Discussion:
2.6.17-rc1-mm1
(too old to reply)
Andrew Morton
2006-04-04 08:45:04 UTC
Permalink
ftp://ftp.kernel.org/pub/linux/kernel/people/akpm/patches/2.6/2.6.17-rc1/2.6.17-rc1-mm1/


- VGA on ia64 is broken - the screen comes up blank. But ia64 otherwise
seems to work OK. I didn't have time to investigate.

- Linus is out of town until the weekend - his net connectedness is
uncertain.

- I dropped the old kgdb patch and picked up a new version for x86 only from
kgdb.linsyssoft.com. It works much better, but disagrees violently with
gcc-4.2 for some reason.




Boilerplate:

- See the `hot-fixes' directory for any important updates to this patchset.

- To fetch an -mm tree using git, use (for example)

git fetch git://git.kernel.org/pub/scm/linux/kernel/git/smurf/linux-trees.git v2.6.16-rc2-mm1

- -mm kernel commit activity can be reviewed by subscribing to the
mm-commits mailing list.

echo "subscribe mm-commits" | mail ***@vger.kernel.org

- If you hit a bug in -mm and it is not obvious which patch caused it, it is
most valuable if you can perform a bisection search to identify which patch
introduced the bug. Instructions for this process are at

http://www.zip.com.au/~akpm/linux/patches/stuff/bisecting-mm-trees.txt

But beware that this process takes some time (around ten rebuilds and
reboots), so consider reporting the bug first and if we cannot immediately
identify the faulty patch, then perform the bisection search.

- When reporting bugs, please try to Cc: the relevant maintainer and mailing
list on any email.



Changes since 2.6.16-mm2:


git-acpi.patch
git-agpgart.patch
git-cfq.patch
git-cpufreq.patch
git-dvb.patch
git-infiniband.patch
git-intelfb.patch
git-libata-all.patch
git-netdev-all.patch
git-ocfs2.patch
git-scsi-misc.patch
git-scsi-target.patch
git-sas-jg.patch
git-viro-bird-m32r.patch
git-viro-bird-m68k.patch
git-viro-bird-frv.patch
git-viro-bird-upf.patch
git-viro-bird-volatile.patch

git trees.

-rtc-remove-rtc-uip-synchronization-on-x86.patch
-rtc-remove-rtc-uip-synchronization-on-x86_64.patch
-rtc-remove-rtc-uip-synchronization-on-sparc64.patch
-rtc-remove-rtc-uip-synchronization-on-ppc-chrp-arch-ppc.patch
-rtc-remove-rtc-uip-synchronization-on-chrp-arch-powerpc.patch
-rtc-remove-rtc-uip-synchronization-on-ppc-maple.patch
-rtc-remove-rtc-uip-synchronization-on-arm.patch
-rtc-remove-rtc-uip-synchronization-on-mips-mc146818.patch
-rtc-remove-rtc-uip-synchronization-on-mips-based-dec.patch
-rtc-remove-rtc-uip-synchronization-on-sh03.patch
-rtc-remove-rtc-uip-synchronization-on-sh-mpc1211.patch
-rtc-remove-rtc-uip-synchronization-on-alpha.patch
-rtc-fix-up-some-rtc-whitespace-and-style.patch
-rtc-remove-some-duplicate-bcd-definitions.patch
-git-acpi-up-fix.patch
-pnpacpi-fix-non-memory-address-space-descriptor-handling.patch
-pnpacpi-remove-some-code-duplication.patch
-pnpacpi-whitespace-cleanup.patch
-acpi-request-correct-fixed-hardware-resource-type-mmio-vs-i-o-port.patch
-acpi-add-acpi-to-motherboard-resources-in-proc-iomemport.patch
-acpi-update-asus_acpi-driver-registration.patch
-acpi-fix-sonypi-acpi-driver-registration.patch
-acpi-make-acpi_bus_register_driver-return-success-failure-not-device-count.patch
-acpi-simplify-scanc-coding.patch
-acpi-print-wakeup-device-list-on-same-line-as-label.patch
-acpi-fix-memory-hotplug-range-length-handling.patch
-hpet-fix-acpi-memory-range-length-handling.patch
-acpi_os_acquire_object-gfp_kernel-called-with-irqs.patch
-acpi-remove-__init-__exit-from-asus-add-remove-methods.patch
-acpi-signedness-fix-2.patch
-acpi-should-depend-on-not-select-pci.patch
-acpi-ec-acpi-ecdt-uid-hack.patch
-remove-entries-in-sys-firmware-acpi-for-processor-also.patch
-remove-unnecessary-lapic-definition-from-acpidefh.patch
-catch-notification-of-memory-add-event-of-acpi-via-container-driver-register-start-func-for-memory-device-tidy.patch
-catch-notification-of-memory-add-event-of-acpi-via-container-driveravoid-redundant-call-add_memory-tidy.patch
-kernel-crash-in-powernow-k8-after-lost-ticks-detected.patch
-drm-sis-fix-compile-warning.patch
-remove-drm_allocfree_pages.patch
-ia64-sn_hwperf-eliminate-use-of-num_online_cpus.patch
-input-adds-snes-mouse-support-to-gamecon.patch
-m25p80-printk-warning-fix.patch
-sem2mutex-mtd-doc2000c.patch
-sem2mutex-drivers-mtd.patch
-drivers-mtd-small-cleanups.patch
-mtd_nand_sharpsl-and-mtd_nand_nandsim-should-be-tristates.patch
-drivers-mtd-use-array_size-macro.patch
-mtd-cmdlinepart-allow-zero-offset-value.patch
-fix-debug-statement-in-inftlcorec.patch
-kill-ifdefs-in-mtdcorec.patch
-kill-ifdefs-in-mtdcorec-fix.patch
-dead-code-in-mtd-maps-pcic.patch
-add-chip-used-in-collie-to-jedec_probe.patch
-mtd-redboot-handle-holes-in-fis-table.patch
-mtd-fix-broken-name_to_dev_t-declaration.patch
-drivers_mtd_maps_vmax301c-fix-off-by-one-vmax_mtd.patch
-natsemi-support-oversized-eeproms.patch
-net-remove-config_net_cbus-conditional-for-ns8390.patch
-com90xx-kmalloc-fix.patch
-netfilter-rename-init-functions.patch
-net-fix-appletalk-compat_ioctl-oopses.patch
-deinline-200-byte-inlines-in-sockh.patch
-deinline-some-larger-functions-from-netdeviceh.patch
-git-powerpc-warn-was-a-dumb-idea.patch
-git-serial.patch
-serial-mystery-kbuild-fix.patch
-git-sym2.patch
-fix-pcmcia_device_remove-oops.patch
-aacraid-fix-the-comparison-with-sizeof.patch
-x86_64-mm-hangcheck-remove-message.patch
-x86_64-mm-lost-cli-debug.patch
-x86_64-mm-sis-agp.patch
-x86_64-mm-acpi-nolapic.patch
-x86_64-mm-extra-nodes_shift-definition.patch
-x86_64-mm-hotadd-reserve-fix.patch
-git-xfs-vn_to_inode-fix.patch
-mm-schedule-find_trylock_page-removal.patch
-remove-long-dead-i386-floppy-asm-code.patch
-i386-kdump-timer-vector-lockup-fix.patch
-uml-hotplug-memory-take-2.patch
-decrapify-asm-generic-localh.patch
-mark-unwind-info-for-signal-trampolines-in-vdsos.patch
-locks-dont-panic.patch
-document-linuxs-memory-barriers.patch
-synclink-remove-dead-code.patch
-synclink_gt-add-gpio-feature.patch
-synclink_gt-remove-uneeded-async-code.patch
-let-blk_dev_ram_count-depend-on-blk_dev_ram.patch
-kill-__init_timer_base-in-favor-of-boot_tvec_bases.patch
-__mod_timer-simplify-base-changing.patch
-paride-register_chrdev-fix.patch
-paride-pt-register_chrdev-fix.patch
-capi-register_chrdev-fix.patch
-symversion-warning-fix.patch
-simplify-proc-devices-and-fix-early-termination-regression.patch
-simplify-proc-devices-and-fix-early-termination-regression-tidy.patch
-simplify-proc-devices-and-fix-early-termination-regression-tidy-2.patch
-simplify-proc-devices-and-fix-early-termination-regression-tidy-3.patch
-dont-pass-boot-parameters-to-argv_init.patch
-add-oprofile_add_ext_sample.patch
-alpha-make-poll-flags-thesame-as-other-architectures.patch
-mqueue-comment-fix-fix.patch
-change-dash2underscore-return-value-to-char.patch
-drivers-block-paride-pdc-fix-an-off-by-one-error.patch
-fs-fat-proper-prototypes-for-two-functions.patch
-philip-gladstone-has-moved.patch
-edac_752x-needs-config_hotplug.patch
-make-tty_insert_flip_char-a-non-gpl-export.patch
-ipmi-fix-startup-race-condition.patch
-ipmi-fix-startup-race-condition-tidy.patch
-ipmi-tidy-up-various-things.patch
-ipmi-convert-from-semaphores-to-mutexes.patch
-mutex-some-cleanups.patch
-remove-relayfs_fsh.patch
-drivers-block-acsi_slmc-size_t-cant-be-0.patch
-autofs4-proper-prototype-for-autofs4_dentry_release.patch
-exec-allow-init-to-exec-from-any-thread.patch
-simplify-exec-from-inits-subthread.patch
-remove-dead-kill_sl-prototype-from-schedh.patch
-do_tty_hangup-use-group_send_sig_info-not.patch
-do_sak-dont-depend-on-session-id-0.patch
-pidhash-kill-switch_exec_pids.patch
-choose_new_parent-remove-unused-arg-sanitize-exit_state-check.patch
-remove-add_parents-parent-argument.patch
-dont-use-remove_links-set_links-for-reparenting.patch
-kill-set_links-remove_links.patch
-pidhash-dont-count-idle-threads.patch
-pidhash-dont-use-zero-pids.patch
-reparent_thread-use-remove_parent-add_parent.patch
-wait_for_helper-trivial-style-cleanup.patch
-release_task-replace-open-coded-ptrace_unlink.patch
-convert-sighand_cache-to-use-slab_destroy_by_rcu.patch
-introduce-lock_task_sighand-helper.patch
-introduce-sig_needs_tasklist-helper.patch
-copy_process-cleanup-bad_fork_cleanup_sighand.patch
-copy_process-cleanup-bad_fork_cleanup_signal.patch
-cleanup-__exit_signal.patch
-rename-__exit_sighand-to-cleanup_sighand.patch
-move-__exit_signal-to-kernel-exitc.patch
-revert-optimize-sys_times-for-a-single-thread-process.patch
-do-__unhash_process-under-siglock.patch
-sys_times-dont-take-tasklist_lock.patch
-relax-sig_needs_tasklist.patch
-do_signal_stop-dont-take-tasklist_lock.patch
-do_group_exit-dont-take-tasklist_lock.patch
-do_sigaction-dont-take-tasklist_lock.patch
-pids-kill-pidtype_tgid.patch
-make-fork-atomic-wrt-pgrp-session-signals.patch
-cleanup-__exit_signal-cleanup_sighand-path.patch
-simplify-do_signal_stop.patch
-finish_stop-dont-check-stop_count-0.patch
-do_notify_parent_cldstop-remove-to_self-param.patch
-send_sigqueue-simplify-and-fix-the-race.patch
-permit-dual-mit-gpl-licenses.patch
-led-class-documentation.patch
-led-add-led-class.patch
-led-add-led-trigger-support.patch
-led-add-led-timer-trigger.patch
-led-add-sharp-charger-status-led-trigger.patch
-led-add-led-device-support-for-the-zaurus-corgi-and.patch
-led-add-led-device-support-for-locomo-devices.patch
-led-add-led-device-support-for-ixp4xx-devices.patch
-led-add-device-support-for-tosa.patch
-led-add-nand-mtd-activity-led-trigger.patch
-led-add-ide-disk-activity-led-trigger.patch
-ensure-ide-taskfile-calls-any-driver-specific.patch
-hrtimer-create-generic-sleeper.patch
-hrtimer-use-generic-sleeper-for-nanosleep.patch
-sched-cleanup_task_activated.patch
-sched-make_task_noninteractive_use_sleep_type.patch
-sched-dont_decrease_idle_sleep_avg.patch
-sched-include_noninteractive_sleep_in_idle_detect.patch
-sched-remove-on-runqueue-requeueing.patch
-sched-activate-sched-batch-expired.patch
-sched-reduce-overhead-of-calc_load.patch
-sched-fix-interactive-task-starvation.patch
-futex-check-and-validate-timevals.patch
-rtc-m41t00-driver-should-use-workqueue-instead-of-tasklet.patch
-rtc-m41t00-driver-cleanup.patch
-rtc-add-support-for-m41t81-m41t85-chips-to-m41t00-driver.patch
-trivial-cleanup-to-proc_check_chroot.patch
-resurrect-__put_task_struct.patch
-task-rcu-protect-task-usage.patch
-make-setsid-more-robust.patch
-dcache-add-helper-d_hash_and_lookup.patch
-pidhash-refactor-the-pid-hash-table.patch
-ide-amd756-no-host-side-cable-detection.patch
-small-fixes-backported-to-old-ide-sis-driver.patch
-ide_generic_all_on-warning-fix.patch
-fbcon-save-current-display-during-initialization.patch
-w100fb-add-acceleration-support-to-ati-imageon.patch
-backlight-backlight-class-improvements.patch
-backlight-hp-jornada-680-backlight-driver-updates-fixes.patch
-backlight-corgi_bl-generalise-to-support-other-sharp.patch
-pxafb-minor-driver-fixes.patch
-fbcon-fix-big-endian-bogosity-in-slow_imageblit.patch
-optimize-select-poll-by-putting-small-data-sets-on-the-stack.patch
-use-fget_light-in-select-poll.patch
-fold-select_bits_alloc-free-into-caller-code.patch
-ia64-const-f_ops-fix.patch
-mark-f_ops-const-in-the-inode.patch
-make-most-file-operations-structs-in-fs-const.patch
-docs-update-missing-files-and-descriptions-for-filesystems-00-index.patch
-arch-i386-kernel-microcodec-remove-the-obsolete-microcode_ioctl.patch
-drivers-block-use-time_after-and-friends.patch
-nvidia-agp-use-time_before_eq.patch
-ide-tape-use-time_after-time_after_eq.patch
-drivers-scsi-use-time_after-and-friends.patch
-replace-0xff-with-correct-dma_xbit_mask.patch
-vfree-null-check-fixup-for-sb_card.patch
-maestro3-vfree-null-check-fixup.patch
-no-need-to-check-vfree-arg-for-null-in-oss-sequencer.patch
-vfree-does-its-own-null-check-no-need-to-be-explicit-in-oss-msndc.patch
-fix-signed-vs-unsigned-in-nmi-watchdog.patch
-trivial-typos-in-documentation-cputopologytxt.patch
-typos-grab-bag-of-the-month.patch
-unexport-get_wchan.patch
-fs-nameic-make-lookup_hash-static.patch
-decrease-number-of-pointer-derefs-in-jsm_ttyc.patch
-sound-remove-unneeded-kmalloc-return-value-casts.patch

Merged into mainline or a subsystem tree

+selinux-build-fix.patch
+sched-fix-interactive-task-starvation.patch
+dont-awaken-rt-tasks-on-expired-array.patch
+select-warning-fixes.patch
+fix-null-pointer-dereference-in-node_read_numastat.patch
+dmi-move-dmi_scanc-from-arch-i386-to-drivers-firmware.patch
+config_net=n-build-fix.patch
+drm_pci-needs-dma-mappingh.patch

2.6.17-rc2 queue (part thereof, many more 2.6.17 patches below)

+acpi-update-asus_acpi-driver-registration-fix.patch

ACPI fix

+for_each_possible_cpu-under-drivers-acpi.patch
+acpi_bus_unregister_driver-make-void.patch
+acpi_os_wait_semaphore-dont-complain-about-timeout.patch

ACPI updates

+for_each_possible_cpu-for-arm.patch

ARM for_each_cpu() conversion.

+powernow-k8-crash-workaround.patch

cpufreq non-fix.

+gregkh-driver-driver-core-safely-unbind-drivers-for-devices-not-on-a-bus.patch
+gregkh-driver-block-delay-all-uevents-until-partition-table-is-scanned.patch

Driver tree updates.

-revert-gregkh-driver-fix-up-the-sysfs-pollable-patch.patch

Dropped.

+spi-whitespace-fixes.patch
+spi-bounce-buffer-has-a-minimum-length.patch
+spi-add-david-as-the-spi-subsystem-maintainer.patch
+spi-renamed-bitbang_transfer_setup-to-spi_bitbang_setup_transfer.patch
+spi-busnum-==-0-needs-to-work.patch
+spi-devices-can-require-lsb-first-encodings.patch

SPI updates.

+driver-core-fix-unnecessary-null-check-in-drivers-base-classc.patch

driver core cleanup.

+drm-fix-issue-reported-by-coverity-in-drivers-char-drm-via_irqc.patch

DRM fix.

+avermedia-6-eyes-avs6eyes-support.patch
+bt866-build-fix.patch

New v4l driver, and a fix.

+gregkh-i2c-rtc-ds1374-convert-tasklet-to-workqueue.patch
+gregkh-i2c-rtc-m41t00-driver-should-use-workqueue-instead-of-tasklet.patch
+gregkh-i2c-hwmon-w83792d-quiet-on-misdetection.patch
+gregkh-i2c-i2c-sis96x-remove-init-log-message.patch
+gregkh-i2c-i2c-parport-require-type-parameter.patch
+gregkh-i2c-hwmon-lm83-add-lm82-support.patch
+gregkh-i2c-hwmon-w83627ehf-add-voltages.patch
+gregkh-i2c-hwmon-w83627ehf-add-alarms.patch
+gregkh-i2c-hwmon-smsc47m192-new-driver.patch
+gregkh-i2c-hwmon-f71805f-no-global-resource.patch
+gregkh-i2c-hwmon-sysfs-interface-individual-alarm-files.patch
+gregkh-i2c-i2c-piix4-add-ati-smbus-support.patch
+gregkh-i2c-rtc-m41t00-driver-cleanup.patch
+gregkh-i2c-w1-added-default-generic-read-write-operations.patch
+gregkh-i2c-w1-replace-dscore-and-ds_w1_bridge-with-ds2490-driver.patch
+gregkh-i2c-w1-userspace-communication-protocol-over-connector.patch
+gregkh-i2c-w1-move-w1-connector-definitions-into-linux-include-connector.h.patch
+gregkh-i2c-w1-netlink-mark-netlink-group-1-as-unused.patch

I2C tree.

+connector-exports.patch
+w1-exports.patch

Fix things in Greg's trees.

+scx200_acb-fix-for-cs5535-eratta.patch
+scx200_acb-use-pci-i-o-resource-when-appropriate.patch

i2c driver updates.

+for_each_possible_cpu-ia64.patch

ia64 for_each_cpu() conversion.

+wistron_btns-add-support-for-amilo-7400m.patch

Additional device support.

+sata_mv-properly-print-hc-registers.patch

SATA fix.

+for_each_possible_cpu-mips.patch

MIPS for_each_cpu() conversion.

-pci-error-recovery-e1000-network-device-driver-tidy.patch

Folded into pci-error-recovery-e1000-network-device-driver.patch

+e1000-prevent-statistics-from-getting-garbled-during-reset.patch
+net-drivers-fix-section-attributes-for-gcc.patch

netdev fixes.

+for_each_possible_cpu-network-codes.patch

net for_each_cpu() conversion.

+netfilter-fix-fragmentation-issues-with-bridge-netfilter.patch
+ebtables-fix-allocation-in-net-bridge-netfilter-ebtablesc.patch

net fixes.

