Skip to contentSkip to conversation

Patch review

PATCH In review Yoann Congal 6 participants 30 messages Jul 18 – Jul 30

lore
Share & tools
lore
#0 lore raw

cover Patch review

Please review this set of changes for wrynose and have comments back by end of day Tuesday, July 21.

Passed a-full on autobuilder:
https://autobuilder.yoctoproject.org/valkyrie/#/builders/29/builds/4229
* qemux86-world and qemux86-64-ptest got impacted by 16316 – AB-INT: tar failure: got *at() syscall for unknown directory, fd 4 retried as

    https://autobuilder.yoctoproject.org/valkyrie/#/builders/73/builds/4062
    https://autobuilder.yoctoproject.org/valkyrie/#/builders/59/builds/4182

* qemuarmv5 got impacted by 16217 – AB-INT: boot hang in qemu (amba chip errors?) retried as https://autobuilder.yoctoproject.org/valkyrie/#/builders/80/builds/3992

The following changes since commit 6f5681afb324aed2cba4e5b819ffc878f74a2bac:

gnutls: fix CVE-2026-33846 (2026-07-16 20:54:12 +0200)

are available in the Git repository at:

https://git.openembedded.org/openembedded-core-contrib stable/wrynose-nut
https://git.openembedded.org/openembedded-core-contrib/log/?h=stable/wrynose-nut

for you to fetch changes up to 551eade6c83483616ea6172336a4cf6c994ce90d:

openssh: set status for CVE-2026-3497 (2026-07-18 01:31:03 +0200)

----------------------------------------------------------------

Adrian Freihofer (1):
oe-selftest: fitimage: Do not expect kernel property in DTBO config
subnodes

Alexander Kanavin (1):
wic: upgrade 0.3.0 -> 0.3.1

Deepak Rathore (1):
util-linux: fix CVE-2026-13595

Gustavo Henrique Nihei (1):
u-boot: re-enable RISC-V compressed (c) ISA extension

Jaipaul Cheernam (2):
gzip: fix CVE-2026-41992
bzip2: Fix CVE-2026-42250

Marek Vasut (2):
kernel-fit-image.bbclass: Do not include kernel property in DTBO
config subnodes
kernel-fit-image.bbclass: Fix operation with KERNEL_DTBVENDORED = "1"

Peter Marko (5):
go: upgrade 1.26.4 -> 1.26.5
vex: remove obsolete semicolon
rootfs: move tasks using image_list_installed_packages to
postuninstall
flac: fix buildpaths with doxygen enabled
flac: make documentation build deterministic

Ryan Eatmon (1):
kernel-fit-image: Add KERNEL_DTBVENDORED support for
FIT_CONF_DEFAULT_DTB

Sudhir Dumbhare (1):
openssh: set status for CVE-2026-3497

Vijay Anusuri (1):
tzdata/tzcode-native: upgrade 2026b -> 2026c

Yoann Congal (1):
scripts/install-buildtools: Update to 6.0.2

 meta/classes-recipe/kernel-fit-image.bbclass  |  22 ++-
 meta/classes-recipe/license_image.bbclass     |   2 +-
 meta/classes/vex.bbclass                      |   2 +-
 meta/lib/oe/fitimage.py                       |   7 +-
 meta/lib/oeqa/selftest/cases/fitimage.py      |   4 +-
 meta/recipes-bsp/u-boot/u-boot_2026.01.bb     |   1 +
 .../openssh/openssh_10.3p1.bb                 |   1 +
 meta/recipes-core/util-linux/util-linux.inc   |   1 +
 .../util-linux/CVE-2026-13595.patch           | 148 ++++++++++++++++++
 .../go/{go-1.26.4.inc => go-1.26.5.inc}       |   2 +-
 ...e_1.26.4.bb => go-binary-native_1.26.5.bb} |   6 +-
 ..._1.26.4.bb => go-cross-canadian_1.26.5.bb} |   0
 ...{go-cross_1.26.4.bb => go-cross_1.26.5.bb} |   0
 ...osssdk_1.26.4.bb => go-crosssdk_1.26.5.bb} |   0
 ...runtime_1.26.4.bb => go-runtime_1.26.5.bb} |   0
 .../go/{go_1.26.4.bb => go_1.26.5.bb}         |   0
 .../bzip2/bzip2/CVE-2026-42250.patch          |  35 +++++
 meta/recipes-extended/bzip2/bzip2_1.0.8.bb    |   1 +
 .../gzip/gzip-1.14/CVE-2026-41992.patch       |  64 ++++++++
 meta/recipes-extended/gzip/gzip_1.14.bb       |   1 +
 meta/recipes-extended/timezone/timezone.inc   |   6 +-
 meta/recipes-multimedia/flac/flac_1.5.0.bb    |   5 +-
 .../wic/{wic_0.3.0.bb => wic_0.3.1.bb}        |   2 +-
 scripts/install-buildtools                    |   4 +-
 24 files changed, 296 insertions(+), 18 deletions(-)
 create mode 100644 meta/recipes-core/util-linux/util-linux/CVE-2026-13595.patch
 rename meta/recipes-devtools/go/{go-1.26.4.inc => go-1.26.5.inc} (90%)
 rename meta/recipes-devtools/go/{go-binary-native_1.26.4.bb => go-binary-native_1.26.5.bb} (80%)
 rename meta/recipes-devtools/go/{go-cross-canadian_1.26.4.bb => go-cross-canadian_1.26.5.bb} (100%)
 rename meta/recipes-devtools/go/{go-cross_1.26.4.bb => go-cross_1.26.5.bb} (100%)
 rename meta/recipes-devtools/go/{go-crosssdk_1.26.4.bb => go-crosssdk_1.26.5.bb} (100%)
 rename meta/recipes-devtools/go/{go-runtime_1.26.4.bb => go-runtime_1.26.5.bb} (100%)
 rename meta/recipes-devtools/go/{go_1.26.4.bb => go_1.26.5.bb} (100%)
 create mode 100644 meta/recipes-extended/bzip2/bzip2/CVE-2026-42250.patch
 create mode 100644 meta/recipes-extended/gzip/gzip-1.14/CVE-2026-41992.patch
 rename meta/recipes-support/wic/{wic_0.3.0.bb => wic_0.3.1.bb} (90%)
#1 lore raw

PATCH 1/17 scripts/install-buildtools: Update to 6.0.2

From: Yoann Congal <[email protected]>

Update to the 6.0.2 release of the 6.0 series for buildtools

---
 scripts/install-buildtools | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
scripts/install-buildtools +2 -2
index f925dfa6b5c..b59cfbb8705 100755
@@ -57,8 +57,8 @@ logger = scriptutils.logger_create(PROGNAME, stream=sys.stdout)
 
 DEFAULT_INSTALL_DIR = os.path.join(os.path.split(scripts_path)[0],'buildtools')
 DEFAULT_BASE_URL = 'https://downloads.yoctoproject.org/releases/yocto'
-DEFAULT_RELEASE = 'yocto-6.0.1'
-DEFAULT_INSTALLER_VERSION = '6.0.1'
+DEFAULT_RELEASE = 'yocto-6.0.2'
+DEFAULT_INSTALLER_VERSION = '6.0.2'
 DEFAULT_BUILDDATE = '202110XX'
 
 # Python version sanity check
#2 lore raw

PATCH 2/17 oe-selftest: fitimage: Do not expect kernel property in DTBO config subnodes

In reply to Yoann Congal

From: Adrian Freihofer <[email protected]>

A DTBO overlay together with only a kernel cannot form a bootable result, so DTBO configuration subnodes should not include the kernel property. Update _get_req_its_fields() accordingly: skip appending 'kernel = "kernel-1"' for .dtbo files, aligning it with _get_req_sections() which already handles this correctly.

(cherry picked from commit 85e0408a81bdca0411a8625061e5e343c05f92d9)
---
 meta/lib/oeqa/selftest/cases/fitimage.py | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)
meta/lib/oeqa/selftest/cases/fitimage.py +3 -1
index 3541c07520a..e8435fb43c1 100644
@@ -659,7 +659,9 @@ class KernelFitImageBase(FitImageTestCase):
         dtb_files, dtb_symlinks = FitImageTestCase._get_dtb_files(bb_vars)
         if dtb_files:
             for dtb in dtb_files:
-                its_field_check.append('kernel = "kernel-1";')
+                # DTBO overlay configs do not include a kernel property
+                if not dtb.endswith('.dtbo'):
+                    its_field_check.append('kernel = "kernel-1";')
                 its_field_check.append('fdt = "fdt-%s";' % dtb)
             for dtb in dtb_symlinks:
                 its_field_check.append('kernel = "kernel-1";')
#3 lore raw

PATCH 3/17 kernel-fit-image.bbclass: Do not include kernel property in DTBO config subnodes

In reply to Yoann Congal

From: Marek Vasut <[email protected]>

The OE 5.0 implementation of fitImage generation did not include the kernel property in DTBO configuration node subnodes, while the current OE 6.0 implementation does. The previous implementation was correct, because kernel together with only a DTBO overlay can not form bootable result. The kernel property should only be part of the configuration subnode which includes both kernel and DTB. Fix this.

