Skip to contentSkip to conversation

kernel-yocto: drop 5.19 recipes and make 6.1 default

PATCH In review [email protected] 4 participants 12 messages Jan 24, 2023 – Jan 26, 2023

lore
Share & tools
lore
#0 lore raw

cover kernel-yocto: drop 5.19 recipes and make 6.1 default

From: Bruce Ashfield <[email protected]>

Richard,

As promised, here are the commits to switch us over to 6.1 as the default.

I'll also send a series for poky to bump things there as well.

Also, if there are any significant problems, we'll just shelve this while I look at any issues. But at this point, it is worth getting the patches out there to move the needle to 6.1

Bruce

The following changes since commit 3c3fd6a65e8103f74ae382d196d486b31a168b39:

insane: Improve patch warning/error handling (2023-01-21 07:46:38 +0000)

are available in the Git repository at:

https://git.yoctoproject.org/poky-contrib zedd/kernel
http://git.yoctoproject.org/cgit.cgi/poky-contrib/log/?h=zedd/kernel

Bruce Ashfield (2):
linux-yocto/5.19: drop recipes
conf/qemu: bump preferred kernel version to 6.1

 meta/conf/machine/include/x86/x86-base.inc    |  2 +-
 meta/conf/machine/qemuarmv5.conf              |  2 +-
 .../linux/linux-yocto-rt_5.19.bb              | 45 ------------
 .../linux/linux-yocto-tiny_5.19.bb            | 30 --------
 meta/recipes-kernel/linux/linux-yocto_5.19.bb | 70 -------------------
 5 files changed, 2 insertions(+), 147 deletions(-)
 delete mode 100644 meta/recipes-kernel/linux/linux-yocto-rt_5.19.bb
 delete mode 100644 meta/recipes-kernel/linux/linux-yocto-tiny_5.19.bb
 delete mode 100644 meta/recipes-kernel/linux/linux-yocto_5.19.bb
Signature · 1 line
2.34.1
#1 lore raw

PATCH 1/2 linux-yocto/5.19: drop recipes

From: Bruce Ashfield <[email protected]>

5.19 is no longer getting upstream -stable support, and we've replaced it with 6.1 as the latest (and default) version. Removing the recipes to make this clear.

Released branches will still get bugfixes as appropriate.