+gregkh-pci-msi-save-restore-for-suspend-resume.patch
+gregkh-pci-pci_ids.h-correct-naming-of-1022-7450.patch
+gregkh-pci-pci-fix-sparse-warning-about-pci_bus_flags.patch
+gregkh-pci-pci-rpaphp-remove-init-error-condition.patch
+gregkh-pci-re-arch-i386-pci-irq.c-new-via-chipsets.patch
+gregkh-pci-pci-add-pci-quirk-for-smbus-on-the-asus-a6va-notebook.patch

PCI tree updates.

+64-bit-resources-drivers-pci-changes-sparc32-fix.patch

Fix 64-bit-resources-drivers-pci-changes.patch.

+64-bit-resources-arch-powerpc-changes.patch
+64-bit-resources-more-drivers-others-changes.patch
+64-bit-resources-more-sound-changes.patch

Updates to the 64-bit resource patches.

+for_each_possible_cpu-sparc.patch
+for_each_possible_cpu-sparc64.patch

for_each_cpu() conversions.

+gregkh-usb-usb-net2282-and-net2280-software-compatibility.patch
+gregkh-usb-usb-cleanups-for-ohci-s3c2410.c.patch
+gregkh-usb-usb-ftdi_sio-add-support-for-eclo-com-to-1-wire-usb-adapter.patch
+gregkh-usb-usb-g_file_storage-set-short_not_ok-for-bulk-out-transfers.patch
+gregkh-usb-usb-g_file_storage-add-comment-about-buffer-allocation.patch
+gregkh-usb-usb-serial-converts-port-semaphore-to-mutexes.patch
+gregkh-usb-usb-pci-quirks.c-proper-prototypes.patch
+gregkh-usb-usb-input-proper-prototypes.patch
+gregkh-usb-usb-add-support-for-papouch-tmu.patch
+gregkh-usb-usb-usbtouchscreen-unified-usb-touchscreen-driver.patch
+gregkh-usb-usb-g_file_storage-use-module_param_array_named-macro.patch
+gregkh-usb-usb-wacom-tablet-driver-update.patch
+gregkh-usb-usb-add-new-wacom-devices-to-usb-hid-core-list.patch
+gregkh-usb-usb-pegasus-driver-bugfix.patch

USB tree updates.

+hid-corec-fix-input-irq-status-32-received-for-silvercrest-usb-keyboard.patch

USB fix.

+softmac-uses-wiress-ext.patch

Wireless fix.

+x86_64-mm-execve-cleanup.patch
+x86_64-mm-clustered-check.patch
+x86_64-mm-nodes-shift-dummy.patch
+x86_64-mm-mce-nmi-watchdog.patch
+x86_64-mm-i386-up-generic-arch.patch
+x86_64-mm-i386-bigsmp-fadt.patch
+x86_64-mm-force-iret.patch
+x86_64-mm-strlen-export.patch
+x86_64-mm-hpet-return.patch
+x86_64-mm-vsmp-cache-boundary.patch
+x86_64-mm-mcfg-check-more-busses.patch
+x86_64-mm-mmconfig-error-value.patch
+x86_64-mm-memset-always-inline.patch
+x86_64-mm-hpet-drift.patch
+x86_64-mm-gs-leak.patch
+x86_64-mm-fix-config_reorder.patch

x86_64 updates.

+arm-add_memory-build-fix.patch

They broke arm.

+for_each_possible_cpu-x86_64.patch

x86_64 for_each_cpu() conversion.

+for_each_possible_cpu-xfs.patch

XFS for_each_cpu() conversion.

+mm-posix-memory-lock.patch
+slab-allocate-node-local-memory-for-off-slab-slabmanagement.patch
+slab-add-statistics-for-alien-cache-overflows.patch
+nommu-use-compound-page-in-slab-allocator.patch
+mm-fix-bug-in-brk.patch

MM updates.

+frv-define-mmu-mode-specific-syscalls-as-cond_syscall-and-clean-up-unneeded-macros.patch

FRV fix.

+x86-clean-up-subarch-definitions.patch
+x86-clean-up-subarch-definitions-update.patch
+arch-i386-kernel-apicc-make-modern_apic-static.patch
+enable-tsc-for-amd-geode-gx-lx.patch
+swsusp-dont-require-bigsmp.patch
+i386-print-eip-esp-last.patch
+menu-relocate-doublefault-option.patch

x86 updates.

-swsusp-resume-parsing-fix.patch

Dropped.

+uml-tls-fixlets.patch

UML fix.

+s390-update-default-configuration.patch
+s390-ebdic-to-ascii-conversion-tables.patch
+s390-invalid-check-after-kzalloc.patch
+s390-wrong-return-codes-in-cio_ignore_proc_init.patch
+s390-increase-cio_trace-debug-event-size.patch
+s390-dasd-device-offline-messages.patch
+s390-fail-fast-requests-on-quiesced-devices.patch
+s390-dasd-proc-entries.patch
+s390-minor-tape-fixes.patch
+s390-fix-implicit-declaration-of-unlikely.patch

s390 updates.

+hdaps-add-support-for-thinkpad-r52.patch
+configurable-nodes_shift.patch
+ext3-block-allocation-reservation-fixes-to-support.patch
+ext3-ext3-in-kernel-block-number-type-fixes.patch
+ext3-ext3-in-kernel-block-number-type-fixes-fix.patch
+unify-pxm_to_node-and-node_to_pxm.patch
+no-arch-specific-strpbrk-implementations.patch
+clean-up-arch-overrides-in-linux-stringh.patch
+sync_file_range-use-unsigned-for-flags.patch
+timer-initialisation-fix.patch
+timer-initialisation-fix-tidy.patch
+avoid-tasklist_lock-at-getrusage-for-multithreaded-case-too.patch
+the-scheduled-unexport-of-panic_timeout.patch
+s3c24xx-gpio-led-support.patch
+s3c24xx-gpio-led-support-tidy.patch
+leds-fix-ide-disk-trigger-name.patch
+leds-reorganise-kconfig.patch
+leds-re-layout-include-linux-ledsh.patch
+vfs-propagate-mnt_flags-into-do_loopback-vfsmount.patch
+build-kernel-irq-migrationc-only-if-config_generic_pending_irq-is-set.patch
+remove-sys_-prefix-of-new-syscalls-from-__nr_sys_.patch
+add-prctl-to-change-endian-of-a-task.patch
+#writeback-fix-range-handling.patch
+make-tty_insert_flip_string_flags-a-non-gpl-export.patch
+memory_hotplugh-cleanup.patch
+9p-handle-sget-failure.patch
+remove-extraneous-n-in-doubletalk-init-printk.patch
+missing-n-in-sc1200wdt-driver.patch
+reinstate-const-in-next_thread.patch
+select-dont-overflow-if-select_stack_alloc-%-sizeoflong-=-0.patch
+silence-a-const-vs-non-const-warning.patch
+kdump-proc-vmcore-size-oveflow-fix.patch
+hdaps-support-new-lenovo-machines.patch
+fix-dcache-race-during-umount.patch
+prune_one_dentry-tweaks.patch
+uniform-pollrdhup-handling-between-epoll-and-poll-select.patch
+add-cpu_relax-to-hrtimer_cancel.patch

Misc.

+introduce-hlist_move_head.patch
+use-hlist_move_head.patch
+use-list_add_tail-instead-of-list_add.patch
+use-list_add_tail-instead-of-list_add-fix.patch
+arch-use-list_move.patch
+core-use-list_move.patch
+net-rxrpc-use-list_move.patch
+drivers-use-list_move.patch
+fs-use-list_move.patch

list.h cleanups.

+rtc-subsystem-ds1672-oscillator-handling.patch
+rtc-subsystem-ds1672-cleanup.patch
+rtc-subsystem-x1205-sysfs-cleanup.patch
+rtc-subsystem-whitespaces-and-error-messages-cleanup.patch
+rtc-subsystem-fix-proc-output.patch
+rtc-subsystem-rs5c372-sysfs-fix.patch
+rtc-subsystem-compact-error-messages.patch
+rtc-subsystem-sa1100-cleanup.patch
+rtc-subsystem-vr41xx-driver.patch
+rtc-subsystem-vr41xx-cleanup.patch

RTC system updates.

+fuse-fix-oops-in-fuse_send_readpages.patch
+fuse-fix-fuse_dev_poll-return-value.patch
+fuse-add-o_async-support-to-fuse-device.patch
+fuse-add-o_nonblock-support-to-fuse-device.patch
+fuse-simplify-locking.patch
+fuse-use-a-per-mount-spinlock.patch
+fuse-consolidate-device-errors.patch
+fuse-clean-up-request-accounting.patch
+fuse-account-background-requests.patch

FUSE updates.

+isdn4linux-siemens-gigaset-drivers-code-cleanup.patch
+isdn4linux-siemens-gigaset-drivers-kconfig-correction.patch
+isdn4linux-siemens-gigaset-drivers-timer-usage.patch
+isdn4linux-siemens-gigaset-drivers-logging-usage.patch
+isdn4linux-siemens-gigaset-drivers-sysfs-usage.patch
+isdn4linux-siemens-gigaset-drivers-remove-ifnull-macros.patch
+isdn4linux-siemens-gigaset-drivers-uninline.patch
+isdn4linux-siemens-gigaset-drivers-elliminate-from_user-argument.patch
+isdn4linux-siemens-gigaset-drivers-mutex-conversion.patch
+isdn4linux-siemens-gigaset-drivers-remove-private-version-of-__skb_put.patch
+isdn4linux-siemens-gigaset-drivers-remove-forward-references.patch
+isdn4linux-siemens-gigaset-drivers-add-readme.patch
+isdn4linux-siemens-gigaset-drivers-make-some-variables-non-atomic.patch

ISDN driver updates.

+knfsd-correct-reserved-reply-space-for-read-requests.patch
+knfsd-locks-flag-nfsv4-owned-locks.patch
+knfsd-nfsd4-wrong-error-handling-in-nfs4acl.patch
+knfsd-nfsd4-better-nfs4acl-errors.patch
+knfsd-nfsd4-fix-acl-xattr-length-return.patch
+knfsd-nfsd-oops-exporting-nonexistent-directory.patch
+knfsd-nfsd-nfsd_setuser-doesnt-really-need-to-modify-rqstp-rq_cred.patch
+knfsd-nfsd4-remove-nfsd_setuser-from-putrootfh.patch
+knfsd-nfsd4-fix-corruption-of-returned-data-when-using-64k-pages.patch
+knfsd-nfsd4-fix-corruption-on-readdir-encoding-with-64k-pages.patch
+knfsd-svcrpc-gss-dont-call-svc_take_page-unnecessarily.patch
+knfsd-svcrpc-warn-instead-of-returning-an-error-from-svc_take_page.patch
+knfsd-nfsd4-fix-laundromat-shutdown-race.patch
+knfsd-nfsd4-nfsd4_probe_callback-cleanup.patch
+knfsd-nfsd4-add-missing-rpciod_down.patch
+knfsd-nfsd4-limit-number-of-delegations-handed-out.patch
+knfsd-nfsd4-limit-number-of-delegations-handed-out-fix.patch
+knfsd-nfsd4-grant-delegations-more-frequently.patch

knfsd updates.

+sched-prevent-high-load-weight-tasks-suppressing-balancing.patch
+sched-improve-stability-of-smpnice-load-balancing.patch
+sched_domain-handle-kmalloc-failure.patch
+sched_domain-dont-use-gfp_atomic.patch
+sched_domain-use-kmalloc_node.patch
+sched_domain-allocate-sched_group-structures-dynamically.patch

CPU scheduler updates.

+pi-futex-futex-code-cleanups-fix.patch

Fix pi-futex-futex-code-cleanups.patch

+pi-futex-rt-mutex-core-fix-timeout-race.patch

Fix pi-futex-rt-mutex-core.patch

+pi-futex-patchset-v4.patch
+pi-futex-patchset-v4-update.patch

PI-futex updates.

-reiser4-sb_sync_inodes-cleanup.patch
-reiser4-include-reiser4.patch
-reiser4-doc.patch
-reiser4-doc-fix-reiser4-links-in-documentation.patch
-reiser4-only.patch
-reiser4-cleanup_init_fake_inode.patch
-reiser4-only-stop-using-__put_page.patch
-reiser4-ver_linux-dont-print-reiser4progs-version-if-none-found.patch
-reiser4-fix-wundef-warnings.patch
-reiser4-swsusp-build-fix.patch
-reiser4-reboot-fix.patch
-reiser4-update-filesystems-for-new-delete_inode-behavior.patch
-reiser4-xattr-build-fix.patch
-reiser4-printk-warning-fix.patch
-reiser4-mm-remove-pg_highmem-fix.patch
-reiser4-kconfig-help-cleanup.patch
-reiser4-update.patch
-reiser4-fix-dependencies.patch
-reiser4-wundef-fix.patch
-reiser4-wundef-fix-2.patch
-reiser4-big-update.patch
-reiser4-big-update-bug-fix-for-readpage.patch
-reiser4-big-update-bug-fix-for-readpage-fix.patch
-reiser4-big-update-rename-print_address.patch
-reiser4-page-private-fixes.patch
-reiser4-big-update-div64-fix.patch
-reiser4-remove-c99isms.patch
-reiser4-use-try_to_freeze.patch
-reiser4-fix-endianess.patch
-reiser4-check-null.patch
-reiser4-fix-built_ptr.patch
-reiser4-key-init-cleanup.patch
-reiser4-fix-list_splice-usage.patch
-reiser4-big-update-spin_macros-fix.patch
-reiser4-spinlock-cleanup.patch
-reiser4-warnings-cleanup.patch
-reiser4_releasepage-gfp_t-fixes.patch
-reiser4-fix-kbuild.patch
-reiser4-remove-rwx-perm-plugin.patch
-reiser4-fix-link_common.patch
-reiser4-fix-readlink.patch
-reiser4-add-crc-sendfile.patch
-reiser4-back-to-one-makefile.patch
-reiser4-rename-cluster-files.patch
-reiser4-crypt2cipher-rename.patch
-reiser4-lock-ordering-fix.patch
-reiser4-improved-comment.patch
-reiser4-try_capture_block-update.patch
-reiser4-fix-zeroing-in-crc-files.patch
-reiser4-atime-update-fix.patch
-reiser4-big-update-update_atime-fixes.patch
-fs-reiser4-possible-cleanups.patch
-reiser4-bugfix-patch.patch
-reiser4-i_sem-mutex-switch.patch
-reiser4-do-not-use-get_user_pages-and-do-not-check.patch
+reiser4.patch
+reiser4fs-use-list_move.patch

The reiser4 patches were all consolidated together.

+ide-claim-extra-dma-ports-regardless-of-channel.patch
+ide-remove-dma_base2-field-form-ide_hwif_t.patch
+ide-always-release-dma-engine.patch
+ide-ati-sb600-ide-support.patch
+ide-error-handling-fixes.patch
+hpt366-fix-segfault-during-init.patch

IDE updates.

+md-dm-reduce-stack-usage-with-stacked-block-devices.patch

MD update.

-kgdb-ga.patch
+kgdb-core-lite.patch
+kgdb-core-lite-add-reboot-command.patch
+kgdb-8250.patch
+kgdb-8250-fix.patch
+kgdb-netpoll_pass_skb_to_rx_hook.patch
+kgdb-eth.patch
+kgdb-i386-lite.patch
+kgdb-cfi_annotations.patch
+kgdb-sysrq_bugfix.patch
+kgdb-module.patch
+kgdb-core.patch
+kgdb-i386.patch

New kgdb patchset.




All 539 patches:


ftp://ftp.kernel.org/pub/linux/kernel/people/akpm/patches/2.6/2.6.17-rc1/2.6.17-rc1-mm1/patch-list
Kumar Gala
2006-04-04 14:31:10 UTC
Permalink
Post by Andrew Morton
- See the `hot-fixes' directory for any important updates to this patchset.
- To fetch an -mm tree using git, use (for example)
git fetch git://git.kernel.org/pub/scm/linux/kernel/git/smurf/
linux-trees.git v2.6.16-rc2-mm1
It doesn't seem like the git tree got updated correctly. The 2.6.17-
rc1-mm1 seems to be just 2.6.17-rc1 (just a quick spot check at
Makefile and a few patches I expected in like the 64-bit resource
change)

- kumar
Adrian Bunk
2006-04-04 16:02:23 UTC
Permalink
Post by Andrew Morton
...
...
+gregkh-i2c-w1-userspace-communication-protocol-over-connector.patch
...
I2C tree.
...
Evgeniy Polyakov did ACK my patch
a9fb1c7b950bed4afe208c9d67e20f086bb6abbb, and I'm therefore really
pissed off seeing him silently reverting it for no good reason as part
of this patch.

Greg, please drop this patch.

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
Evgeniy Polyakov
2006-04-04 16:35:22 UTC
Permalink
Post by Adrian Bunk
Post by Andrew Morton
...
...
+gregkh-i2c-w1-userspace-communication-protocol-over-connector.patch
...
I2C tree.
...
Evgeniy Polyakov did ACK my patch
a9fb1c7b950bed4afe208c9d67e20f086bb6abbb, and I'm therefore really
pissed off seeing him silently reverting it for no good reason as part
of this patch.
Greg, please drop this patch.
I'm really sorry, Adrian, if you feel it this way.
I converted your patch to the new tree.

Peace? :)

----
Nice cleanup spotted by Adrian Bunk, which was lost due to moving to the
completely new functionality.
Shame-shame-shame on me.

Signed-off-by: Evgeniy Polyakov <***@2ka.mipt.ru>

diff --git a/drivers/w1/w1.c b/drivers/w1/w1.c
index 6c23b38..174ac2b 100644
--- a/drivers/w1/w1.c
+++ b/drivers/w1/w1.c
@@ -214,11 +214,12 @@ struct device w1_master_device = {
.release = &w1_master_release
};

-struct device_driver w1_slave_driver = {
+static struct device_driver w1_slave_driver = {
.name = "w1_slave_driver",
.bus = &w1_bus_type,
};

+#if 0
struct device w1_slave_device = {
.parent = NULL,
.bus = &w1_bus_type,
@@ -226,6 +227,7 @@ struct device w1_slave_device = {
.driver = &w1_slave_driver,
.release = &w1_slave_release
};
+#endif /* 0 */

static ssize_t w1_master_attribute_show_name(struct device *dev, struct device_attribute *attr, char *buf)
{
@@ -383,7 +385,7 @@ int w1_create_master_attributes(struct w
return sysfs_create_group(&master->dev.kobj, &w1_master_defattr_group);
}

-void w1_destroy_master_attributes(struct w1_master *master)
+static void w1_destroy_master_attributes(struct w1_master *master)
{
sysfs_remove_group(&master->dev.kobj, &w1_master_defattr_group);
}
diff --git a/drivers/w1/w1.h b/drivers/w1/w1.h
index f5ad81d..deb3c13 100644
--- a/drivers/w1/w1.h
+++ b/drivers/w1/w1.h
@@ -213,6 +213,16 @@ static inline struct w1_master* dev_to_w
return container_of(dev, struct w1_master, dev);
}

+extern struct device_driver w1_master_driver;
+extern struct bus_type w1_bus_type;
+extern struct device w1_master_device;
+extern int w1_max_slave_count;
+extern int w1_max_slave_ttl;
+extern struct list_head w1_masters;
+extern struct mutex w1_mlock;
+
+extern int w1_process(void *);
+
#endif /* __KERNEL__ */

#endif /* __W1_H */
diff --git a/drivers/w1/w1_int.c b/drivers/w1/w1_int.c
index 24e7c10..475996c 100644
--- a/drivers/w1/w1_int.c
+++ b/drivers/w1/w1_int.c
@@ -30,16 +30,6 @@

static u32 w1_ids = 1;

-extern struct device_driver w1_master_driver;
-extern struct bus_type w1_bus_type;
-extern struct device w1_master_device;
-extern int w1_max_slave_count;
-extern int w1_max_slave_ttl;
-extern struct list_head w1_masters;
-extern struct mutex w1_mlock;
-
-extern int w1_process(void *);
-
static struct w1_master * w1_alloc_dev(u32 id, int slave_count, int slave_ttl,
struct device_driver *driver,
struct device *device)
@@ -96,7 +86,7 @@ static struct w1_master * w1_alloc_dev(u
return dev;
}

-void w1_free_dev(struct w1_master *dev)
+static void w1_free_dev(struct w1_master *dev)
{
device_unregister(&dev->dev);
}
Post by Adrian Bunk
cu
Adrian
--
Evgeniy Polyakov
Adrian Bunk
2006-04-04 16:29:21 UTC
Permalink
Post by Andrew Morton
...
...
+x86-clean-up-subarch-definitions.patch
...
x86 updates.
...
The following looks bogus:

config KEXEC
bool "kexec system call (EXPERIMENTAL)"
- depends on EXPERIMENTAL
+ depends on EXPERIMENTAL && (!X86_VOYAGER && SMP)

The dependencies do now say that KEXEC is only offered for machines that
are _both_ non-Voyager and SMP.

