Tim Gardner [Tue, 31 Jan 2017 13:49:08 +0000 (06:49 -0700)]
UBUNTU: [Config] CONFIG_NET_DROP_MONITOR=m
<zioproto> hello all. I am new here. I would like some feedback
about CONFIG_NET_DROP_MONITOR=n in the Ubuntu Kernel. It would
be of great help to have it set to module. We use ubuntu for the
openstack network node.
Signed-off-by: Tim Gardner <tim.gardner@canonical.com>
Kamal Mostafa [Tue, 24 Jan 2017 20:05:20 +0000 (12:05 -0800)]
UBUNTU: [debian] derive indep_hdrs_pkg_name from src_pkg_name
This long-standing oversight in our debian rules hardcodes the string "linux"
instead of using the $(src_pkg_name) for just one of the generated .deb package
names: linux-headers-x.x.x-x. Lets fix it in the generic branches
(T,X,Y,Z,unstable) so that we won't have to keep applying this patch to each of
the derivative/custom kernels.
-----8<-----
Ignore: yes
Signed-off-by: Kamal Mostafa <kamal@canonical.com> Signed-off-by: Tim Gardner <tim.gardner@canonical.com>
Manoj Iyer [Tue, 24 Jan 2017 17:05:12 +0000 (11:05 -0600)]
UBUNTU: d-i: initrd needs msm_emac on amberwing platform.
Amberwing systems has an onboard two port nic that uses msm_emac
driver. This module is needed in d-i's initrd so that these nics
can be used to d-i install the system. Tested on the amberwing
system at canonical.
Signed-off-by: Manoj Iyer <manoj.iyer@canonical.com> Signed-off-by: Tim Gardner <tim.gardner@canonical.com>
Seth Forshee [Thu, 19 Jan 2017 22:05:12 +0000 (16:05 -0600)]
UBUNTU: [Config] Fix up s390x config options changed during 4.10 rebase
Fix the following options mistakenly changed during the rebase
from 4.9 to 4.10.
- CONFIG_I2C was selected by CONFIG_SFC_FALCON but should be
disabled because s390x lacks hw support. Revert these and
related options, and enforce the CONFIG_I2C values to prevent
this in the future.
Ming Lei [Sat, 17 Dec 2016 10:49:09 +0000 (18:49 +0800)]
block: relax check on sg gap
BugLink: http://bugs.launchpad.net/bugs/1657539
If the last bvec of the 1st bio and the 1st bvec of the next
bio are physically contigious, and the latter can be merged
to last segment of the 1st bio, we should think they don't
violate sg gap(or virt boundary) limit.
Both Vitaly and Dexuan reported lots of unmergeable small bios
are observed when running mkfs on Hyper-V virtual storage, and
performance becomes quite low. This patch fixes that performance
issue.
The same issue should exist on NVMe, since it sets virt boundary too.
Reported-by: Vitaly Kuznetsov <vkuznets@redhat.com> Reported-by: Dexuan Cui <decui@microsoft.com> Tested-by: Dexuan Cui <decui@microsoft.com> Cc: Keith Busch <keith.busch@intel.com> Signed-off-by: Ming Lei <ming.lei@canonical.com> Signed-off-by: Jens Axboe <axboe@fb.com>
(cherry picked from linux-next commit 729204ef49ec00b788ce23deb9eb922a5769f55d) Signed-off-by: Tim Gardner <tim.gardner@canonical.com>
Tim Gardner [Wed, 18 Jan 2017 18:43:34 +0000 (11:43 -0700)]
UBUNTU: [Config] CONFIG_DEFAULT_IOSCHED=cfq
Hi there,
after several days of running (way too) many tests, I've got some data
to show that it may be a good idea to drop the DEADLINE I/O scheduler
for Zesty and move to CFQ with buffered writeback throttling (WBT) +
WBT_MQ (WBT multi-queu) enabled.