---
 .../linux/linux-yocto-rt_5.19.bb              | 45 ------------
 .../linux/linux-yocto-tiny_5.19.bb            | 30 --------
 meta/recipes-kernel/linux/linux-yocto_5.19.bb | 70 -------------------
 3 files changed, 145 deletions(-)
 delete mode 100644 meta/recipes-kernel/linux/linux-yocto-rt_5.19.bb
 delete mode 100644 meta/recipes-kernel/linux/linux-yocto-tiny_5.19.bb
 delete mode 100644 meta/recipes-kernel/linux/linux-yocto_5.19.bb
meta/recipes-kernel/linux/linux-yocto-rt_5.19.bb deleted +0 -45
deleted file mode 100644
index f4d205e613..0000000000
@@ -1,45 +0,0 @@
-KBRANCH ?= "v5.19/standard/preempt-rt/base"
-
-require recipes-kernel/linux/linux-yocto.inc
-
-# Skip processing of this recipe if it is not explicitly specified as the
-# PREFERRED_PROVIDER for virtual/kernel. This avoids errors when trying
-# to build multiple virtual/kernel providers, e.g. as dependency of
-# core-image-rt-sdk, core-image-rt.
-python () {
-    if d.getVar("KERNEL_PACKAGE_NAME") == "kernel" and d.getVar("PREFERRED_PROVIDER_virtual/kernel") != "linux-yocto-rt":
-        raise bb.parse.SkipRecipe("Set PREFERRED_PROVIDER_virtual/kernel to linux-yocto-rt to enable it")
-}
-
-SRCREV_machine ?= "3101c367eab8952721086b545ad37c301b2a7452"
-SRCREV_meta ?= "239a6c0d3c3b046971909f1e066380465b0c331d"
-
-SRC_URI = "git://git.yoctoproject.org/linux-yocto.git;branch=${KBRANCH};name=machine \
-           git://git.yoctoproject.org/yocto-kernel-cache;type=kmeta;name=meta;branch=yocto-5.19;destsuffix=${KMETA}"
-
-LINUX_VERSION ?= "5.19.17"
-
-LIC_FILES_CHKSUM = "file://COPYING;md5=6bc538ed5bd9a7fc9398086aedcd7e46"
-
-DEPENDS += "${@bb.utils.contains('ARCH', 'x86', 'elfutils-native', '', d)}"
-DEPENDS += "openssl-native util-linux-native"
-
-PV = "${LINUX_VERSION}+git${SRCPV}"
-
-KMETA = "kernel-meta"
-KCONF_BSP_AUDIT_LEVEL = "1"
-
-LINUX_KERNEL_TYPE = "preempt-rt"
-
-COMPATIBLE_MACHINE = "^(qemux86|qemux86-64|qemuarm|qemuarmv5|qemuarm64|qemuppc|qemumips)$"
-
-KERNEL_DEVICETREE:qemuarmv5 = "versatile-pb.dtb"
-
-# Functionality flags
-KERNEL_EXTRA_FEATURES ?= "features/netfilter/netfilter.scc features/taskstats/taskstats.scc"
-KERNEL_FEATURES:append = " ${KERNEL_EXTRA_FEATURES}"
-KERNEL_FEATURES:append:qemuall=" cfg/virtio.scc features/drm-bochs/drm-bochs.scc"
-KERNEL_FEATURES:append:qemux86=" cfg/sound.scc cfg/paravirt_kvm.scc"
-KERNEL_FEATURES:append:qemux86-64=" cfg/sound.scc cfg/paravirt_kvm.scc"
-KERNEL_FEATURES:append = "${@bb.utils.contains("DISTRO_FEATURES", "ptest", " features/scsi/scsi-debug.scc", "", d)}"
-KERNEL_FEATURES:append = "${@bb.utils.contains("DISTRO_FEATURES", "ptest", " features/gpio/mockup.scc", "", d)}"
meta/recipes-kernel/linux/linux-yocto-tiny_5.19.bb deleted +0 -30
deleted file mode 100644
index 95a8a46bbc..0000000000
@@ -1,30 +0,0 @@
-KBRANCH ?= "v5.19/standard/tiny/base"
-
-LINUX_KERNEL_TYPE = "tiny"
-KCONFIG_MODE = "--allnoconfig"
-
-require recipes-kernel/linux/linux-yocto.inc
-
-LINUX_VERSION ?= "5.19.17"
-LIC_FILES_CHKSUM = "file://COPYING;md5=6bc538ed5bd9a7fc9398086aedcd7e46"
-
-DEPENDS += "${@bb.utils.contains('ARCH', 'x86', 'elfutils-native', '', d)}"
-DEPENDS += "openssl-native util-linux-native"
-
-KMETA = "kernel-meta"
-KCONF_BSP_AUDIT_LEVEL = "2"
-
-SRCREV_machine ?= "84f2f8e7a625aae0fa9e7027a2e774b99b646cf7"
-SRCREV_meta ?= "239a6c0d3c3b046971909f1e066380465b0c331d"
-
-PV = "${LINUX_VERSION}+git${SRCPV}"
-
-SRC_URI = "git://git.yoctoproject.org/linux-yocto.git;branch=${KBRANCH};name=machine \
-           git://git.yoctoproject.org/yocto-kernel-cache;type=kmeta;name=meta;branch=yocto-5.19;destsuffix=${KMETA}"
-
-COMPATIBLE_MACHINE = "^(qemux86|qemux86-64|qemuarm64|qemuarm|qemuarmv5)$"
-
-# Functionality flags
-KERNEL_FEATURES = ""
-
-KERNEL_DEVICETREE:qemuarmv5 = "versatile-pb.dtb"
meta/recipes-kernel/linux/linux-yocto_5.19.bb deleted +0 -70

70 lines not on this page · raw message

Signature · 1 line
2.34.1
#2 lore raw

PATCH 2/2 conf/qemu: bump preferred kernel version to 6.1

In reply to [email protected]

From: Bruce Ashfield <[email protected]>

6.1 has been soaking for a while now, so it is time to bump the default qemu version to 6.1 and prepare to remove 5.19.

