I just sifted through what we had in rawhide, after noticing that ls *.patch was starting to scroll my terminal (which is never a good sign). In doing so, I found a bunch of patches that weren't applied any more that we forgot to remove, and a bunch that were applied that shouldn't have been. Sifting through the remnants gave me a list that I've added to cvs as the file 'TODO' in devel. Hopefully we can keep this up to date as patches are introduced/removed. At the least it should serve as reasoning for why the hell we're carrying some patches for years (some of the CVS changelogs are pretty crappy, including some from yours truly).
Here's what it looks like today..
Dave
drm-modesetting-i915.patch drm-modesetting-radeon.patch linux-2.6-export-shmem-bits-for-gem.patch Intel/Radeon kernel mode-setting. Won't go upstream for a while.
drm-nouveau.patch Nouveau DRM driver. Won't go upstream until ABI confirmed.
linux-2.6-acpi-clear-wake-status.patch linux-2.6-acpi-video-dos.patch linux-2.6-defaults-acpi-video.patch linux-2.6-input-dell-keyboard-keyup.patch linux-2.6-eeepc-laptop-update.patch mjg59 ACPI/laptop bits. Upstreamable for 2.6.28 ?
linux-2.6-at76.patch linux-2.6-iwlwifi-use-dma_alloc_coherent.patch linux-2.6-wireless.patch linux-2.6-wireless-pending.patch Linville. Wireless bits. Most should be upstream for 2.6.28
linux-2.6-ata-quirk.patch IA64 oddness. Query sent to f-k-l
linux-2.6-build-nonintconfig.patch linux-2.6-debug-nmi-timeout.patch linux-2.6-debug-spinlock-taint.patch linux-2.6-debug-taint-vm.patch linux-2.6-debug-vm-would-have-oomkilled.patch linux-2.6-scsi-cpqarray-set-master.patch linux-2.6-usb-ehci-hcd-respect-nousb.patch Push for 2.6.28
linux-2.6-compile-fixes.patch linux-2.6-hotfixes.patch Empty
linux-2.6-crash-driver.patch Not upstreamable.
linux-2.6-debug-always-inline-kzalloc.patch Sent upstream Sep 25 2008
linux-2.6-debug-sizeof-structs.patch Fedora local debug stuff.
linux-2.6-default-mmf_dump_elf_headers.patch linux-2.6-utrace.patch linux-2.6-x86-tracehook.patch Roland magick & utrace
linux-2.6-defaults-fat-utf8.patch Drop?
linux-2.6-defaults-pci_no_msi.patch linux-2.6-input-kill-stupid-messages.patch linux-2.6-x86-tune-generic.patch Fedora local choices uninteresting to upstream
linux-2.6-e1000e-add-support-for-82567LM-3-and-82567LF-3-ICH10D-parts.patch linux-2.6-e1000e-add-support-for-new-82574L-part.patch linux-2.6-e1000e-add-support-for-the-82567LM-4-device.patch linux-2.6-e1000-ich9.patch linux-2.6-firewire-git-update.patch linux-2.6-netdev-atl2.patch Should go upstream for .28
linux-2.6-efika-not-chrp.patch linux-2.6-g5-therm-shutdown.patch linux-2.6-imac-transparent-bridge.patch linux-2.6-ps3-ehci-iso.patch linux-2.6-ps3-legacy-bootloader-hack.patch linux-2.6-ps3-storage-alias.patch linux-2.6-vio-modalias.patch ppc bits. dwmw2.
linux-2.6-execshield.patch linux-2.6-xen-execshield-add-xen-specific-load_user_cs_desc.patch linux-2.6-xen-execshield-fix-endless-gpf-fault-loop.patch linux-2.6-xen-execshield-only-define-load_user_cs_desc-on-32-bit.patch Not interesting to upstream.
linux-2.6-lirc.patch jarod working on upstreaming
linux-2.6-merge-efifb-imacfb.patch pjones. merge for 2.6.28 ?
linux-2.6-nfs-client-mounts-hang.patch SteveD. Sent ping on Sep 25 to find out status.
linux-2.6-net-silence-noisy-printks.patch linux-2.6-piix3-silence-quirk.patch linux-2.6-quiet-iommu.patch linux-2.6-silence-acpi-blacklist.patch linux-2.6-silence-fbcon-logo.patch linux-2.6-silence-noise.patch Fedora local 'hush' patches.
linux-2.6-selinux-mprotect-checks.patch linux-2.6-sparc-selinux-mprotect-checks.patch Not upstreamable.
linux-2.6-serial-460800.patch Probably not upstreamable. http://marc.theaimsgroup.com/?l=linux-kernel&m=112687270832687&w=2 https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=126403 http://lkml.org/lkml/2006/8/2/208
linux-2.6-squashfs.patch Sigh. Who the hell knows when this will go upstream.
linux-2.6-sysrq-c.patch Que?
Dave Jones (davej@redhat.com) said:
linux-2.6-defaults-fat-utf8.patch Drop?