Is the problem you wanted to express that a non-SMP Voyager config
didn't compile?

AFAIR I recently sent a patch disallowing non-SMP Voyager configurations
that wasn't yet applied.

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
Zachary Amsden
2006-04-04 17:22:31 UTC
Permalink
Post by Adrian Bunk
Post by Andrew Morton
...
...
+x86-clean-up-subarch-definitions.patch
...
x86 updates.
...
config KEXEC
bool "kexec system call (EXPERIMENTAL)"
- depends on EXPERIMENTAL
+ depends on EXPERIMENTAL && (!X86_VOYAGER && SMP)
The dependencies do now say that KEXEC is only offered for machines that
are _both_ non-Voyager and SMP.
Is the problem you wanted to express that a non-SMP Voyager config
didn't compile?
Whoops, that should be

depends on EXPERIMENTAL && !(X86_VOYAGER && SMP)

Voyager SMP builds don't compile with kexec(), and it isn't clear how to
shootdown CPUs using NMIs without an APIC.
Eric W. Biederman
2006-04-04 17:50:49 UTC
Permalink
Post by Zachary Amsden
Post by Adrian Bunk
Post by Andrew Morton
...
...
+x86-clean-up-subarch-definitions.patch
...
x86 updates.
...
config KEXEC
bool "kexec system call (EXPERIMENTAL)"
- depends on EXPERIMENTAL
+ depends on EXPERIMENTAL && (!X86_VOYAGER && SMP)
The dependencies do now say that KEXEC is only offered for machines that are
_both_ non-Voyager and SMP.
Is the problem you wanted to express that a non-SMP Voyager config didn't
compile?
Whoops, that should be
depends on EXPERIMENTAL && !(X86_VOYAGER && SMP)
Voyager SMP builds don't compile with kexec(), and it isn't clear how to
shootdown CPUs using NMIs without an APIC.
Well unless you need the crash dump functionality you don't need to
shot down CPUs using NMIs.

So I expect machine_crash_shutdown or at least a part of it
should be a call into the subarchitecture code. Having
it be a noop on voyager would be perfectly fine.


Eric
Adrian Bunk
2006-04-06 22:37:16 UTC
Permalink
Post by Zachary Amsden
...
Voyager SMP builds don't compile with kexec(), and it isn't clear how to
shootdown CPUs using NMIs without an APIC.
...
config KEXEC
bool "kexec system call (EXPERIMENTAL)"
- depends on EXPERIMENTAL && (!X86_VOYAGER && SMP)
+ depends on EXPERIMENTAL && !(X86_VOYAGER && SMP)
...
Unless James disagrees with me, I'd prefer to let X86_VOYAGER depend on
SMP (the CONFIG_SMP=n case is anyways not compiling), and you could
express this simply as "&& !X86_VOYAGER".

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
Eric W. Biederman
2006-04-04 17:43:01 UTC
Permalink
Post by Adrian Bunk
Post by Andrew Morton
...
...
+x86-clean-up-subarch-definitions.patch
...
x86 updates.
...
It is.
Post by Adrian Bunk
config KEXEC
bool "kexec system call (EXPERIMENTAL)"
- depends on EXPERIMENTAL
+ depends on EXPERIMENTAL && (!X86_VOYAGER && SMP)
The dependencies do now say that KEXEC is only offered for machines that
are _both_ non-Voyager and SMP.
Is the problem you wanted to express that a non-SMP Voyager config
didn't compile?
AFAIR I recently sent a patch disallowing non-SMP Voyager configurations
that wasn't yet applied.
I think this cleanup patch is even going in the wrong direction. The
subarch code right now is a real pain because it is never clear when
you are calling a function with multiple definitions. Which makes it
really easy to break.

If we are going to refactor this can we please move in the direction
of a machine vector like alpha, ppc, and arm. I don't see the current
this cleanup making it any easier to tell there is code in a subarch.

Eric
Zachary Amsden
2006-04-04 17:51:43 UTC
Permalink
Post by Eric W. Biederman
Post by Adrian Bunk
Post by Andrew Morton
...
...
+x86-clean-up-subarch-definitions.patch
...
x86 updates.
...
It is.
Post by Adrian Bunk
config KEXEC
bool "kexec system call (EXPERIMENTAL)"
- depends on EXPERIMENTAL
+ depends on EXPERIMENTAL && (!X86_VOYAGER && SMP)
The dependencies do now say that KEXEC is only offered for machines that
are _both_ non-Voyager and SMP.
Is the problem you wanted to express that a non-SMP Voyager config
didn't compile?
AFAIR I recently sent a patch disallowing non-SMP Voyager configurations
that wasn't yet applied.
I think this cleanup patch is even going in the wrong direction. The
subarch code right now is a real pain because it is never clear when
you are calling a function with multiple definitions. Which makes it
really easy to break.
If we are going to refactor this can we please move in the direction
of a machine vector like alpha, ppc, and arm. I don't see the current
this cleanup making it any easier to tell there is code in a subarch.
No, this cleanup only eliminates the need to duplicate redundant code.
How does a machine vector make it any harder to break? You still have a
function with multiple definitions. Duplicating code makes things
really easy to break - twice.
Eric W. Biederman
2006-04-04 18:43:26 UTC
Permalink
No, this cleanup only eliminates the need to duplicate redundant code. How
does a machine vector make it any harder to break? You still have a function
with multiple definitions. Duplicating code makes things really easy to break -
twice.
Sharing functions is good, but if you don't share things carefully your code
becomes very brittle because it depends in non obvious ways on other
code.

A machine vector isn't exactly what is needed (although that allows building
all of the subarches at the same time). What is needed are clear places
where the sub architectures get called.

The lack of visibility is what makes subarch code so easy to break right now.

I have had times where I have made a global change and fixed up the entire
kernel and all that broke was the i386 subarchitectures because there
was a dependency but it was totally invisible. Despite testing on and
being most familiar with i386.

For example every other arch only has one implementation
of machine_restart, machine_halt, machine_power_off, and if they
have subarchitectures they have calls to subarch_restart, subarch_halt,
and subarch_power_off. At which point it is trivial to see that
the code lives in a subarchitecture.

Even if that code turns right around and calls a common function on
all but one of the subarchitectures, at least the logic is visible when
you read the code.

If all you are doing is this one little clean up we can probably stop here.
But this looks like a start on getting a vmi or xen subarch working.

If this is really a prelude to introducing more subarchitectures we
need to fix the infrastructure, so it is obvious what is going on.
I would really like to see a machine vector, so we could compile in
multiple subarchitectures at the same time. That makes building
a generic kernel easier, and the requirement that the we need
to build a generic kernel makes the structure of the subarchiteture
hooks hierarchical and you wind up with code whose dependencies
are visible. Instead of the current linker and preprocessor magic.
Functions named hook are impossible to comprehend what they
are supposed to do while reading through the code.

Eric
Zachary Amsden
2006-04-04 19:23:24 UTC
Permalink
Post by Eric W. Biederman
If all you are doing is this one little clean up we can probably stop here.
But this looks like a start on getting a vmi or xen subarch working.
Yes, that is certainly part of the purpose. But the subarch layer
really should be cleaned up, and getting rid of code duplication seems
like a good first step.
Post by Eric W. Biederman
If this is really a prelude to introducing more subarchitectures we
need to fix the infrastructure, so it is obvious what is going on.
I would really like to see a machine vector, so we could compile in
multiple subarchitectures at the same time. That makes building
a generic kernel easier, and the requirement that the we need
to build a generic kernel makes the structure of the subarchiteture
hooks hierarchical and you wind up with code whose dependencies
are visible. Instead of the current linker and preprocessor magic.
Functions named hook are impossible to comprehend what they
are supposed to do while reading through the code.
I see your point. Are you thinking of something like:

struct subarch_hooks subarch_hook_vector = {
.machine_power_off = machine_power_off,
.machine_halt = machine_halt,
.machine_irq_setup = machine_irq_setup,
.machine_subarch_setup = machine_subarch_probe
...
};

And machine_subarch_probe can dynamically change this vector if it
confirms that the platform is appropriate?

This gets rid of both the code duplication and makes it somewhat more
obvious what is going on - plus it is easy to extend to other functions,
and as a bonus feature, you don't need to change any code in other
subarchitectures if you need to add a new hook.


Zach
Muli Ben-Yehuda
2006-04-04 19:38:55 UTC
Permalink
Post by Zachary Amsden
This gets rid of both the code duplication and makes it somewhat more
obvious what is going on - plus it is easy to extend to other functions,
and as a bonus feature, you don't need to change any code in other
subarchitectures if you need to add a new hook.
ppc has 'struct machdep_calls' (asm-powerpc/machdep.h) which serves
the same purpose, and it's indeed a clean way of doing this sort of
thing.

Cheers,
Muli
--
Muli Ben-Yehuda
http://www.mulix.org | http://mulix.livejournal.com/
Andrew Morton
2006-04-04 20:25:46 UTC
Permalink
Post by Zachary Amsden
Post by Eric W. Biederman
If this is really a prelude to introducing more subarchitectures we
need to fix the infrastructure, so it is obvious what is going on.
I would really like to see a machine vector, so we could compile in
multiple subarchitectures at the same time. That makes building
a generic kernel easier, and the requirement that the we need
to build a generic kernel makes the structure of the subarchiteture
hooks hierarchical and you wind up with code whose dependencies
are visible. Instead of the current linker and preprocessor magic.
Functions named hook are impossible to comprehend what they
are supposed to do while reading through the code.
struct subarch_hooks subarch_hook_vector = {
.machine_power_off = machine_power_off,
.machine_halt = machine_halt,
.machine_irq_setup = machine_irq_setup,
.machine_subarch_setup = machine_subarch_probe
...
};
And machine_subarch_probe can dynamically change this vector if it
confirms that the platform is appropriate?
I don't recall anyone expressing any desire for the ability to set these
things at runtime. Unless there is such a requirement I'd suggest that the
best way to address Eric's point is to simply rename the relevant functions
from foo() to subarch_foo().
Zachary Amsden
2006-04-04 22:02:25 UTC
Permalink
Post by Andrew Morton
Post by Zachary Amsden
struct subarch_hooks subarch_hook_vector = {
.machine_power_off = machine_power_off,
.machine_halt = machine_halt,
.machine_irq_setup = machine_irq_setup,
.machine_subarch_setup = machine_subarch_probe
...
};
And machine_subarch_probe can dynamically change this vector if it
confirms that the platform is appropriate?
I don't recall anyone expressing any desire for the ability to set these
things at runtime. Unless there is such a requirement I'd suggest that the
best way to address Eric's point is to simply rename the relevant functions
from foo() to subarch_foo().
Avoiding the runtime assignment isn't possible if you want a generic
subarch that truly can run on multiple different platforms.

I prefer runtime assignment not for this reason, but simply because it
also eliminates two artifacts:

1) You can add new subarch hooks without breaking every other
sub-architecture
2) You don't need #ifdef SUBARCH_FUNC_FOO goo to do this (renaming
voyager_halt -> default)

Zach
Andrew Morton
2006-04-04 22:19:04 UTC
Permalink
Post by Zachary Amsden
Post by Andrew Morton
Post by Zachary Amsden
struct subarch_hooks subarch_hook_vector = {
.machine_power_off = machine_power_off,
.machine_halt = machine_halt,
.machine_irq_setup = machine_irq_setup,
.machine_subarch_setup = machine_subarch_probe
...
};
And machine_subarch_probe can dynamically change this vector if it
confirms that the platform is appropriate?
I don't recall anyone expressing any desire for the ability to set these
things at runtime. Unless there is such a requirement I'd suggest that the
best way to address Eric's point is to simply rename the relevant functions
from foo() to subarch_foo().
Avoiding the runtime assignment isn't possible if you want a generic
subarch that truly can run on multiple different platforms.
Well as I said - I haven't seen any requirement for this expressed. That
doesn't mean that such a requirements doesn't exist, of course.
Post by Zachary Amsden
I prefer runtime assignment not for this reason, but simply because it
1) You can add new subarch hooks without breaking every other
sub-architecture
Is that useful? If you need a new subarch_bar() then

a) Implement it in the subarch which needs it
b) Implement an attribute(weak) stub in a new subarch-stubs.c
c) call it.

That's a little more costly than a static inline stub, but not much. Are
there likely to be any subarch calls which are a) called frequently and b)
not required on some subarchs?
Post by Zachary Amsden
2) You don't need #ifdef SUBARCH_FUNC_FOO goo to do this (renaming
voyager_halt -> default)
Why would one need that? Isn't it simply a matter of renaming
machine_halt() to subarch_machine_halt() everywhere?

I'm just looking for the simplest option here. Eric has identified a code
maintainability problem - it'd be good to fix that, but we shouldn't add
runtime cost/complexity unless we actually gain something from it.
Zachary Amsden
2006-04-04 22:34:17 UTC
Permalink
Post by Andrew Morton
Post by Zachary Amsden
Post by Andrew Morton
Post by Zachary Amsden
struct subarch_hooks subarch_hook_vector = {
.machine_power_off = machine_power_off,
.machine_halt = machine_halt,
.machine_irq_setup = machine_irq_setup,
.machine_subarch_setup = machine_subarch_probe
...
};
And machine_subarch_probe can dynamically change this vector if it
confirms that the platform is appropriate?
I don't recall anyone expressing any desire for the ability to set these
things at runtime. Unless there is such a requirement I'd suggest that the
best way to address Eric's point is to simply rename the relevant functions
from foo() to subarch_foo().
Avoiding the runtime assignment isn't possible if you want a generic
subarch that truly can run on multiple different platforms.
Well as I said - I haven't seen any requirement for this expressed. That
doesn't mean that such a requirements doesn't exist, of course.
Post by Zachary Amsden
I prefer runtime assignment not for this reason, but simply because it
1) You can add new subarch hooks without breaking every other
sub-architecture
Is that useful? If you need a new subarch_bar() then
a) Implement it in the subarch which needs it
b) Implement an attribute(weak) stub in a new subarch-stubs.c
c) call it.
That's a little more costly than a static inline stub, but not much. Are
there likely to be any subarch calls which are a) called frequently and b)
not required on some subarchs?
No, most of these are one time init calls. The problem before was the
default subarch couldn't define weak symbols, since setup.c was in the
subarch itself and not in arch/i386/kernel. Do weak symbols work with
all tool chains?
Post by Andrew Morton
Post by Zachary Amsden
2) You don't need #ifdef SUBARCH_FUNC_FOO goo to do this (renaming
voyager_halt -> default)
Why would one need that? Isn't it simply a matter of renaming
machine_halt() to subarch_machine_halt() everywhere?
No - if you rename machine_halt to subarch_machine_halt, you again can't
add a new subarch interface without changing all subarchitectures. If I
add voyager_smp_bless_voyage(), I now need to add #define
visws_smp_bless_voyage default_smp_bless_voyage, ... or did you mean
subarch_machine_halt literally?
Post by Andrew Morton
I'm just looking for the simplest option here. Eric has identified a code
maintainability problem - it'd be good to fix that, but we shouldn't add
runtime cost/complexity unless we actually gain something from it.
I think weak symbols are the best approach, if they indeed work with all
tool chains.

Zach
Andrew Morton
2006-04-04 22:38:48 UTC
Permalink
Do weak symbols work with all tool chains?
I hope so. `grep weak */*.c'.
Post by Andrew Morton
Post by Zachary Amsden
2) You don't need #ifdef SUBARCH_FUNC_FOO goo to do this (renaming
voyager_halt -> default)
Why would one need that? Isn't it simply a matter of renaming
machine_halt() to subarch_machine_halt() everywhere?
No - if you rename machine_halt to subarch_machine_halt, you again can't
add a new subarch interface without changing all subarchitectures. If I
add voyager_smp_bless_voyage(), I now need to add #define
visws_smp_bless_voyage default_smp_bless_voyage, ... or did you mean
subarch_machine_halt literally?
Yes, I meant subarch_machine_halt() literally.
Post by Andrew Morton
I'm just looking for the simplest option here. Eric has identified a code
maintainability problem - it'd be good to fix that, but we shouldn't add
runtime cost/complexity unless we actually gain something from it.
I think weak symbols are the best approach, if they indeed work with all
tool chains.
They do.
Martin Bligh
2006-04-05 00:21:19 UTC
Permalink
Post by Andrew Morton
Post by Zachary Amsden
Post by Andrew Morton
I don't recall anyone expressing any desire for the ability to set these
things at runtime. Unless there is such a requirement I'd suggest that the
best way to address Eric's point is to simply rename the relevant functions
from foo() to subarch_foo().
Avoiding the runtime assignment isn't possible if you want a generic
subarch that truly can run on multiple different platforms.
Well as I said - I haven't seen any requirement for this expressed. That
doesn't mean that such a requirements doesn't exist, of course.
I think there is a real requirement to do this at boot-time, yes. We
don't want a proliferation of different kernel builds in distros,
but one kernel that installs and boots everywhere (think installer
kernels on boot CDs, etc ... plus testing requirements). Autoswitching,
without magic user-flags. That's what the generic subarch was always
for, and frankly the others all ought to die (apart from possibly really
specialised non-mainstream stuff like voyager and NUMA-Q).

M.
Eric W. Biederman
2006-04-05 02:45:21 UTC
Permalink
Post by Andrew Morton
Post by Andrew Morton
I don't recall anyone expressing any desire for the ability to set these
things at runtime. Unless there is such a requirement I'd suggest that the
best way to address Eric's point is to simply rename the relevant functions
from foo() to subarch_foo().
Avoiding the runtime assignment isn't possible if you want a generic subarch
that truly can run on multiple different platforms.
Well as I said - I haven't seen any requirement for this expressed. That
doesn't mean that such a requirements doesn't exist, of course.
I think there is a real requirement to do this at boot-time, yes. We don't want
a proliferation of different kernel builds in distros,
but one kernel that installs and boots everywhere (think installer
kernels on boot CDs, etc ... plus testing requirements). Autoswitching,
without magic user-flags. That's what the generic subarch was always
for, and frankly the others all ought to die (apart from possibly really
specialised non-mainstream stuff like voyager and NUMA-Q).
Plus it is pretty simple if you compile for a single architecture to
kill the function pointers. That is how the alpha is currently structured.

So for the people who don't want it switchable the code can just compile
out.

Getting this infrastructure now before i386 becomes primarily and embedded
architecture will be nice.

