Hi,
Here's 2.6.10-as3; it's way overdue. Lots of alsa, usb, and ia64 fixes,
as well as numerous others all over the place.
The kernel patches can be grabbed from here:
http://www.acm.cs.rpi.edu/~dilinger/patches/2.6.10/as2/
07dc484313cd97dea357b988ba615994 ChangeLog
7457076a2c54af9f534ed053fb3e42de linux-2.6.10-as3.tar.gz
87ce1ab9312837b4bb35d490abbfba7e patch-2.6.10-as3.gz
Changes from 2.6.10-as2:
2005-02-01 07:48:41 GMT Andres Salomon <[email protected]> patch-90
Summary:
tag 2.6.10-as3
Revision:
linux--dilinger--0--patch-90
modified files:
000-extraversion.patch
2005-02-01 06:45:20 GMT Andres Salomon <[email protected]> patch-89
Summary:
087-ext3_graceful_corruption_fixes.patch
Revision:
linux--dilinger--0--patch-89
[EXT3] This patch makes ext3 handle instances of corruption more gracefully.
It also changes __ext3_journal_forget (which 078-jbd_journal_revoke*.patch
needs).
new files:
.arch-ids/087-ext3_graceful_corruption_fixes.patch.id
087-ext3_graceful_corruption_fixes.patch
2005-01-31 05:35:48 GMT Andres Salomon <[email protected]> patch-88
Summary:
086-i386_cpufreq_powernow_k8_acpi_error.patch
Revision:
linux--dilinger--0--patch-88
[I386] If powernowk8_cpu_init() has a problem after it has initted acpi
stuff, make sure to call the appropriate acpi exit routines before returning.
new files:
.arch-ids/086-i386_cpufreq_powernow_k8_acpi_error.patch.id
086-i386_cpufreq_powernow_k8_acpi_error.patch
2005-01-31 05:21:23 GMT Andres Salomon <[email protected]> patch-87
Summary:
085-ia64_irq_reg_typo.patch
Revision:
linux--dilinger--0--patch-87
[IA64] Fix typo of register name; s/_IA64_REG_AR_SP/_IA64_REG_SP/.
new files:
.arch-ids/085-ia64_irq_reg_typo.patch.id
085-ia64_irq_reg_typo.patch
2005-01-31 05:05:34 GMT Andres Salomon <[email protected]> patch-86
Summary:
084-smp_nmi_watchdog_race.patch
Revision:
linux--dilinger--0--patch-86
[I386/X86-64] Ensure that if a cpu takes too long to come up, the NMI
watchdog check doesn't think it's having problems.
new files:
.arch-ids/084-smp_nmi_watchdog_race.patch.id
084-smp_nmi_watchdog_race.patch
2005-01-31 04:56:59 GMT Andres Salomon <[email protected]> patch-85
Summary:
083-x86_64_switch_mm_context_race.patch
Revision:
linux--dilinger--0--patch-85
[X86-64] Fix race in x86-64 SMP TLB handling, when context switching.
new files:
.arch-ids/083-x86_64_switch_mm_context_race.patch.id
083-x86_64_switch_mm_context_race.patch
2005-01-30 09:18:29 GMT Andres Salomon <[email protected]> patch-84
Summary:
082-ide_it8172_init_return_failure.patch
Revision:
linux--dilinger--0--patch-84
[IDE] it8172_init_one returns 1 instead of -ENODEV in probe function. The
pci subsystem considers that to be a success, when it should be a failure.
new files:
.arch-ids/082-ide_it8172_init_return_failure.patch.id
082-ide_it8172_init_return_failure.patch
2005-01-30 08:23:48 GMT Andres Salomon <[email protected]> patch-83
Summary:
081-ia64_early_sn_setup_nested_loop.patch
Revision:
linux--dilinger--0--patch-83
[IA64] early_sn_setup has a nested loop that uses the same loop variable
as the outer loop. This patch fixes that.
new files:
.arch-ids/081-ia64_early_sn_setup_nested_loop.patch.id
081-ia64_early_sn_setup_nested_loop.patch
2005-01-30 06:18:34 GMT Andres Salomon <[email protected]> patch-82
Summary:
080-kernel_sched_might_sleep_ignore_if_oopsed.patch
Revision:
linux--dilinger--0--patch-82
Make __might_sleep() a no-op if the kernel has oopsed.
