel.org/pub/linux/kernel/people/akpm/patches/2.6/2.6.8-rc2/2.6.8-rc2-mm1/
...
> gcc35-pppoe.c.patch
>   gcc-3.5 fixes

Andrew,

I'm not sure if my problem is related to this patch, but in
2.6.8-rc2-mm1, PPPoE doesn't work for me, kernel compiled with gcc (GCC)
3.3.1 (SuSE Linux), a vanilla 2.6.7 is fine with the same compiler.

pppd[5685]: Plugin /usr/lib/pppd/2.4.1/pppoe.so loaded.
pppd[5685]: PPPoE Plugin Initialized
pppd[5685]: pppd 2.4.1 started by root, uid 0
pppd[5685]: Sending PADI
pppd[5685]: HOST_UNIQ successful match 
pppd[5685]: Failed to negotiate PPPoE connection: 25 Inappropriate ioctl for device
pppd[5685]: Exit.

A successful pppd session start, with 2.6.7, looks like this:

pppd[5070]: PPPoE Plugin Initialized
pppd[5070]: pppd 2.4.1 started by root, uid 0
pppd[5070]: Sending PADI
pppd[5070]: HOST_UNIQ successful match 
pppd[5070]: HOST_UNIQ successful match 
pppd[5070]: Got connection: 164b
pppd[5070]: Connecting PPPoE socket: 00:90:1a:XX:XX:XX 4b16 eth1 0x808a560
/sbin/hotplug[5166]: INTERFACE=ppp0
pppd[5070]: Using interface ppp0
pppd[5070]: Connect: ppp0 <--> eth1
pppd[5070]: Setting MTU to 1492.
pppd[5070]: Couldn't increase MRU to 1500
pppd[5070]: Setting MTU to 1492.
pppd[5070]: local  IP address 217.81.XXX.XXX
pppd[5070]: remote IP address 217.5.XXX.XXX
pppd[5070]: Script /etc/ppp/ip-up finished (pid 5180), status = 0x0

HTH,

-- 
Matthias Andree

Encrypted mail welcome: my GnuPG key ID is 0x052E7D95 (PGP/MIME preferred)

Re: 2.6.8-rc2-mm1

[Zwane Mwaikambo]

Ingo i believe you have a patch for this, could you push it to Andrew? I
reckon it's provoked by CONFIG_PREEMPT.

Unable to handle kernel paging request at virtual address dd27dfa0
 printing eip:
c01f1576
*pde = 00076067
Oops: 0000 [#1]
PREEMPT SMP DEBUG_PAGEALLOC
Modules linked in:
CPU:    1
EIP:    0060:[<c01f1576>]    Not tainted VLI
EFLAGS: 00010246   (2.6.8-rc2-mm1)
EIP is at __journal_clean_checkpoint_list+0x136/0x1b0
eax: 00000000   ebx: de6e1a3c   ecx: dd27df78   edx: 04000000
esi: df6ea000   edi: de6e1a3c   ebp: df6ebdb0   esp: df6ebd8c
ds: 007b   es: 007b   ss: 0068
Process kjournald (pid: 273, threadinfo=df6ea000 task=dfa8ba40)
Stack: df6ea000 de6e1a3c 0000007f dce48f78 d7afcf78 dd27df78 df6ea000 df639df8
       00000000 df6ebf50 c01ee617 df639df8 df639edc df639e8c 5a5a5a5a 5a5a5a5a
       d7afcfd0 5a5a5a5a d74d3fb0 df639e0c 00000000 00000000 00000000 00000000
Call Trace:
 [<c0108795>] show_stack+0x75/0x90
 [<c01088f5>] show_registers+0x125/0x190
 [<c0108aca>] die+0xda/0x1c0
 [<c011c4c8>] do_page_fault+0x1e8/0x565
 [<c01083ed>] error_code+0x2d/0x40
 [<c01ee617>] journal_commit_transaction+0x3c7/0x1c40
 [<c01f27d8>] kjournald+0x118/0x3d0
 [<c0105355>] kernel_thread_helper+0x5/0x10
Code: 63 8b 4d 08 8a 81 e4 00 00 00 84 c0 7f 4c 8b 45 08 86 90 e4 00 00 00
8b 55 dc 8b 4a 14 8b 42 08 49 a8 08 89 4a 14 75 2b 8b 4d f0 <8b> 59 28 85
db 74 06 8b 43 28 89 41 28 8b 45 08 8b 40 40 89 45