Eric
Adrian Bunk
2006-04-04 16:30:01 UTC
Permalink
Post by Andrew Morton
...
...
+avermedia-6-eyes-avs6eyes-support.patch
...
New v4l driver, and a fix.
...
This patch contains the following fixes:
- no need to hide MODULE_LICENSE() behind an #ifdef
- use module_{init,exit}; this also fixes the bug that bt866_init()
was never called

Signed-off-by: Adrian Bunk <***@stusta.de>

---

drivers/media/video/bt866.c | 15 +++++----------
1 file changed, 5 insertions(+), 10 deletions(-)

--- linux-2.6.17-rc1-mm1-full/drivers/media/video/bt866.c.old 2006-04-04 17:31:55.000000000 +0200
+++ linux-2.6.17-rc1-mm1-full/drivers/media/video/bt866.c 2006-04-04 17:33:38.000000000 +0200
@@ -51,9 +51,7 @@

#include <linux/video_encoder.h>

-#ifdef MODULE_LICENSE
MODULE_LICENSE("GPL");
-#endif

#define DEBUG(x) /* Debug driver */

@@ -372,22 +370,19 @@
/****************************************************************************
* linux kernel module api
****************************************************************************/
-#ifdef MODULE
-int init_module(void)
-#else
-int bt866_init(void)
-#endif
+static int __devinit bt866_init(void)
{
i2c_add_driver(&i2c_driver_bt866);
return 0;
}

-#ifdef MODULE
-void cleanup_module(void)
+static void __devexit bt866_exit(void)
{
i2c_del_driver(&i2c_driver_bt866);
}
-#endif
+
+module_init(bt866_init);
+module_exit(bt866_exit);

/*
* Overrides for Emacs so that we follow Linus's tabbing style.
Martin Samuelsson
2006-04-04 18:32:19 UTC
Permalink
An extension to Adrian's patch:
- Numerous coding style compliance fixes.
- Obscure defines moved to prominent places.
- sizeof(init) replaced by ARRAY_SIZE(init).
- schedule_timeout() replaced by schedule_timeout_interruptible().

Signed-off-by: Martin Samuelsson <***@home.se>

---

bt866.c | 67 +++++++++++++++++++++++++++-------------------------------------
1 files changed, 29 insertions(+), 38 deletions(-)

This should fix all things Andrew pointed out when I first submitted the
avs6eyes driver.

--- linux-2.6.17-rc1-mm1-ab-avs6eyes/drivers/media/video/bt866.c 2006-04-04 22:13:01.000000000 +0200
+++ linux-2.6.16-git15-avs6eyes/drivers/media/video/bt866.c 2006-04-04 22:00:32.000000000 +0200
@@ -51,9 +51,12 @@

#include <linux/video_encoder.h>

+#define BT866_DEVNAME "bt866"
+#define I2C_BT866 0x88
+
MODULE_LICENSE("GPL");

-#define DEBUG(x) /* Debug driver */
+#define DEBUG(x) /* Debug driver */

/* ----------------------------------------------------------------------- */

@@ -70,8 +73,6 @@ struct bt866 {
int sat;
};

-#define I2C_BT866 0x88
-
static int bt866_write(struct bt866 *dev,
unsigned char subaddr, unsigned char data);

@@ -152,9 +153,8 @@ static int bt866_do_command(struct bt866
int i;
u8 val;

- for (i = 0; i < sizeof(init) / 2; i += 2) {
+ for (i = 0; i < ARRAY_SIZE(init) / 2; i += 2)
bt866_write(encoder, init[i], init[i+1]);
- }

val = encoder->reg[0xdc];

@@ -162,7 +162,6 @@ static int bt866_do_command(struct bt866
val |= 0x40; /* CBSWAP */
else
val &= ~0x40; /* !CBSWAP */
- //debug printk("0xdc = 0x%02x\n", val);

bt866_write(encoder, 0xdc, val);

@@ -196,9 +195,8 @@ static int bt866_do_command(struct bt866
encoder->i2c->name, *iarg));

/* not much choice of outputs */
- if (*iarg != 0) {
+ if (*iarg != 0)
return -EINVAL;
- }
}
break;

@@ -236,8 +234,6 @@ static int bt866_do_command(struct bt866
return 0;
}

-#define BT866_DEVNAME "bt866"
-
static int bt866_write(struct bt866 *encoder,
unsigned char subaddr, unsigned char data)
{
@@ -250,19 +246,20 @@ static int bt866_write(struct bt866 *enc
encoder->reg[subaddr] = data;

DEBUG(printk
- ( "%s: write 0x%02X = 0x%02X\n",
+ ("%s: write 0x%02X = 0x%02X\n",
encoder->i2c->name, subaddr, data));

for (err = 0; err < 3;) {
- if (2 == i2c_master_send(encoder->i2c, buffer, 2))
+ if (i2c_master_send(encoder->i2c, buffer, 2) == 2)
break;
err++;
- printk(KERN_WARNING "%s: I/O error #%d (write 0x%02x/0x%02x)\n",
+ printk(KERN_WARNING "%s: I/O error #%d "
+ "(write 0x%02x/0x%02x)\n",
encoder->i2c->name, err, encoder->addr, subaddr);
current->state = TASK_INTERRUPTIBLE;
- schedule_timeout(HZ/10);
+ schedule_timeout_interruptible(HZ/10);
}
- if (3 == err) {
+ if (err == 3) {
printk(KERN_WARNING "%s: giving up\n",
encoder->i2c->name);
return -1;
@@ -273,13 +270,14 @@ static int bt866_write(struct bt866 *enc

static int bt866_attach(struct i2c_adapter *adapter);
static int bt866_detach(struct i2c_client *client);
-static int bt866_command(struct i2c_client *client, unsigned int cmd, void *arg );
+static int bt866_command(struct i2c_client *client,
+ unsigned int cmd, void *arg);


/* Addresses to scan */
-static unsigned short normal_i2c[] = { I2C_BT866>>1, I2C_CLIENT_END };
-static unsigned short probe[2] = { I2C_CLIENT_END, I2C_CLIENT_END };
-static unsigned short ignore[2] = { I2C_CLIENT_END, I2C_CLIENT_END };
+static unsigned short normal_i2c[] = {I2C_BT866>>1, I2C_CLIENT_END};
+static unsigned short probe[2] = {I2C_CLIENT_END, I2C_CLIENT_END};
+static unsigned short ignore[2] = {I2C_CLIENT_END, I2C_CLIENT_END};

static struct i2c_client_address_data addr_data = {
normal_i2c,
@@ -288,12 +286,8 @@ static struct i2c_client_address_data ad
};

static struct i2c_driver i2c_driver_bt866 = {
- .driver = {
- .name = BT866_DEVNAME,
- },
-
+ .driver.name = BT866_DEVNAME,
.id = I2C_DRIVERID_BT866,
-
.attach_adapter = bt866_attach,
.detach_client = bt866_detach,
.command = bt866_command
@@ -302,11 +296,11 @@ static struct i2c_driver i2c_driver_bt86

static struct i2c_client bt866_client_tmpl =
{
- name: "(nil)",
- addr: 0,
- adapter: NULL,
- driver: &i2c_driver_bt866,
- usage_count: 0
+ .name = "(nil)",
+ .addr = 0,
+ .adapter = NULL,
+ .driver = &i2c_driver_bt866,
+ .usage_count = 0
};

static int bt866_found_proc(struct i2c_adapter *adapter,
@@ -316,13 +310,13 @@ static int bt866_found_proc(struct i2c_a
struct i2c_client *client;

client = kzalloc(sizeof(*client), GFP_KERNEL);
- if(client == NULL)
+ if (client == NULL)
return -ENOMEM;
- memcpy( client, &bt866_client_tmpl, sizeof(*client) );
+ memcpy(client, &bt866_client_tmpl, sizeof(*client));

- encoder = kzalloc( sizeof(*encoder), GFP_KERNEL );
- if( encoder == NULL ) {
- kfree( client );
+ encoder = kzalloc(sizeof(*encoder), GFP_KERNEL);
+ if (encoder == NULL) {
+ kfree(client);
return -ENOMEM;
}

@@ -344,7 +338,7 @@ static int bt866_found_proc(struct i2c_a

static int bt866_attach(struct i2c_adapter *adapter)
{
- if ( adapter->id == I2C_HW_B_ZR36067 )
+ if (adapter->id == I2C_HW_B_ZR36067)
return i2c_probe(adapter, &addr_data, bt866_found_proc);
return 0;
}
@@ -367,9 +361,6 @@ static int bt866_command(struct i2c_clie
return bt866_do_command(encoder, cmd, arg);
}

-/****************************************************************************
-* linux kernel module api
-****************************************************************************/
static int __devinit bt866_init(void)
{
i2c_add_driver(&i2c_driver_bt866);
Andrew Morton
2006-04-05 07:42:12 UTC
Permalink
Post by Martin Samuelsson
This should fix all things Andrew pointed out when I first submitted the
avs6eyes driver.
We still have all that #ifdef MODULE stuff at the end of bt866.c.
(Shouldn't it have module_init() and module_exit() handlers?)

All those ".input_mux = 0," lines you added to the struct initialisation
are unneeded - the compiler did that for you.
Adrian Bunk
2006-04-05 09:02:18 UTC
Permalink
Post by Andrew Morton
Post by Martin Samuelsson
This should fix all things Andrew pointed out when I first submitted the
avs6eyes driver.
We still have all that #ifdef MODULE stuff at the end of bt866.c.
(Shouldn't it have module_init() and module_exit() handlers?)
...
This is what my patch does.

Martin's patch does not include my patch, it is on top of it.

You should have seen a merge conflict when merging Martin's patch, and
the reason is that you hadn't applied my patch before his patch.

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
Martin Samuelsson
2006-04-05 13:44:45 UTC
Permalink
On Wed, 5 Apr 2006 00:42:12 -0700
Post by Andrew Morton
Post by Martin Samuelsson
This should fix all things Andrew pointed out when I first submitted the
avs6eyes driver.
We still have all that #ifdef MODULE stuff at the end of bt866.c.
(Shouldn't it have module_init() and module_exit() handlers?)
I'm sorry for the confusion. I simply didn't want to cause a conflict with Adrian's patch, and canceled out those things from my version.
Post by Andrew Morton
All those ".input_mux = 0," lines you added to the struct initialisation
are unneeded - the compiler did that for you.
Ah. I didn't know that. It's noted, and will be used in the future. I added them mostly in an attempt to aim for clarity and completeness.

Regards,
/Sam
Johannes Stezenbach
2006-04-05 08:32:04 UTC
Permalink
Post by Martin Samuelsson
current->state = TASK_INTERRUPTIBLE;
- schedule_timeout(HZ/10);
+ schedule_timeout_interruptible(HZ/10);
schedule_timeout_interruptible() already sets current->state.

Johannes
Martin Samuelsson
2006-04-05 13:54:42 UTC
Permalink
On Wed, 5 Apr 2006 10:32:04 +0200
Post by Johannes Stezenbach
Post by Martin Samuelsson
current->state = TASK_INTERRUPTIBLE;
- schedule_timeout(HZ/10);
+ schedule_timeout_interruptible(HZ/10);
schedule_timeout_interruptible() already sets current->state.
I have much to learn, it seems. I'll trim that off, too. Thanks!

Regards,
/Sam
Adrian Bunk
2006-04-04 16:29:55 UTC
Permalink
Post by Andrew Morton
...
git-acpi.patch
...
git trees.
...
acpi_ns_build_external_path() became global but isn't used outside the
file it's defined in.

Was this accidental or is a usage pending?

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
Adrian Bunk
2006-04-04 16:29:35 UTC
Permalink
Post by Andrew Morton
...
...
+x86-clean-up-subarch-definitions.patch
...
x86 updates.
...
Let's make the kernel smaller for everyone.

Signed-off-by: Adrian Bunk <***@stusta.de>

---

arch/i386/kernel/i8259.c | 13 ++++++++++++-
include/asm-i386/arch_hooks.h | 14 --------------
2 files changed, 12 insertions(+), 15 deletions(-)

--- linux-2.6.17-rc1-mm1-full/include/asm-i386/arch_hooks.h.old 2006-04-04 16:54:15.000000000 +0200
+++ linux-2.6.17-rc1-mm1-full/include/asm-i386/arch_hooks.h 2006-04-04 16:56:14.000000000 +0200
@@ -44,20 +44,6 @@
/* these are the default hooks */

/**
- * pre_intr_init_hook - initialisation prior to setting up interrupt vectors
- *
- * Description:
- * Perform any necessary interrupt initialisation prior to setting up
- * the "ordinary" interrupt call gates. For legacy reasons, the ISA
- * interrupts should be initialised here if the machine emulates a PC
- * in any way.
- **/
-#ifndef pre_intr_init_hook
-#define pre_intr_init_hook() init_ISA_irqs()
-#endif
-
-
-/**
* intr_init_hook - post gate setup interrupt initialisation
*
* Description:
--- linux-2.6.17-rc1-mm1-full/arch/i386/kernel/i8259.c.old 2006-04-04 16:55:01.000000000 +0200
+++ linux-2.6.17-rc1-mm1-full/arch/i386/kernel/i8259.c 2006-04-04 16:56:08.000000000 +0200
@@ -368,7 +368,17 @@
*/
static struct irqaction fpu_irq = { math_error_irq, 0, CPU_MASK_NONE, "fpu", NULL, NULL };

-void __init init_ISA_irqs (void)
+/**
+ * pre_intr_init_hook - initialisation prior to setting up interrupt vectors
+ *
+ * Description:
+ * Perform any necessary interrupt initialisation prior to setting up
+ * the "ordinary" interrupt call gates. For legacy reasons, the ISA
+ * interrupts should be initialised here if the machine emulates a PC
+ * in any way.
+ **/
+#ifndef pre_intr_init_hook
+static void __init pre_intr_init_hook(void)
{
int i;

@@ -395,6 +405,7 @@
}
}
}
+#endif /* pre_intr_init_hook */

void __init init_IRQ(void)
{
Zachary Amsden
2006-04-04 17:16:02 UTC
Permalink
Post by Adrian Bunk
Post by Andrew Morton
...
...
+x86-clean-up-subarch-definitions.patch
...
x86 updates.
...
Let's make the kernel smaller for everyone.
Looks fine.
Adrian Bunk
2006-04-04 16:30:10 UTC
Permalink
Post by Andrew Morton
...
...
+knfsd-locks-flag-nfsv4-owned-locks.patch
...
knfsd updates.
...
This patch makes the needlessly global struct nfsd_posix_mng_ops static.

Signed-off-by: Adrian Bunk <***@stusta.de>

--- linux-2.6.17-rc1-mm1-full/fs/nfsd/nfs4state.c.old 2006-04-04 18:03:25.000000000 +0200
+++ linux-2.6.17-rc1-mm1-full/fs/nfsd/nfs4state.c 2006-04-04 18:03:38.000000000 +0200
@@ -2507,7 +2507,7 @@

/* Hack!: For now, we're defining this just so we can use a pointer to it
* as a unique cookie to identify our (NFSv4's) posix locks. */
-struct lock_manager_operations nfsd_posix_mng_ops = {
+static struct lock_manager_operations nfsd_posix_mng_ops = {
};

static inline void


-------------------------------------------------------
This SF.Net email is sponsored by xPML, a groundbreaking scripting language
that extends applications into web and mobile media. Attend the live webcast
and join the prime developer group breaking into this new coding territory!
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=110944&bid=241720&dat=121642
_______________________________________________
NFS maillist - ***@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/nfs
J. Bruce Fields
2006-04-04 16:50:45 UTC
Permalink
Post by Adrian Bunk
Post by Andrew Morton
...
...
+knfsd-locks-flag-nfsv4-owned-locks.patch
...
knfsd updates.
...
This patch makes the needlessly global struct nfsd_posix_mng_ops static.
Whoops, thanks, looks good.

Do you have a script to look for patches that add non-statics?

--b.
Adrian Bunk
2006-04-04 17:29:13 UTC
Permalink
Post by J. Bruce Fields
Post by Adrian Bunk
Post by Andrew Morton
...
...
+knfsd-locks-flag-nfsv4-owned-locks.patch
...
knfsd updates.
...
This patch makes the needlessly global struct nfsd_posix_mng_ops static.
Whoops, thanks, looks good.
Do you have a script to look for patches that add non-statics?
I diff the output of "make namespacecheck" with the output from the
-mm kernel before.
Post by J. Bruce Fields
--b.
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
Moore, Robert
2006-04-04 20:07:14 UTC
Permalink
This interface is public now because it is used in common/dsrestag.c

This is the problem as I see it:

We have a continuing issue where we have a bunch of ACPI components
(interpreter, table manager, resource manager, etc.) Depending on how
the pieces are put together, various sub-parts of the components are
used or not used, depending on the application. The kernel-level code is
integrated into many different operating systems. Each OS used a
different subset of the available external interfaces. Further, each of
the user-level applications (iASL compiler, AcpiExec utility, etc.) use
different interfaces.

In this case, it is the iASL compiler that is using build_external_path.

We could certainly go so far as to put each public interface into a
separate file, allowing each application (and host OS) to pick and
choose the exact interfaces it needs (and I have seen it done this
way.), but this leads to a huge number of small files and I'm not sure
we want this.

However, attempting to track all of this with extensive #ifdefs, etc. is
awkward and ugly at best and very difficult to maintain.

I'm open to suggestions.

Bob
-----Original Message-----
Sent: Tuesday, April 04, 2006 9:30 AM
To: Andrew Morton; Brown, Len
Subject: 2.6.17-rc1-mm1: why did acpi_ns_build_external_path() become
global?
Post by Andrew Morton
...
git-acpi.patch
...
git trees.
...
acpi_ns_build_external_path() became global but isn't used outside the
file it's defined in.
Was this accidental or is a usage pending?
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
-
To unsubscribe from this list: send the line "unsubscribe linux-acpi"
in
More majordomo info at http://vger.kernel.org/majordomo-info.html
-
To unsubscribe from this list: send the line "unsubscribe linux-acpi" in
the body of a message to ***@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Rafael J. Wysocki
2006-04-04 20:53:05 UTC
Permalink
Post by Andrew Morton
ftp://ftp.kernel.org/pub/linux/kernel/people/akpm/patches/2.6/2.6.17-rc1/2.6.17-rc1-mm1/
With this kernel mlockall(MCL_CURRENT | MCL_FUTURE) returns -EFAULT if called
by root (unconditionally, it seems) on x86_64.

On my box the output of:

#include <sys/mman.h>
#include <stdio.h>
#include <string.h>
#include <errno.h>

int main()
{
int ret;

ret = mlockall(MCL_CURRENT | MCL_FUTURE);
if (ret < 0)
printf("%s\n", strerror(errno));

return 0;
}

is "Bad address", if run by root.

Greetings,
Rafael
Andrew Morton
2006-04-04 22:24:54 UTC
Permalink
Post by Rafael J. Wysocki
Post by Andrew Morton
ftp://ftp.kernel.org/pub/linux/kernel/people/akpm/patches/2.6/2.6.17-rc1/2.6.17-rc1-mm1/
With this kernel mlockall(MCL_CURRENT | MCL_FUTURE) returns -EFAULT if called
by root (unconditionally, it seems) on x86_64.
#include <sys/mman.h>
#include <stdio.h>
#include <string.h>
#include <errno.h>
int main()
{
int ret;
ret = mlockall(MCL_CURRENT | MCL_FUTURE);
if (ret < 0)
printf("%s\n", strerror(errno));
return 0;
}
is "Bad address", if run by root.
Yes, me too. It works OK on x86_32.

It's due to mm-posix-memory-lock.patch. I'd guess that
make_pages_present() is returning -EFAULT for some reason.
Zan Lynx
2006-04-04 21:53:37 UTC
Permalink
Post by Andrew Morton
ftp://ftp.kernel.org/pub/linux/kernel/people/akpm/patches/2.6/2.6.17-rc1/2.6.17-rc1-mm1/
Has anyone seen this yet?