new files:
.arch-ids/080-kernel_sched_might_sleep_ignore_if_oopsed.patch.id
080-kernel_sched_might_sleep_ignore_if_oopsed.patch
2005-01-30 06:14:48 GMT Andres Salomon <[email protected]> patch-81
Summary:
079-i386_timer_resume_slowdown.patch
Revision:
linux--dilinger--0--patch-81
[I386] After resume, make time run at normal speed. Thanks to
Daniel Drake <[email protected]> for pointing this one out.
new files:
.arch-ids/079-i386_timer_resume_slowdown.patch.id
079-i386_timer_resume_slowdown.patch
2005-01-30 05:43:06 GMT Andres Salomon <[email protected]> patch-80
Summary:
078-jbd_journal_revoke_graceful_double_delete.patch
Revision:
linux--dilinger--0--patch-80
[JBD] journal_revoke may try to delete a metadata block twice; this patch
allows a graceful recovery, instead of a BUG().
new files:
.arch-ids/078-jbd_journal_revoke_graceful_double_delete.patch.id
078-jbd_journal_revoke_graceful_double_delete.patch
2005-01-30 05:34:58 GMT Andres Salomon <[email protected]> patch-79
Summary:
077-ext3_journal_abort_handle_silence.patch
Revision:
linux--dilinger--0--patch-79
[EXT3] For aborted transaction, only spew errors the first time to avoid
a flood of errors.
new files:
.arch-ids/077-ext3_journal_abort_handle_silence.patch.id
077-ext3_journal_abort_handle_silence.patch
2005-01-30 05:10:58 GMT Andres Salomon <[email protected]> patch-78
Summary:
076-i2c_ali1563_devinit_fixes.patch
Revision:
linux--dilinger--0--patch-78
Fix up probe and exit functions in i2c-ali1563 driver to use proper __devinit
and __devexit attribs.
new files:
.arch-ids/076-i2c_ali1563_devinit_fixes.patch.id
076-i2c_ali1563_devinit_fixes.patch
2005-01-30 03:35:39 GMT Andres Salomon <[email protected]> patch-77
Summary:
075-acpiphp_ibm_module_parm_perm.patch
Revision:
linux--dilinger--0--patch-77
Fix permissions passed to module_param() in IBM pci hotplug driver.
new files:
.arch-ids/075-acpiphp_ibm_module_parm_perm.patch.id
075-acpiphp_ibm_module_parm_perm.patch
2005-01-30 03:23:10 GMT Andres Salomon <[email protected]> patch-76
Summary:
074-usb_makefile_fixes.patch
Revision:
linux--dilinger--0--patch-76
[USB] Update makefile for sl811(-hcd) and dc2xx drivers.
new files:
.arch-ids/074-usb_makefile_fixes.patch.id
074-usb_makefile_fixes.patch
2005-01-30 03:20:23 GMT Andres Salomon <[email protected]> patch-75
Summary:
073-usb_gadget_serial_free_ports.patch
Revision:
linux--dilinger--0--patch-75
[USB] Inside gs_free_ports: don't deref memory that's already been freed.
new files:
.arch-ids/073-usb_gadget_serial_free_ports.patch.id
073-usb_gadget_serial_free_ports.patch
2005-01-30 03:12:33 GMT Andres Salomon <[email protected]> patch-74
Summary:
072-sctp_do_bind_random_port.patch
Revision:
linux--dilinger--0--patch-74
Fix bug within sctp_do_bind(); for ephemeral ports, set bp->port earlier,
so that addr actually gets the correct port (inside sctp_add_bind_addr).
new files:
.arch-ids/072-sctp_do_bind_random_port.patch.id
072-sctp_do_bind_random_port.patch
2005-01-30 02:57:58 GMT Andres Salomon <[email protected]> patch-73
Summary:
071-alsa_creation_order_fixes.patch
Revision:
linux--dilinger--0--patch-73
[ALSA] This fixes the creation order for a bunch of sound drivers. Some
drivers were creating /proc entries and registering callback functions that
required driver structs that had not yet been initialized.
new files:
.arch-ids/071-alsa_creation_order_fixes.patch.id
071-alsa_creation_order_fixes.patch
2005-01-30 02:44:36 GMT Andres Salomon <[email protected]> patch-72
Summary:
070-alsa_rme9652_hdsp_get_autosync.patch
Revision:
linux--dilinger--0--patch-72
[ALSA] rme9652's snd_hdsp_get_autosync_ref() should be getting the value
of hdsp_autosync_ref, not hdsp_pref_sync_ref.