Fixes 05d0c7342d76 ("kernel-fit-image.bbclass: add a new FIT image implementation")
(cherry picked from commit 06ed34005957a6afb88270603df5e545941546b0)
---
 meta/lib/oe/fitimage.py | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)
meta/lib/oe/fitimage.py +6 -1
index 881d0eae0ab..d7e21171ab9 100644
@@ -425,8 +425,13 @@ class ItsNodeRootKernel(ItsNode):
         opt_props = {}
         conf_desc = []
         sign_entries = []
+        include_kernel = 1
 
-        if self._kernel:
+        # DTBO configuration sections do not include kernel property
+        if dtb and os.path.splitext(dtb.name)[1] == ".dtbo":
+            include_kernel = 0
+
+        if self._kernel and include_kernel:
             conf_desc.append("Linux kernel")
             opt_props["kernel"] = self._kernel.name
             if self._sign_enable:
#4 lore raw

PATCH 4/17 kernel-fit-image.bbclass: Fix operation with KERNEL_DTBVENDORED = "1"

In reply to Yoann Congal

From: Marek Vasut <[email protected]>

In the case KERNEL_DTBVENDORED = "1" , the fitImage configuration subnode names used to contain the vendor prefix in OE 5.0 , e.g.:
"

$ mkimage  -l fitImage
...
 Configuration 0 (conf-freescale_imx8mn-board.dtb)
                       ^^^^^^^^^^
...
"

With OE 6.0 this is no longer the case and KERNEL_DTBVENDORED = "1" and KERNEL_DTBVENDORED = "0" behave exactly the same way:
"

$ mkimage  -l fitImage
...
 Configuration 0 (conf-imx8mn-board.dtb)
...
"

Users depend on the correct configuration subnode names as those are used in the U-Boot "bootm" command when booting fitImages.

Fix this by converting each entry in the KERNEL_DEVICETREE list into the configuration subnode name by replacing all of its path separators with underscores.

Fixes 05d0c7342d76 ("kernel-fit-image.bbclass: add a new FIT image implementation")
(cherry picked from commit 4297b94c3728cd2e320d75b68c508e12ab127719)
---
 meta/classes-recipe/kernel-fit-image.bbclass | 9 ++++++++-
 1 file changed, 8 insertions(+), 1 deletion(-)
meta/classes-recipe/kernel-fit-image.bbclass +8 -1
index ae8f3c66883..448a88ccb11 100644
@@ -94,6 +94,13 @@ python do_compile() {
         for dtb in kernel_devicetree.split():
             # In deploy_dir the DTBs are without sub-directories also with KERNEL_DTBVENDORED = "1"
             dtb_name = os.path.basename(dtb)
+            # With vendored DTs, use the DT name as listed in KERNEL_DEVICETREE
+            # and replace any path separators with underscores, this behaves the
+            # same way as KERNEL_DTBVENDORED = "1" did in OE 5.0 and older.
+            if d.getVar('KERNEL_DTBVENDORED') != "1":
+                conf_name = dtb_name
+            else:
+                conf_name = dtb.replace('/', '_')
 
             # Skip DTB if it's also provided in EXTERNAL_KERNEL_DEVICETREE directory
             if external_kernel_devicetree:
@@ -103,7 +110,7 @@ python do_compile() {
 
             # Copy the dtb or dtbo file into the FIT image assembly directory
             shutil.copyfile(os.path.join(kernel_deploydir, dtb_name), dtb_name)
-            root_node.fitimage_emit_section_dtb(dtb_name, dtb_name,
+            root_node.fitimage_emit_section_dtb(conf_name, dtb_name,
                 d.getVar("UBOOT_DTB_LOADADDRESS"), d.getVar("UBOOT_DTBO_LOADADDRESS"))
 
     if external_kernel_devicetree:
#5 lore raw

PATCH 5/17 kernel-fit-image: Add KERNEL_DTBVENDORED support for FIT_CONF_DEFAULT_DTB

In reply to Yoann Congal

From: Ryan Eatmon <[email protected]>

When specifying a FIT_CONF_DEFAULT_DTB for a machine, you have to exactly align the name with what will be in the fitImage file or you will get a build error. If you also turn on KERNEL_DTBVENDORED then you must also specify the vendor directory as part of the dtb name that you want for the default, but you must manually do the same mapping that the kernel-fit-image class is doing when it generates the fit-image.its file.

This patch just adds the same logic to figure out the value for the requested default dtb and eliminate the need to understand the internal mapping of the class. It should make specifying the value more intuitive. The same value that you put in the KERNEL_DEVICETREE can be used in the FIT_CONF_DEFAULT_DTB and the new code will correctly honor the KERNEL_DTBVENDORED setting.

Before:

  KERNEL_DEVICETREE = "
    ti/k3-am62p5-sk.dtb \
    ... \
  "
  FIT_CONF_DEFAULT_DTB = "ti_k3-am62p5-sk.dtb"

After:

  KERNEL_DEVICETREE = "
    ti/k3-am62p5-sk.dtb \
    ... \
  "
  FIT_CONF_DEFAULT_DTB = "ti/k3-am62p5-sk.dtb"
(cherry picked from commit 3bceb2dabeee13c0a80ddd74ea7ae991606d6772)
---
 meta/classes-recipe/kernel-fit-image.bbclass | 21 +++++++++++++++-----
 1 file changed, 16 insertions(+), 5 deletions(-)
meta/classes-recipe/kernel-fit-image.bbclass +16 -5
index 448a88ccb11..cfe43a5a306 100644
@@ -55,6 +55,13 @@ FIT_KERNEL_SIGN_ENABLE ?= "${UBOOT_SIGN_ENABLE}"
 FIT_KERNEL_SIGN_KEYNAME ?= "${UBOOT_SIGN_KEYNAME}"
 FIT_KERNEL_SIGN_KEYDIR ?= "${UBOOT_SIGN_KEYDIR}"
 
+def apply_dtbvendored(d,dtb):
+    if d.getVar('KERNEL_DTBVENDORED') != "1":
+        mapped_name = os.path.basename(dtb)
+    else:
+        mapped_name = dtb.replace('/', '_')
+    return mapped_name
+
 python do_compile() {
     import shutil
     import oe.fitimage
@@ -97,10 +104,7 @@ python do_compile() {
             # With vendored DTs, use the DT name as listed in KERNEL_DEVICETREE
             # and replace any path separators with underscores, this behaves the
             # same way as KERNEL_DTBVENDORED = "1" did in OE 5.0 and older.
-            if d.getVar('KERNEL_DTBVENDORED') != "1":
-                conf_name = dtb_name
-            else:
-                conf_name = dtb.replace('/', '_')
+            conf_name = apply_dtbvendored(d, dtb)
 
             # Skip DTB if it's also provided in EXTERNAL_KERNEL_DEVICETREE directory
             if external_kernel_devicetree:
@@ -195,8 +199,15 @@ python do_compile() {
                                                  loadable_loadaddress,
                                                  loadable_entrypoint)
 
+    # Figure out if we have a default dtb and if we need to honor the
+    # KERNEL_DTBVENDORED variable to tweak the name to match what will be
+    # in the fitImage.
+    default_dtb = d.getVar("FIT_CONF_DEFAULT_DTB")
+    if default_dtb:
+        default_dtb = apply_dtbvendored(d, default_dtb)
+
     # Generate the configuration section
-    root_node.fitimage_emit_section_config(d.getVar("FIT_CONF_DEFAULT_DTB"), d.getVar("FIT_CONF_MAPPINGS"))
+    root_node.fitimage_emit_section_config(default_dtb, d.getVar("FIT_CONF_MAPPINGS"))
 
     # Write the its file
     root_node.write_its_file(itsfile)
#6 lore raw

Re: [OE-core][wrynose 05/17] kernel-fit-image: Add KERNEL_DTBVENDORED support for FIT_CONF_DEFAULT_DTB

Quoted 32 lines · Yoann Congal

From: Ryan Eatmon <[email protected]>

When specifying a FIT_CONF_DEFAULT_DTB for a machine, you have to exactly align the name with what will be in the fitImage file or you will get a build error. If you also turn on KERNEL_DTBVENDORED then you must also specify the vendor directory as part of the dtb name that you want for the default, but you must manually do the same mapping that the kernel-fit-image class is doing when it generates the fit-image.its file.

This patch just adds the same logic to figure out the value for the requested default dtb and eliminate the need to understand the internal mapping of the class. It should make specifying the value more intuitive. The same value that you put in the KERNEL_DEVICETREE can be used in the FIT_CONF_DEFAULT_DTB and the new code will correctly honor the KERNEL_DTBVENDORED setting.

Before:

  KERNEL_DEVICETREE = "
    ti/k3-am62p5-sk.dtb \
    ... \
  "
  FIT_CONF_DEFAULT_DTB = "ti_k3-am62p5-sk.dtb"

After:

  KERNEL_DEVICETREE = "
    ti/k3-am62p5-sk.dtb \
    ... \
  "
  FIT_CONF_DEFAULT_DTB = "ti/k3-am62p5-sk.dtb"
Quoted 5 lines
(cherry picked from commit 3bceb2dabeee13c0a80ddd74ea7ae991606d6772)

Hi Yoann,

This looks like an enhancement rather than a bug fix. If there is a strong reason to backport it though we could take it as it's small, self contained and not invasive.