---
 meta/conf/machine/include/x86/x86-base.inc | 2 +-
 meta/conf/machine/qemuarmv5.conf           | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)
meta/conf/machine/include/x86/x86-base.inc +1 -1
index 4052eacdb5..c3cc0dbff6 100644
@@ -18,7 +18,7 @@ SERIAL_CONSOLES ?= "115200;ttyS0"
 # kernel-related variables
 #
 PREFERRED_PROVIDER_virtual/kernel ??= "linux-yocto"
-PREFERRED_VERSION_linux-yocto ??= "5.19%"
+PREFERRED_VERSION_linux-yocto ??= "6.1%"
 
 #
 # XSERVER subcomponents, used to build the XSERVER variable
meta/conf/machine/qemuarmv5.conf +1 -1
index fe6c0117a8..6e59e42c3a 100644
@@ -17,5 +17,5 @@ QB_GRAPHICS = "-device virtio-gpu-pci"
 QB_OPT_APPEND = "-device qemu-xhci -device usb-tablet -device usb-kbd"
 QB_DTB = "${@oe.utils.version_less_or_equal('PREFERRED_VERSION_linux-yocto', '4.7', '', 'zImage-versatile-pb.dtb', d)}"
 
-PREFERRED_VERSION_linux-yocto ??= "5.19%"
+PREFERRED_VERSION_linux-yocto ??= "6.1%"
 KMACHINE:qemuarmv5 = "arm-versatile-926ejs"
Signature · 1 line
2.34.1
#3 lore raw
In reply to [email protected]
Quoted 4 lines · On Tue, 2023-01-24 at 16:59 -0500, [email protected] wrote:

From: Bruce Ashfield <[email protected]>

Richard,

Quoted 8 lines

As promised, here are the commits to switch us over to 6.1 as the default.

I'll also send a series for poky to bump things there as well.

Also, if there are any significant problems, we'll just shelve this while I look at any issues. But at this point, it is worth getting the patches out there to move the needle to 6.1

Thanks!

I think the issues are:

a) two selftest failures:

oe-selftest -r devtool.DevtoolUpgradeTests.test_devtool_virtual_kernel_modify oe-selftest -r efibootpartition.GenericEFITest.test_boot_efi

b) meta-arm issues

c) meta-virt issue (which you have a patch pending for in -next)

I suspect we can switch poky-tiny and defer the removal patches for a few days to give meta-arm time to sort itself out. The only question then becomes the selftest issues. I'm taking a quick look to see if they're simple to sort.

Cheers,

Richard

#4 lore raw
Quoted 18 lines · Richard Purdie
Quoted 4 lines · On Tue, 2023-01-24 at 16:59 -0500, [email protected] wrote:

From: Bruce Ashfield <[email protected]>

Richard,

Quoted 8 lines

As promised, here are the commits to switch us over to 6.1 as the default.

I'll also send a series for poky to bump things there as well.

Also, if there are any significant problems, we'll just shelve this while I look at any issues. But at this point, it is worth getting the patches out there to move the needle to 6.1

Thanks!

I think the issues are:

Quoted 9 lines

a) two selftest failures:

oe-selftest -r devtool.DevtoolUpgradeTests.test_devtool_virtual_kernel_modify oe-selftest -r efibootpartition.GenericEFITest.test_boot_efi

b) meta-arm issues

c) meta-virt issue (which you have a patch pending for in -next)

Unfortunately, I have to add

d) perf reproducibility issues:
https://autobuilder.yoctoproject.org/typhoon/#/builders/117/builds/2294/steps/13/logs/stdio

Quoted 18 lines

I suspect we can switch poky-tiny and defer the removal patches for a few days to give meta-arm time to sort itself out. The only question then becomes the selftest issues. I'm taking a quick look to see if they're simple to sort.

Cheers,

Richard

-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#176392): https://lists.openembedded.org/g/openembedded-core/message/176392
Mute This Topic: https://lists.openembedded.org/mt/96508797/3617179
Group Owner: [email protected]
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub [[email protected]]
-=-=-=-=-=-=-=-=-=-=-=-

Signature · 3 lines
Alexandre Belloni, co-owner and COO, Bootlin
Embedded Linux and Kernel engineering
https://bootlin.com
#5 lore raw
Quoted 26 lines · Alexandre Belloni
Quoted 25 lines · Richard Purdie
Quoted 4 lines · On Tue, 2023-01-24 at 16:59 -0500, [email protected] wrote:

From: Bruce Ashfield <[email protected]>

Richard,

Quoted 8 lines

As promised, here are the commits to switch us over to 6.1 as the default.

I'll also send a series for poky to bump things there as well.

Also, if there are any significant problems, we'll just shelve this while I look at any issues. But at this point, it is worth getting the patches out there to move the needle to 6.1

Thanks!

