From: Justin M. Forbes on gitlab.com
https://gitlab.com/cki-project/kernel-ark/-/merge_requests/1501#note_739392253
Now upstream, closing
___
kernel mailing list -- kernel@lists.fedoraproject.org
To unsubscribe send an email to kernel-le...@lists.fedorap
From: Justin M. Forbes on gitlab.com
https://gitlab.com/cki-project/kernel-ark/-/merge_requests/1284#note_746658332
Since it was set to allow maintainers to commit to the MR, I did the rebase.
___
kernel mailing list -- kernel@lists.fedoraproject.org
To
From: Justin M. Forbes on gitlab.com
https://gitlab.com/cki-project/kernel-ark/-/merge_requests/1284#note_749677987
@omos Done.
___
kernel mailing list -- kernel@lists.fedoraproject.org
To unsubscribe send an email to kernel-le...@lists.fedoraproject.org
From: Justin M. Forbes on gitlab.com
https://gitlab.com/cki-project/kernel-ark/-/merge_requests/1459#note_758223323
From the first rawhide build after this merged (ELN hasn't started yet):
drivers/pci/pci-driver.c:299:5: warning: "CONFIG_RHEL_DIFFERENCES" is not
defined, evaluates to 0 [-Wundef]
From: Justin M. Forbes on gitlab.com
https://gitlab.com/cki-project/kernel-ark/-/merge_requests/1408#note_802369351
> (2) are we turning this on because there is a Customer need, or because the
script for configs can't handle it? The former makes sense, but the latter
would be puzzling to me.
Th
From: Justin M. Forbes
Turn CONFIG_DEVMEM back off for aarch64
We have ended up with a config mismatch for the RHEL configs in a way
that was unclear and untracked. CONFIG_STRICT_DEVMEM is enabled in
configs/common/generic, but CONFIG_DEVMEM was disabled for aarch64 in
RHEL (though not in Fedora
From: Justin M. Forbes on gitlab.com
https://gitlab.com/cki-project/kernel-ark/-/merge_requests/1408#note_802380446
With the new push, we end up with:
kernel-aarch64-debug-rhel.config
```
CONFIG_DEVMEM=y
# CONFIG_IO_STRICT_DEVMEM is not set
# CONFIG_STRICT_DEVMEM is not set
```
kernel-aarch64-rh
From: Justin M. Forbes on gitlab.com
https://gitlab.com/cki-project/kernel-ark/-/merge_requests/1537#note_803671210
I am okay with this for Fedora, but given the way the rules work, I won't
approve until it has a RHEL approval since it is changing the config for
CONFIG_DM_MULTIPATH_HST in RHEL too
From: Justin M. Forbes on gitlab.com
https://gitlab.com/cki-project/kernel-ark/-/merge_requests/1537#note_803673297
@bmarzins We aren't that fancy here. AFAIK nothing here will automatically
touch bugzilla at all.
___
kernel mailing list -- kernel@lists
From: Justin M. Forbes
lib/crypto: add prompts back to crypto libraries
Commit 6048fdcc5f269 ("lib/crypto: blake2s: include as built-in") took
away a number of prompt texts from other crypto libraries. This makes
values flip from built-in to module when oldconfig runs, and causes
problems when t
From: Justin M. Forbes on gitlab.com
https://gitlab.com/cki-project/kernel-ark/-/merge_requests/1552#note_808649182
This MR isn't going to be merged, it is simply a placeholder until the latest
version of this patch is merged upstream. By keeping it here and "include in
releases" it ends up in th
From: Justin M. Forbes
Add dev_addr_lists_test to mod-internal.list
With the 5.17 merge window, we are now building the dev_addr_lists_test
kunit test module. This needs to be added to the mod-internal.list or we
end up with depmod failures.
Signed-off-by: Justin M. Forbes
diff --git a/redhat
From: Justin M. Forbes
Add packaged but empty /lib/modules//systemtap/
It was requested in rhbz 1477295 quite some time ago and got missed, but
a new bug has been opened against RHEL 9 requesting the same. This is the
default location for systemtap to pull in taps for stapusr executed
scripts.