(gdb) list *__journal_clean_checkpoint_list+0x136
0xc01f1576 is in __journal_clean_checkpoint_list (fs/jbd/checkpoint.c:514).
509                             /*
510                              * We need to schedule away.  Rotate both this
511                              * transaction's buffer list and the checkpoint list to
512                              * try to avoid quadratic behaviour.
513                              */
514                             jh = transaction->t_checkpoint_list;
515                             if (jh)
516                                     transaction->t_checkpoint_list = jh->b_cpnext;
517
518                             transaction = journal->j_checkpoint_transactions;

Re: 2.6.8-rc2-mm1

[Andrew Morton]

Zwane Mwaikambo <zwane@linuxpower.ca> wrote:
>
> Ingo i believe you have a patch for this, could you push it to Andrew?

I suspect Ingo's patch will be livelockable under some circumstances.

I suspect mine is too, only less so.

> I reckon it's provoked by CONFIG_PREEMPT.

This should fix.


diff -puN fs/jbd/checkpoint.c~journal_clean_checkpoint_list-latency-fix-fix fs/jbd/checkpoint.c
--- 25/fs/jbd/checkpoint.c~journal_clean_checkpoint_list-latency-fix-fix	2004-07-31 11:43:39.320530424 -0700
+++ 25-akpm/fs/jbd/checkpoint.c	2004-07-31 11:44:11.859583736 -0700
@@ -497,8 +497,7 @@ int __journal_clean_checkpoint_list(jour
 		 * We don't test cond_resched() here because another CPU could
 		 * be waiting on j_list_lock() while holding a different lock.
 		 */
-		if ((ret & 127) == 127) {
-			spin_unlock(&journal->j_list_lock);
+		if (transaction && (ret & 127) == 127) {
 			/*
 			 * We need to schedule away.  Rotate both this
 			 * transaction's buffer list and the checkpoint list to
@@ -512,6 +511,7 @@ int __journal_clean_checkpoint_list(jour
 			if (transaction)
 				journal->j_checkpoint_transactions =
 					transaction->t_cpnext;
+			spin_unlock(&journal->j_list_lock);
 			return ret;
 		}
 #endif
_

Re: 2.6.8-rc2-mm1

[Zwane Mwaikambo]

On Sat, 31 Jul 2004, Andrew Morton wrote:

> Zwane Mwaikambo <zwane@linuxpower.ca> wrote:
> >
> > Ingo i believe you have a patch for this, could you push it to Andrew?
>
> I suspect Ingo's patch will be livelockable under some circumstances.
> I suspect mine is too, only less so.
>
> > I reckon it's provoked by CONFIG_PREEMPT.
>
> This should fix.

Thanks that took care of it.

Re: 2.6.8-rc2-mm1

[Zwane Mwaikambo]

On Sat, 31 Jul 2004, Zwane Mwaikambo wrote:

> On Sat, 31 Jul 2004, Andrew Morton wrote:
>
> > Zwane Mwaikambo <zwane@linuxpower.ca> wrote:
> > >
> > > Ingo i believe you have a patch for this, could you push it to Andrew?
> >
> > I suspect Ingo's patch will be livelockable under some circumstances.
> > I suspect mine is too, only less so.
> >
> > > I reckon it's provoked by CONFIG_PREEMPT.
> >
> > This should fix.
>
> Thanks that took care of it.

Oh bugger, spoke too soon, it took a bit longer this time.

Re: 2.6.8-rc2-mm1

[Andrew Morton]

Zwane Mwaikambo <zwane@linuxpower.ca> wrote:
>
> Oh bugger, spoke too soon, it took a bit longer this time.