BUG: scheduling while atomic: mii-tool/0x00000001/2968
<c02db7f7> schedule+0x43/0x540
<c02dc617> schedule_timeout+0x7a/0x95
<c011d687> process_timeout+0x0/0x5
<c011e8a4> msleep+0x1a/0x1f
<e09100c9> rhine_disable_linkmon+0x40/0xf1 [via_rhine]
<e09101a6> mdio_read+0x1f/0xab [via_rhine]
<e08c7443> generic_mii_ioctl+0x6c/0x13f [mii]
<e0911900> netdev_ioctl+0x2e/0x4e [via_rhine]
<e09118d2> netdev_ioctl+0x0/0x4e [via_rhine]
<c02890a7> dev_ifsioc+0x3b8/0x3d2
<c0289438> dev_ioctl+0x34e/0x47a
<c027fc63> sock_ioctl+0x0/0x1c0
<c015b694> do_ioctl+0x1c/0x5d
<c015b917> vfs_ioctl+0x242/0x255
<c015b954> sys_ioctl+0x2a/0x42
<c02dd80f> sysenter_past_esp+0x54/0x75

The system also has Intel Ethernet Pro 100 and SiS900 Ethernet
controllers, but only the VIA Rhine driver does this.
--
Zan Lynx <***@acm.org>
Andrew Morton
2006-04-04 22:09:53 UTC
Permalink
Post by Zan Lynx
Post by Andrew Morton
ftp://ftp.kernel.org/pub/linux/kernel/people/akpm/patches/2.6/2.6.17-rc1/2.6.17-rc1-mm1/
Has anyone seen this yet?
BUG: scheduling while atomic: mii-tool/0x00000001/2968
<c02db7f7> schedule+0x43/0x540
<c02dc617> schedule_timeout+0x7a/0x95
<c011d687> process_timeout+0x0/0x5
<c011e8a4> msleep+0x1a/0x1f
<e09100c9> rhine_disable_linkmon+0x40/0xf1 [via_rhine]
<e09101a6> mdio_read+0x1f/0xab [via_rhine]
<e08c7443> generic_mii_ioctl+0x6c/0x13f [mii]
<e0911900> netdev_ioctl+0x2e/0x4e [via_rhine]
<e09118d2> netdev_ioctl+0x0/0x4e [via_rhine]
<c02890a7> dev_ifsioc+0x3b8/0x3d2
<c0289438> dev_ioctl+0x34e/0x47a
<c027fc63> sock_ioctl+0x0/0x1c0
<c015b694> do_ioctl+0x1c/0x5d
<c015b917> vfs_ioctl+0x242/0x255
<c015b954> sys_ioctl+0x2a/0x42
<c02dd80f> sysenter_past_esp+0x54/0x75
The system also has Intel Ethernet Pro 100 and SiS900 Ethernet
controllers, but only the VIA Rhine driver does this.
netdev_ioctl() does spin_lock_irq(), so the mdelay->msleep conversion (now
in mainline) was insufficient.

Someone needs to turn on those nice debugging options...

hmm, according to git-whatchanged, this bug has been in there since October
last year. Weird that it hasn't been spotted before now.
Roger Luethi
2006-04-05 07:01:50 UTC
Permalink
Post by Andrew Morton
Post by Zan Lynx
Has anyone seen this yet?
BUG: scheduling while atomic: mii-tool/0x00000001/2968
<c02db7f7> schedule+0x43/0x540
<c02dc617> schedule_timeout+0x7a/0x95
<c011d687> process_timeout+0x0/0x5
<c011e8a4> msleep+0x1a/0x1f
<e09100c9> rhine_disable_linkmon+0x40/0xf1 [via_rhine]
[...]
Post by Andrew Morton
hmm, according to git-whatchanged, this bug has been in there since October
last year. Weird that it hasn't been spotted before now.
It has been spotted [1] and diagnosed [2] about two weeks ago. I guess the
reason it took so long is that in addition to the debugging options, you
need a kernel that does call the spinlock code _and_ you need to look at
the kernel log even though the behavior is unchanged.

Anyhow, the mdelay to msleep conversion is useful only for a corner case: a
user with Rhine-I hardware (ancient) who needs low latency even while
fiddling with the NIC's media settings. I am not sure it's worth the
trouble. Any suggestions for an elegant solution?

Roger

[1] http://marc.theaimsgroup.com/?l=linux-netdev&m=114321570402396
[2] http://marc.theaimsgroup.com/?l=linux-netdev&m=114349201223976
Andrew Morton
2006-04-05 07:29:09 UTC
Permalink
Post by Roger Luethi
Any suggestions for an elegant solution?
Move the locking down lower, so it just locks the stuff which needs locking?

It all depends on what the lock's role is, and so often that's a big secret.

If the intention is to prevent concurrent execution of mdio_read()
(reasonable) and we really need that 1 msec delay between writing the
registers and reading back the result then we're somewhat screwed. Either
use a sleeping lock to protect that hardware state or go back to using
udelay().
Roger Luethi
2006-04-05 22:01:18 UTC
Permalink
Post by Andrew Morton
Post by Roger Luethi
Any suggestions for an elegant solution?
Move the locking down lower, so it just locks the stuff which needs locking?
It all depends on what the lock's role is, and so often that's a big secret.
Indeed.

Over a dozen drivers now use calls like mii_ethtool_gset or
generic_mii_ioctl, and only one (sis190) makes the call without grabbing
the lock to the driver private data.

Pushing that lock down is not something I'd do lightly.
Post by Andrew Morton
If the intention is to prevent concurrent execution of mdio_read()
(reasonable) and we really need that 1 msec delay between writing the
I'm afraid that delay is not negotiable -- unless we want to go back to
polling the chip for link state changes.

Roger
Luck, Tony
2006-04-04 23:38:51 UTC
Permalink
Post by Andrew Morton
- VGA on ia64 is broken - the screen comes up blank. But ia64 otherwise
seems to work OK. I didn't have time to investigate.
Broken in base 2.6.17-rc1 too :-( VGA comes up and prints a
few messages, and then goes wonky and dies. Comparing
what I _think_ I saw with the dmesg output, it appears to
die here:

[ 6.416740] ACPI: PCI Interrupt 0000:01:00.0[A] -> GSI 18 (level, low) -> IRQ 48
[ 6.708439] e1000: 0000:01:00.0: e1000_probe: (PCI:33MHz:32-bit) 00:03:47:fd:bb:42
[ 6.754439] e1000: eth0: e1000_probe: Intel(R) PRO/1000 Network Connection
[ 6.761547] netconsole: not configured, aborting
[ 6.766195] initcall at 0xa0000001007c4c30: init_netconsole+0x0/0x140(): returned with error code -22
[ 6.775520] Uniform Multi-Platform E-IDE driver Revision: 7.00alpha2

DEAD DEAD DEAD

Should have gone on to say:
[ 6.781924] ide: Assuming 33MHz system bus speed for PIO modes; override with idebus=xx
[ 6.790052] ICH4: IDE controller at PCI slot 0000:00:1f.1
[ 6.795513] PCI: Device 0000:00:1f.1 not available because of resource collisions
[ 6.803100] ACPI: PCI Interrupt 0000:00:1f.1[A]: no GSI
[ 6.808403] ICH4: BIOS configuration fixed.
[ 6.812646] ICH4: chipset revision 1
[ 6.816271] ICH4: not 100% native mode: will probe irqs later


But I might be off by a line or two, the last bit flashed by quite quickly.

I'll start bisecting tomorrow to see when it was broken.

-Tony
-
To unsubscribe from this list: send the line "unsubscribe linux-ia64" in
the body of a message to ***@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Zou Nan hai
2006-04-05 02:05:06 UTC
Permalink
Post by Luck, Tony
Post by Andrew Morton
- VGA on ia64 is broken - the screen comes up blank. But ia64 otherwise
seems to work OK. I didn't have time to investigate.
Broken in base 2.6.17-rc1 too :-( VGA comes up and prints a
few messages, and then goes wonky and dies. Comparing
what I _think_ I saw with the dmesg output, it appears to
The wild VGA comes from the patch which changed ioremap.

Now ioremap would not remap memory to region 6 unless that memory is
marked as EFI_MEMORY_UC by EFI.

Unfortunately on the Tiger Machine, VGA ram base was marked as
EFI_MEMORY_WB instead of EFI_MEMORY_UC...

So you can see the problem disappear if change VGA_MAP_MEM to use
ioremap_nocache.

But I am not quite sure if we can fully trust EFI on this attribute.

Zou Nan hai
-
To unsubscribe from this list: send the line "unsubscribe linux-ia64" in
the body of a message to ***@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Bjorn Helgaas
2006-04-05 16:15:34 UTC
Permalink
Post by Zou Nan hai
Post by Luck, Tony
Post by Andrew Morton
- VGA on ia64 is broken - the screen comes up blank. But ia64 otherwise
seems to work OK. I didn't have time to investigate.
Broken in base 2.6.17-rc1 too :-( VGA comes up and prints a
few messages, and then goes wonky and dies. Comparing
what I _think_ I saw with the dmesg output, it appears to
The wild VGA comes from the patch which changed ioremap.
Now ioremap would not remap memory to region 6 unless that memory is
marked as EFI_MEMORY_UC by EFI.
Unfortunately on the Tiger Machine, VGA ram base was marked as
EFI_MEMORY_WB instead of EFI_MEMORY_UC...
So you can see the problem disappear if change VGA_MAP_MEM to use
ioremap_nocache.
But I am not quite sure if we can fully trust EFI on this attribute.
Huh. Intel firmware used to just not mention the VGA framebuffer
(0xa0000-0xc0000) at all in the EFI memory map. I think that was
clearly a bug. So maybe they fixed that by marking it WB (and
hopefully UC as well).

Tiger might support WB to the VGA framebuffer, but I don't think it
make much sense for us to use it that way. If we did, writes to the
framebuffer would just sit in the cache until forced out by something
else. Probably using WC would be the best, but we don't have a good
interface for that yet.

Tony, if you agree with this analysis and haven't fixed it yet, here's
a trivial patch.


[IA64] always map VGA framebuffer UC, even if it supports WB

EFI on some machines, e.g., Intel Tiger, reports that the VGA framebuffer
supports WB access. ioremap() prefers WB when possible, so it can work
when mapping main memory.

But it doesn't make sense to map a framebuffer WB, because the driver
doesn't flush explicitly, so updates won't make it to the device
immediately.

This is due to Zou Nan hai <***@intel.com>.

Signed-off-by: Bjorn Helgaas <***@hp.com>

Index: work-mm5/include/asm-ia64/vga.h
===================================================================
--- work-mm5.orig/include/asm-ia64/vga.h 2006-01-02 20:21:10.000000000 -0700
+++ work-mm5/include/asm-ia64/vga.h 2006-04-05 09:57:55.000000000 -0600
@@ -17,7 +17,7 @@
extern unsigned long vga_console_iobase;
extern unsigned long vga_console_membase;

-#define VGA_MAP_MEM(x) ((unsigned long) ioremap(vga_console_membase + (x), 0))
+#define VGA_MAP_MEM(x) ((unsigned long) ioremap_nocache(vga_console_membase + (x), 0))

#define vga_readb(x) (*(x))
#define vga_writeb(x,y) (*(y) = (x))
-
To unsubscribe from this list: send the line "unsubscribe linux-ia64" in
the body of a message to ***@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Luck, Tony
2006-04-05 21:17:57 UTC
Permalink
Post by Bjorn Helgaas
Huh. Intel firmware used to just not mention the VGA framebuffer
(0xa0000-0xc0000) at all in the EFI memory map. I think that was
clearly a bug. So maybe they fixed that by marking it WB (and
hopefully UC as well).
Nope ... not fixed (at least not in the f/w that I'm running). The
VGA buffer is still simply not mentioned in the EFI memory map.

The problem looks to come from this code in vgacon.c:

vga_vram_base = VGA_MAP_MEM(vga_vram_base);
vga_vram_end = VGA_MAP_MEM(vga_vram_end);
vga_vram_size = vga_vram_end - vga_vram_base;

vga_vram_base is 0xb8000, and this call gets a UC return of
c0000000000b8000. But vga_vram_end is 0xc0000 ... which is
the address of the start of a block of memory that is both
WB and UC capable. So ioremap() gives us e0000000000c0000
(which means that vga_vram_size is 2000000000008000, surely
the biggest, baddest video card in the history of the world!).

Perhaps the right fix is to subtract 1 from vga_vram_end and pass
that into VGA_MAP_MEM(), and then add the 1 byte back when computing
the size? But I don't know whether that might do something bad on
some other architecture that uses vgacon.c. If this is not
acceptable, then we can fall back and use the Nanhai/Bjorn fix
of using ioremap_nocache().

Signed-off-by: Tony Luck <***@intel.com>

---

diff --git a/drivers/video/console/vgacon.c b/drivers/video/console/vgacon.c
index d5a04b6..4ca9877 100644
--- a/drivers/video/console/vgacon.c
+++ b/drivers/video/console/vgacon.c
@@ -484,8 +484,8 @@ #endif
}

vga_vram_base = VGA_MAP_MEM(vga_vram_base);
- vga_vram_end = VGA_MAP_MEM(vga_vram_end);
- vga_vram_size = vga_vram_end - vga_vram_base;
+ vga_vram_end = VGA_MAP_MEM(vga_vram_end - 1);
+ vga_vram_size = vga_vram_end - vga_vram_base + 1;

/*
* Find out if there is a graphics card present.
Andrew Morton
2006-04-05 21:37:27 UTC
Permalink
Post by Luck, Tony
Post by Bjorn Helgaas
Huh. Intel firmware used to just not mention the VGA framebuffer
(0xa0000-0xc0000) at all in the EFI memory map. I think that was
clearly a bug. So maybe they fixed that by marking it WB (and
hopefully UC as well).
Nope ... not fixed (at least not in the f/w that I'm running). The
VGA buffer is still simply not mentioned in the EFI memory map.
vga_vram_base = VGA_MAP_MEM(vga_vram_base);
vga_vram_end = VGA_MAP_MEM(vga_vram_end);
vga_vram_size = vga_vram_end - vga_vram_base;
vga_vram_base is 0xb8000, and this call gets a UC return of
c0000000000b8000. But vga_vram_end is 0xc0000 ... which is
the address of the start of a block of memory that is both
WB and UC capable.
OK, so it's really an off-by-one error.
Post by Luck, Tony
So ioremap() gives us e0000000000c0000
(which means that vga_vram_size is 2000000000008000, surely
the biggest, baddest video card in the history of the world!).
Perhaps the right fix is to subtract 1 from vga_vram_end and pass
that into VGA_MAP_MEM(), and then add the 1 byte back when computing
the size? But I don't know whether that might do something bad on
some other architecture that uses vgacon.c. If this is not
acceptable, then we can fall back and use the Nanhai/Bjorn fix
of using ioremap_nocache().
---
diff --git a/drivers/video/console/vgacon.c b/drivers/video/console/vgacon.c
index d5a04b6..4ca9877 100644
--- a/drivers/video/console/vgacon.c
+++ b/drivers/video/console/vgacon.c
@@ -484,8 +484,8 @@ #endif
}
vga_vram_base = VGA_MAP_MEM(vga_vram_base);
- vga_vram_end = VGA_MAP_MEM(vga_vram_end);
- vga_vram_size = vga_vram_end - vga_vram_base;
+ vga_vram_end = VGA_MAP_MEM(vga_vram_end - 1);
+ vga_vram_size = vga_vram_end - vga_vram_base + 1;
/*
* Find out if there is a graphics card present.
Looks like the correct fix to me.

Tony (D), can you think of any problems with that approach?
-
To unsubscribe from this list: send the line "unsubscribe linux-ia64" in
the body of a message to ***@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Andreas Schwab
2006-04-05 21:39:16 UTC
Permalink
diff --git a/drivers/video/console/vgacon.c b/drivers/video/console/v=
gacon.c
index d5a04b6..4ca9877 100644
--- a/drivers/video/console/vgacon.c
+++ b/drivers/video/console/vgacon.c
@@ -484,8 +484,8 @@ #endif
}
=20
vga_vram_base =3D VGA_MAP_MEM(vga_vram_base);
- vga_vram_end =3D VGA_MAP_MEM(vga_vram_end);
- vga_vram_size =3D vga_vram_end - vga_vram_base;
+ vga_vram_end =3D VGA_MAP_MEM(vga_vram_end - 1);
+ vga_vram_size =3D vga_vram_end - vga_vram_base + 1;
Better use vga_vram_end =3D VGA_MAP_MEM(vga_vram_end - 1) + 1, or you'l=
l
screw up the other computations using vga_vram_end.

Andreas.

--=20
Andreas Schwab, SuSE Labs, ***@suse.de
SuSE Linux Products GmbH, Maxfeldstra=DFe 5, 90409 N=FCrnberg, Germany
PGP key fingerprint =3D 58CA 54C7 6D53 942B 1756 01D3 44D5 214B 8276 4=
ED5
"And now for something completely different."
-
To unsubscribe from this list: send the line "unsubscribe linux-ia64" i=
n
the body of a message to ***@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Bjorn Helgaas
2006-04-05 22:01:08 UTC
Permalink
Post by Luck, Tony
Post by Bjorn Helgaas
Huh. Intel firmware used to just not mention the VGA framebuffer
(0xa0000-0xc0000) at all in the EFI memory map. I think that was
clearly a bug. So maybe they fixed that by marking it WB (and
hopefully UC as well).
Nope ... not fixed (at least not in the f/w that I'm running). The
VGA buffer is still simply not mentioned in the EFI memory map.
vga_vram_base = VGA_MAP_MEM(vga_vram_base);
vga_vram_end = VGA_MAP_MEM(vga_vram_end);
vga_vram_size = vga_vram_end - vga_vram_base;
vga_vram_base is 0xb8000, and this call gets a UC return of
c0000000000b8000. But vga_vram_end is 0xc0000 ... which is
the address of the start of a block of memory that is both
WB and UC capable. So ioremap() gives us e0000000000c0000
(which means that vga_vram_size is 2000000000008000, surely
the biggest, baddest video card in the history of the world!).
Perhaps the right fix is to subtract 1 from vga_vram_end and pass
that into VGA_MAP_MEM(), and then add the 1 byte back when computing
the size? But I don't know whether that might do something bad on
some other architecture that uses vgacon.c.
I think the VGA_MAP_MEM(vga_vram_end) is just bogus -- it's not
that we need to map the memory starting at vga_vram_end. I think
it would cleaner (though much more intrusive) to do something like
the patch below, which basically just hard-codes (base, size)
rather than (base, end).
Post by Luck, Tony
If this is not
acceptable, then we can fall back and use the Nanhai/Bjorn fix
of using ioremap_nocache().
Regardless of how we solve the vga_vram_size issue, I still think
ioremap_nocache() is more appropriate in this case. A framebuffer
won't work like we expect if it's mapped WB, will it?

(The patch below assumes the ioremap_nocache change precedes it.)


[PATCH] vgacon: make VGA_MAP_MEM take size, remove extra use

VGA_MAP_MEM translates to ioremap() on some architectures. It
makes sense to do this to vga_vram_base, because we're going to
access memory between vga_vram_base and vga_vram_end.

But it doesn't really make sense to map starting at vga_vram_end,
because we aren't going to access memory starting there. On ia64,
which always has to be different, ioremapping vga_vram_end gives
you something completely incompatible with ioremapped vga_vram_start,
so vga_vram_size ends up being nonsense.

As a bonus, we often know the size up front, so we can use ioremap()
correctly, rather than giving it a zero size.