From: Justin M. Forbes
lib/crypto: add prompts back to crypto libraries
Commit 6048fdcc5f269 ("lib/crypto: blake2s: include as built-in") took
away a number of prompt texts from other crypto libraries. This makes
values flip from built-in to module when oldconfig runs, and causes
problems when t
From: Justin M. Forbes on gitlab.com
https://gitlab.com/cki-project/kernel-ark/-/merge_requests/1552#note_812855905
Updated with the final version that is in linux-next right now. We could go
ahead and merge this version (with acks) if we want os-build to work while
waiting for it to make Linus' t
From: Justin M. Forbes on gitlab.com
https://gitlab.com/cki-project/kernel-ark/-/merge_requests/1552#note_812950236
I do not expect any further changes to this one.
___
kernel mailing list -- kernel@lists.fedoraproject.org
To unsubscribe send an email to
From: Justin M. Forbes on gitlab.com
https://gitlab.com/cki-project/kernel-ark/-/merge_requests/1535#note_814101795
Blocking thread until a RHEL review comes in.
___
kernel mailing list -- kernel@lists.fedoraproject.org
To unsubscribe send an email to ke
From: Justin M. Forbes
Config consolidation into common
This is the output of the 'evaluate_configs -j' run. Configs which were
the same in both ark and fedora are consolidated to common. I have
verified that this makes absolutely no changes to the generated config
files, there is specifically n
From: Justin M. Forbes on gitlab.com
https://gitlab.com/cki-project/kernel-ark/-/merge_requests/1583#note_815515160
Just as a side note, this is something we plan to run regularly. If the
generated config files are 100% identical to the ones we get before the script
run, do we really need to gene
From: Justin M. Forbes on gitlab.com
https://gitlab.com/cki-project/kernel-ark/-/merge_requests/1583#note_816592682
Not exactly, without a lot more manual work. The problem is, new config items
are introduced into both pending-fedora and pending-common. The MR is created
to move pending-common to
From: Sergei Trofimovich
objtool: check: give big enough buffer for pv_ops
On gcc-12 build fails flagging possible buffer overflow:
check.c: In function 'validate_call':
check.c:2865:58: error: '%d' directive output may be truncated writing
between 1 and 10 bytes into a region of size
From: Justin M. Forbes on gitlab.com
https://gitlab.com/cki-project/kernel-ark/-/merge_requests/1582#note_819456390
I did. https://gitlab.com/cki-project/kernel-ark/-/merge_requests/1583
___
kernel mailing list -- kernel@lists.fedoraproject.org
To unsubs
From: Justin M. Forbes on gitlab.com
https://gitlab.com/cki-project/kernel-ark/-/merge_requests/1579#note_819750244
> That's not entirely true. koji builds do not execute a s390x+zfcpdump build
and do not exceed the 64 char limit. This only occurs on brew bulds because of
the s390x+zfcpdump target
From: Justin M. Forbes on gitlab.com
https://gitlab.com/cki-project/kernel-ark/-/merge_requests/1586#note_819847030
upstream discussion: https://lkml.org/lkml/2022/1/14/88
___
kernel mailing list -- kernel@lists.fedoraproject.org
To unsubscribe send an e
From: Justin M. Forbes on gitlab.com
https://gitlab.com/cki-project/kernel-ark/-/merge_requests/1589#note_822826988
I think it will be okay, but I can always turn it back on for Fedora if users
complain. I just didn't want to ack it until it had a RHEL ack.
From: Justin M. Forbes
Add test_hash to the mod-internal.list
Upstream commit 0acc968f352336a459f27ba1f23745a174933c9c refactored
test_hash.c to use the KUnit framework. As such it depends on kunit
symbols and needs to be filtered to mod-internal.
Signed-off-by: Justin M. Forbes
diff --git a
From: Justin M. Forbes
Workaround for gcc12 compile issues in ubcmd-util.h
While upstream has made no move around this and I am not positive that
this is the "proper" fix for upstream, it does get us building again for
several arches. This is being added as "include in release" until
upstream we
From: Justin M. Forbes
Fix gcc 12 warning address of 'mem_section' will never be NULL
Signed-off-by: Justin M. Forbes
diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h
index blahblah..blahblah 100644
--- a/include/linux/mmzone.h
+++ b/include/linux/mmzone.h
@@ -1390,11 +1390,9 @@ st
From: Waiman Long
mm/sparsemem: Fix 'mem_section' will never be NULL gcc 12 warning
The gcc 12 compiler reports a warning on the following code:
static inline struct mem_section *__nr_to_section(unsigned long nr)
{
#ifdef CONFIG_SPARSEMEM_EXTREME
if (!mem_section)
From: Justin M. Forbes on gitlab.com
https://gitlab.com/cki-project/kernel-ark/-/merge_requests/1598#note_828575658
Updated with your patch which Andrew just pulled into mm.