new files:
.arch-ids/070-alsa_rme9652_hdsp_get_autosync.patch.id
070-alsa_rme9652_hdsp_get_autosync.patch
2005-01-30 02:21:58 GMT Andres Salomon <[email protected]> patch-71
Summary:
069-alsa_sscape_upload_firmware_len.patch
Revision:
linux--dilinger--0--patch-71
[ALSA] This patch does two things; the first is a cleanup of calls to
verify_area (replaced w/ access_oks) in the sscape driver that I was
too lazy to strip out; the second is a fixup of the length handling
of firmware in upload_dma_data. If __copy_from_user fails to copy data,
the number of failed bytes wasn't kept track of.
new files:
.arch-ids/069-alsa_sscape_upload_firmware_len.patch.id
069-alsa_sscape_upload_firmware_len.patch
2005-01-30 01:59:47 GMT Andres Salomon <[email protected]> patch-70
Summary:
068-scsi_sd_read_capacity_LBD_bail.patch
Revision:
linux--dilinger--0--patch-70
For scsi disks, if the disk capacity is too large for what's configured
in the kernel, don't even attempt to use the disk.
new files:
.arch-ids/068-scsi_sd_read_capacity_LBD_bail.patch.id
068-scsi_sd_read_capacity_LBD_bail.patch
2005-01-30 01:09:31 GMT Andres Salomon <[email protected]> patch-69
Summary:
067-scsi_gdth_pci_map_sg.patch
Revision:
linux--dilinger--0--patch-69
In gdth driver, drop DMA/page mapping from gdth_copy_internal_data(); it's unnecessary, and buggy.
new files:
.arch-ids/067-scsi_gdth_pci_map_sg.patch.id
067-scsi_gdth_pci_map_sg.patch
2005-01-30 00:08:23 GMT Andres Salomon <[email protected]> patch-68
Summary:
066-ibmvscsi_probe_delay_loop_fix.patch
Revision:
linux--dilinger--0--patch-68
The ibmvscsi driver's probe function could potentially exit the delay loop
before the adapter has finished initializing (hostdata->request_limit == 0),
and thus end up not scanning for drives.
new files:
.arch-ids/066-ibmvscsi_probe_delay_loop_fix.patch.id
066-ibmvscsi_probe_delay_loop_fix.patch
2005-01-27 07:06:37 GMT Andres Salomon <[email protected]> patch-67
Summary:
065-alsa_vx_kcalloc.patch
Revision:
linux--dilinger--0--patch-67
[ALSA] the vx driver allocates enough size for a pointer, when it should
be allocating enough size for a struct vx_core_t.
new files:
.arch-ids/065-alsa_vx_kcalloc.patch.id
065-alsa_vx_kcalloc.patch
2005-01-27 07:03:58 GMT Andres Salomon <[email protected]> patch-66
Summary:
064-alsa_usbaudio_disconnect.patch
Revision:
linux--dilinger--0--patch-66
[ALSA] return -EBADFD when a usbaudio device is disconnected.
new files:
.arch-ids/064-alsa_usbaudio_disconnect.patch.id
064-alsa_usbaudio_disconnect.patch
2005-01-27 07:01:41 GMT Andres Salomon <[email protected]> patch-65
Summary:
063-alsa_opl4_build_fix.patch
Revision:
linux--dilinger--0--patch-65
[ALSA] build snd-opl3-lib when SND_OPL4_LIB is enabled.
new files:
.arch-ids/063-alsa_opl4_build_fix.patch.id
063-alsa_opl4_build_fix.patch
2005-01-27 06:47:48 GMT Andres Salomon <[email protected]> patch-64
Summary:
062-alsa_sscape_user_copy_check.patch
Revision:
linux--dilinger--0--patch-64
[ALSA] the sscape driver does a __copy_to_user without actually checking to
see whether it succeeds. This patch makes it check the result.
new files:
.arch-ids/062-alsa_sscape_user_copy_check.patch.id
062-alsa_sscape_user_copy_check.patch
2005-01-27 06:45:52 GMT Andres Salomon <[email protected]> patch-63
Summary:
061-alsa_es18xx_ifdef_typo.patch
Revision:
linux--dilinger--0--patch-63
[ALSA] es18xx.c also has an ifdef typo; s/CONFIG_PNP_/CONFIG_PNP/.
new files:
.arch-ids/061-alsa_es18xx_ifdef_typo.patch.id
061-alsa_es18xx_ifdef_typo.patch
2005-01-27 06:43:26 GMT Andres Salomon <[email protected]> patch-62
Summary:
060-alsa_cs4231_lib_ifdef_typo.patch
Revision:
linux--dilinger--0--patch-62
[ALSA] cs4231_lib.c has an #ifdef typo; s/SNDRV_DEBUGq_MCE/SNDRV_DEBUG_MCE/.