Does the previous syntax (FIT_CONF_DEFAULT_DTB = "ti_k3-am62p5-sk.dtb") still work after this patch?

If we do take this, should we take the accompanying test case [1]?

[1]: https://lore.kernel.org/openembedded-core/[email protected]/

Best regards,

Signature · 1 line
Paul Barker
#7 lore raw

Re: [OE-core][wrynose 05/17] kernel-fit-image: Add KERNEL_DTBVENDORED support for FIT_CONF_DEFAULT_DTB

Quoted 39 lines · Paul Barker
Quoted 32 lines · Yoann Congal

From: Ryan Eatmon <[email protected]>

When specifying a FIT_CONF_DEFAULT_DTB for a machine, you have to exactly align the name with what will be in the fitImage file or you will get a build error. If you also turn on KERNEL_DTBVENDORED then you must also specify the vendor directory as part of the dtb name that you want for the default, but you must manually do the same mapping that the kernel-fit-image class is doing when it generates the fit-image.its file.

This patch just adds the same logic to figure out the value for the requested default dtb and eliminate the need to understand the internal mapping of the class. It should make specifying the value more intuitive. The same value that you put in the KERNEL_DEVICETREE can be used in the FIT_CONF_DEFAULT_DTB and the new code will correctly honor the KERNEL_DTBVENDORED setting.

Before:

  KERNEL_DEVICETREE = "
    ti/k3-am62p5-sk.dtb \
    ... \
  "
  FIT_CONF_DEFAULT_DTB = "ti_k3-am62p5-sk.dtb"

After:

  KERNEL_DEVICETREE = "
    ti/k3-am62p5-sk.dtb \
    ... \
  "
  FIT_CONF_DEFAULT_DTB = "ti/k3-am62p5-sk.dtb"
Quoted 5 lines
(cherry picked from commit 3bceb2dabeee13c0a80ddd74ea7ae991606d6772)
Quoted 5 lines

Hi Yoann,

This looks like an enhancement rather than a bug fix. If there is a strong reason to backport it though we could take it as it's small, self contained and not invasive.

I interpreted the mismatch between KERNEL_DEVICETREE and FIT_CONF_DEFAULT_DTB as a bug to solve.
But one could argue that each variable has its syntax, a bit weird but not worthy of a stable fix.

I still think this is worth fixing though.

Quoted 3 lines

Does the previous syntax (FIT_CONF_DEFAULT_DTB = "ti_k3-am62p5-sk.dtb") still work after this patch?

As I read it, the patch only add a .replace('/', '_') (when KERNEL_DTBVENDORED="1") so the old syntax should be preserved.

Quoted 3 lines

If we do take this, should we take the accompanying test case [1]?

[1]: https://lore.kernel.org/openembedded-core/[email protected]/

Yes, Ryan suggested it but it was not yet merged when I reviewed it.

Now it has merged, I think I should hold the series:
* oe-selftest: fitimage: Do not expect kernel property in DTBO config subnodes
* kernel-fit-image.bbclass: Do not include kernel property in DTBO config subnodes
* kernel-fit-image.bbclass: Fix operation with KERNEL_DTBVENDORED = "1"
* kernel-fit-image: Add KERNEL_DTBVENDORED support for FIT_CONF_DEFAULT_DTB ... until next wrynose review cycle, and add
* de2e11e6362 - oe-selftest: fitimage: Add tests for KERNEL_DTBVENDORED from master to test it.

Thanks!

Signature · 2 lines
Yoann Congal
Smile ECS
#8 lore raw

Re: [OE-core][wrynose 05/17] kernel-fit-image: Add KERNEL_DTBVENDORED support for FIT_CONF_DEFAULT_DTB

Quoted 66 lines · Yoann Congal
Quoted 39 lines · Paul Barker
Quoted 32 lines · Yoann Congal

From: Ryan Eatmon <[email protected]>

When specifying a FIT_CONF_DEFAULT_DTB for a machine, you have to exactly align the name with what will be in the fitImage file or you will get a build error. If you also turn on KERNEL_DTBVENDORED then you must also specify the vendor directory as part of the dtb name that you want for the default, but you must manually do the same mapping that the kernel-fit-image class is doing when it generates the fit-image.its file.

This patch just adds the same logic to figure out the value for the requested default dtb and eliminate the need to understand the internal mapping of the class. It should make specifying the value more intuitive. The same value that you put in the KERNEL_DEVICETREE can be used in the FIT_CONF_DEFAULT_DTB and the new code will correctly honor the KERNEL_DTBVENDORED setting.

Before:

  KERNEL_DEVICETREE = "
    ti/k3-am62p5-sk.dtb \
    ... \
  "
  FIT_CONF_DEFAULT_DTB = "ti_k3-am62p5-sk.dtb"

After:

  KERNEL_DEVICETREE = "
    ti/k3-am62p5-sk.dtb \
    ... \
  "
  FIT_CONF_DEFAULT_DTB = "ti/k3-am62p5-sk.dtb"
Quoted 5 lines
(cherry picked from commit 3bceb2dabeee13c0a80ddd74ea7ae991606d6772)
Quoted 5 lines

Hi Yoann,

This looks like an enhancement rather than a bug fix. If there is a strong reason to backport it though we could take it as it's small, self contained and not invasive.

I interpreted the mismatch between KERNEL_DEVICETREE and FIT_CONF_DEFAULT_DTB as a bug to solve.
But one could argue that each variable has its syntax, a bit weird but not worthy of a stable fix.

I still think this is worth fixing though.

Quoted 3 lines

Does the previous syntax (FIT_CONF_DEFAULT_DTB = "ti_k3-am62p5-sk.dtb") still work after this patch?

As I read it, the patch only add a .replace('/', '_') (when KERNEL_DTBVENDORED="1") so the old syntax should be preserved.

Quoted 3 lines

If we do take this, should we take the accompanying test case [1]?

[1]: https://lore.kernel.org/openembedded-core/[email protected]/

Yes, Ryan suggested it but it was not yet merged when I reviewed it.

Quoted 8 lines

Now it has merged, I think I should hold the series:
* oe-selftest: fitimage: Do not expect kernel property in DTBO config subnodes
* kernel-fit-image.bbclass: Do not include kernel property in DTBO config subnodes
* kernel-fit-image.bbclass: Fix operation with KERNEL_DTBVENDORED = "1"
* kernel-fit-image: Add KERNEL_DTBVENDORED support for FIT_CONF_DEFAULT_DTB ... until next wrynose review cycle, and add
* de2e11e6362 - oe-selftest: fitimage: Add tests for KERNEL_DTBVENDORED from master to test it.

To clarify, I expect a backport patch of "oe-selftest: fitimage: Add tests for KERNEL_DTBVENDORED" for wrynose from Ryan. The cherry-pick is not trivial.

Signature · 2 lines
Yoann Congal
Smile ECS
#9 lore raw

Re: [OE-core][wrynose 05/17] kernel-fit-image: Add KERNEL_DTBVENDORED support for FIT_CONF_DEFAULT_DTB

Hi Yoann

Just a confirmation from my side: Taking the 4 patches is the best we can do. Considering them as a series of bug fixes is right. Getting the test back-ported is of course nice to have.

Regards
Adrian

Quoted 11 lines

-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group. View/Reply Online (#241413):
https://lists.openembedded.org/g/openembedded-core/message/241413
Mute This Topic: https://lists.openembedded.org/mt/120329199/3616858
Group Owner: [email protected] Unsubscribe:
https://lists.openembedded.org/g/openembedded-core/unsub [
[email protected]]
-=-=-=-=-=-=-=-=-=-=-=-

#10 lore raw

Re: [OE-core][wrynose 05/17] kernel-fit-image: Add KERNEL_DTBVENDORED support for FIT_CONF_DEFAULT_DTB

Quoted 9 lines · Adrian

Hi Yoann

Just a confirmation from my side: Taking the 4 patches is the best we can do. Considering them as a series of bug fixes is right. Getting the test back-ported is of course nice to have.