Signed-off-by: Bjorn Helgaas <***@hp.com>

Index: work-mm5/drivers/video/console/vgacon.c
===================================================================
--- work-mm5.orig/drivers/video/console/vgacon.c 2006-04-03 15:04:48.000000000 -0600
+++ work-mm5/drivers/video/console/vgacon.c 2006-04-05 15:48:37.000000000 -0600
@@ -391,7 +391,7 @@
static struct resource ega_console_resource =
{ .name = "ega", .start = 0x3B0, .end = 0x3BF };
vga_video_type = VIDEO_TYPE_EGAM;
- vga_vram_end = 0xb8000;
+ vga_vram_size = 0x8000;
display_desc = "EGA+";
request_resource(&ioport_resource,
&ega_console_resource);
@@ -401,7 +401,7 @@
static struct resource mda2_console_resource =
{ .name = "mda", .start = 0x3BF, .end = 0x3BF };
vga_video_type = VIDEO_TYPE_MDA;
- vga_vram_end = 0xb2000;
+ vga_vram_size = 0x2000;
display_desc = "*MDA";
request_resource(&ioport_resource,
&mda1_console_resource);
@@ -418,7 +418,7 @@
if ((ORIG_VIDEO_EGA_BX & 0xff) != 0x10) {
int i;

- vga_vram_end = 0xc0000;
+ vga_vram_size = 0x8000;

if (!ORIG_VIDEO_ISVGA) {
static struct resource ega_console_resource
@@ -443,7 +443,7 @@
* and COE=1 isn't necessarily a good idea)
*/
vga_vram_base = 0xa0000;
- vga_vram_end = 0xb0000;
+ vga_vram_size = 0x10000;
outb_p(6, VGA_GFX_I);
outb_p(6, VGA_GFX_D);
#endif
@@ -475,7 +475,7 @@
static struct resource cga_console_resource =
{ .name = "cga", .start = 0x3D4, .end = 0x3D5 };
vga_video_type = VIDEO_TYPE_CGA;
- vga_vram_end = 0xba000;
+ vga_vram_size = 0x2000;
display_desc = "*CGA";
request_resource(&ioport_resource,
&cga_console_resource);
@@ -483,9 +483,8 @@
}
}

- vga_vram_base = VGA_MAP_MEM(vga_vram_base);
- vga_vram_end = VGA_MAP_MEM(vga_vram_end);
- vga_vram_size = vga_vram_end - vga_vram_base;
+ vga_vram_base = VGA_MAP_MEM(vga_vram_base, vga_vram_size);
+ vga_vram_end = vga_vram_base + vga_vram_size;

/*
* Find out if there is a graphics card present.
@@ -1020,14 +1019,14 @@
char *charmap;

if (vga_video_type != VIDEO_TYPE_EGAM) {
- charmap = (char *) VGA_MAP_MEM(colourmap);
+ charmap = (char *) VGA_MAP_MEM(colourmap, 0);
beg = 0x0e;
#ifdef VGA_CAN_DO_64KB
if (vga_video_type == VIDEO_TYPE_VGAC)
beg = 0x06;
#endif
} else {
- charmap = (char *) VGA_MAP_MEM(blackwmap);
+ charmap = (char *) VGA_MAP_MEM(blackwmap, 0);
beg = 0x0a;
}

Index: work-mm5/include/asm-alpha/vga.h
===================================================================
--- work-mm5.orig/include/asm-alpha/vga.h 2006-01-02 20:21:10.000000000 -0700
+++ work-mm5/include/asm-alpha/vga.h 2006-04-05 15:42:35.000000000 -0600
@@ -46,6 +46,6 @@
#define vga_readb(a) readb((u8 __iomem *)(a))
#define vga_writeb(v,a) writeb(v, (u8 __iomem *)(a))

-#define VGA_MAP_MEM(x) ((unsigned long) ioremap(x, 0))
+#define VGA_MAP_MEM(x,s) ((unsigned long) ioremap(x, s))

#endif
Index: work-mm5/include/asm-arm/vga.h
===================================================================
--- work-mm5.orig/include/asm-arm/vga.h 2006-01-02 20:21:10.000000000 -0700
+++ work-mm5/include/asm-arm/vga.h 2006-04-05 15:42:21.000000000 -0600
@@ -4,7 +4,7 @@
#include <asm/hardware.h>
#include <asm/io.h>

-#define VGA_MAP_MEM(x) (PCIMEM_BASE + (x))
+#define VGA_MAP_MEM(x,s) (PCIMEM_BASE + (x))

#define vga_readb(x) (*((volatile unsigned char *)x))
#define vga_writeb(x,y) (*((volatile unsigned char *)y) = (x))
Index: work-mm5/include/asm-i386/vga.h
===================================================================
--- work-mm5.orig/include/asm-i386/vga.h 2006-01-02 20:21:10.000000000 -0700
+++ work-mm5/include/asm-i386/vga.h 2006-04-05 15:42:49.000000000 -0600
@@ -12,7 +12,7 @@
* access the videoram directly without any black magic.
*/

-#define VGA_MAP_MEM(x) (unsigned long)phys_to_virt(x)
+#define VGA_MAP_MEM(x,s) (unsigned long)phys_to_virt(x)

#define vga_readb(x) (*(x))
#define vga_writeb(x,y) (*(y) = (x))
Index: work-mm5/include/asm-ia64/vga.h
===================================================================
--- work-mm5.orig/include/asm-ia64/vga.h 2006-04-05 09:57:55.000000000 -0600
+++ work-mm5/include/asm-ia64/vga.h 2006-04-05 15:43:09.000000000 -0600
@@ -17,7 +17,7 @@
extern unsigned long vga_console_iobase;
extern unsigned long vga_console_membase;

-#define VGA_MAP_MEM(x) ((unsigned long) ioremap_nocache(vga_console_membase + (x), 0))
+#define VGA_MAP_MEM(x,s) ((unsigned long) ioremap_nocache(vga_console_membase + (x), s))

#define vga_readb(x) (*(x))
#define vga_writeb(x,y) (*(y) = (x))
Index: work-mm5/include/asm-m32r/vga.h
===================================================================
--- work-mm5.orig/include/asm-m32r/vga.h 2006-01-02 20:21:10.000000000 -0700
+++ work-mm5/include/asm-m32r/vga.h 2006-04-05 15:43:22.000000000 -0600
@@ -14,7 +14,7 @@
* access the videoram directly without any black magic.
*/

-#define VGA_MAP_MEM(x) (unsigned long)phys_to_virt(x)
+#define VGA_MAP_MEM(x,s) (unsigned long)phys_to_virt(x)

#define vga_readb(x) (*(x))
#define vga_writeb(x,y) (*(y) = (x))
Index: work-mm5/include/asm-mips/vga.h
===================================================================
--- work-mm5.orig/include/asm-mips/vga.h 2006-03-23 10:22:17.000000000 -0700
+++ work-mm5/include/asm-mips/vga.h 2006-04-05 15:43:32.000000000 -0600
@@ -13,7 +13,7 @@
* access the videoram directly without any black magic.
*/

-#define VGA_MAP_MEM(x) (0xb0000000L + (unsigned long)(x))
+#define VGA_MAP_MEM(x,s) (0xb0000000L + (unsigned long)(x))

#define vga_readb(x) (*(x))
#define vga_writeb(x,y) (*(y) = (x))
Index: work-mm5/include/asm-powerpc/vga.h
===================================================================
--- work-mm5.orig/include/asm-powerpc/vga.h 2006-01-02 20:21:10.000000000 -0700
+++ work-mm5/include/asm-powerpc/vga.h 2006-04-05 15:43:57.000000000 -0600
@@ -42,9 +42,9 @@
extern unsigned long vgacon_remap_base;

#ifdef __powerpc64__
-#define VGA_MAP_MEM(x) ((unsigned long) ioremap((x), 0))
+#define VGA_MAP_MEM(x,s) ((unsigned long) ioremap((x), s))
#else
-#define VGA_MAP_MEM(x) (x + vgacon_remap_base)
+#define VGA_MAP_MEM(x,s) (x + vgacon_remap_base)
#endif

#define vga_readb(x) (*(x))
Index: work-mm5/include/asm-sparc64/vga.h
===================================================================
--- work-mm5.orig/include/asm-sparc64/vga.h 2006-01-02 20:21:10.000000000 -0700
+++ work-mm5/include/asm-sparc64/vga.h 2006-04-05 15:44:08.000000000 -0600
@@ -28,6 +28,6 @@
return *addr;
}

-#define VGA_MAP_MEM(x) (x)
+#define VGA_MAP_MEM(x,s) (x)

#endif
Index: work-mm5/include/asm-x86_64/vga.h
===================================================================
--- work-mm5.orig/include/asm-x86_64/vga.h 2006-01-02 20:21:10.000000000 -0700
+++ work-mm5/include/asm-x86_64/vga.h 2006-04-05 15:44:18.000000000 -0600
@@ -12,7 +12,7 @@
* access the videoram directly without any black magic.
*/

-#define VGA_MAP_MEM(x) (unsigned long)phys_to_virt(x)
+#define VGA_MAP_MEM(x,s) (unsigned long)phys_to_virt(x)

#define vga_readb(x) (*(x))
#define vga_writeb(x,y) (*(y) = (x))
Index: work-mm5/include/asm-xtensa/vga.h
===================================================================
--- work-mm5.orig/include/asm-xtensa/vga.h 2006-01-02 20:21:10.000000000 -0700
+++ work-mm5/include/asm-xtensa/vga.h 2006-04-05 15:44:31.000000000 -0600
@@ -11,7 +11,7 @@
#ifndef _XTENSA_VGA_H
#define _XTENSA_VGA_H

-#define VGA_MAP_MEM(x) (unsigned long)phys_to_virt(x)
+#define VGA_MAP_MEM(x,s) (unsigned long)phys_to_virt(x)

#define vga_readb(x) (*(x))
#define vga_writeb(x,y) (*(y) = (x))
Index: work-mm5/drivers/video/console/mdacon.c
===================================================================
--- work-mm5.orig/drivers/video/console/mdacon.c 2006-01-02 20:21:10.000000000 -0700
+++ work-mm5/drivers/video/console/mdacon.c 2006-04-05 15:46:33.000000000 -0600
@@ -313,8 +313,8 @@
mda_num_columns = 80;
mda_num_lines = 25;

- mda_vram_base = VGA_MAP_MEM(0xb0000);
mda_vram_len = 0x01000;
+ mda_vram_base = VGA_MAP_MEM(0xb0000, mda_vram_len);

mda_index_port = 0x3b4;
mda_value_port = 0x3b5;
Index: work-mm5/drivers/video/vga16fb.c
===================================================================
--- work-mm5.orig/drivers/video/vga16fb.c 2006-03-23 10:22:16.000000000 -0700
+++ work-mm5/drivers/video/vga16fb.c 2006-04-05 15:49:34.000000000 -0600
@@ -1351,7 +1351,7 @@
}

/* XXX share VGA_FB_PHYS and I/O region with vgacon and others */
- info->screen_base = (void __iomem *)VGA_MAP_MEM(VGA_FB_PHYS);
+ info->screen_base = (void __iomem *)VGA_MAP_MEM(VGA_FB_PHYS, 0);

if (!info->screen_base) {
printk(KERN_ERR "vga16fb: unable to map device\n");
-
To unsubscribe from this list: send the line "unsubscribe linux-ia64" in
the body of a message to ***@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Antonino A. Daplas
2006-04-06 01:49:55 UTC
Permalink
Post by Bjorn Helgaas
[PATCH] vgacon: make VGA_MAP_MEM take size, remove extra use
VGA_MAP_MEM translates to ioremap() on some architectures. It
makes sense to do this to vga_vram_base, because we're going to
access memory between vga_vram_base and vga_vram_end.
But it doesn't really make sense to map starting at vga_vram_end,
because we aren't going to access memory starting there. On ia64,
which always has to be different, ioremapping vga_vram_end gives
you something completely incompatible with ioremapped vga_vram_start,
so vga_vram_size ends up being nonsense.
As a bonus, we often know the size up front, so we can use ioremap()
correctly, rather than giving it a zero size.
I definitely prefer this patch.
Acked-by: Antonino Daplas <***@pol.net>

Tony
-
To unsubscribe from this list: send the line "unsubscribe linux-ia64" in
the body of a message to ***@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Russell King
2006-04-06 10:21:54 UTC
Permalink
Post by Bjorn Helgaas
[PATCH] vgacon: make VGA_MAP_MEM take size, remove extra use
Ah, seems to be what I just suggested...
Post by Bjorn Helgaas
@@ -1020,14 +1019,14 @@
char *charmap;
if (vga_video_type != VIDEO_TYPE_EGAM) {
- charmap = (char *) VGA_MAP_MEM(colourmap);
+ charmap = (char *) VGA_MAP_MEM(colourmap, 0);
Don't like this though - can't we pass a real size here rather than zero?
There seems to be several clues as to the maximum size:

#define cmapsz 8192

if (!vga_font_is_default)
charmap += 4 * cmapsz;

charmap += 2 * cmapsz;
for (i = 0; i < cmapsz; i++)
vga_writeb(arg[i], charmap + i);

so that's about 7 * cmapsz - call that 8 for completeness, which is 64K.
--
Russell King
Linux kernel 2.6 ARM Linux - http://www.arm.linux.org.uk/
maintainer of: 2.6 Serial core
-
To unsubscribe from this list: send the line "unsubscribe linux-ia64" in
the body of a message to ***@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Russell King
2006-04-06 10:34:27 UTC
Permalink
Post by Russell King
Post by Bjorn Helgaas
[PATCH] vgacon: make VGA_MAP_MEM take size, remove extra use
Ah, seems to be what I just suggested...
Post by Bjorn Helgaas
@@ -1020,14 +1019,14 @@
char *charmap;
if (vga_video_type != VIDEO_TYPE_EGAM) {
- charmap = (char *) VGA_MAP_MEM(colourmap);
+ charmap = (char *) VGA_MAP_MEM(colourmap, 0);
Don't like this though - can't we pass a real size here rather than zero?
#define cmapsz 8192
if (!vga_font_is_default)
charmap += 4 * cmapsz;
charmap += 2 * cmapsz;
for (i = 0; i < cmapsz; i++)
vga_writeb(arg[i], charmap + i);
so that's about 7 * cmapsz - call that 8 for completeness, which is 64K.
Oh, and obviously, can we also have a VGA_UNMAP_MEM() macro as well please?
8) IOW, something like this (which I cobbled together from your patch, some
of it by hand-edits - couldn't get it to apply cleanly to current -git.)

diff --git a/drivers/video/console/vgacon.c b/drivers/video/console/vgacon.c
--- a/drivers/video/console/vgacon.c
+++ b/drivers/video/console/vgacon.c
@@ -391,7 +391,7 @@ static const char __init *vgacon_startup
static struct resource ega_console_resource =
{ .name = "ega", .start = 0x3B0, .end = 0x3BF };
vga_video_type = VIDEO_TYPE_EGAM;
- vga_vram_end = 0xb8000;
+ vga_vram_size = 0x8000;
display_desc = "EGA+";
request_resource(&ioport_resource,
&ega_console_resource);
@@ -401,7 +401,7 @@ static const char __init *vgacon_startup
static struct resource mda2_console_resource =
{ .name = "mda", .start = 0x3BF, .end = 0x3BF };
vga_video_type = VIDEO_TYPE_MDA;
- vga_vram_end = 0xb2000;
+ vga_vram_size = 0x2000;
display_desc = "*MDA";
request_resource(&ioport_resource,
&mda1_console_resource);
@@ -418,7 +418,7 @@ static const char __init *vgacon_startup
if ((ORIG_VIDEO_EGA_BX & 0xff) != 0x10) {
int i;

- vga_vram_end = 0xc0000;
+ vga_vram_size = 0x8000;

if (!ORIG_VIDEO_ISVGA) {
static struct resource ega_console_resource
@@ -443,7 +443,7 @@ static const char __init *vgacon_startup
* and COE=1 isn't necessarily a good idea)
*/
vga_vram_base = 0xa0000;
- vga_vram_end = 0xb0000;
+ vga_vram_size = 0x10000;
outb_p(6, VGA_GFX_I);
outb_p(6, VGA_GFX_D);
#endif
@@ -475,7 +475,7 @@ static const char __init *vgacon_startup
static struct resource cga_console_resource =
{ .name = "cga", .start = 0x3D4, .end = 0x3D5 };
vga_video_type = VIDEO_TYPE_CGA;
- vga_vram_end = 0xba000;
+ vga_vram_size = 0x2000;
display_desc = "*CGA";
request_resource(&ioport_resource,
&cga_console_resource);
@@ -483,9 +483,8 @@ static const char __init *vgacon_startup
}
}

- vga_vram_base = VGA_MAP_MEM(vga_vram_base);
- vga_vram_end = VGA_MAP_MEM(vga_vram_end);
- vga_vram_size = vga_vram_end - vga_vram_base;
+ vga_vram_base = VGA_MAP_MEM(vga_vram_base, vga_vram_size);
+ vga_vram_end = vga_vram_base + vga_vram_size;

/*
* Find out if there is a graphics card present.
@@ -1020,14 +1019,14 @@ static int vgacon_do_font_op(struct vgas
char *charmap;

if (vga_video_type != VIDEO_TYPE_EGAM) {
- charmap = (char *) VGA_MAP_MEM(colourmap);
+ charmap = (char *) VGA_MAP_MEM(colourmap, 8 * cmapsz);
beg = 0x0e;
#ifdef VGA_CAN_DO_64KB
if (vga_video_type == VIDEO_TYPE_VGAC)
beg = 0x06;
#endif
} else {
- charmap = (char *) VGA_MAP_MEM(blackwmap);
+ charmap = (char *) VGA_MAP_MEM(blackwmap, 8 * cmapsz);
beg = 0x0a;
}

@@ -1102,6 +1101,8 @@ static int vgacon_do_font_op(struct vgas
}
}

+ VGA_UNMAP_MEM(charmap, 8 * cmapsz);
+
spin_lock_irq(&vga_lock);
/* First, the sequencer, Synchronous reset */
vga_wseq(state->vgabase, VGA_SEQ_RESET, 0x01);
Index: work-mm5/include/asm-alpha/vga.h
===================================================================
--- work-mm5.orig/include/asm-alpha/vga.h 2006-01-02 20:21:10.000000000 -0700
+++ work-mm5/include/asm-alpha/vga.h 2006-04-05 15:42:35.000000000 -0600
@@ -46,6 +46,7 @@
#define vga_readb(a) readb((u8 __iomem *)(a))
#define vga_writeb(v,a) writeb(v, (u8 __iomem *)(a))

-#define VGA_MAP_MEM(x) ((unsigned long) ioremap(x, 0))
+#define VGA_MAP_MEM(x,s) ((unsigned long) ioremap(x, s))
+#define VGA_UNMAP_MEM(x,s) do { } while (0)

#endif
Index: work-mm5/include/asm-arm/vga.h
===================================================================
--- work-mm5.orig/include/asm-arm/vga.h 2006-01-02 20:21:10.000000000 -0700
+++ work-mm5/include/asm-arm/vga.h 2006-04-05 15:42:21.000000000 -0600
@@ -4,7 +4,8 @@
#include <asm/hardware.h>
#include <asm/io.h>

-#define VGA_MAP_MEM(x) (PCIMEM_BASE + (x))
+#define VGA_MAP_MEM(x,s) (PCIMEM_BASE + (x))
+#define VGA_UNMAP_MEM(x,s) do { } while (0)

#define vga_readb(x) (*((volatile unsigned char *)x))
#define vga_writeb(x,y) (*((volatile unsigned char *)y) = (x))
Index: work-mm5/include/asm-i386/vga.h
===================================================================
--- work-mm5.orig/include/asm-i386/vga.h 2006-01-02 20:21:10.000000000 -0700
+++ work-mm5/include/asm-i386/vga.h 2006-04-05 15:42:49.000000000 -0600
@@ -12,7 +12,8 @@
* access the videoram directly without any black magic.
*/

-#define VGA_MAP_MEM(x) (unsigned long)phys_to_virt(x)
+#define VGA_MAP_MEM(x,s) (unsigned long)phys_to_virt(x)
+#define VGA_UNMAP_MEM(x,s) do { } while (0)