___
kernel mailing list -- kernel@lists.fedoraproject.org
To unsubscribe send an
From: Justin M. Forbes on gitlab.com
https://gitlab.com/cki-project/kernel-ark/-/merge_requests/1598#note_828596655
Actually, that doesn't solve the whole issue, and creates a new warning for
infinite recursion:
https://kojipkgs.fedoraproject.org/work/tasks/2682/82252682/build.log
___
From: Justin M. Forbes on gitlab.com
https://gitlab.com/cki-project/kernel-ark/-/merge_requests/1598#note_828687222
Thanks, yes, it is on for x86_64 and off for everything else. The only thing
that fails is armv7 though.
___
kernel mailing list -- kernel
From: Waiman Long
mm/sparsemem: Fix 'mem_section' will never be NULL gcc 12 warning
The gcc 12 compiler reports a "'mem_section' will never be NULL"
warning on the following code:
static inline struct mem_section *__nr_to_section(unsigned long nr)
{
#ifdef CONFIG_SPARSEMEM_EXTREME
From: Justin M. Forbes on gitlab.com
https://gitlab.com/cki-project/kernel-ark/-/merge_requests/1604#note_836288800
This does not work as =y and the CKI testing exposed that. It should be =m
___
kernel mailing list -- kernel@lists.fedoraproject.org
To un
From: Justin M. Forbes
Change CONFIG_TEST_BPF to a module
Due to dependencies, CONFIG_TEST_BPF fails the config check when set to
inline as it gets forced to be a module. Changing it to a module passed
the config check, and I believe meets the original intent of MR 1604 as
an entry for it was al
From: Kees Cook
tools: Fix use-after-free for realloc(..., 0)
GCC 12 was correctly reporting a potential use-after-free condition in
the xrealloc helper. Fix the warning by avoiding an implicit "free(ptr)"
when size == 0:
In file included from help.c:12:
In function 'xrealloc',
inlined from
From: Justin M. Forbes on gitlab.com
https://gitlab.com/cki-project/kernel-ark/-/merge_requests/1594#note_842557272
Updated with a new patch from Kees upstream.
___
kernel mailing list -- kernel@lists.fedoraproject.org
To unsubscribe send an email to ker
From: Justin M. Forbes on gitlab.com
https://gitlab.com/cki-project/kernel-ark/-/merge_requests/1619#note_846439422
It seems the Kconfig entry for this really belongs in Kconfig.redhat perhaps
even better if it had a dep on RHEL_DIFFERENCES as well.
___
From: Justin M. Forbes on gitlab.com
https://gitlab.com/cki-project/kernel-ark/-/merge_requests/1594#note_848024026
This is upstream now, closing.
___
kernel mailing list -- kernel@lists.fedoraproject.org
To unsubscribe send an email to kernel-le...@list
From: Justin M. Forbes on gitlab.com
https://gitlab.com/cki-project/kernel-ark/-/merge_requests/1621#note_850065513
Right, so by default, new items get created in both pending-common and
pending-fedora, but it is a manual run that only happens every few months. As
they are reviewed and processed,
From: Justin M. Forbes on gitlab.com
https://gitlab.com/cki-project/kernel-ark/-/merge_requests/1636#note_857554328
That should probably be fixed, and we do need the .0 in both places. It is
fairly useless in the rawhide case, but that last digit has meaning in stable
releases.
__
From: Justin M. Forbes on gitlab.com
https://gitlab.com/cki-project/kernel-ark/-/merge_requests/1637#note_858675240
This seems as though it might break the kernel-rt and similar efforts at least
until rt is upstream and it simply becomes another flavor of kernel built from
the same spec/srpm.
From: Justin M. Forbes on gitlab.com
https://gitlab.com/cki-project/kernel-ark/-/merge_requests/1637#note_859350285
@prarit yes, that would make sense.
___
kernel mailing list -- kernel@lists.fedoraproject.org
To unsubscribe send an email to kernel-le...