Regards
Adrian

Backporting the test will be challenging as I built the test on top of the 7 patch series from you that overhauled a lot of the fitImage testing.

https://lore.kernel.org/openembedded-core/[email protected]/
https://lore.kernel.org/openembedded-core/[email protected]/
https://lore.kernel.org/openembedded-core/[email protected]/
https://lore.kernel.org/openembedded-core/[email protected]/
https://lore.kernel.org/openembedded-core/[email protected]/
https://lore.kernel.org/openembedded-core/[email protected]/

That means that my backport would need to be making a bunch of changes that these patches may have done. I can do that, if everyone wants me to. I know specifically that this patch is a big one that mine is making use of:

https://git.openembedded.org/openembedded-core/commit/meta/lib/oeqa/selftest/cases/fitimage.py?id=4b08cf6048b6575c39aa9b3bea282cdeae228e8a

Quoted 11 lines
Quoted 11 lines

-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group. View/Reply Online (#241413):
https://lists.openembedded.org/g/openembedded-core/message/241413
Mute This Topic: https://lists.openembedded.org/mt/120329199/3616858
Group Owner: [email protected] Unsubscribe:
https://lists.openembedded.org/g/openembedded-core/unsub [
[email protected]]
-=-=-=-=-=-=-=-=-=-=-=-

Signature · 3 lines
Ryan Eatmon                [email protected]
-----------------------------------------
Texas Instruments, Inc.  -  LCPD  -  MGTS
#11 lore raw

Re: [OE-core][wrynose 05/17] kernel-fit-image: Add KERNEL_DTBVENDORED support for FIT_CONF_DEFAULT_DTB

Quoted 28 lines · Ryan Eatmon via lists.openembedded.org
Quoted 11 lines · Adrian

Hi Yoann

Just a confirmation from my side: Taking the 4 patches is the best we
can do. Considering them as a series of bug fixes is right. Getting the
test back-ported is of course nice to have.

Regards
Adrian

Backporting the test will be challenging as I built the test on top of
the 7 patch series from you that overhauled a lot of the fitImage testing.

https://lore.kernel.org/openembedded-core/[email protected]/
https://lore.kernel.org/openembedded-core/[email protected]/
https://lore.kernel.org/openembedded-core/[email protected]/
https://lore.kernel.org/openembedded-core/[email protected]/
https://lore.kernel.org/openembedded-core/[email protected]/
https://lore.kernel.org/openembedded-core/[email protected]/

Quoted 9 lines

That means that my backport would need to be making a bunch of changes
that these patches may have done.  I can do that, if everyone wants me
to.  I know specifically that this patch is a big one that mine is making use of:

https://git.openembedded.org/openembedded-core/commit/meta/lib/oeqa/selftest/cases/fitimage.py?id=4b08cf6048b6575c39aa9b3bea282cdeae228e8a

I agree that this is not very easy. It would require a pretty wrynose specific patch. This is also sort of developing on stable.

On the existing tests on wrynose there is coverage for FIT_CONF_DEFAULT_DTB for DTB names without a path. Since the beaglebone is the reference, there is only
FIT_CONF_DEFAULT_DTB = "am335x-bonegreen.dtb"
https://git.openembedded.org/openembedded-core/tree/meta/lib/oeqa/selftest/cases/fitimage.py?h=wrynose#n983

Adding test coverage on master is much simpler because there we can also find a reference machine with a default DTB like "arm/foundation- v8.dtb"
https://git.openembedded.org/openembedded-core/tree/meta/lib/oeqa/selftest/cases/fitimage.py#n120

@Yoann: The missing detail is test coverage for default DTBs with a directory in the name. All the rest seems to be covered by the existing tests.

Alternative would be to go with only three patches. From a users perspective that would mean a breaking change with FIT_CONF_DEFAULT_DTB when updating from wrynose to a later version. With the four patches it will be future proof from now onward.

So my opinion is: Let's go with four patches and 99% coverage for wrynose. That's good enough.

Regards,
Adrian

Quoted 14 lines
Quoted 2 lines
Quoted 2 lines

-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group. View/Reply Online (#242366):
https://lists.openembedded.org/g/openembedded-core/message/242366
Mute This Topic: https://lists.openembedded.org/mt/120329199/4454582
Group Owner: [email protected] Unsubscribe:
https://lists.openembedded.org/g/openembedded-core/unsub [
[email protected]]
-=-=-=-=-=-=-=-=-=-=-=-

#12 lore raw

Re: [OE-core][wrynose 05/17] kernel-fit-image: Add KERNEL_DTBVENDORED support for FIT_CONF_DEFAULT_DTB

In reply to Yoann Congal

Hi Yoann

Just a confirmation from my side: Taking the 4 patches is the best we can do. Considering them as a series of bug fixes is right. Getting the test back-ported is of course nice to have.

Regards
Adrian

Quoted 11 lines

-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group. View/Reply Online (#241413):
https://lists.openembedded.org/g/openembedded-core/message/241413
Mute This Topic: https://lists.openembedded.org/mt/120329199/3616858
Group Owner: [email protected] Unsubscribe:
https://lists.openembedded.org/g/openembedded-core/unsub [
[email protected]]
-=-=-=-=-=-=-=-=-=-=-=-

#13 lore raw

Re: [OE-core][wrynose 05/17] kernel-fit-image: Add KERNEL_DTBVENDORED support for FIT_CONF_DEFAULT_DTB

In reply to Yoann Congal

Hi Yoann

Just a confirmation from my side: Taking the 4 patches is the best we can do. Considering them as a series of bug fixes is right. Getting the test back-ported is of course nice to have.

Regards
Adrian

Quoted 11 lines

-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group. View/Reply Online (#241413):
https://lists.openembedded.org/g/openembedded-core/message/241413
Mute This Topic: https://lists.openembedded.org/mt/120329199/3616858
Group Owner: [email protected] Unsubscribe:
https://lists.openembedded.org/g/openembedded-core/unsub [
[email protected]]
-=-=-=-=-=-=-=-=-=-=-=-

#14 lore raw

PATCH 6/17 gzip: fix CVE-2026-41992

In reply to Yoann Congal

From: Jaipaul Cheernam <[email protected]>

Backport upstream fix for a global buffer overflow in the LZH decompression logic (unlzh.c). The left[] and right[] global arrays shared across LZW and LZH decompression routines are not reinitialized between files processed in the same invocation, allowing an out-of-bounds read in the LZH decoder.

Adapted for gzip 1.14:
- Refreshed NEWS and THANKS hunks to match 1.14 release context.

Reference: https://nvd.nist.gov/vuln/detail/CVE-2026-41992

(cherry picked from commit deaaaacabbf8d21fb9271e3f6f83055893510cff)
---
 .../gzip/gzip-1.14/CVE-2026-41992.patch       | 64 +++++++++++++++++++
 meta/recipes-extended/gzip/gzip_1.14.bb       |  1 +
 2 files changed, 65 insertions(+)
 create mode 100644 meta/recipes-extended/gzip/gzip-1.14/CVE-2026-41992.patch
meta/recipes-extended/gzip/gzip-1.14/CVE-2026-41992.patch new file +64 -0
new file mode 100644
index 00000000000..f55c89978d6
@@ -0,0 +1,64 @@
+From 63dbf6b3b9e6e781df1a6a64e609b10e23969681 Mon Sep 17 00:00:00 2001
+From: Paul Eggert <[email protected]>
+Date: Wed, 15 Apr 2026 12:00:17 -0700
+Subject: =?UTF-8?q?gzip:=20don=E2=80=99t=20mishandle=20.lzh=20after=20.Z?=
+MIME-Version: 1.0
+Content-Type: text/plain; charset=UTF-8
+Content-Transfer-Encoding: 8bit
+
+Problem reported by Michał Majchrowicz.
+* unlzh.c (read_c_len): Clear left and right when n == 0.
+
+CVE: CVE-2026-41992
+Upstream-Status: Backport [https://cgit.git.savannah.gnu.org/cgit/gzip.git/commit/?id=63dbf6b3b9e6e781df1a6a64e609b10e23969681]
+Signed-off-by: Jaipaul Cheernam <[email protected]>
+---
+ NEWS    | 4 ++++
+ THANKS  | 1 +
+ unlzh.c | 6 ++++++
+ 3 files changed, 11 insertions(+)
+
+diff --git a/NEWS b/NEWS
+index 6388227..8fb8918 100644
+--- a/NEWS
++++ b/NEWS
+@@ -4,6 +4,10 @@ GNU gzip NEWS                                    -*- outline -*-
+ 
+ ** Bug fixes
+ 
++  A buffer overflow has been fixed when decompressing an .lzh file
++  after decompressing a .Z file.
++  [bug present since the beginning]
++
+   'gzip -d' no longer omits the last partial output buffer when the
+   input ends unexpectedly on an IBM Z platform.
+   [bug introduced in gzip-1.11]
+diff --git a/THANKS b/THANKS
+index 4e545d9..a7d25e4 100644
+--- a/THANKS
++++ b/THANKS
+@@ -186,6 +186,7 @@ Jamie Lokier            [email protected]
+ Richard Lloyd           [email protected]
+ David J. MacKenzie	[email protected]
+ John R MacMillan        [email protected]
++Michał Majchrowicz	[email protected]
+ Ron Male                [email protected]
+ Jakub Martisko		[email protected]
+ Don R. Maszle           [email protected]
+diff --git a/unlzh.c b/unlzh.c
+index 3320196..a6cf109 100644
+--- a/unlzh.c
++++ b/unlzh.c
+@@ -232,6 +232,12 @@ read_c_len ()
+         c = getbits(CBIT);
+         for (i = 0; i < NC; i++) c_len[i] = 0;
+         for (i = 0; i < 4096; i++) c_table[i] = c;
++
++        /* Needed in case LEFT and RIGHT are reused from a previous
++           LZW decompression.  It may be overkill to clear all of both
++           arrays, but nobody has had time to analyze this carefully.  */
++        memzero(left, (2 * NC - 1) * sizeof *left);
++        memzero(right, (2 * NC - 1) * sizeof *left);
+     } else {
+         i = 0;
+         while (i < n) {
meta/recipes-extended/gzip/gzip_1.14.bb +1 -0
index c7837cdae01..99174a58c53 100644
@@ -6,6 +6,7 @@ LICENSE = "GPL-3.0-or-later"
 
 SRC_URI = "${GNU_MIRROR}/gzip/${BP}.tar.gz \
            file://run-ptest \
+           file://CVE-2026-41992.patch \
            "
 SRC_URI:append:class-target = " file://wrong-path-fix.patch"
 
#15 lore raw

PATCH 7/17 tzdata/tzcode-native: upgrade 2026b -> 2026c

In reply to Yoann Congal

From: Vijay Anusuri <[email protected]>

This release contains the following changes:

Briefly:

     Alberta moved to permanent -06 on 2026-06-18.
     Morocco moves to permanent +00 on 2026-09-20.
     More integer overflow bugs have been fixed in zic.