We originally moved to DEADLINE because of the issues with slow I/O (say
to flash drives) causing applications to hang while blocked on the slow
I/O being flushed out. It seems that with the recent 4.10 WBT driver
and (possibly other block driver changes) we see some performance
benefits also with CFQ, namely:
1. Faster boots. On a 8 thread Xeon CPU E3-1275 I'm seeing a reduction
in usertime boots from 33.92s (Deadline) to ~24.5s (CFQ)
There are some places where CFQ + MQ is less performant than CFQ + MQ +
SQ, and vice-versa. However, my general feeling for Zesty is that we
should give this a try as it seems to work well. The config changes are:
This will give us plenty of time to give this a good test in the next
few months and revert them if we find any problematic corner cases.
(The win on boot time, build times and writes to slow devices) is
probably the most compelling choice for these changes IMHO.
Colin King
Signed-off-by: Tim Gardner <tim.gardner@canonical.com>
Seth Forshee [Tue, 17 Jan 2017 21:19:39 +0000 (15:19 -0600)]
UBUNTU: SAUCE: (no-up) i915: Remove MODULE_FIRMWARE statements for unreleased firmware
BugLink: http://bugs.launchpad.net/bugs/1626740
Intel has added MODULE_FIRMWARE statements to i915 which refer to
firmware files that they have not yet pushed out to upstream
linux-firmware. This causes the following warnings when
generating the initrd:
W: Possible missing firmware /lib/firmware/i915/kbl_guc_ver9_14.bin for module i915
W: Possible missing firmware /lib/firmware/i915/bxt_guc_ver8_7.bin for module i915
This firmware is clearly optional, and the warnings have been
generating a lot of confusion for users. Remove the offending
MODULE_FIRMWARE statements until Intel makes these files
available.
Tim Gardner [Wed, 11 Jan 2017 14:12:13 +0000 (07:12 -0700)]
UBUNTU: [Config] linux-source Provides should not be a macro
Addresses a review comment from Adam Conrad:
debian.master/control.stub.in:
- when changing linux-source to SRCPKGNAME-source, you also changed the
Provides, which doesn't make sense. That should change back.
Signed-off-by: Tim Gardner <tim.gardner@canonical.com>
Seth Forshee [Fri, 6 Jan 2017 17:18:41 +0000 (11:18 -0600)]
UBUNTU: [Config] Disble stack protector for powerpc-smp
Fixes FTBFS:
arch/powerpc/platforms/built-in.o: In function `bootx_add_display_props':
/<<PKGBUILDDIR>>/arch/powerpc/platforms/powermac/bootx_init.c:211: undefined reference to `__stack_chk_fail_local'
arch/powerpc/platforms/built-in.o: In function `bootx_scan_dt_build_struct':
/<<PKGBUILDDIR>>/arch/powerpc/platforms/powermac/bootx_init.c:350: undefined reference to `__stack_chk_fail_local'
arch/powerpc/platforms/built-in.o: In function `bootx_init':
/<<PKGBUILDDIR>>/arch/powerpc/platforms/powermac/bootx_init.c:596: undefined reference to `__stack_chk_fail_local'
ld: .tmp_vmlinux1: hidden symbol `__stack_chk_fail_local' isn't defined
ld: final link failed: Bad value
Seth Forshee [Fri, 6 Jan 2017 15:58:11 +0000 (09:58 -0600)]
UBUNTU: [Config] Update and enforce IMA options
BugLink: http://bugs.launchpad.net/bugs/1643652
Set CONFIG_IMA_KEXEC=y for supported powerpc architectures.
Update the annotations for this option, and mark all IMA options
for enforcment. Remove notes for options which have no prompt and
are thus not included in the annotations file.
Seth Forshee [Wed, 4 Jan 2017 16:08:38 +0000 (10:08 -0600)]
UBUNTU: [Config] CONFIG_IPDDP=n
BugLink: http://bugs.launchpad.net/bugs/1559772
This module isn't being actively maintained, and when it is
enabled it prevents a newer userspace implementation from
working.