From: Justin M. Forbes on gitlab.com
https://gitlab.com/cki-project/kernel-ark/-/merge_requests/1612#note_859350785
I am not clicking until a RHEL reviewer does considering this changes RHEL
more than Fedora, and it only needs one ack from me. (Though I will ack it)
___
From: Justin M. Forbes on gitlab.com
https://gitlab.com/cki-project/kernel-ark/-/merge_requests/1635#note_863252874
Let's hold off on merging this for a bit, I added it as include in releases,
and today's build showed:
```
make[1]: *** [Makefile:148: dist-configs-check] Error 1
make[1]: Leaving di
From: Justin M. Forbes on gitlab.com
https://gitlab.com/cki-project/kernel-ark/-/merge_requests/1635#note_863271340
Actually, weirder, 'make dist-configs-check' works fine in ark-latest, but
'make dist-git' fails as above. If I do a 'make dist-srpm', that succeeds,
and afterwards 'make dist-git'
From: Justin M. Forbes on gitlab.com
https://gitlab.com/cki-project/kernel-ark/-/merge_requests/1598#note_867053758
Yes, it was backed out from the queue upstream because of a similar problem on
m1 macs (support not upstreamed yet). @llong1 is there an updated version in
the works?
__
From: Justin M. Forbes on gitlab.com
https://gitlab.com/cki-project/kernel-ark/-/merge_requests/1635#note_867434770
It ran today, include in release put the new version in ark-latest this
morning. It worked fine.
___
kernel mailing list -- kernel@lists.f
From: Justin M. Forbes on gitlab.com
Merge Request: https://gitlab.com/cki-project/kernel-ark/-/merge_requests/1643
Now that we are capable of processing multiple config files in parallel, we do
not know exactly which arch a failure is seen on. This is important as
frequently errors are specific t
From: Justin M. Forbes
Pass RHJOBS to process_configs for dist-configs-check as well
Signed-off-by: Justin M. Forbes
diff --git a/redhat/Makefile b/redhat/Makefile
index blahblah..blahblah 100644
--- a/redhat/Makefile
+++ b/redhat/Makefile
@@ -145,7 +145,7 @@ dist-rhel-configs: dist-configs
r
From: Justin M. Forbes
Print arch with process_configs errors
Now that we are capable of processing multiple config files in parallel,
we do not know exactly which arch a failure is seen on. This is
important as frequently errors are specific to a single arch. This patch
just prints the arch bef
From: Justin M. Forbes on gitlab.com
https://gitlab.com/cki-project/kernel-ark/-/merge_requests/1641#note_873929542
Given the nature of the changes, I want to run these as "include in release"
once ready to make sure that things are working with the typical work flows.
From: Justin M. Forbes on gitlab.com
https://gitlab.com/cki-project/kernel-ark/-/merge_requests/1610#note_878394578
Just a heads up, the rebase did not happen this week as Linus released a
surprise rc8 (the rc7 announcement implied that there would not be one). This
means the rebase will happen n
From: Waiman Long
mm/sparsemem: Fix 'mem_section' will never be NULL gcc 12 warning
The gcc 12 compiler reports a "'mem_section' will never be NULL"
warning on the following code:
static inline struct mem_section *__nr_to_section(unsigned long nr)
{
#ifdef CONFIG_SPARSEMEM_EXTREME
From: Justin M. Forbes
Define SNAPSHOT correctly when VERSION_ON_UPSTREAM is 0
The Makefile changes cleaned things up quite a bit, but left out the
else to define SNAPSHOT when VERSION_ON_UPSTREAM:=0. The result was as
specfile with debugbuildsenabled set incorrectly.
Signed-off-by: Justin M.
From: Justin M. Forbes
Add new kunit tests for 5.18 to mod-internal.list
Two new kunit tests came in during the merge window so far,
stackinit_kunit and overflow_kunit. Add them to mod-internal.list so
that depmod works correctly and they can find their needed symbols.
Signed-off-by: Justin M.
From: Justin M. Forbes
Turn on CONFIG_RANDOM_TRUST_BOOTLOADER
While doing the rng rewrites, the upstream maintainer was looking to see
which distros enabled RANDOM_TRUST_BOOTLOADER, and reached out to us to
ask why not. This came in during the 5.4 cycle, and I did not review it,
but do agree tha
From: Justin M. Forbes
Fix RHDISTGIT for Fedora
While the existing setup worked fine for users not setting up an
RHDISTGIT_CACHE, it fails once you do because there was no entry to
properly push to the Fedora dist-git.