Changes to future timestamps

Alberta’s 2026-03-08 spring forward was its last foreseeable clock change, as it moved to permanent -06 thereafter. (Thanks to Roozbeh Pournader and others.) Model this with its traditional abbreviation CST. Although the change to permanent -06 legally took place on 2026-06-18, temporarily model the change to occur on 2026-11-01 at 02:00 instead, for the same reason we introduced a similarly temporary hack for British Columbia in 2026b.

Although another TZDB release will likely be needed soon because Northwest Territories will likely follow Alberta, the legal formalities have not yet taken place.

Morocco plans to move back to permanent UTC, without daylight saving time transitions, on 2026-09-20 at 02:00. This also affects Western Sahara.

Changes to code

zic no longer overflows integers when processing outlandish input like ‘Zone Ouch 0 - LMT 9223372036854775807’, ‘Zone Ouch 0 2562047788015215 LMT’, ‘Zone Ouch -2562047788015215:30:08 - LMT’, and ‘Zone Ouch -2562047788015215:30:08 - %%z’. This avoids undefined behavior in C. (Problems reported by Naveed Khan.)

On platforms that have EFTYPE, tzalloc now fails with errno set to EFTYPE, not EINVAL, if it detects that the TZif file has an invalid format or is not a regular file. Formerly it did this only on NetBSD, and only when the file was not a regular file.

Unprivileged programs no longer require TZif files to be regular files or reject relative names containing ".." components. This reverts to the more-permissive 2025b behavior, as the stricter behavior did not catch on in FreeBSD.

zic now reports any failure to remove a temporary file when cleaning up after a previous failure. (Problem reported by Tom Lane.)

Changes to commentary

Northwest Territories is expected to move to permanent -06 prior to 2026-11-01 02:00, when clocks would otherwise fall back. (Thanks to Tim Parenti and James Bellaire.) Model this with its traditional abbreviation CST. Unfortunately the change is not yet official, so it is currently present only as comments that can be uncommented as needed.

Changes to build procedure

The undocumented ‘typecheck’ Makefile check rule has been removed. It stopped working in 2025a and evidently nobody noticed. The rule was superseded by ‘check_time_t_alternatives’ in 2013d.

Ref: https://lists.iana.org/hyperkitty/list/[email protected]/thread/NVHSX2PAQIT44U5FCCEVNJJYXQMMTJSA/

(cherry picked from commit 33a7e1170b0c8ba83cdb4c7d6d9f83f6c194baed)
---
 meta/recipes-extended/timezone/timezone.inc | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)
meta/recipes-extended/timezone/timezone.inc +3 -3
index ff02416ed3e..47f9ac2855f 100644
@@ -6,7 +6,7 @@ SECTION = "base"
 LICENSE = "PD & BSD-3-Clause"
 LIC_FILES_CHKSUM = "file://LICENSE;md5=c679c9d6b02bc2757b3eaf8f53c43fba"
 
-PV = "2026b"
+PV = "2026c"
 
 SRC_URI = "http://www.iana.org/time-zones/repository/releases/tzcode${PV}.tar.gz;name=tzcode;subdir=tz \
            http://www.iana.org/time-zones/repository/releases/tzdata${PV}.tar.gz;name=tzdata;subdir=tz \
@@ -16,5 +16,5 @@ S = "${UNPACKDIR}/tz"
 
 UPSTREAM_CHECK_URI = "http://www.iana.org/time-zones"
 
-SRC_URI[tzcode.sha256sum] = "37e9ed8427f5d3521c22fc58e293cbfb043d70eedf1003870b33f363f61ca344"
-SRC_URI[tzdata.sha256sum] = "114543d9f19a6bfeb5bca43686aea173d38755a3db1f2eec112647ae92c6f544"
+SRC_URI[tzcode.sha256sum] = "b1cffc3ace4c4c7cd0efba2f7add86ec3d0b79da48bcf03582671fd3c8feace8"
+SRC_URI[tzdata.sha256sum] = "e4a178a4477f3d0ea77cc31828ff72aa38feff8d61aa13e7e99e142e9d902be4"
#16 lore raw

PATCH 8/17 util-linux: fix CVE-2026-13595

In reply to Yoann Congal

From: Deepak Rathore <[email protected]>

This patch applies the upstream stable/v2.41 backport for CVE-2026-13595. The upstream fix merge or commit is referenced in [1], and the public CVE advisory is referenced in [2]. The individual backported commit links are recorded in the embedded patch headers when the fix expands to multiple commits.

[1] https://github.com/util-linux/util-linux/commit/132d9c8aa15a8efd0a23d8ca7ed8b98f365e84fa
[2] https://access.redhat.com/security/cve/CVE-2026-13595

---
 meta/recipes-core/util-linux/util-linux.inc   |   1 +
 .../util-linux/CVE-2026-13595.patch           | 148 ++++++++++++++++++
 2 files changed, 149 insertions(+)
 create mode 100644 meta/recipes-core/util-linux/util-linux/CVE-2026-13595.patch
meta/recipes-core/util-linux/util-linux.inc +1 -0
index 02358626669..2a994d7a386 100644
@@ -21,6 +21,7 @@ SRC_URI = "${KERNELORG_MIRROR}/linux/utils/util-linux/v${MAJOR_VERSION}/util-lin
            file://0001-ts-kill-decode-use-RTMIN-from-kill-L-instead-of-hard.patch \
            file://0001-tests-script-Disable-size-option-test.patch \
            file://0001-loopdev-add-LOOPDEV_FL_NOFOLLOW-to-prevent-symlink-a.patch \
+           file://CVE-2026-13595.patch \
            "
 
 SRC_URI[sha256sum] = "3330d873f0fceb5560b89a7dc14e4f3288bbd880e96903ed9b50ec2b5799e58b"
meta/recipes-core/util-linux/util-linux/CVE-2026-13595.patch new file +148 -0

148 lines not on this page · raw message

#17 lore raw

PATCH 9/17 bzip2: Fix CVE-2026-42250

In reply to Yoann Congal

From: Jaipaul Cheernam <[email protected]>

This patch applies the upstream fix as referenced in [1], using the commit shown in [2].

[1] https://nvd.nist.gov/vuln/detail/CVE-2026-42250
[2] https://sourceware.org/cgit/bzip2/commit/?id=35d122a3df8b0cc4082a4d89fdc6ee99f375fe67

(cherry picked from commit bf39a3c0497023e96de11444579ffef31f968bcd)
---
 .../bzip2/bzip2/CVE-2026-42250.patch          | 35 +++++++++++++++++++
 meta/recipes-extended/bzip2/bzip2_1.0.8.bb    |  1 +
 2 files changed, 36 insertions(+)
 create mode 100644 meta/recipes-extended/bzip2/bzip2/CVE-2026-42250.patch
meta/recipes-extended/bzip2/bzip2/CVE-2026-42250.patch new file +35 -0
new file mode 100644
index 00000000000..1679e744478
@@ -0,0 +1,35 @@
+From 71bf61d2e664c754ae5b1eb04018c8eaf5f99ae3 Mon Sep 17 00:00:00 2001
+From: Mark Wielaard <[email protected]>
+Date: Thu, 28 May 2026 16:15:45 +0200
+Subject: [PATCH] bzip2recover: Make sure to not process more than
+ BZ_MAX_HANDLED_BLOCKS
+
+There is an off-by-one in the check before calling tooManyBlocks. This
+causes the scanning loop to run one more time and cause a possible
+read or write one past the global bStart, bEnd, rbStart and rbEnd
+buffers. There are no known exploits of this issue and you will need
+to compile with something like gcc -fsanitize=address (ASAN
+AddressSanitizer) to observe the faulty read/write.
+
+This has been assigned CVE-2026-42250.
+
+CVE: CVE-2026-42250
+Upstream-Status: Backport [https://sourceware.org/cgit/bzip2/commit/?id=35d122a3df8b0cc4082a4d89fdc6ee99f375fe67]
+Signed-off-by: Jaipaul Cheernam <[email protected]>
+---
+ bzip2recover.c | 2 +-
+ 1 file changed, 1 insertion(+), 1 deletion(-)
+
+diff --git a/bzip2recover.c b/bzip2recover.c
+index a8131e0..4b1c219 100644
+--- a/bzip2recover.c
++++ b/bzip2recover.c
+@@ -402,7 +402,7 @@ Int32 main ( Int32 argc, Char** argv )
+             rbEnd[rbCtr] = bEnd[currBlock];
+             rbCtr++;
+          }
+-         if (currBlock >= BZ_MAX_HANDLED_BLOCKS)
++         if (currBlock >= BZ_MAX_HANDLED_BLOCKS - 1)
+             tooManyBlocks(BZ_MAX_HANDLED_BLOCKS);
+          currBlock++;
+ 
meta/recipes-extended/bzip2/bzip2_1.0.8.bb +1 -0
index 9cecf6a331b..36267e5073a 100644
@@ -27,6 +27,7 @@ SRC_URI = "https://sourceware.org/pub/${BPN}/${BPN}-${PV}.tar.gz \
            file://Makefile.am;subdir=${BP} \
            file://run-ptest \
            file://0001-fix-bzip2-version-tmp-aaa-will-hang.patch;subdir=${BP} \
+           file://CVE-2026-42250.patch;subdir=${BP} \
            "
 SRC_URI[sha256sum] = "ab5a03176ee106d3f0fa90e381da478ddae405918153cca248e682cd0c4a2269"
 