#define vga_readb(x) (*(x))
#define vga_writeb(x,y) (*(y) = (x))
Index: work-mm5/include/asm-ia64/vga.h
===================================================================
--- work-mm5.orig/include/asm-ia64/vga.h 2006-04-05 09:57:55.000000000 -0600
+++ work-mm5/include/asm-ia64/vga.h 2006-04-05 15:43:09.000000000 -0600
@@ -17,7 +17,8 @@
extern unsigned long vga_console_iobase;
extern unsigned long vga_console_membase;

-#define VGA_MAP_MEM(x) ((unsigned long) ioremap_nocache(vga_console_membase + (x), 0))
+#define VGA_MAP_MEM(x,s) ((unsigned long) ioremap_nocache(vga_console_membase + (x), s))
+#define VGA_UNMAP_MEM(x,s) do { } while (0)

#define vga_readb(x) (*(x))
#define vga_writeb(x,y) (*(y) = (x))
Index: work-mm5/include/asm-m32r/vga.h
===================================================================
--- work-mm5.orig/include/asm-m32r/vga.h 2006-01-02 20:21:10.000000000 -0700
+++ work-mm5/include/asm-m32r/vga.h 2006-04-05 15:43:22.000000000 -0600
@@ -14,7 +14,8 @@
* access the videoram directly without any black magic.
*/

-#define VGA_MAP_MEM(x) (unsigned long)phys_to_virt(x)
+#define VGA_MAP_MEM(x,s) (unsigned long)phys_to_virt(x)
+#define VGA_UNMAP_MEM(x,s) do { } while (0)

#define vga_readb(x) (*(x))
#define vga_writeb(x,y) (*(y) = (x))
Index: work-mm5/include/asm-mips/vga.h
===================================================================
--- work-mm5.orig/include/asm-mips/vga.h 2006-03-23 10:22:17.000000000 -0700
+++ work-mm5/include/asm-mips/vga.h 2006-04-05 15:43:32.000000000 -0600
@@ -13,7 +13,8 @@
* access the videoram directly without any black magic.
*/

-#define VGA_MAP_MEM(x) (0xb0000000L + (unsigned long)(x))
+#define VGA_MAP_MEM(x,s) (0xb0000000L + (unsigned long)(x))
+#define VGA_UNMAP_MEM(x,s) do { } while (0)

#define vga_readb(x) (*(x))
#define vga_writeb(x,y) (*(y) = (x))
Index: work-mm5/include/asm-powerpc/vga.h
===================================================================
--- work-mm5.orig/include/asm-powerpc/vga.h 2006-01-02 20:21:10.000000000 -0700
+++ work-mm5/include/asm-powerpc/vga.h 2006-04-05 15:43:57.000000000 -0600
@@ -42,9 +42,11 @@
extern unsigned long vgacon_remap_base;

#ifdef __powerpc64__
-#define VGA_MAP_MEM(x) ((unsigned long) ioremap((x), 0))
+#define VGA_MAP_MEM(x,s) ((unsigned long) ioremap((x), s))
+#define VGA_UNMAP_MEM(x,s) do { } while (0)
#else
-#define VGA_MAP_MEM(x) (x + vgacon_remap_base)
+#define VGA_MAP_MEM(x,s) (x + vgacon_remap_base)
+#define VGA_UNMAP_MEM(x,s) do { } while (0)
#endif

#define vga_readb(x) (*(x))
Index: work-mm5/include/asm-sparc64/vga.h
===================================================================
--- work-mm5.orig/include/asm-sparc64/vga.h 2006-01-02 20:21:10.000000000 -0700
+++ work-mm5/include/asm-sparc64/vga.h 2006-04-05 15:44:08.000000000 -0600
@@ -28,6 +28,7 @@
return *addr;
}

-#define VGA_MAP_MEM(x) (x)
+#define VGA_MAP_MEM(x,s) (x)
+#define VGA_UNMAP_MEM(x,s) do { } while (0)

#endif
Index: work-mm5/include/asm-x86_64/vga.h
===================================================================
--- work-mm5.orig/include/asm-x86_64/vga.h 2006-01-02 20:21:10.000000000 -0700
+++ work-mm5/include/asm-x86_64/vga.h 2006-04-05 15:44:18.000000000 -0600
@@ -12,7 +12,8 @@
* access the videoram directly without any black magic.
*/

-#define VGA_MAP_MEM(x) (unsigned long)phys_to_virt(x)
+#define VGA_MAP_MEM(x,s) (unsigned long)phys_to_virt(x)
+#define VGA_UNMAP_MEM(x,s) do { } while (0)

#define vga_readb(x) (*(x))
#define vga_writeb(x,y) (*(y) = (x))
Index: work-mm5/include/asm-xtensa/vga.h
===================================================================
--- work-mm5.orig/include/asm-xtensa/vga.h 2006-01-02 20:21:10.000000000 -0700
+++ work-mm5/include/asm-xtensa/vga.h 2006-04-05 15:44:31.000000000 -0600
@@ -11,7 +11,8 @@
#ifndef _XTENSA_VGA_H
#define _XTENSA_VGA_H

-#define VGA_MAP_MEM(x) (unsigned long)phys_to_virt(x)
+#define VGA_MAP_MEM(x,s) (unsigned long)phys_to_virt(x)
+#define VGA_UNMAP_MEM(x,s) do { } while (0)

#define vga_readb(x) (*(x))
#define vga_writeb(x,y) (*(y) = (x))
Index: work-mm5/drivers/video/console/mdacon.c
===================================================================
--- work-mm5.orig/drivers/video/console/mdacon.c 2006-01-02 20:21:10.000000000 -0700
+++ work-mm5/drivers/video/console/mdacon.c 2006-04-05 15:46:33.000000000 -0600
@@ -313,8 +313,8 @@
mda_num_columns = 80;
mda_num_lines = 25;

- mda_vram_base = VGA_MAP_MEM(0xb0000);
mda_vram_len = 0x01000;
+ mda_vram_base = VGA_MAP_MEM(0xb0000, mda_vram_len);

mda_index_port = 0x3b4;
mda_value_port = 0x3b5;
Index: work-mm5/drivers/video/vga16fb.c
===================================================================
--- work-mm5.orig/drivers/video/vga16fb.c 2006-03-23 10:22:16.000000000 -0700
+++ work-mm5/drivers/video/vga16fb.c 2006-04-05 15:49:34.000000000 -0600
@@ -1351,7 +1351,7 @@
}

/* XXX share VGA_FB_PHYS and I/O region with vgacon and others */
- info->screen_base = (void __iomem *)VGA_MAP_MEM(VGA_FB_PHYS);
+ info->screen_base = (void __iomem *)VGA_MAP_MEM(VGA_FB_PHYS, VGA_FB_PHYS_LEN);

if (!info->screen_base) {
printk(KERN_ERR "vga16fb: unable to map device\n");
--
Russell King
Linux kernel 2.6 ARM Linux - http://www.arm.linux.org.uk/
maintainer of: 2.6 Serial core
-
To unsubscribe from this list: send the line "unsubscribe linux-ia64" in
the body of a message to ***@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Bjorn Helgaas
2006-04-06 14:55:27 UTC
Permalink
Post by Russell King
Post by Bjorn Helgaas
[PATCH] vgacon: make VGA_MAP_MEM take size, remove extra use
Ah, seems to be what I just suggested...
Post by Bjorn Helgaas
@@ -1020,14 +1019,14 @@
char *charmap;
if (vga_video_type != VIDEO_TYPE_EGAM) {
- charmap = (char *) VGA_MAP_MEM(colourmap);
+ charmap = (char *) VGA_MAP_MEM(colourmap, 0);
Don't like this though - can't we pass a real size here rather than zero?
I didn't like it either, but was too lazy to work out the actual size,
so I just preserved the previous behavior.

Andrew's put my first patch in -mm already, so I'll put this size
issue and your unmap suggestion on my to-do list.
-
To unsubscribe from this list: send the line "unsubscribe linux-ia64" in
the body of a message to ***@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Russell King
2006-04-06 10:16:05 UTC
Permalink
Post by Luck, Tony
Post by Bjorn Helgaas
Huh. Intel firmware used to just not mention the VGA framebuffer
(0xa0000-0xc0000) at all in the EFI memory map. I think that was
clearly a bug. So maybe they fixed that by marking it WB (and
hopefully UC as well).
Nope ... not fixed (at least not in the f/w that I'm running). The
VGA buffer is still simply not mentioned in the EFI memory map.
vga_vram_base = VGA_MAP_MEM(vga_vram_base);
vga_vram_end = VGA_MAP_MEM(vga_vram_end);
vga_vram_size = vga_vram_end - vga_vram_base;
Wouldn't it be better to do:

vga_vram_size = vga_vram_end - vga_vram_base;
vga_vram_base = VGA_IOREMAP(vga_vram_base, vga_vram_size);
vga_vram_end = vga_vram_base + vga_vram_size;

and for compatibility:

#define VGA_IOREMAP(base,size) VGA_MAP_MEM(base)

?
--
Russell King
Linux kernel 2.6 ARM Linux - http://www.arm.linux.org.uk/
maintainer of: 2.6 Serial core
-
To unsubscribe from this list: send the line "unsubscribe linux-ia64" in
the body of a message to ***@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Zan Lynx
2006-04-05 02:27:16 UTC
Permalink
Post by Andrew Morton
ftp://ftp.kernel.org/pub/linux/kernel/people/akpm/patches/2.6/2.6.17-rc1/2.6.17-rc1-mm1/
While running the following command on a Reiser4 partition mounted on a
MD RAID-0 device on top of two SCSI disks:

find -type f -print0 | xargs -0 cat > /dev/null

I was running this on the local machine, not on the NFS mounts. In
fact, the NFS mounts should have had none, or very little, activity.

I started getting the following error messages. I had been running
2.6.16-rc5-mm3 and hadn't noticed anything like that, so when I get some
time I may try to isolate it, if I can reproduce it.

But, in case anyone else knows the problem, here it is

BUG: unable to handle kernel NULL pointer dereference at virtual address 00000000
printing eip:
e0a9a3d2
*pde = 00000000
Oops: 0002 [#1]
PREEMPT
last sysfs file: /devices/pci0000:00/0000:00:00.0/class
Modules linked in: it87 hwmon_vid hwmon i2c_isa parport_pc lp parport nfsd exportfs lockd autofs4 sunrpc iptable_filter ip_tables x_tables binfmt_misc dm_mod ohci_hcd eepro100 uhci_hcd ehci_hcd sis_agp agpgart i2c_sis630 i2c_core snd_intel8x0 snd_ac97_codec snd_ac97_bus snd_pcm_oss snd_mixer_oss snd_pcm snd_timer snd soundcore snd_page_alloc e100 via_rhine sis900 mii crc32 ide_cd cdrom usbcore
CPU: 0
EIP: 0060:[<e0a9a3d2>] Not tainted VLI
EFLAGS: 00010246 (2.6.17-rc1-mm1 #1)
EIP is at nfsd_cache_lookup+0x169/0x2e3 [nfsd]
eax: 00000000 ebx: dd05a120 ecx: ddc3da38 edx: 00000000
esi: c815319e edi: 00000000 ebp: 00000007 esp: ddfe1f40
ds: 007b es: 007b ss: 0068
Process nfsd (pid: 2170, threadinfo=ddfe1000 task=c148a070)
Stack: <0>de05d000 ddc3d9e0 00000011 00000003 e0ac9d00 de05d000 e0aa80bc e0aa7fa0
e0aa80bc e0a93049 c51b9014 de05d064 e0aa80bc e0aa7fa0 de05d000 e0ab27d2
0000003d dd7eaba0 e0aa7fa0 de05d040 c51b9014 000186a3 00000007 c51b9008
Call Trace:
<e0a93049> nfsd_dispatch+0x39/0x16f [nfsd] <e0ab27d2> svc_process+0x38c/0x5c6 [sunrpc]
<e0a93509> nfsd+0x19a/0x315 [nfsd] <e0a9336f> nfsd+0x0/0x315 [nfsd]
<c0101005> kernel_thread_helper+0x5/0xb
Code: 28 8b 54 24 0c 89 53 30 8b 53 24 a1 e8 ba 3b c0 89 43 34 89 d0 c1 e8 18 31 d0 8b 54 24 04 83 e0 3f 8d 0c 82 8b 03 8b 53 04 85 c0 <89> 02 74 03 89 50 04 8b 01 85 c0 89 03 74 03 89 58 04 89 19 80
EIP: [<e0a9a3d2>] nfsd_cache_lookup+0x169/0x2e3 [nfsd] SS:ESP 0068:ddfe1f40
<6>note: nfsd[2170] exited with preempt_count 1
BUG: unable to handle kernel NULL pointer dereference at virtual address 00000000
printing eip:
e0a9a3d2
*pde = 00000000
Oops: 0002 [#2]
PREEMPT
last sysfs file: /devices/pci0000:00/0000:00:00.0/class
Modules linked in: it87 hwmon_vid hwmon i2c_isa parport_pc lp parport nfsd exportfs lockd autofs4 sunrpc iptable_filter ip_tables x_tables binfmt_misc dm_mod ohci_hcd eepro100 uhci_hcd ehci_hcd sis_agp agpgart i2c_sis630 i2c_core snd_intel8x0 snd_ac97_codec snd_ac97_bus snd_pcm_oss snd_mixer_oss snd_pcm snd_timer snd soundcore snd_page_alloc e100 via_rhine sis900 mii crc32 ide_cd cdrom usbcore
CPU: 0
EIP: 0060:[<e0a9a3d2>] Not tainted VLI
EFLAGS: 00010246 (2.6.17-rc1-mm1 #1)
EIP is at nfsd_cache_lookup+0x169/0x2e3 [nfsd]
eax: 00000000 ebx: dd05a160 ecx: ddc3da38 edx: 00000000
esi: c815319e edi: 00000000 ebp: 00000007 esp: dff43f40
ds: 007b es: 007b ss: 0068
Process nfsd (pid: 2167, threadinfo=dff43000 task=df599550)
Stack: <0>de7f5000 ddc3d9e0 00000011 00000003 e0ac9d00 de7f5000 e0aa80bc e0aa7fa0
e0aa80bc e0a93049 d7196014 de7f5064 e0aa80bc e0aa7fa0 de7f5000 e0ab27d2
0000003d dd7eaba0 e0aa7fa0 de7f5040 d7196014 000186a3 00000007 d7196008
Call Trace:
<e0a93049> nfsd_dispatch+0x39/0x16f [nfsd] <e0ab27d2> svc_process+0x38c/0x5c6 [sunrpc]
<e0a93509> nfsd+0x19a/0x315 [nfsd] <e0a9336f> nfsd+0x0/0x315 [nfsd]
<c0101005> kernel_thread_helper+0x5/0xb
Code: 28 8b 54 24 0c 89 53 30 8b 53 24 a1 e8 ba 3b c0 89 43 34 89 d0 c1 e8 18 31 d0 8b 54 24 04 83 e0 3f 8d 0c 82 8b 03 8b 53 04 85 c0 <89> 02 74 03 89 50 04 8b 01 85 c0 89 03 74 03 89 58 04 89 19 80
EIP: [<e0a9a3d2>] nfsd_cache_lookup+0x169/0x2e3 [nfsd] SS:ESP 0068:dff43f40
<6>note: nfsd[2167] exited with preempt_count 1
BUG: unable to handle kernel NULL pointer dereference at virtual address 00000000
printing eip:
e0a9a3d2
*pde = 00000000
Oops: 0002 [#3]
PREEMPT
last sysfs file: /devices/pci0000:00/0000:00:00.0/class
Modules linked in: it87 hwmon_vid hwmon i2c_isa parport_pc lp parport nfsd exportfs lockd autofs4 sunrpc iptable_filter ip_tables x_tables binfmt_misc dm_mod ohci_hcd eepro100 uhci_hcd ehci_hcd sis_agp agpgart i2c_sis630 i2c_core snd_intel8x0 snd_ac97_codec snd_ac97_bus snd_pcm_oss snd_mixer_oss snd_pcm snd_timer snd soundcore snd_page_alloc e100 via_rhine sis900 mii crc32 ide_cd cdrom usbcore
CPU: 0
EIP: 0060:[<e0a9a3d2>] Not tainted VLI
EFLAGS: 00010246 (2.6.17-rc1-mm1 #1)
EIP is at nfsd_cache_lookup+0x169/0x2e3 [nfsd]
eax: 00000000 ebx: dd05a1a0 ecx: ddc3da38 edx: 00000000
esi: c815319e edi: 00000000 ebp: 00000007 esp: de555f40
ds: 007b es: 007b ss: 0068
Process nfsd (pid: 2171, threadinfo=de555000 task=dfe23030)
Stack: <0>dd7cda00 ddc3d9e0 00000011 00000003 e0ac9d00 dd7cda00 e0aa80bc e0aa7fa0
e0aa80bc e0a93049 c54bb014 dd7cda64 e0aa80bc e0aa7fa0 dd7cda00 e0ab27d2
0000003d dd7eaba0 e0aa7fa0 dd7cda40 c54bb014 000186a3 00000007 c54bb008
Call Trace:
<e0a93049> nfsd_dispatch+0x39/0x16f [nfsd] <e0ab27d2> svc_process+0x38c/0x5c6 [sunrpc]
<e0a93509> nfsd+0x19a/0x315 [nfsd] <e0a9336f> nfsd+0x0/0x315 [nfsd]
<c0101005> kernel_thread_helper+0x5/0xb
Code: 28 8b 54 24 0c 89 53 30 8b 53 24 a1 e8 ba 3b c0 89 43 34 89 d0 c1 e8 18 31 d0 8b 54 24 04 83 e0 3f 8d 0c 82 8b 03 8b 53 04 85 c0 <89> 02 74 03 89 50 04 8b 01 85 c0 89 03 74 03 89 58 04 89 19 80
EIP: [<e0a9a3d2>] nfsd_cache_lookup+0x169/0x2e3 [nfsd] SS:ESP 0068:de555f40
<6>note: nfsd[2171] exited with preempt_count 1
BUG: unable to handle kernel NULL pointer dereference at virtual address 00000000
printing eip:
e0a9a3d2
*pde = 00000000
Oops: 0002 [#4]
PREEMPT
last sysfs file: /devices/pci0000:00/0000:00:00.0/class
Modules linked in: it87 hwmon_vid hwmon i2c_isa parport_pc lp parport nfsd exportfs lockd autofs4 sunrpc iptable_filter ip_tables x_tables binfmt_misc dm_mod ohci_hcd eepro100 uhci_hcd ehci_hcd sis_agp agpgart i2c_sis630 i2c_core snd_intel8x0 snd_ac97_codec snd_ac97_bus snd_pcm_oss snd_mixer_oss snd_pcm snd_timer snd soundcore snd_page_alloc e100 via_rhine sis900 mii crc32 ide_cd cdrom usbcore
CPU: 0
EIP: 0060:[<e0a9a3d2>] Not tainted VLI
EFLAGS: 00010246 (2.6.17-rc1-mm1 #1)
EIP is at nfsd_cache_lookup+0x169/0x2e3 [nfsd]
eax: 00000000 ebx: dd05a1e0 ecx: ddc3da38 edx: 00000000
esi: c815319e edi: 00000000 ebp: 00000007 esp: ddfccf40
ds: 007b es: 007b ss: 0068
Process nfsd (pid: 2166, threadinfo=ddfcc000 task=df488a90)
Stack: <0>ddfebe00 ddc3d9e0 00000011 00000003 e0ac9d00 ddfebe00 e0aa80bc e0aa7fa0
e0aa80bc e0a93049 c5629014 ddfebe64 e0aa80bc e0aa7fa0 ddfebe00 e0ab27d2
0000003d dd7eaba0 e0aa7fa0 ddfebe40 c5629014 000186a3 00000007 c5629008
Call Trace:
<e0a93049> nfsd_dispatch+0x39/0x16f [nfsd] <e0ab27d2> svc_process+0x38c/0x5c6 [sunrpc]
<e0a93509> nfsd+0x19a/0x315 [nfsd] <e0a9336f> nfsd+0x0/0x315 [nfsd]
<c0101005> kernel_thread_helper+0x5/0xb
Code: 28 8b 54 24 0c 89 53 30 8b 53 24 a1 e8 ba 3b c0 89 43 34 89 d0 c1 e8 18 31 d0 8b 54 24 04 83 e0 3f 8d 0c 82 8b 03 8b 53 04 85 c0 <89> 02 74 03 89 50 04 8b 01 85 c0 89 03 74 03 89 58 04 89 19 80
EIP: [<e0a9a3d2>] nfsd_cache_lookup+0x169/0x2e3 [nfsd] SS:ESP 0068:ddfccf40
<6>note: nfsd[2166] exited with preempt_count 1
BUG: unable to handle kernel NULL pointer dereference at virtual address 00000000
printing eip:
e0a9a3d2
*pde = 00000000
Oops: 0002 [#5]
PREEMPT
last sysfs file: /devices/pci0000:00/0000:00:00.0/class
Modules linked in: it87 hwmon_vid hwmon i2c_isa parport_pc lp parport nfsd exportfs lockd autofs4 sunrpc iptable_filter ip_tables x_tables binfmt_misc dm_mod ohci_hcd eepro100 uhci_hcd ehci_hcd sis_agp agpgart i2c_sis630 i2c_core snd_intel8x0 snd_ac97_codec snd_ac97_bus snd_pcm_oss snd_mixer_oss snd_pcm snd_timer snd soundcore snd_page_alloc e100 via_rhine sis900 mii crc32 ide_cd cdrom usbcore
CPU: 0
EIP: 0060:[<e0a9a3d2>] Not tainted VLI
EFLAGS: 00010246 (2.6.17-rc1-mm1 #1)
EIP is at nfsd_cache_lookup+0x169/0x2e3 [nfsd]
eax: 00000000 ebx: dd05a220 ecx: ddc3da38 edx: 00000000
esi: c815319e edi: 00000000 ebp: 00000007 esp: de477f40
ds: 007b es: 007b ss: 0068
Process nfsd (pid: 2169, threadinfo=de477000 task=debdba70)
Stack: <0>de05d600 ddc3d9e0 00000011 00000003 e0ac9d00 de05d600 e0aa80bc e0aa7fa0
e0aa80bc e0a93049 db0ed014 de05d664 e0aa80bc e0aa7fa0 de05d600 e0ab27d2
0000003d dd7eaba0 e0aa7fa0 de05d640 db0ed014 000186a3 00000007 db0ed008
Call Trace:
<e0a93049> nfsd_dispatch+0x39/0x16f [nfsd] <e0ab27d2> svc_process+0x38c/0x5c6 [sunrpc]
<e0a93509> nfsd+0x19a/0x315 [nfsd] <e0a9336f> nfsd+0x0/0x315 [nfsd]
<c0101005> kernel_thread_helper+0x5/0xb
Code: 28 8b 54 24 0c 89 53 30 8b 53 24 a1 e8 ba 3b c0 89 43 34 89 d0 c1 e8 18 31 d0 8b 54 24 04 83 e0 3f 8d 0c 82 8b 03 8b 53 04 85 c0 <89> 02 74 03 89 50 04 8b 01 85 c0 89 03 74 03 89 58 04 89 19 80
EIP: [<e0a9a3d2>] nfsd_cache_lookup+0x169/0x2e3 [nfsd] SS:ESP 0068:de477f40
<6>note: nfsd[2169] exited with preempt_count 1
BUG: unable to handle kernel NULL pointer dereference at virtual address 00000000
printing eip:
e0a9a3d2
*pde = 00000000
Oops: 0002 [#6]
PREEMPT
last sysfs file: /devices/pci0000:00/0000:00:00.0/class
Modules linked in: it87 hwmon_vid hwmon i2c_isa parport_pc lp parport nfsd exportfs lockd autofs4 sunrpc iptable_filter ip_tables x_tables binfmt_misc dm_mod ohci_hcd eepro100 uhci_hcd ehci_hcd sis_agp agpgart i2c_sis630 i2c_core snd_intel8x0 snd_ac97_codec snd_ac97_bus snd_pcm_oss snd_mixer_oss snd_pcm snd_timer snd soundcore snd_page_alloc e100 via_rhine sis900 mii crc32 ide_cd cdrom usbcore
CPU: 0
EIP: 0060:[<e0a9a3d2>] Not tainted VLI
EFLAGS: 00010246 (2.6.17-rc1-mm1 #1)
EIP is at nfsd_cache_lookup+0x169/0x2e3 [nfsd]
eax: 00000000 ebx: dd05a260 ecx: ddc3da38 edx: 00000000
esi: c815319e edi: 00000000 ebp: 00000007 esp: ddf70f40
ds: 007b es: 007b ss: 0068
Process nfsd (pid: 2165, threadinfo=ddf70000 task=dfe1aa90)
Stack: <0>ddfeb200 ddc3d9e0 00000011 00000003 e0ac9d00 ddfeb200 e0aa80bc e0aa7fa0
e0aa80bc e0a93049 db575014 ddfeb264 e0aa80bc e0aa7fa0 ddfeb200 e0ab27d2
0000003d dd7eaba0 e0aa7fa0 ddfeb240 db575014 000186a3 00000007 db575008
Call Trace:
<e0a93049> nfsd_dispatch+0x39/0x16f [nfsd] <e0ab27d2> svc_process+0x38c/0x5c6 [sunrpc]
<e0a93509> nfsd+0x19a/0x315 [nfsd] <e0a9336f> nfsd+0x0/0x315 [nfsd]
<c0101005> kernel_thread_helper+0x5/0xb
Code: 28 8b 54 24 0c 89 53 30 8b 53 24 a1 e8 ba 3b c0 89 43 34 89 d0 c1 e8 18 31 d0 8b 54 24 04 83 e0 3f 8d 0c 82 8b 03 8b 53 04 85 c0 <89> 02 74 03 89 50 04 8b 01 85 c0 89 03 74 03 89 58 04 89 19 80
EIP: [<e0a9a3d2>] nfsd_cache_lookup+0x169/0x2e3 [nfsd] SS:ESP 0068:ddf70f40
<6>note: nfsd[2165] exited with preempt_count 1
--
Zan Lynx <***@acm.org>
Luck, Tony
2006-04-05 22:50:30 UTC
Permalink
Post by Bjorn Helgaas
I think the VGA_MAP_MEM(vga_vram_end) is just bogus -- it's not
that we need to map the memory starting at vga_vram_end. I think
it would cleaner (though much more intrusive) to do something like
the patch below, which basically just hard-codes (base, size)
rather than (base, end).
This patch works for me on ia64/Tiger