Signed-off-by: Justin M. Forbes
diff --git a/redhat/Makefile.rhpkg b/redha
From: Justin M. Forbes
Remove i686 configs and filters
Fedora has not built an i686 kernel in quite some time. There is no more
point in trying to maintain configs for a kernel that we do not build.
Signed-off-by: Justin M. Forbes
diff --git a/redhat/configs/build_configs.sh b/redhat/configs/
From: Justin M. Forbes on gitlab.com
https://gitlab.com/cki-project/kernel-ark/-/merge_requests/1603#note_891415723
I would much rather it not. While things may build and function, it just makes
it easier for patches like this to get lost on the Fedora stable rebases,
where they are all ripped out
From: Justin M. Forbes on gitlab.com
https://gitlab.com/cki-project/kernel-ark/-/merge_requests/1603#note_891516554
I mean all of the RHELisms are ripped out every time I do a rebased fedora-5.x
branch. This ends up being several times a year, and since os-build is
rebased, I can't simply keep a l
From: Justin M. Forbes on gitlab.com
https://gitlab.com/cki-project/kernel-ark/-/merge_requests/1603#note_891563350
That should do it.
___
kernel mailing list -- kernel@lists.fedoraproject.org
To unsubscribe send an email to kernel-le...@lists.fedoraproj
From: Justin M. Forbes on gitlab.com
https://gitlab.com/cki-project/kernel-ark/-/merge_requests/1695#note_892814613
Right, as it stands, they are enabled on Fedora, but not eln/RHEL. Given what
these are used for, I doubt they have Fedora users, though I could be wrong.
Going to ack this, and if a
From: Justin M. Forbes on gitlab.com
https://gitlab.com/cki-project/kernel-ark/-/merge_requests/1695#note_893095650
The patch needs to be updated either way. This change has no impact on the
RHEL configs at all because the ark config directory has priority over the
common directory for RHEL, meani
From: Justin M. Forbes on gitlab.com
https://gitlab.com/cki-project/kernel-ark/-/merge_requests/1721#note_894758127
This does not work, testing another fix and will update the MR if that works.
___
kernel mailing list -- kernel@lists.fedoraproject.org
To
From: Waiman Long
mm/sparsemem: Fix 'mem_section' will never be NULL gcc 12 warning
The gcc 12 compiler reports a "'mem_section' will never be NULL"
warning on the following code:
static inline struct mem_section *__nr_to_section(unsigned long nr)
{
#ifdef CONFIG_SPARSEMEM_EXTREME
From: Justin M. Forbes on gitlab.com
https://gitlab.com/cki-project/kernel-ark/-/merge_requests/1721#note_896016507
Right, so I did
`%{make} CFLAGS="${RPM_OPT_FLAGS} -D_GNU_SOURCE -I/usr/include/libnl3"
LDFLAGS="%{__global_ldflags} -lnl-genl-3 -lnl-3`
In the kernel spec and that got around both
From: Justin M. Forbes on gitlab.com
https://gitlab.com/cki-project/kernel-ark/-/merge_requests/1721#note_896035285
That's pretty much what I did, but not a whole lot of point in pushing it
until things actually build.
___
kernel mailing list -- kernel@l
From: Justin M. Forbes on gitlab.com
https://gitlab.com/cki-project/kernel-ark/-/merge_requests/1721#note_896201128
It does work with the version I have locally, but until I can get a build
through to verify, I was not planning to send the update. The other issue is
in iio_event_monitor, which wo
From: Justin M. Forbes
Add clk_test and clk-gate_test to mod-internal.list
These are new kunit tests added in 5.18 that need to be in
mod-internal.list for depmod to succeed.
Signed-off-by: Justin M. Forbes
diff --git a/redhat/mod-internal.list b/redhat/mod-internal.list
index blahblah..blahb
From: Justin M. Forbes on gitlab.com
https://gitlab.com/cki-project/kernel-ark/-/merge_requests/1721#note_896710175
Yes, it seems my fix is not actually a fix when building in koji.
___
kernel mailing list -- kernel@lists.fedoraproject.org
To unsubscribe
From: Waiman Long
mm/sparsemem: Fix 'mem_section' will never be NULL gcc 12 warning
The gcc 12 compiler reports a "'mem_section' will never be NULL"
warning on the following code:
static inline struct mem_section *__nr_to_section(unsigned long nr)
{
#ifdef CONFIG_SPARSEMEM_EXTREME
From: Justin M. Forbes on gitlab.com
https://gitlab.com/cki-project/kernel-ark/-/merge_requests/1721#note_897518778
Does centos9 even need this? It wasn't required before the 5.18 merge window,
and the issues are a combination of the added features there and gcc 12
issues.