#18 lore raw

PATCH 10/17 u-boot: re-enable RISC-V compressed (c) ISA extension

In reply to Yoann Congal

From: Gustavo Henrique Nihei <[email protected]>

SRC_URI_RISCV clears every RISC-V ISA extension via u-boot-riscv-isa_clear.cfg, then conditionally re-adds a fragment per TUNE_FEATURES bit: a, f, d, b/zbb, zicbom. There is no line for c, even though u-boot-riscv-isa_c.cfg (CONFIG_RISCV_ISA_C=y) already ships in the recipe. So for any tune with c in TUNE_FEATURES (e.g. the default rv64gc/rv32gc tunes), compressed instructions stay disabled after the clear fragment runs, and .config ends up with:

# CONFIG_RISCV_ISA_C is not set

Building without RVC noticeably inflates .text: on qemuriscv64 (tune-riscv64, rv64gc), the resulting u-boot binary is 1312625 bytes of .text without the fix vs 1082853 bytes with CONFIG_RISCV_ISA_C=y correctly set, 229772 bytes (17.5%) smaller. On size-constrained RISC-V SPL targets this .text growth can make the SPL .bss VMA overlap the .text VMA and fail the link.

Add the missing "c" mapping line, mirroring the existing per-extension entries (a, f, d, b/zbb, zicbom) already present in SRC_URI_RISCV.

Tested on oe-core master, MACHINE=qemuriscv64, via the oe-nodistro-master bitbake-setup config:

  bitbake -c cleansstate u-boot && bitbake u-boot
  grep CONFIG_RISCV_ISA_C .../u-boot/2026.04/build/.config
  # before: "# CONFIG_RISCV_ISA_C is not set"
  # after:  "CONFIG_RISCV_ISA_C=y"

do_package_qa passes both before and after; only the ISA config and resulting .text size change.

Fixes cd9e7304481b ("u-boot: Overhaul UBOOT_CONFIG flow")

AI-Generated: Uses Cursor

(cherry picked from commit fef027f2350cec0e6b890b636d2529a13f1fab58)
---
 meta/recipes-bsp/u-boot/u-boot_2026.01.bb | 1 +
 1 file changed, 1 insertion(+)
meta/recipes-bsp/u-boot/u-boot_2026.01.bb +1 -0
index ac1b0b9b2b4..6d9bc126a16 100644
@@ -14,6 +14,7 @@ SRC_URI_RISCV = "\
     ${@bb.utils.contains    ("TUNE_FEATURES", "a",      "file://u-boot-riscv-isa_a.cfg", "", d)} \
     ${@bb.utils.contains    ("TUNE_FEATURES", "f",      "file://u-boot-riscv-isa_f.cfg", "", d)} \
     ${@bb.utils.contains    ("TUNE_FEATURES", "d",      "file://u-boot-riscv-isa_d.cfg", "", d)} \
+    ${@bb.utils.contains    ("TUNE_FEATURES", "c",      "file://u-boot-riscv-isa_c.cfg", "", d)} \
     ${@bb.utils.contains_any("TUNE_FEATURES", "b zbb",  "file://u-boot-riscv-isa_zbb.cfg", "", d)} \
     ${@bb.utils.contains    ("TUNE_FEATURES", "zicbom", "file://u-boot-riscv-isa_zicbom.cfg", "", d)} \
     "
#19 lore raw

Re: [OE-core][wrynose 10/17] u-boot: re-enable RISC-V compressed (c) ISA extension

Quoted 30 lines · Yoann Congal via lists.openembedded.org

From: Gustavo Henrique Nihei <[email protected]>

SRC_URI_RISCV clears every RISC-V ISA extension via u-boot-riscv-isa_clear.cfg, then conditionally re-adds a fragment per TUNE_FEATURES bit: a, f, d, b/zbb, zicbom. There is no line for c, even though u-boot-riscv-isa_c.cfg (CONFIG_RISCV_ISA_C=y) already ships in the recipe. So for any tune with c in TUNE_FEATURES (e.g. the default rv64gc/rv32gc tunes), compressed instructions stay disabled after the clear fragment runs, and .config ends up with:

# CONFIG_RISCV_ISA_C is not set

Building without RVC noticeably inflates .text: on qemuriscv64 (tune-riscv64, rv64gc), the resulting u-boot binary is 1312625 bytes of .text without the fix vs 1082853 bytes with CONFIG_RISCV_ISA_C=y correctly set, 229772 bytes (17.5%) smaller. On size-constrained RISC-V SPL targets this .text growth can make the SPL .bss VMA overlap the .text VMA and fail the link.

Add the missing "c" mapping line, mirroring the existing per-extension entries (a, f, d, b/zbb, zicbom) already present in SRC_URI_RISCV.

Tested on oe-core master, MACHINE=qemuriscv64, via the oe-nodistro-master bitbake-setup config:

  bitbake -c cleansstate u-boot && bitbake u-boot
  grep CONFIG_RISCV_ISA_C .../u-boot/2026.04/build/.config
  # before: "# CONFIG_RISCV_ISA_C is not set"
  # after:  "CONFIG_RISCV_ISA_C=y"
Quoted 4 lines

do_package_qa passes both before and after; only the ISA config and resulting .text size change.

Fixes cd9e7304481b ("u-boot: Overhaul UBOOT_CONFIG flow")

Should this have been de890297b392 ("u-boot: Dynamic RISC-V ISA configuration")?

Best regards,

Signature · 1 line
Paul Barker
#20 lore raw

Re: [OE-core][wrynose 10/17] u-boot: re-enable RISC-V compressed (c) ISA extension

You're right, that was my mistake.
Should I send a v2 of this backport with the corrected "Fixes:" tag? The already-merged master commit (fef027f2350c) carries the same wrong tag. Let me know if you'd like a follow-up fixup there too, or if it's fine to leave as-is.

Best regards,
Gustavo.
________________________________
From: Paul Barker <[email protected]>
Sent: Monday, July 20, 2026 04:48
To: [email protected] <[email protected]>; [email protected] <[email protected]>
Cc: Gustavo Nihei <[email protected]>
Subject: Re: [OE-core][wrynose 10/17] u-boot: re-enable RISC-V compressed (c) ISA extension

[You don't often get email from [email protected]. Learn why this is important at https://aka.ms/LearnAboutSenderIdentification ]

[External: This email originated outside Espressif]

#21 lore raw

Re: [OE-core][wrynose 10/17] u-boot: re-enable RISC-V compressed (c) ISA extension

Quoted 2 lines · Gustavo Nihei

You're right, that was my mistake.
Should I send a v2 of this backport with the corrected "Fixes:" tag?

Yes, please, send a v2 for wrynose :)

Quoted 2 lines

The already-merged master commit (fef027f2350c) carries the same wrong tag. Let me know if you'd like a follow-up fixup there too, or if it's fine to leave as-is.

We won't be able to change the commit in master so nothing to do there.

Quoted 13 lines

Best regards,
Gustavo.
________________________________
From: Paul Barker <[email protected]>
Sent: Monday, July 20, 2026 04:48
To: [email protected] <[email protected]>; [email protected] <[email protected]>
Cc: Gustavo Nihei <[email protected]>
Subject: Re: [OE-core][wrynose 10/17] u-boot: re-enable RISC-V compressed (c) ISA extension

[You don't often get email from [email protected]. Learn why this is important at https://aka.ms/LearnAboutSenderIdentification ]

[External: This email originated outside Espressif]

Signature · 2 lines
Yoann Congal
Smile ECS
#22 lore raw

Re: [OE-core][wrynose 10/17] u-boot: re-enable RISC-V compressed (c) ISA extension

Hi Yoann,

Thanks for the quick feedback.
I've just sent a v2 patch: https://lists.openembedded.org/g/openembedded-core/message/241393

Best regards,
Gustavo.
________________________________
From: [email protected] <[email protected]> on behalf of Yoann Congal via lists.openembedded.org <[email protected]>
Sent: Monday, July 20, 2026 09:07
To: Gustavo Nihei <[email protected]>; Paul Barker <[email protected]>; [email protected] <[email protected]>
Subject: Re: [OE-core][wrynose 10/17] u-boot: re-enable RISC-V compressed (c) ISA extension

[External: This email originated outside Espressif]

Quoted 2 lines · Gustavo Nihei

You're right, that was my mistake.
Should I send a v2 of this backport with the corrected "Fixes:" tag?