-Tony
-
To unsubscribe from this list: send the line "unsubscribe linux-ia64" in
the body of a message to ***@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Adrian Bunk
2006-04-07 12:27:57 UTC
Permalink
I'm getting the following compile error with CONFIG_ACPI_NUMA=y:

<-- snip -->

...
CC drivers/acpi/numa.o
drivers/acpi/numa.c: In function 'acpi_numa_init':
drivers/acpi/numa.c:231: error: 'NR_NODE_MEMBLKS' undeclared (first use in this function)
drivers/acpi/numa.c:231: error: (Each undeclared identifier is reported only once
drivers/acpi/numa.c:231: error: for each function it appears in.)
make[2]: *** [drivers/acpi/numa.o] 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

-
To unsubscribe from this list: send the line "unsubscribe linux-acpi" in
the body of a message to ***@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Andrew Morton
2006-04-07 20:59:37 UTC
Permalink
Post by Adrian Bunk
<-- snip -->
...
CC drivers/acpi/numa.o
drivers/acpi/numa.c:231: error: 'NR_NODE_MEMBLKS' undeclared (first use in this function)
I'm not quite sure how we managed that, but I guess
unify-pxm_to_node-and-node_to_pxm.patch triggered it?
-
To unsubscribe from this list: send the line "unsubscribe linux-acpi" in
the body of a message to ***@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Adrian Bunk
2006-04-09 15:44:46 UTC
Permalink
Post by Andrew Morton
Post by Adrian Bunk
<-- snip -->
...
CC drivers/acpi/numa.o
drivers/acpi/numa.c:231: error: 'NR_NODE_MEMBLKS' undeclared (first use in this function)
I'm not quite sure how we managed that, but I guess
unify-pxm_to_node-and-node_to_pxm.patch triggered it?
Yes, obviously (I got this error on i386):

config ACPI_NUMA
bool "NUMA support"
depends on NUMA
- depends on (IA64 || X86_64)
+ depends on (X86_32 || IA64 || X86_64)
default y if IA64_GENERIC || IA64_SGI_SN2


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

-
To unsubscribe from this list: send the line "unsubscribe linux-acpi" in
the body of a message to ***@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Randy.Dunlap
2006-04-09 15:57:08 UTC
Permalink
Post by Adrian Bunk
Post by Andrew Morton
Post by Adrian Bunk
<-- snip -->
...
CC drivers/acpi/numa.o
drivers/acpi/numa.c:231: error: 'NR_NODE_MEMBLKS' undeclared (first use in this function)
I'm not quite sure how we managed that, but I guess
unify-pxm_to_node-and-node_to_pxm.patch triggered it?
config ACPI_NUMA
bool "NUMA support"
depends on NUMA
- depends on (IA64 || X86_64)
+ depends on (X86_32 || IA64 || X86_64)
default y if IA64_GENERIC || IA64_SGI_SN2
The new/fixed line should just be
depends on X86 || IA64
since X86_32 and X86_64 both set X86.

---
~Randy
-
To unsubscribe from this list: send the line "unsubscribe linux-acpi" in
the body of a message to ***@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Jim Cromie
2006-04-07 20:58:29 UTC
Permalink
FYI,

as the 2 syslog extracts show;
1. the new kernel is now detecting the buggy TSC on the GEODE-sc1100
2. the bug is apparently correctable by passing 'idle=poll' on kernel
boot-line.

Heres one vendor's bug/erratta description:
http://soekris.com/Issue0003.htm


Apr 7 11:42:01 truck kernel: [ 19.160016] Kernel command line:
console=ttyS0,115200n81 root=/dev/nfs
nfsroot=192.168.42.1:/nfshost/soekris
nfsaddrs=192.168.42.100:192.168.42.1:192.168.42.1:255.255.255.0:soekris:eth0
panic=5 BOOT_IMAGE=vmlinuz-2.6.17-rc1-mm1-sk
Apr 7 11:42:01 truck kernel: [ 24.314851] Time: tsc clocksource has
been installed.
Apr 7 11:42:01 truck kernel: [ 29.977802] TSC appears to be running
slowly. Marking it as unstable
Apr 7 11:42:01 truck kernel: [ 20.460000] Time: pit clocksource has
been installed.


Apr 7 12:35:56 truck kernel: [ 21.562573] Kernel command line:
console=ttyS0,115200n81 root=/dev/nfs
nfsroot=192.168.42.1:/nfshost/soekris
nfsaddrs=192.168.42.100:192.168.42.1:192.168.42.1:255.255.255.0:soekris:eth0
panic=5 idle=poll BOOT_IMAGE=vmlinuz-2.6.17-rc1-mm1-sk
Apr 7 12:35:56 truck kernel: [ 21.563049] using polling idle threads.
Apr 7 12:35:56 truck kernel: [ 28.393469] Time: tsc clocksource has
been installed.


Its nice to see the buggy TSC detector detect, and the work-around work.
Andrew Morton
2006-04-08 00:07:06 UTC
Permalink
Post by Jim Cromie
FYI,
as the 2 syslog extracts show;
1. the new kernel is now detecting the buggy TSC on the GEODE-sc1100
2. the bug is apparently correctable by passing 'idle=poll' on kernel
boot-line.
http://soekris.com/Issue0003.htm
console=ttyS0,115200n81 root=/dev/nfs
nfsroot=192.168.42.1:/nfshost/soekris
nfsaddrs=192.168.42.100:192.168.42.1:192.168.42.1:255.255.255.0:soekris:eth0
panic=5 BOOT_IMAGE=vmlinuz-2.6.17-rc1-mm1-sk
Apr 7 11:42:01 truck kernel: [ 24.314851] Time: tsc clocksource has
been installed.
Apr 7 11:42:01 truck kernel: [ 29.977802] TSC appears to be running
slowly. Marking it as unstable
Apr 7 11:42:01 truck kernel: [ 20.460000] Time: pit clocksource has
been installed.
console=ttyS0,115200n81 root=/dev/nfs
nfsroot=192.168.42.1:/nfshost/soekris
nfsaddrs=192.168.42.100:192.168.42.1:192.168.42.1:255.255.255.0:soekris:eth0
panic=5 idle=poll BOOT_IMAGE=vmlinuz-2.6.17-rc1-mm1-sk
Apr 7 12:35:56 truck kernel: [ 21.563049] using polling idle threads.
Apr 7 12:35:56 truck kernel: [ 28.393469] Time: tsc clocksource has
been installed.
Its nice to see the buggy TSC detector detect, and the work-around work.
hm.

John, does this mean that enable-tsc-for-amd-geode-gx-lx.patch is only safe
to merge after all your time-management patches have gone in?
john stultz
2006-04-08 00:25:44 UTC
Permalink
Post by Andrew Morton
Post by Jim Cromie
as the 2 syslog extracts show;
1. the new kernel is now detecting the buggy TSC on the GEODE-sc1100
2. the bug is apparently correctable by passing 'idle=poll' on kernel
boot-line.
John, does this mean that enable-tsc-for-amd-geode-gx-lx.patch is only safe
to merge after all your time-management patches have gone in?
Hmmm. That would look to be the case from Jim's mail, although I'm not
very familiar with the hardware in question, so I could be wrong.

thanks
-john
Jim Cromie
2006-04-08 01:15:07 UTC
Permalink
Post by Andrew Morton
Post by Jim Cromie
FYI,
as the 2 syslog extracts show;
1. the new kernel is now detecting the buggy TSC on the GEODE-sc1100
2. the bug is apparently correctable by passing 'idle=poll' on kernel
boot-line.
http://soekris.com/Issue0003.htm
console=ttyS0,115200n81 root=/dev/nfs
nfsroot=192.168.42.1:/nfshost/soekris
nfsaddrs=192.168.42.100:192.168.42.1:192.168.42.1:255.255.255.0:soekris:eth0
panic=5 BOOT_IMAGE=vmlinuz-2.6.17-rc1-mm1-sk
Apr 7 11:42:01 truck kernel: [ 24.314851] Time: tsc clocksource has
been installed.
Apr 7 11:42:01 truck kernel: [ 29.977802] TSC appears to be running
slowly. Marking it as unstable
Apr 7 11:42:01 truck kernel: [ 20.460000] Time: pit clocksource has
been installed.
console=ttyS0,115200n81 root=/dev/nfs
nfsroot=192.168.42.1:/nfshost/soekris
nfsaddrs=192.168.42.100:192.168.42.1:192.168.42.1:255.255.255.0:soekris:eth0
panic=5 idle=poll BOOT_IMAGE=vmlinuz-2.6.17-rc1-mm1-sk
Apr 7 12:35:56 truck kernel: [ 21.563049] using polling idle threads.
Apr 7 12:35:56 truck kernel: [ 28.393469] Time: tsc clocksource has
been installed.
Its nice to see the buggy TSC detector detect, and the work-around work.
hm.
John, does this mean that enable-tsc-for-amd-geode-gx-lx.patch is only safe
to merge after all your time-management patches have gone in?
that patch adds only MGEODE_LX, MGEODEGX1 was already there.

per the soekris link:

The net4801 board use a new single chip x86 processor from National
Semiconductor, the SC1100. It is based on the Cyrix GX1 core and the
CS5530 support chip, but has some difference. So far we have identified
the following issues that might need a patch to the operating system:

So, by a narrow reading, the current Kconfig already enables the TSC for
my board.
IOW, the patch doesnt worsen the situation. I dont know whether the bug
affects MGEODE_LX,
but it sounds like it could be a different core, w/o the bug.

The only folks possibly hurt by this patch are those who have
mis-selected MGEODE_LX
when they have a MGEODEGX1, and are currently protected from using the
buggy TSC.

my $.02, keep it in.
Jordan Crouse
2006-04-08 16:43:22 UTC
Permalink
Sorry, I should have cc'ed you on this :(
No worries.
Post by Andrew Morton
Post by Jim Cromie
1. the new kernel is now detecting the buggy TSC on the GEODE-sc1100
2. the bug is apparently correctable by passing 'idle=poll' on kernel
boot-line.
I am assuming that this is the problem where the TSC doesn't roll over
correctly on the SC1100 and SC1200.
Post by Andrew Morton
John, does this mean that enable-tsc-for-amd-geode-gx-lx.patch is only safe
to merge after all your time-management patches have gone in?
If that is the case, then the enable-tsc-for-amd-geode-gx-lx.patch is still
safe since the TSC on the GX and LX doesn't have that same problem.

For what its worth, we used to have a fix for the TSC issue back
in the 2.4 days that looked something like this:

#ifdef CONFIG_GEODE_SC1200
#define rdtsc(low,high) \
__asm__ __volatile__("rdtsc" : "=a" (low), "=d" (high)); \
if ((unsigned long) low > 0xFFFFFFFC) high--
#else
#define rdtsc(low,high) \
__asm__ __volatile__("rdtsc" : "=a" (low), "=d" (high))
#endif

That seemed to work, but it is ugly, and I didn't bring it forward
when we moved to 2.6. Perhaps we can revive it if there are SC1100/SC1200
users who need it.

Thanks,
Jordan
--
Jordan Crouse
Senior Linux Engineer
AMD - Personal Connectivity Solutions Group
<www.amd.com/embeddedprocessors>
Heiko Carstens
2006-04-13 07:39:58 UTC
Permalink
Post by Andrew Morton
+md-dm-reduce-stack-usage-with-stacked-block-devices.patch
MD update.
Any chance to see this merged? I think this one is pending for quite a while.
Andrew Morton
2006-04-13 08:13:03 UTC
Permalink
Post by Heiko Carstens
Post by Andrew Morton
+md-dm-reduce-stack-usage-with-stacked-block-devices.patch
MD update.
Any chance to see this merged? I think this one is pending for quite a while.
Last time I broached it with Alasdair (10 Jan) he said

I can see nothing wrong with this in principle.

For device-mapper at the moment though it's essential that, while the
bio mappings may now get delayed, they still get processed in exactly the
same order as they were passed to generic_make_request().

My main concern is whether the timing changes implicit in this patch
will make the rare data-corrupting races in the existing snapshot code
more likely. (I'm working on a fix for these races, but the unfinished
patch is already several hundred lines long.)

It would be helpful if some people on this mailing list would test this
patch in various scenarios and report back.


yes, it has been around for rather a while.
Alasdair G Kergon
2006-04-14 16:07:56 UTC
Permalink
Post by Heiko Carstens
Post by Andrew Morton
+md-dm-reduce-stack-usage-with-stacked-block-devices.patch
MD update.
Any chance to see this merged? I think this one is pending for quite a while.
I looked at this in detail last month, hoping to get it
out of the way, but unfortunately found that, in dm-crypt:

Instead of the stack growing, we'll be allocating successive
bios out of the *same* mempool and, without the recursion, we'll
have lost the possibility of a later allocation waiting for an
earlier one to be released after its endio.

In other words, I think part of dm-crypt needs re-writing to avoid problems
under memory pressure after this patch is applied. And on the face of it,
__clone_and_map() may suffer from similar problems should a bio need to be
split into several pieces.

Alasdair
--
***@redhat.com
Continue reading on narkive:
Loading...