Sorry.  Try this one instead.

--- 25/fs/jbd/checkpoint.c~journal_clean_checkpoint_list-latency-fix-fix	2004-07-31 11:43:39.320530424 -0700
+++ 25-akpm/fs/jbd/checkpoint.c	2004-07-31 13:20:22.562303576 -0700
@@ -497,8 +497,8 @@ int __journal_clean_checkpoint_list(jour
 		 * We don't test cond_resched() here because another CPU could
 		 * be waiting on j_list_lock() while holding a different lock.
 		 */
-		if ((ret & 127) == 127) {
-			spin_unlock(&journal->j_list_lock);
+		transaction = journal->j_checkpoint_transactions
+		if (transaction && (ret & 127) == 127) {
 			/*
 			 * We need to schedule away.  Rotate both this
 			 * transaction's buffer list and the checkpoint list to
@@ -508,10 +508,9 @@ int __journal_clean_checkpoint_list(jour
 			if (jh)
 				transaction->t_checkpoint_list = jh->b_cpnext;
 
-			transaction = journal->j_checkpoint_transactions;
-			if (transaction)
-				journal->j_checkpoint_transactions =
+			journal->j_checkpoint_transactions =
 					transaction->t_cpnext;
+			spin_unlock(&journal->j_list_lock);
 			return ret;
 		}
 #endif
_

Re: 2.6.8-rc2-mm1

[William Lee Irwin III]

On Wed, Jul 28, 2004 at 02:04:44AM -0700, Andrew Morton wrote:
> ftp://ftp.kernel.org/pub/linux/kernel/people/akpm/patches/2.6/2.6.8-rc2/2.6.8-rc2-mm1/
[...]

There's trouble here with the link checking; it pukes all over
sparc32's btfixup stuff. Not entirely sure what the proper form of a
solution is.


-- wli

Re: 2.6.8-rc2-mm1

[Zwane Mwaikambo]

On Sat, 31 Jul 2004, Andrew Morton wrote:

> Zwane Mwaikambo <zwane@linuxpower.ca> wrote:
> >
> > Oh bugger, spoke too soon, it took a bit longer this time.
>
> Sorry.  Try this one instead.

Thanks again, i think we have a winner, this one got substantially more
runtime.

Re: 2.6.8-rc2-mm1

[Andrew Morton]

William Lee Irwin III <wli@holomorphy.com> wrote:
>
> On Wed, Jul 28, 2004 at 02:04:44AM -0700, Andrew Morton wrote:
> > ftp://ftp.kernel.org/pub/linux/kernel/people/akpm/patches/2.6/2.6.8-rc2/2.6.8-rc2-mm1/
> [...]
> 
> There's trouble here with the link checking; it pukes all over
> sparc32's btfixup stuff. Not entirely sure what the proper form of a
> solution is.
> 

Do you mean the "check vmlinux for undefined symbols" thing?

That's proving to be a royal pain, although rmk's arguments for needing it
are good.  Could you find a way of fixing it up?

Re: 2.6.8-rc2-mm1

[Ingo Molnar]

On Sat, 31 Jul 2004, Andrew Morton wrote:

> > Ingo i believe you have a patch for this, could you push it to Andrew?
> 
> I suspect Ingo's patch will be livelockable under some circumstances.

the first versions were - i could even reproduce it. Fixed it up by doing
less work in this function. But i like your solution of rotating the list
too. Anyway, the -M5 patch shouldnt be livelockable. (but it might have
the crash problem).

btw., breaking the outer loop here is not enough for latencies, it needs
to happen in the inner loop. (which can take thousands of iterations too.)  
See the checkpoint.c bits of the -M5 patch:

 http://people.redhat.com/mingo/voluntary-preempt/voluntary-preempt-2.6.8-rc2-mm1-M5

	Ingo

Re: 2.6.8-rc2-mm1

[Ingo Molnar]

On Sat, 31 Jul 2004, Andrew Morton wrote:

> Sorry.  Try this one instead.
> 
> --- 25/fs/jbd/checkpoint.c~journal_clean_checkpoint_list-latency-fix-fix	2004-07-31 11:43:39.320530424 -0700
> +++ 25-akpm/fs/jbd/checkpoint.c	2004-07-31 13:20:22.562303576 -0700
> @@ -497,8 +497,8 @@ int __journal_clean_checkpoint_list(jour
>  		 * We don't test cond_resched() here because another CPU could
>  		 * be waiting on j_list_lock() while holding a different lock.
>  		 */
> -		if ((ret & 127) == 127) {
> -			spin_unlock(&journal->j_list_lock);
> +		transaction = journal->j_checkpoint_transactions
> +		if (transaction && (ret & 127) == 127) {

this should cause a crash later on - 'transaction' is also the loop
variable here.

	Ingo

Re: 2.6.8-rc2-mm1

[William Lee Irwin III]

William Lee Irwin III <wli@holomorphy.com> wrote:
>> There's trouble here with the link checking; it pukes all over
>> sparc32's btfixup stuff. Not entirely sure what the proper form of a
>> solution is.

On Sun, Aug 01, 2004 at 01:05:32AM -0700, Andrew Morton wrote:
> Do you mean the "check vmlinux for undefined symbols" thing?
> That's proving to be a royal pain, although rmk's arguments for needing it
> are good.  Could you find a way of fixing it up?

I may need core help. The executable is postprocessed by a program in
arch/sparc/boot/ and so some kind of hook to give it a chance to
properly fix up the symbol table (which I'll have to add afresh), for
instance, an extra stage of .tmp_vmlinux*, seems to be needed.

It treats vmlinux as a throwaway, and does the real linking pass in
arch/sparc/boot/; exchanging the roles of the top-level and arch
linking phases in the makefiles is all that's needed to fix this up.

Once an optional extra pass is okayed, it's a SMOP to deal with the
rest (i.e. the real undefined symbol).

-- wli

Re: 2.6.8-rc2-mm1

[Sam Ravnborg]

On Sun, Aug 01, 2004 at 05:33:34AM -0700, William Lee Irwin III wrote:
> William Lee Irwin III <wli@holomorphy.com> wrote:
> >> There's trouble here with the link checking; it pukes all over
> >> sparc32's btfixup stuff. Not entirely sure what the proper form of a
> >> solution is.
> 
> On Sun, Aug 01, 2004 at 01:05:32AM -0700, Andrew Morton wrote:
> > Do you mean the "check vmlinux for undefined symbols" thing?
> > That's proving to be a royal pain, although rmk's arguments for needing it
> > are good.  Could you find a way of fixing it up?
> 
> I may need core help. The executable is postprocessed by a program in
> arch/sparc/boot/ and so some kind of hook to give it a chance to
> properly fix up the symbol table (which I'll have to add afresh), for
> instance, an extra stage of .tmp_vmlinux*, seems to be needed.

Took a look at this and atm compiling a sparc tool-chain to try it out.
What about moving the check added by rmk to kallsyms.c?
This would remove the extra pass on vmlinux which is for no use for
most people anyway. On the other hand an error could go unpassed
because we (for now) do not do the kallsyms stuff if not configured in.

We could make the kallsyms run independent on the configuration, but
only link in the symbols if required to do so.
This would also allow us to have architecture specific final-linking
rules in one place if sparc needs special rules.
Today kallsyms already knows about 'SDA_BASE*_' only valid for ppc.

wli - can you post the output of a failing sparc compile?

	Sam

Re: 2.6.8-rc2-mm1

[William Lee Irwin III]

[-- Attachment #1: brief message --]
[-- Type: text/plain, Size: 898 bytes --]

On Sun, Aug 01, 2004 at 11:11:46PM +0200, Sam Ravnborg wrote:
> Took a look at this and atm compiling a sparc tool-chain to try it out.
> What about moving the check added by rmk to kallsyms.c?
> This would remove the extra pass on vmlinux which is for no use for
> most people anyway. On the other hand an error could go unpassed
> because we (for now) do not do the kallsyms stuff if not configured in.
> We could make the kallsyms run independent on the configuration, but
> only link in the symbols if required to do so.
> This would also allow us to have architecture specific final-linking
> rules in one place if sparc needs special rules.
> Today kallsyms already knows about 'SDA_BASE*_' only valid for ppc.
> wli - can you post the output of a failing sparc compile?