Yes, please, send a v2 for wrynose :)

Quoted 2 lines

The already-merged master commit (fef027f2350c) carries the same wrong tag. Let me know if you'd like a follow-up fixup there too, or if it's fine to leave as-is.

We won't be able to change the commit in master so nothing to do there.

Quoted 10 lines

Best regards,
Gustavo.
________________________________
From: Paul Barker <[email protected]>
Sent: Monday, July 20, 2026 04:48
To: [email protected] <[email protected]>; [email protected] <[email protected]>
Cc: Gustavo Nihei <[email protected]>
Subject: Re: [OE-core][wrynose 10/17] u-boot: re-enable RISC-V compressed (c) ISA extension

Quoted 3 lines

[You don't often get email from [email protected]. Learn why this is important at https://aka.ms/LearnAboutSenderIdentification ]

[External: This email originated outside Espressif]

--
Yoann Congal
Smile ECS

#23 lore raw

PATCH 11/17 go: upgrade 1.26.4 -> 1.26.5

In reply to Yoann Congal

From: Peter Marko <[email protected]>

Upgrade to latest 1.26.x release [1]:

$ git --no-pager log --oneline go1.26.4..go1.26.5
c19862e5f8 (tag: go1.26.5) [release-branch.go1.26] go1.26.5
f9ef7f5598 [release-branch.go1.26] os: properly handle trailing slashes in paths in Root
ca8ca590cc [release-branch.go1.26] crypto/tls: omit PSK in ECH outer client hello
7397dd7495 [release-branch.go1.26] net: fix TestLookupCNAME
044c995af7 [release-branch.go1.26] runtime: tolerate vendor suffixes in Linux kernel release strings
814591410b [release-branch.go1.26] cmd/compile,sync/atomic: make Add And & Or SQCST on PPC64
2055b1a15d [release-branch.go1.26] syscall: add //go:norace to rawSyscall on darwin
d7c9c818f9 [release-branch.go1.26] runtime: in moveSliceNoCap, round size down to a whole multiple of element size
dc80258a97 [release-branch.go1.26] internal/runtime/gc: require AVX512DQ for greentea
6efcff6602 [release-branch.go1.26] os/signal: make NotifyContext Cause match context.Canceled
fe91cea4f2 [release-branch.go1.26] cmd/go: use fsys.ReadDir for IsStandardPackage
ec988a9af1 [release-branch.go1.26] os: notify testlog from (*File).Chdir

Fixes CVE-2026-39822 and CVE-2026-42505.

Release information: [2]

[1] https://github.com/golang/go/compare/go1.26.4...go1.26.5
[2] https://groups.google.com/g/golang-announce/c/OrmQE_Yp5Sc

(From OE-Core rev: b498f6d6d93e0adb3f0e9439aa55a2abfcb9b063)

---
 meta/recipes-devtools/go/{go-1.26.4.inc => go-1.26.5.inc}   | 2 +-
 ...o-binary-native_1.26.4.bb => go-binary-native_1.26.5.bb} | 6 +++---
 ...cross-canadian_1.26.4.bb => go-cross-canadian_1.26.5.bb} | 0
 .../go/{go-cross_1.26.4.bb => go-cross_1.26.5.bb}           | 0
 .../go/{go-crosssdk_1.26.4.bb => go-crosssdk_1.26.5.bb}     | 0
 .../go/{go-runtime_1.26.4.bb => go-runtime_1.26.5.bb}       | 0
 meta/recipes-devtools/go/{go_1.26.4.bb => go_1.26.5.bb}     | 0
 7 files changed, 4 insertions(+), 4 deletions(-)
 rename meta/recipes-devtools/go/{go-1.26.4.inc => go-1.26.5.inc} (90%)
 rename meta/recipes-devtools/go/{go-binary-native_1.26.4.bb => go-binary-native_1.26.5.bb} (80%)
 rename meta/recipes-devtools/go/{go-cross-canadian_1.26.4.bb => go-cross-canadian_1.26.5.bb} (100%)
 rename meta/recipes-devtools/go/{go-cross_1.26.4.bb => go-cross_1.26.5.bb} (100%)
 rename meta/recipes-devtools/go/{go-crosssdk_1.26.4.bb => go-crosssdk_1.26.5.bb} (100%)
 rename meta/recipes-devtools/go/{go-runtime_1.26.4.bb => go-runtime_1.26.5.bb} (100%)
 rename meta/recipes-devtools/go/{go_1.26.4.bb => go_1.26.5.bb} (100%)
meta/recipes-devtools/go/go-1.26.5.inc renamed from meta/recipes-devtools/go/go-1.26.4.inc +1 -1
similarity index 90%
rename from meta/recipes-devtools/go/go-1.26.4.inc
rename to meta/recipes-devtools/go/go-1.26.5.inc
index 8bc6f4d4def..a4302b2790a 100644
@@ -16,4 +16,4 @@ SRC_URI += "\
     file://0009-go-Filter-build-paths-on-staticly-linked-arches.patch \
     file://0010-cmd-go-clear-GOROOT-for-func-ldShared-when-trimpath-.patch \
 "
-SRC_URI[main.sha256sum] = "4f668a32fbfc1132e6a881fb968c2f1dada631492a339211735fbb255a42602d"
+SRC_URI[main.sha256sum] = "495be4bc87176ac567392e5b4116abd98466d33d7b49d41e764ccc6976b2dc42"
meta/recipes-devtools/go/go-binary-native_1.26.5.bb renamed from meta/recipes-devtools/go/go-binary-native_1.26.4.bb +3 -3
similarity index 80%
rename from meta/recipes-devtools/go/go-binary-native_1.26.4.bb
rename to meta/recipes-devtools/go/go-binary-native_1.26.5.bb
index 3757b9f6a52..97a8270be2d 100644
@@ -9,9 +9,9 @@ PROVIDES = "go-native"
 
 # Checksums available at https://go.dev/dl/
 SRC_URI = "https://dl.google.com/go/go${PV}.${BUILD_GOOS}-${BUILD_GOARCH}.tar.gz;name=go_${BUILD_GOTUPLE}"
-SRC_URI[go_linux_amd64.sha256sum] = "1153d3d50e0ac764b447adfe05c2bcf08e889d42a02e0fe0259bd47f6733ad7f"
-SRC_URI[go_linux_arm64.sha256sum] = "ef758ae7c6cf9267c9c0ef080b8965f453d89ab2d25d9eb22de4405925238768"
-SRC_URI[go_linux_ppc64le.sha256sum] = "53f49b8c7eace2d30389327b4a516b13321f90377fdf5929a6b63174609bc22e"
+SRC_URI[go_linux_amd64.sha256sum] = "5c2c3b16caefa1d968a94c1daca04a7ca301a496d9b086e17ad77bb81393f053"
+SRC_URI[go_linux_arm64.sha256sum] = "fe4789e92b1f33358680864bbe8704289e7bb5fc207d80623c308935bd696d49"
+SRC_URI[go_linux_ppc64le.sha256sum] = "c5d60e2b303bb612f20cd82786594b64874e73b35134025e27d3390bf284ae43"
 
 UPSTREAM_CHECK_URI = "https://golang.org/dl/"
 UPSTREAM_CHECK_REGEX = "go(?P<pver>\d+(\.\d+)+)\.linux"
meta/recipes-devtools/go/go-cross-canadian_1.26.5.bb renamed from meta/recipes-devtools/go/go-cross-canadian_1.26.4.bb +0 -0
similarity index 100%
rename from meta/recipes-devtools/go/go-cross-canadian_1.26.4.bb
rename to meta/recipes-devtools/go/go-cross-canadian_1.26.5.bb
meta/recipes-devtools/go/go-cross_1.26.5.bb renamed from meta/recipes-devtools/go/go-cross_1.26.4.bb +0 -0
similarity index 100%
rename from meta/recipes-devtools/go/go-cross_1.26.4.bb
rename to meta/recipes-devtools/go/go-cross_1.26.5.bb
meta/recipes-devtools/go/go-crosssdk_1.26.5.bb renamed from meta/recipes-devtools/go/go-crosssdk_1.26.4.bb +0 -0
similarity index 100%
rename from meta/recipes-devtools/go/go-crosssdk_1.26.4.bb
rename to meta/recipes-devtools/go/go-crosssdk_1.26.5.bb
meta/recipes-devtools/go/go-runtime_1.26.5.bb renamed from meta/recipes-devtools/go/go-runtime_1.26.4.bb +0 -0
similarity index 100%
rename from meta/recipes-devtools/go/go-runtime_1.26.4.bb
rename to meta/recipes-devtools/go/go-runtime_1.26.5.bb
meta/recipes-devtools/go/go_1.26.5.bb renamed from meta/recipes-devtools/go/go_1.26.4.bb +0 -0
similarity index 100%
rename from meta/recipes-devtools/go/go_1.26.4.bb
rename to meta/recipes-devtools/go/go_1.26.5.bb
#24 lore raw

PATCH 12/17 vex: remove obsolete semicolon

In reply to Yoann Congal

From: Peter Marko <[email protected]>

Usage of semicolon as separator in ROOTFS/IMAGE_*COMMAND was deprecated long time ago.
Remove it.