Craig Magina [Thu, 8 Dec 2016 10:39:31 +0000 (11:39 +0100)]
UBUNTU: d-i initrd needs additional usb modules to support the merlin platform
BugLink: https://launchpad.net/bugs/1625222
Ubuntu builds the uas and xhci-plat-hcd usb drivers as modules, these
module are needed for d-i to use usb in the installer on the merlin platform.
Signed-off-by: Craig Magina <craig.magina@canonical.com> Signed-off-by: Tim Gardner <tim.gardner@canonical.com>
Seth Forshee [Tue, 6 Dec 2016 15:51:01 +0000 (09:51 -0600)]
UBUNTU: SAUCE: aufs -- Add flags argument to aufs_rename()
Starting with Linux 4.9-rc1 the rename2 inode operation has
replaced the rename op, so filesystem rename implementations
require an extra flags argument. Add the argument to fix the
FTBFS with Linux 4.9, but since aufs doesn't support any of the
flags return an error if it is non-zero.
Seth Forshee [Tue, 6 Dec 2016 15:27:46 +0000 (09:27 -0600)]
UBUNTU: SAUCE: aufs -- Convert to use xattr handlers
Starting with Linux 4.9-rc1 the {get,set,remove}xattr inode
operations have been removed, and filesystems are required to use
xattr handlers instead. There's some partially implemented xattr
handlers commented out in aufs already. Finish those handlers and
convert aufs over to use them.
Ming Lei [Tue, 6 Dec 2016 17:20:03 +0000 (01:20 +0800)]
UBUNTU: [Config] CONFIG_ARM64_ERRATUM_845719=y
BugLink: http://bugs.launchpad.net/bugs/1647793
Looks this option is missed in Yakkety, and should have been
enabled. It is enabled on Xenial, and used for workaround
the issue of "Cortex-A53: 845719: a load might read incorrect data".
Signed-off-by: Ming Lei <ming.lei@canonical.com> Signed-off-by: Tim Gardner <tim.gardner@canonical.com>
Seth Forshee [Thu, 1 Dec 2016 18:18:28 +0000 (12:18 -0600)]
UBUNTU: Set build_arch=x86 for amd64 and x32
The kernel build infrastructure expects the value of ARCH to be
the same as the architecture subdirectory in arch/*. The values
for amd64 and x32 are not, so fix them to be what the kernel
expects.
Seth Forshee [Tue, 29 Nov 2016 04:02:48 +0000 (22:02 -0600)]
UBUNTU: [Config] Update annotations
Update the annotations file based on the current configurations.
Add bug numbers for some options, and flag some options for
review which require another look.
Seth Forshee [Fri, 18 Nov 2016 19:37:40 +0000 (13:37 -0600)]
UBUNTU: [Config] Fix up CONFIG_I2C_SLAVE values
The annotations file says this should be =n on all architectures
except for armhf, but it is currently =y on several others. This
is because it's being selected by CONFIG_I2C_EMEV2, which is only
really relevant for armhf, and CONFIG_I2C_RCAR, which is only
relevant on armhf and arm64. Disable CONFIG_I2C_EMEV2 and
CONFIG_I2C_SLAVE for all other architectures, and update the
annotations file entry for CONFIG_I2C_EMEV2 accordingly.
Seth Forshee [Tue, 15 Nov 2016 16:36:49 +0000 (10:36 -0600)]
UBUNTU: [Config] Fix s390x config carnage
In all the config carnage during the yakkety cycle some of the
s390x config options got changed by mistake. Try to fix this
based on xenial's config and the git history from yakkety.
Seth Forshee [Tue, 22 Nov 2016 14:32:17 +0000 (08:32 -0600)]
UBUNTU: [Debian] config-check -- Make it easier to find annotations syntax errors
The errors from eval due to malformed annotations values often
are not very helpful in tracking down the source of the error.
Make this easier by printing the name of the confguration option
along with the error string.