From: Justin M. Forbes on gitlab.com
https://gitlab.com/cki-project/kernel-ark/-/merge_requests/1721#note_898479057
I have not found a successful way to get this to build in koji without your
Makefile changes Herton. Want to send a patch upstream and create an MR to get
it in ark?
From: Justin M. Forbes on gitlab.com
https://gitlab.com/cki-project/kernel-ark/-/merge_requests/1721#note_899643862
Friday I just fixed things up by hand in dist-git, and I had planed to do the
same for rc1, but would like to just get things include-in-releases at least
until we get proper acks.
_
From: Justin M. Forbes on gitlab.com
https://gitlab.com/cki-project/kernel-ark/-/merge_requests/1721#note_900288988
Thanks, just used that to build kernel-tools for Fedora. We do have a
different make call there though, and I had to append the include, so my lines
are:
%{tools_make} CFLAGS+="-D_G
From: Justin M. Forbes on gitlab.com
https://gitlab.com/cki-project/kernel-ark/-/merge_requests/1735#note_901266926
Correct, I would rather make the BUILD go away even for the KABI tarballs, so
for instance kernel-kabi-dw-5.17.1-300.tar.bz2 becomes kernel-kabi-
dw-5.17.1.tar.bz2 as there is no nee
From: Justin M. Forbes
Spec fixes for intel-speed-select
Now that Herton's patch is in, we still need the includes added to the
make commands from the spec. This patch makes intel-speed-select build
in current koji.
Signed-off-by: Justin M. Forbes
diff --git a/redhat/kernel.spec.template b/re
From: Justin M. Forbes on gitlab.com
https://gitlab.com/cki-project/kernel-ark/-/merge_requests/1713#note_905957214
I have a feeling that someone in Fedora might complain, but happy to turn it
off and see. I can always flip it back if needed.
___
kernel
From: Justin M. Forbes on gitlab.com
https://gitlab.com/cki-project/kernel-ark/-/merge_requests/1713#note_906307366
@jsnitsel Actually, you are correct. I misread the common and ark bits, so no
change there. I have been toying around with some sort of a configdiff make
target to make some of thi
From: Justin M. Forbes on gitlab.com
https://gitlab.com/cki-project/kernel-ark/-/merge_requests/1710#note_908015595
So it seems this might not be such a good idea after all. Some benchmarking
posted over the weekend shows that the AMD_PSTATE driver under-performs when
compared to ACPI_CPUFREQ on 5
From: Justin M. Forbes on gitlab.com
https://gitlab.com/cki-project/kernel-ark/-/merge_requests/1739#note_908249022
This does not build as is:
+ /usr/bin/make -s 'HOSTCFLAGS=-O2 -fexceptions -g -grecord-gcc-switches
-pipe -Wall -Werror=format-security -Wp,-D_FORTIFY_SOURCE=2
-Wp,-D_GLIBCXX_ASSERT
From: Justin M. Forbes
Build intel_sdsi with %{tools_make}
Building with %{make} leaves off some CFLAGS and LDFLAGS which make this
build fail. Building with %{tools_make} fixes this.