I think the issues are:

a) two selftest failures:

oe-selftest -r devtool.DevtoolUpgradeTests.test_devtool_virtual_kernel_modify oe-selftest -r efibootpartition.GenericEFITest.test_boot_efi

b) meta-arm issues

Quoted 7 lines
Quoted 2 lines

c) meta-virt issue (which you have a patch pending for in -next)

Unfortunately, I have to add

d) perf reproducibility issues:
https://autobuilder.yoctoproject.org/typhoon/#/builders/117/builds/2294/steps/13/logs/stdio

Ahrrg.

http://autobuilder.yocto.io/pub/repro-fail/oe-reproducible-20230126-poma_o8v/packages/diff-html/

/lib/modules/6.1.5-yocto-standard/build/include/generated/compile.h

from

#define·LINUX_COMPILE_HOST» "alma8-ty-2.yocto.io

to

#define·LINUX_COMPILE_HOST» "ubuntu2004-ty-1"

Bruce: Any idea if we can set that to some specific value?

We probably should start by fixing that and then see if that improves things...

Cheers,

Richard

#6 lore raw
Quoted 45 lines · Richard Purdie
Quoted 26 lines · Alexandre Belloni
Quoted 25 lines · Richard Purdie
Quoted 4 lines · On Tue, 2023-01-24 at 16:59 -0500, [email protected] wrote:

From: Bruce Ashfield <[email protected]>

Richard,

Quoted 8 lines

As promised, here are the commits to switch us over to 6.1 as the default.

I'll also send a series for poky to bump things there as well.

Also, if there are any significant problems, we'll just shelve this while I look at any issues. But at this point, it is worth getting the patches out there to move the needle to 6.1

Thanks!

I think the issues are:

a) two selftest failures:

oe-selftest -r devtool.DevtoolUpgradeTests.test_devtool_virtual_kernel_modify oe-selftest -r efibootpartition.GenericEFITest.test_boot_efi

b) meta-arm issues

Quoted 7 lines
Quoted 2 lines

c) meta-virt issue (which you have a patch pending for in -next)

Unfortunately, I have to add

d) perf reproducibility issues:
https://autobuilder.yoctoproject.org/typhoon/#/builders/117/builds/2294/steps/13/logs/stdio

Ahrrg.

http://autobuilder.yocto.io/pub/repro-fail/oe-reproducible-20230126-poma_o8v/packages/diff-html/

/lib/modules/6.1.5-yocto-standard/build/include/generated/compile.h

from

Quoted 8 lines

#define·LINUX_COMPILE_HOST» "alma8-ty-2.yocto.io

to

#define·LINUX_COMPILE_HOST» "ubuntu2004-ty-1"

Bruce: Any idea if we can set that to some specific value?

My perf build just finished and I was just starting to poke into it.

I'll have a look now.

Let me know if you come up with anything in the meantime, I'll do the same.

At the worst, I imagine we can sed the value to something consistent.

Bruce

Quoted 10 lines

We probably should start by fixing that and then see if that improves things...

Cheers,

Richard

Signature · 3 lines
- Thou shalt not follow the NULL pointer, for chaos and madness await
thee at its end
- "Use the force Harry" - Gandalf, Star Trek II
#7 lore raw
In reply to Richard Purdie
Quoted 47 lines · Richard Purdie
Quoted 26 lines · Alexandre Belloni
Quoted 25 lines · Richard Purdie
Quoted 4 lines · On Tue, 2023-01-24 at 16:59 -0500, [email protected] wrote:

From: Bruce Ashfield <[email protected]>

Richard,

Quoted 8 lines

As promised, here are the commits to switch us over to 6.1 as the default.

I'll also send a series for poky to bump things there as well.

Also, if there are any significant problems, we'll just shelve this while I look at any issues. But at this point, it is worth getting the patches out there to move the needle to 6.1

Thanks!