Isn't this a local choice similar to the later ones?
linux-2.6-net-silence-noisy-printks.patch linux-2.6-piix3-silence-quirk.patch linux-2.6-quiet-iommu.patch linux-2.6-silence-acpi-blacklist.patch linux-2.6-silence-fbcon-logo.patch linux-2.6-silence-noise.patch Fedora local 'hush' patches.
Speaking of 'hush' patches -
... ALSA sound/pci/hda/hda_intel.c:1404: azx_pcm_prepare: bufsize=0x10000, format=0x4011 ALSA sound/pci/hda/hda_codec.c:716: hda_codec_setup_stream: NID=0x2, stream=0x5, channel=0, format=0x4011 ...
Is this a config option, or do we need to patch this stuff out?
Bill
On Thu, Sep 25, 2008 at 03:07:04PM -0400, Bill Nottingham wrote:
Dave Jones (davej@redhat.com) said:
linux-2.6-defaults-fat-utf8.patch Drop?
Isn't this a local choice similar to the later ones?
The problem is this is a "who do we want to screw over" patch. Some people have disks which aren't UTF8, and get crazy moon language instead of their expected charset.
linux-2.6-net-silence-noisy-printks.patch linux-2.6-piix3-silence-quirk.patch linux-2.6-quiet-iommu.patch linux-2.6-silence-acpi-blacklist.patch linux-2.6-silence-fbcon-logo.patch linux-2.6-silence-noise.patch Fedora local 'hush' patches.
Speaking of 'hush' patches -
.. ALSA sound/pci/hda/hda_intel.c:1404: azx_pcm_prepare: bufsize=0x10000, format=0x4011 ALSA sound/pci/hda/hda_codec.c:716: hda_codec_setup_stream: NID=0x2, stream=0x5, channel=0, format=0x4011 ..
Is this a config option, or do we need to patch this stuff out?
Probably one of the many ALSA debug options.
CONFIG_SND_VERBOSE_PRINTK=y CONFIG_SND_DEBUG=y CONFIG_SND_DEBUG_DETECT=y CONFIG_SND_DEBUG_VERBOSE=y CONFIG_SND_PCM_XRUN_DEBUG=y
Dave
On Thu, Sep 25, 2008 at 02:57:22PM -0400, Dave Jones wrote:
linux-2.6-acpi-clear-wake-status.patch
Upstream.
linux-2.6-acpi-video-dos.patch linux-2.6-defaults-acpi-video.patch
These are policy decisions. Probably not going usptream.
linux-2.6-input-dell-keyboard-keyup.patch
Upstream.
linux-2.6-eeepc-laptop-update.patch
Upstream.
On Mon, Sep 29, 2008 at 10:47:05AM -0700, Roland McGrath wrote:
linux-2.6-acpi-video-dos.patch linux-2.6-defaults-acpi-video.patch
These are policy decisions. Probably not going usptream.
Can they be turned into upstream config options?
They're already runtime settable, so it's primarily a convenience thing - they're the sensible defaults given our userland.
On Mon, Sep 29, 2008 at 06:59:00PM +0100, Matthew Garrett wrote:
On Mon, Sep 29, 2008 at 10:47:05AM -0700, Roland McGrath wrote:
linux-2.6-acpi-video-dos.patch linux-2.6-defaults-acpi-video.patch
These are policy decisions. Probably not going usptream.
Can they be turned into upstream config options?
They're already runtime settable, so it's primarily a convenience thing
- they're the sensible defaults given our userland.
I think Rolands point was that if we had for eg CONFIG_ACPI_VIDEO_DEFAULT we'd get to carry one less patch.
Dave
These are policy decisions. Probably not going usptream.
Can they be turned into upstream config options?
They're already runtime settable, so it's primarily a convenience thing
- they're the sensible defaults given our userland.
Can the boot defaults be turned into upstream config options?
If it's worthwhile to patch the kernel to make it boot this way by default, it's worthwhile to get that default without a patch. Fewer patches to keep track of, same defaults for kernel-vanilla rpms, etc.
Thanks, Roland
On Thu, 2008-09-25 at 14:57 -0400, Dave Jones wrote:
linux-2.6-net-silence-noisy-printks.patch linux-2.6-piix3-silence-quirk.patch linux-2.6-quiet-iommu.patch linux-2.6-silence-acpi-blacklist.patch linux-2.6-silence-fbcon-logo.patch linux-2.6-silence-noise.patch Fedora local 'hush' patches.
I did a good number of these (and added another one today). The PIIX3 and ACPI ones are plausible for upstream, I'll resend them.
The IOMMU one is possibly controversial, but I've seriously never seen any machine ever where there's a BIOS option to make those messages go away. I suppose I should fire it at lkml just to generate flames.
I also added one today to shut up the KERNEL ALIVE on amd64. I think notting had a better version of this? I just ripped them out, they could probably be redone in such a way that they honor 'quiet' on the command line but I really can't see the point of having them at all.
- ajax
kernel@lists.fedoraproject.org
-
Adam Jackson -
Bill Nottingham -
Dave Jones -
Matthew Garrett -
Roland McGrath