Signed-off-by: Justin M. Forbes
diff --git a/redhat/kernel.spec.template b/redhat/kernel.spec.template
index
From: Justin M. Forbes on gitlab.com
https://gitlab.com/cki-project/kernel-ark/-/merge_requests/1748#note_908605382
Curious on CONFIG_KFENCE_STATIC_KEYS, was there benchmarking done both enabled
and disabled? Just asking because the help text seemed to indicate it was
"only recommended when using
From: Justin M. Forbes on gitlab.com
https://gitlab.com/cki-project/kernel-ark/-/merge_requests/1735#note_908610999
It seems that it would work as we do not set SINGLE_TARBALL:=1
___
kernel mailing list -- kernel@lists.fedoraproject.org
To unsubscribe se
From: Justin M. Forbes on gitlab.com
https://gitlab.com/cki-project/kernel-ark/-/merge_requests/1748#note_908617944
Actually commit 4f612ed3f748962cbef1316ff3d323e2b9055b6e changed it to depend
on CONFIG_EXPERT, which is not set, so there is no need for an entry on either
CONFIG_KFENCE_STATIC_KEYS
From: Justin M. Forbes on gitlab.com
https://gitlab.com/cki-project/kernel-ark/-/merge_requests/1750#note_910729241
Please enable this in the ark directory. We do not want KASAN enabled for
debug kernels in Rawhide on Fedora. By policy, we build only debug kernels
other than -rcX and the final .0,
From: Justin M. Forbes on gitlab.com
https://gitlab.com/cki-project/kernel-ark/-/merge_requests/1750#note_911462288
You are correct, it does not, though it seems somewhat illogical. It is
expected that fedora overrides common for Fedora configs, but it does seem
that fedora/generic should not over
From: Justin M. Forbes on gitlab.com
https://gitlab.com/cki-project/kernel-ark/-/merge_requests/1647#note_913334565
The FLAVOR parameter is still defined, though given as empty unless you call
one of the make targets that define it. The script itself has behavior around
empty as well though. It i
From: Justin M. Forbes on gitlab.com
https://gitlab.com/cki-project/kernel-ark/-/merge_requests/1647#note_913340492
Do we really want to key this off of Fedora? That limits the ability for SIGs
to split out their patches. It can also be more difficult for multi tier
setups where the upstream is ac
From: Justin M. Forbes on gitlab.com
https://gitlab.com/cki-project/kernel-ark/-/merge_requests/3416#note_2132082003
This is turned off for Fedora in another chunk of code later in the spec.
--
___
kernel mailing list -- kernel@lists.fedoraproject.org
From: Justin M. Forbes on gitlab.com
https://gitlab.com/cki-project/kernel-ark/-/merge_requests/3223#note_2129176765
Reverting this as Jan's proper upstream solution is in place now.
--
___
kernel mailing list -- kernel@lists.fedoraproject.org
To unsub
From: Justin M. Forbes
Revert "Merge branch 'drop_engine_api' into 'os-build'"
This reverts merge request !3223
diff --git a/certs/extract-cert.c b/certs/extract-cert.c
index blahblah..blahblah 100644
--- a/certs/extract-cert.c
+++ b/certs/extract-cert.c
@@ -21,6 +21,7 @@
#include
#include
From: Justin M. Forbes on gitlab.com
https://gitlab.com/cki-project/kernel-ark/-/merge_requests/3409#note_2129188193
Pipeline will fail because the new fix is not in tree. We need this one out
and then the merge job will bring in the proper fix.
--
___
From: Justin M. Forbes on gitlab.com
https://gitlab.com/cki-project/kernel-ark/-/merge_requests/3442#note_2158346233
Well, we still need an MR to flip the config in common and remove the no
longer needed debug entry:
```
diff --git a/redhat/configs/common/debug/CONFIG_RCU_TRACE
b/redhat/configs/c
From: Justin M. Forbes on gitlab.com
https://gitlab.com/cki-project/kernel-ark/-/merge_requests/3442#note_2158177745
We should actually just flip the config in
redhat/configs/common/generic/CONFIG_RCU_TRACE as I have had a request for
this in Fedora as well.
--
__
From: Justin M. Forbes on gitlab.com
https://gitlab.com/cki-project/kernel-ark/-/merge_requests/3440#note_2158956485
The De-dup script gets weird when things are not y, m, or not set. I can fix
up Fedora though.
--
___
kernel mailing list -- kernel@lis
From: Justin M. Forbes on gitlab.com
https://gitlab.com/cki-project/kernel-ark/-/merge_requests/3429#note_2146034514
I fixed this up in a different way, moving the unset to fedora/generic. That
way it won't get accidentally turned off if another arch decides to support
it. It is currently explici
From: Justin M. Forbes on gitlab.com
https://gitlab.com/cki-project/kernel-ark/-/merge_requests/3429#note_2146095663
We did not have CONFIG_I2C_DESIGNWARE_CORE set for ppc previously. Anyway,
should be fixed up now.
--
___
kernel mailing list -- kernel
From: Justin M. Forbes on gitlab.com
https://gitlab.com/cki-project/kernel-ark/-/merge_requests/3346#note_2170054806
I am just going to revert all of this. It turns out it is a bit of a mess from
a Fedora standpoint. How about you figure out which ones you actually need to
change and work with me
501 - 600 of 605 matches
Mail list logo