I think the issues are:

a) two selftest failures:

oe-selftest -r devtool.DevtoolUpgradeTests.test_devtool_virtual_kernel_modify oe-selftest -r efibootpartition.GenericEFITest.test_boot_efi

b) meta-arm issues

Quoted 7 lines
Quoted 2 lines

c) meta-virt issue (which you have a patch pending for in -next)

Unfortunately, I have to add

d) perf reproducibility issues:
https://autobuilder.yoctoproject.org/typhoon/#/builders/117/builds/2294/steps/13/logs/stdio

Ahrrg.

http://autobuilder.yocto.io/pub/repro-fail/oe-reproducible-20230126-poma_o8v/packages/diff-html/

/lib/modules/6.1.5-yocto-standard/build/include/generated/compile.h

from

#define·LINUX_COMPILE_HOST» "alma8-ty-2.yocto.io

Quoted 8 lines

to

#define·LINUX_COMPILE_HOST» "ubuntu2004-ty-1"

Bruce: Any idea if we can set that to some specific value?

We probably should start by fixing that and then see if that improves things...

Interesting. In my local builds:

#define LINUX_COMPILE_HOST "oe-host"

I'm building 6.1.7, but that shouldn't be the difference.

The value comes from: classes-recipe/kernel.bbclass:export KBUILD_BUILD_HOST ?= "oe-host",
and has been that way for a while.

I'm also grepping for other BUILDPATHS references, but the QA test should have caught
those if that was the issue.

Bruce

Quoted 8 lines

Cheers,

Richard

Signature · 3 lines
- Thou shalt not follow the NULL pointer, for chaos and madness await
thee at its end
- "Use the force Harry" - Gandalf, Star Trek II
#8 lore raw
Quoted 64 lines · Bruce Ashfield
Quoted 47 lines · Richard Purdie
Quoted 26 lines · Alexandre Belloni
Quoted 25 lines · Richard Purdie
Quoted 4 lines · On Tue, 2023-01-24 at 16:59 -0500, [email protected] wrote:

From: Bruce Ashfield <[email protected]>

Richard,

Quoted 8 lines

As promised, here are the commits to switch us over to 6.1 as the default.

I'll also send a series for poky to bump things there as well.

Also, if there are any significant problems, we'll just shelve this while I look at any issues. But at this point, it is worth getting the patches out there to move the needle to 6.1

Thanks!

I think the issues are:

a) two selftest failures:

oe-selftest -r devtool.DevtoolUpgradeTests.test_devtool_virtual_kernel_modify oe-selftest -r efibootpartition.GenericEFITest.test_boot_efi

b) meta-arm issues

Quoted 7 lines
Quoted 2 lines

c) meta-virt issue (which you have a patch pending for in -next)

Unfortunately, I have to add

d) perf reproducibility issues:
https://autobuilder.yoctoproject.org/typhoon/#/builders/117/builds/2294/steps/13/logs/stdio

Ahrrg.

http://autobuilder.yocto.io/pub/repro-fail/oe-reproducible-20230126-poma_o8v/packages/diff-html/

/lib/modules/6.1.5-yocto-standard/build/include/generated/compile.h

from

#define·LINUX_COMPILE_HOST» "alma8-ty-2.yocto.io

Quoted 8 lines

to

#define·LINUX_COMPILE_HOST» "ubuntu2004-ty-1"

Bruce: Any idea if we can set that to some specific value?

We probably should start by fixing that and then see if that improves things...

Interesting. In my local builds:

#define LINUX_COMPILE_HOST "oe-host"

I'm building 6.1.7, but that shouldn't be the difference.

Quoted 7 lines

The value comes from: classes-recipe/kernel.bbclass:export KBUILD_BUILD_HOST ?= "oe-host",
and has been that way for a while.

I'm also grepping for other BUILDPATHS references, but the QA test should have caught
those if that was the issue.

I just realised that there are two issues, one is perf, the other is kernel-devsrc. The compile.h issue is in kernel-devsrc and that did reproduce for my locally just now. I'm not sure what is wrong with perf...

Cheers,

Richard

#9 lore raw

we need to define

KBUILD_BUILD_USER="yocto"
KBUILD_BUILD_HOST="yocto"

in linux-yocto defconfigs

You can use my name too :)

Quoted 95 lines · Richard Purdie
Quoted 64 lines · Bruce Ashfield
Quoted 47 lines · Richard Purdie
Quoted 26 lines · Alexandre Belloni
Quoted 25 lines · Richard Purdie
Quoted 4 lines · On Tue, 2023-01-24 at 16:59 -0500, [email protected] wrote:

From: Bruce Ashfield <[email protected]>

Richard,

Quoted 8 lines

As promised, here are the commits to switch us over to 6.1 as the default.

I'll also send a series for poky to bump things there as well.