new files:
.arch-ids/060-alsa_cs4231_lib_ifdef_typo.patch.id
060-alsa_cs4231_lib_ifdef_typo.patch
2005-01-27 06:39:19 GMT Andres Salomon <[email protected]> patch-61
Summary:
059-ia64_unwind_stack_check.patch
Revision:
linux--dilinger--0--patch-61
[IA64] Add a an additional sanity check to ia64's unw_unwind_to_user().
new files:
.arch-ids/059-ia64_unwind_stack_check.patch.id
059-ia64_unwind_stack_check.patch
2005-01-27 06:35:54 GMT Andres Salomon <[email protected]> patch-60
Summary:
058-ia64_binfmt_elf_bug_out.patch
Revision:
linux--dilinger--0--patch-60
[IA64] Returning early from ia64_elf32_init in the case of failures
could be problematic. Instead of doing that, call BUG().
new files:
.arch-ids/058-ia64_binfmt_elf_bug_out.patch.id
058-ia64_binfmt_elf_bug_out.patch
2005-01-27 06:31:58 GMT Andres Salomon <[email protected]> patch-59
Summary:
057-scsi_ioctl_unknown_opcode.patch
Revision:
linux--dilinger--0--patch-59
scsi_ioctl's verify_command() had a check added to print debugging info
about unknown scsi opcodes; however, the check was incorrect. This fixes
it.
new files:
.arch-ids/057-scsi_ioctl_unknown_opcode.patch.id
057-scsi_ioctl_unknown_opcode.patch
2005-01-22 10:11:06 GMT Andres Salomon <[email protected]> patch-58
Summary:
056-x86_64_acpi_do_suspend_lowlevel_arg.patch
Revision:
linux--dilinger--0--patch-58
[x86_64] ACPI's do_suspend_lowlevel for x86_64 was never updated to no longer
take an argument. This patch removes the test checking the (non-existent)
arg.
new files:
.arch-ids/056-x86_64_acpi_do_suspend_lowlevel_arg.patch.id
056-x86_64_acpi_do_suspend_lowlevel_arg.patch
2005-01-22 09:50:13 GMT Andres Salomon <[email protected]> patch-57
Summary:
055-uml_new_thread_race.patch
Revision:
linux--dilinger--0--patch-57
[UML] When creating a new thread, fix potential signal race that could
cause stack corruption.
new files:
.arch-ids/055-uml_new_thread_race.patch.id
055-uml_new_thread_race.patch
2005-01-22 09:34:18 GMT Andres Salomon <[email protected]> patch-56
Summary:
054-ext3_journal_abort_before_panic.patch
Revision:
linux--dilinger--0--patch-56
When an error occurs in ext3, if we're going to panic, do it *after* aborting
the journal.
new files:
.arch-ids/054-ext3_journal_abort_before_panic.patch.id
054-ext3_journal_abort_before_panic.patch
2005-01-22 09:26:18 GMT Andres Salomon <[email protected]> patch-55
Summary:
053-ipmi_unhandled_message_counting.patch
Revision:
linux--dilinger--0--patch-55
This fixes some IPMI issues:
- fix counting of handled and unhandled messages
- decode_dmi should've been handling 16 byte boundaries instead of
16 bit boundaries
new files:
.arch-ids/053-ipmi_unhandled_message_counting.patch.id
053-ipmi_unhandled_message_counting.patch
2005-01-22 09:08:03 GMT Andres Salomon <[email protected]> patch-54
Summary:
052-vfat_valid_longname_proper_return.patch
Revision:
linux--dilinger--0--patch-54
Return proper error values from vfat_valid_longname(), and pass them back
to calling functions (vfat_mkdir, vfat_create).
new files:
.arch-ids/052-vfat_valid_longname_proper_return.patch.id
052-vfat_valid_longname_proper_return.patch
2005-01-22 08:19:23 GMT Andres Salomon <[email protected]> patch-53
Summary:
051-md_sync_page_io_max_vecs.patch
Revision:
linux--dilinger--0--patch-53
Fix oops in md driver; when calling sync_page_io, correctly initialize
a bio, ensuring bi_max_vecs is set.
new files:
.arch-ids/051-md_sync_page_io_max_vecs.patch.id
051-md_sync_page_io_max_vecs.patch
--
Andres Salomon <[email protected]>