The only part that fails is the ldchk part (unique to -mm), which is
just happening prematurely. Here's a log.


-- wli

[-- Attachment #2: sparc.compile.log.1.gz --]
[-- Type: application/octet-stream, Size: 19302 bytes --]

Re: 2.6.8-rc2-mm1

[Sam Ravnborg]

On Sun, Aug 01, 2004 at 04:57:29PM -0700, William Lee Irwin III wrote:
Content-Description: brief message
> On Sun, Aug 01, 2004 at 11:11:46PM +0200, Sam Ravnborg wrote:
> > Took a look at this and atm compiling a sparc tool-chain to try it out.
> > What about moving the check added by rmk to kallsyms.c?
> > This would remove the extra pass on vmlinux which is for no use for
> > most people anyway. On the other hand an error could go unpassed
> > because we (for now) do not do the kallsyms stuff if not configured in.
> > We could make the kallsyms run independent on the configuration, but
> > only link in the symbols if required to do so.
> > This would also allow us to have architecture specific final-linking
> > rules in one place if sparc needs special rules.
> > Today kallsyms already knows about 'SDA_BASE*_' only valid for ppc.
> > wli - can you post the output of a failing sparc compile?
> 
> The only part that fails is the ldchk part (unique to -mm), which is
> just happening prematurely. Here's a log.
Thanks.

I have been looking into this - and got sidetracked by coding an Elf
parser to replace usage of nm & friends. I abandoned this idea
again - there was not much gain in the end.

For sparc the rule seems to be:
symbols starting with ___ are legal to be undefined, they will
be fixed up later.
Following the code in btfixupprep.c gave me an headace..
A few comments with sample output from objdump would have helped a lot.


For now I'm working with the idea to move generation of System.map to a
separate shell script 'scripts/mksystemmap'.
As a sideeffect let this script check for potential undefined symbols.
Either ignore ___* symbols always or make this a sparc special case.
Compared to rmk's original patch this would only execute the check once.

sparc can use the same script when generating the final image.

I did not use sed to make a nice output of undefined symbols to avoid
only seeing a bunch of empty lines as Dave Hansen reported in one case.

	Sam
	

First version of the (simple) script. 
Direct reference to nm only for testing purpose.
Too late so no patch..

#!/bin/sh -e
# From the vmlinux file create the System.map file
# System.map is used by various debugging tools to retreive the actual
# addresses of symbols in the kernel.
# The optional kallsyms file in /proc/kallsyms serve the same purpose
# While creating the System.map file as a sideeffect check for
# undefined symbols.
# At least one version of the ARM bin-utils did not error out on
# undefined symbols, so catch them here instead.
# Usage
# mksystemmap vmlinux System.map

# $(NM) produces the following output:
# f0081e80 T alloc_vfsmnt
# The second row specify the type f the symbol:
# A = Absolute
# B = Uninitialised data (.bss)
# C = Comon symbol
# D = Initialised data
# G = Initialised data for small objects
# I = Indirect reference to another symbol
# N = Debugging symbol
# R = Read only
# S = Uninitialised data for small objects
# T = Text code symbol
# U = Undefined symbol
# V = Weak symbol
# W = Weak symbol
# Corresponding small letters are local symbols

# For System.map filter away:
# a - local absolute symbols
# U - undefined global symbols
# w - local weak symbols

nm $1 | grep -v ' [aUw] ' > $2

# Any undefined symbols?
UNDEF=`nm -u $1 | grep -v ' ____'`

UNDEFCNT=`echo $UNDEF | wc -l`

if [ $UNDEFCNT ]; then
	echo $0: Found $UNDEFCNT undefined symbols:
	nm -u $1
	echo This may signal in error in bin-utils
	exit 1
fi
exit 0