Also, if there are any significant problems, we'll just shelve this while I look at any issues. But at this point, it is worth getting the patches out there to move the needle to 6.1

Thanks!

I think the issues are:

a) two selftest failures:

oe-selftest -r devtool.DevtoolUpgradeTests.test_devtool_virtual_kernel_modify oe-selftest -r efibootpartition.GenericEFITest.test_boot_efi

b) meta-arm issues

Quoted 7 lines
Quoted 2 lines

c) meta-virt issue (which you have a patch pending for in -next)

Unfortunately, I have to add

d) perf reproducibility issues:
https://autobuilder.yoctoproject.org/typhoon/#/builders/117/builds/2294/steps/13/logs/stdio

Ahrrg.

http://autobuilder.yocto.io/pub/repro-fail/oe-reproducible-20230126-poma_o8v/packages/diff-html/

/lib/modules/6.1.5-yocto-standard/build/include/generated/compile.h

from

#define·LINUX_COMPILE_HOST» "alma8-ty-2.yocto.io

Quoted 8 lines

to

#define·LINUX_COMPILE_HOST» "ubuntu2004-ty-1"

Bruce: Any idea if we can set that to some specific value?

We probably should start by fixing that and then see if that improves things...

Interesting. In my local builds:

#define LINUX_COMPILE_HOST "oe-host"

I'm building 6.1.7, but that shouldn't be the difference.

Quoted 7 lines

The value comes from: classes-recipe/kernel.bbclass:export KBUILD_BUILD_HOST ?= "oe-host",
and has been that way for a while.

I'm also grepping for other BUILDPATHS references, but the QA test should have caught
those if that was the issue.

I just realised that there are two issues, one is perf, the other is kernel-devsrc. The compile.h issue is in kernel-devsrc and that did reproduce for my locally just now. I'm not sure what is wrong with perf...

Cheers,

Richard

-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#176414): https://lists.openembedded.org/g/openembedded-core/message/176414
Mute This Topic: https://lists.openembedded.org/mt/96508797/1997914
Group Owner: [email protected]
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub [[email protected]]
-=-=-=-=-=-=-=-=-=-=-=-

#10 lore raw
Quoted 8 lines · Khem Raj

we need to define

KBUILD_BUILD_USER="yocto"
KBUILD_BUILD_HOST="yocto"

in linux-yocto defconfigs

No, we really don't want to do that.

There's an environment variable that is supposed to be reflected in the build. Which we already set.

If that has changed in 6.1, then I'll come up with a patch as required.

Bruce

Quoted 99 lines

You can use my name too :)

Quoted 95 lines · Richard Purdie
Quoted 64 lines · Bruce Ashfield
Quoted 47 lines · Richard Purdie
Quoted 26 lines · Alexandre Belloni
Quoted 25 lines · Richard Purdie
Quoted 4 lines · On Tue, 2023-01-24 at 16:59 -0500, [email protected] wrote:

From: Bruce Ashfield <[email protected]>

Richard,

Quoted 8 lines

As promised, here are the commits to switch us over to 6.1 as the default.

I'll also send a series for poky to bump things there as well.

Also, if there are any significant problems, we'll just shelve this while I look at any issues. But at this point, it is worth getting the patches out there to move the needle to 6.1

Thanks!

I think the issues are:

a) two selftest failures:

oe-selftest -r devtool.DevtoolUpgradeTests.test_devtool_virtual_kernel_modify oe-selftest -r efibootpartition.GenericEFITest.test_boot_efi

b) meta-arm issues

Quoted 7 lines
Quoted 2 lines

c) meta-virt issue (which you have a patch pending for in -next)

Unfortunately, I have to add

d) perf reproducibility issues:
https://autobuilder.yoctoproject.org/typhoon/#/builders/117/builds/2294/steps/13/logs/stdio

Ahrrg.

http://autobuilder.yocto.io/pub/repro-fail/oe-reproducible-20230126-poma_o8v/packages/diff-html/

/lib/modules/6.1.5-yocto-standard/build/include/generated/compile.h

from

#define·LINUX_COMPILE_HOST» "alma8-ty-2.yocto.io

Quoted 8 lines

to

#define·LINUX_COMPILE_HOST» "ubuntu2004-ty-1"

Bruce: Any idea if we can set that to some specific value?

We probably should start by fixing that and then see if that improves things...

Interesting. In my local builds:

#define LINUX_COMPILE_HOST "oe-host"

I'm building 6.1.7, but that shouldn't be the difference.