(From OE-Core rev: 311d418d22a609fb54b87bfc909bdd1861892228)

---
 meta/classes/vex.bbclass | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
meta/classes/vex.bbclass +1 -1
index c57b8209c23..4c36cab5815 100644
@@ -211,7 +211,7 @@ python vex_write_rootfs_manifest () {
     bb.plain("Image VEX JSON report stored in: %s" % manifest_name)
 }
 
-ROOTFS_POSTPROCESS_COMMAND:prepend = "vex_write_rootfs_manifest; "
+ROOTFS_POSTPROCESS_COMMAND:prepend = "vex_write_rootfs_manifest "
 do_rootfs[recrdeptask] += "do_generate_vex "
 do_populate_sdk[recrdeptask] += "do_generate_vex "
 
#25 lore raw

PATCH 13/17 rootfs: move tasks using image_list_installed_packages to postuninstall

In reply to Yoann Congal

From: Peter Marko <[email protected]>

Since some packages can be uninstalled, any task querying installed packages should be run only after both installation and uninstallation is completed.

(From OE-Core rev: c3097962ac925538e99b17b771c541950a8b8c26)

---
 meta/classes-recipe/license_image.bbclass | 2 +-
 meta/classes/vex.bbclass                  | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)
meta/classes-recipe/license_image.bbclass +1 -1
index 95e0a9ba2e0..1ecd8f88bc7 100644
@@ -353,7 +353,7 @@ def get_deployed_files(man_file):
             dep_files.append(os.path.basename(f))
     return dep_files
 
-ROOTFS_POSTPROCESS_COMMAND:prepend = "write_package_manifest license_create_manifest "
+ROOTFS_POSTUNINSTALL_COMMAND:prepend = "write_package_manifest license_create_manifest "
 POPULATE_SDK_POST_HOST_COMMAND:prepend = "write_sdk_host_package_manifest license_create_sdk_host_manifest "
 POPULATE_SDK_POST_TARGET_COMMAND:prepend = "write_sdk_target_package_manifest license_create_sdk_target_manifest "
 do_rootfs[recrdeptask] += "do_populate_lic"
meta/classes/vex.bbclass +1 -1
index 4c36cab5815..71be705fc10 100644
@@ -211,7 +211,7 @@ python vex_write_rootfs_manifest () {
     bb.plain("Image VEX JSON report stored in: %s" % manifest_name)
 }
 
-ROOTFS_POSTPROCESS_COMMAND:prepend = "vex_write_rootfs_manifest "
+ROOTFS_POSTUNINSTALL_COMMAND:prepend = "vex_write_rootfs_manifest "
 do_rootfs[recrdeptask] += "do_generate_vex "
 do_populate_sdk[recrdeptask] += "do_generate_vex "
 
#26 lore raw

PATCH 14/17 flac: fix buildpaths with doxygen enabled

In reply to Yoann Congal

From: Peter Marko <[email protected]>

When documentation is built, buildpaths errors occur. There is one path to S and one to UNPACKDIR. We can have a single sed rule to fix both.

(From OE-Core rev: 9905d3d66c13045e2c422a36cafc8ce11db37c01)

---
 meta/recipes-multimedia/flac/flac_1.5.0.bb | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
meta/recipes-multimedia/flac/flac_1.5.0.bb +2 -2
index 2f43b4e84b2..a667ea9f625 100644
@@ -44,6 +44,6 @@ FILES:libflac++ = "${libdir}/libFLAC++.so.*"
 do_install:append() {
     # make the links in documentation relative to avoid buildpaths reproducibility problem
     sed -i "s#${S}/include#${includedir}#g" ${D}${docdir}/flac/FLAC.tag ${D}${docdir}/flac/api/*.html
-    # there is also one root path without trailing slash
-    sed -i "s#${S}#/#g" ${D}${docdir}/flac/api/*.html
+    # there are also some root paths in API titles
+    sed -i "s#${UNPACKDIR}#sources#g" ${D}${docdir}/flac/api/*.html
 }
#27 lore raw

PATCH 15/17 flac: make documentation build deterministic

In reply to Yoann Congal

From: Peter Marko <[email protected]>

doxygen is autodetected (e.g. from HOSTTOOLS or a transient dependency). Add a knob to make it deterministic when the documentation is being built. Keep it disabled by default to preserve current default behavior.

(From OE-Core rev: facd4a7561f9e4c8b4b36329115634a586a0f7e1)

---
 meta/recipes-multimedia/flac/flac_1.5.0.bb | 1 +
 1 file changed, 1 insertion(+)
meta/recipes-multimedia/flac/flac_1.5.0.bb +1 -0
index a667ea9f625..d55e157a9d1 100644
@@ -34,6 +34,7 @@ PACKAGECONFIG ??= " \
     ogg \
 "
 PACKAGECONFIG[avx] = "--enable-avx,--disable-avx"
+PACKAGECONFIG[doxygen-docs] = "--enable-doxygen-docs,--disable-doxygen-docs,doxygen-native"
 PACKAGECONFIG[ogg] = "--enable-ogg --with-ogg=${STAGING_DIR_HOST},--disable-ogg,libogg"
 
 PACKAGES += "libflac libflac++"
#28 lore raw

PATCH 16/17 wic: upgrade 0.3.0 -> 0.3.1

In reply to Yoann Congal

From: Alexander Kanavin <[email protected]>

(cherry picked from commit b24c4f46bccba16eb6fb2b494d351644704b5371)
[YC: Changelog URL: https://git.yoctoproject.org/wic/log/?h=v0.3.1]
---
 meta/recipes-support/wic/{wic_0.3.0.bb => wic_0.3.1.bb} | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
 rename meta/recipes-support/wic/{wic_0.3.0.bb => wic_0.3.1.bb} (90%)
meta/recipes-support/wic/wic_0.3.1.bb renamed from meta/recipes-support/wic/wic_0.3.0.bb +1 -1
similarity index 90%
rename from meta/recipes-support/wic/wic_0.3.0.bb
rename to meta/recipes-support/wic/wic_0.3.1.bb
index 7dbf84b039a..d9b4cc05c4b 100644
@@ -4,7 +4,7 @@ LICENSE = "GPL-2.0-only"
 LIC_FILES_CHKSUM = "file://LICENSE;md5=4ee23c52855c222cba72583d301d2338"
 
 SRC_URI = "git://git.yoctoproject.org/wic.git;branch=master;protocol=https;tag=v${PV}"
-SRCREV = "5974ade11032f218841d9f449ef0efeee3f9a2ca"
+SRCREV = "6c66da65bf2aac618092811c13792dcbd453d140"
 
 CVE_PRODUCT = "yoctoproject:wic"
 
#29 lore raw

PATCH 17/17 openssh: set status for CVE-2026-3497

In reply to Yoann Congal

From: Sudhir Dumbhare <[email protected]>

Analysis:
- CVE-2026-3497 affects downstream OpenSSH GSSAPI Key Exchange patches.
- The vulnerable code uses sshpkt_disconnect() in the GSSAPI KEX server path.
- Upstream OpenSSH/OE-Core does not carry the vulnerable GSSAPI key-exchange delta.
- Hence ignoring the CVE for this version.

Reference:
https://nvd.nist.gov/vuln/detail/CVE-2026-3497
https://github.com/advisories/ghsa-wcpp-3x59-h8vp
https://ubuntu.com/security/CVE-2026-3497
https://security-tracker.debian.org/tracker/CVE-2026-3497
https://www.openwall.com/lists/oss-security/2026/03/12/3

(cherry picked from commit c2bd43b373d65d717e606cab3793b8a64facd946)
---
 meta/recipes-connectivity/openssh/openssh_10.3p1.bb | 1 +
 1 file changed, 1 insertion(+)
meta/recipes-connectivity/openssh/openssh_10.3p1.bb +1 -0
index a050475532b..8669d080b6e 100644
@@ -36,6 +36,7 @@ Red Hat Enterprise Linux 7 and when running in a Kerberos environment"
 
 CVE_STATUS[CVE-2008-3844] = "not-applicable-platform: Only applies to some distributed RHEL binaries."
 CVE_STATUS[CVE-2023-51767] = "upstream-wontfix: It was demonstrated on modified sshd and does not exist in upstream openssh https://bugzilla.mindrot.org/show_bug.cgi?id=3656#c1."
+CVE_STATUS[CVE-2026-3497] = "not-applicable-platform: Only affects GSSAPI Key Exchange patches used by some Linux distributions and does not exist in upstream openssh."
 
 PAM_SRC_URI = "file://sshd"