Quoted 7 lines

The value comes from: classes-recipe/kernel.bbclass:export KBUILD_BUILD_HOST ?= "oe-host",
and has been that way for a while.

I'm also grepping for other BUILDPATHS references, but the QA test should have caught
those if that was the issue.

I just realised that there are two issues, one is perf, the other is kernel-devsrc. The compile.h issue is in kernel-devsrc and that did reproduce for my locally just now. I'm not sure what is wrong with perf...

Cheers,

Richard

-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#176414): https://lists.openembedded.org/g/openembedded-core/message/176414
Mute This Topic: https://lists.openembedded.org/mt/96508797/1997914
Group Owner: [email protected]
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub [[email protected]]
-=-=-=-=-=-=-=-=-=-=-=-

Signature · 3 lines
- Thou shalt not follow the NULL pointer, for chaos and madness await
thee at its end
- "Use the force Harry" - Gandalf, Star Trek II
#11 lore raw
Quoted 8 lines · Bruce Ashfield
Quoted 6 lines · Khem Raj

we need to define

KBUILD_BUILD_USER="yocto"
KBUILD_BUILD_HOST="yocto"

Quoted 7 lines
Quoted 2 lines

in linux-yocto defconfigs

No, we really don't want to do that.

There's an environment variable that is supposed to be reflected in the build. Which we already set.

a better location would be distro config defining it and derived distros can override it if needed.

Quoted 111 lines

If that has changed in 6.1, then I'll come up with a patch as required.

Bruce

Quoted 99 lines

You can use my name too :)

Quoted 95 lines · Richard Purdie
Quoted 64 lines · Bruce Ashfield
Quoted 47 lines · Richard Purdie
Quoted 26 lines · Alexandre Belloni
Quoted 25 lines · Richard Purdie
Quoted 4 lines · On Tue, 2023-01-24 at 16:59 -0500, [email protected] wrote:

From: Bruce Ashfield <[email protected]>

Richard,

Quoted 8 lines

As promised, here are the commits to switch us over to 6.1 as the default.

I'll also send a series for poky to bump things there as well.

Also, if there are any significant problems, we'll just shelve this while I look at any issues. But at this point, it is worth getting the patches out there to move the needle to 6.1

Thanks!

I think the issues are:

a) two selftest failures:

oe-selftest -r devtool.DevtoolUpgradeTests.test_devtool_virtual_kernel_modify oe-selftest -r efibootpartition.GenericEFITest.test_boot_efi

b) meta-arm issues

Quoted 7 lines
Quoted 2 lines

c) meta-virt issue (which you have a patch pending for in -next)

Unfortunately, I have to add

d) perf reproducibility issues:
https://autobuilder.yoctoproject.org/typhoon/#/builders/117/builds/2294/steps/13/logs/stdio

Ahrrg.

http://autobuilder.yocto.io/pub/repro-fail/oe-reproducible-20230126-poma_o8v/packages/diff-html/

/lib/modules/6.1.5-yocto-standard/build/include/generated/compile.h

from

#define·LINUX_COMPILE_HOST» "alma8-ty-2.yocto.io

Quoted 8 lines

to

#define·LINUX_COMPILE_HOST» "ubuntu2004-ty-1"

Bruce: Any idea if we can set that to some specific value?

We probably should start by fixing that and then see if that improves things...

Interesting. In my local builds:

#define LINUX_COMPILE_HOST "oe-host"

I'm building 6.1.7, but that shouldn't be the difference.

Quoted 7 lines

The value comes from: classes-recipe/kernel.bbclass:export KBUILD_BUILD_HOST ?= "oe-host",
and has been that way for a while.

I'm also grepping for other BUILDPATHS references, but the QA test should have caught
those if that was the issue.

I just realised that there are two issues, one is perf, the other is kernel-devsrc. The compile.h issue is in kernel-devsrc and that did reproduce for my locally just now. I'm not sure what is wrong with perf...

Cheers,

Richard

-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#176414): https://lists.openembedded.org/g/openembedded-core/message/176414
Mute This Topic: https://lists.openembedded.org/mt/96508797/1997914
Group Owner: [email protected]
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub [[email protected]]
-=-=-=-=-=-=-=-=-=-=-=-

--
- Thou shalt not follow the NULL pointer, for chaos and madness await thee at its end
- "Use the force Harry" - Gandalf, Star Trek II