* [Buildroot] [PATCH v2 1/1] toolchain/toolchain-external: introduce BR2_TOOLCHAIN_EXTERNAL_UPDATE_LINUX_HEADERS
@ 2025-12-27 21:14 Rustam Abdullaev
2025-12-27 22:24 ` Thomas Petazzoni via buildroot
0 siblings, 1 reply; 6+ messages in thread
From: Rustam Abdullaev @ 2025-12-27 21:14 UTC (permalink / raw)
To: buildroot
Cc: Andreas Ziegler, Giulio Benetti, Romain Naour, Thomas Petazzoni,
Rustam Abdullaev
This change makes it possible to update kernel headers in the external
toolchain with the headers from the current kernel.
It is important to have matching kernel headers when building packages
that make use of kernel features such as seccomp, as they depend on
system call definitions in linux/unistd.h. A typical example of this is
openssh - building it with the official ARM GNU toolchain and kernel 6.x
results in a non-working sshd.
Currently restricted to glibc-based toolchains as glibc is backward- and
forward-compatible with different kernel versions while uClibc and musl
are not.
Signed-off-by: Rustam Abdullaev <rustamabd@gmail.com>
---
Changes v1 -> v2:
- Restrict the option to glibc-based toolchains.
- Improve commit message.
- Restore the check-kernel-headers call in linux-headers.mk.
- Install the headers into the toolchain explicitly instead of copying
over all staging headers.
- Remove the loose version check option as it doesn't seem necessary.
---
| 2 ++
| 3 ++-
toolchain/Config.in | 4 ++++
toolchain/helpers.mk | 4 +---
toolchain/toolchain-buildroot/Config.in | 2 --
toolchain/toolchain-external/Config.in | 12 ++++++++++++
.../pkg-toolchain-external.mk | 19 ++++++++++++++++++-
7 files changed, 39 insertions(+), 7 deletions(-)
--git a/package/linux-headers/Config.in b/package/linux-headers/Config.in
index 0ea95164d6..ff04529325 100644
--- a/package/linux-headers/Config.in
+++ b/package/linux-headers/Config.in
@@ -1,2 +1,4 @@
config BR2_PACKAGE_LINUX_HEADERS
bool
+ # Can be selected by internal toolchains or external toolchains needing updated headers
+ default y if BR2_TOOLCHAIN_BUILDROOT
--git a/package/linux-headers/linux-headers.mk b/package/linux-headers/linux-headers.mk
index b4b5a5b47d..10ad7c4630 100644
--- a/package/linux-headers/linux-headers.mk
+++ b/package/linux-headers/linux-headers.mk
@@ -5,7 +5,8 @@
################################################################################
# This package is used to provide Linux kernel headers for the
-# internal toolchain backend.
+# internal toolchain backend and for external toolchains that need
+# updated kernel headers.
# Set variables depending on whether we are using headers from a kernel
# build or a standalone header package.
diff --git a/toolchain/Config.in b/toolchain/Config.in
index 99337a1873..35d1245061 100644
--- a/toolchain/Config.in
+++ b/toolchain/Config.in
@@ -68,6 +68,10 @@ source "toolchain/toolchain-external/Config.in"
# so put it here instead
source "package/gdb/Config.in.host"
+# Linux headers can be used by both internal and external toolchains
+source "package/linux-headers/Config.in.host"
+source "package/linux-headers/Config.in"
+
comment "Toolchain Generic Options"
# https://sourceware.org/bugzilla/show_bug.cgi?id=30730
diff --git a/toolchain/helpers.mk b/toolchain/helpers.mk
index 21c710d0fb..1802d57421 100644
--- a/toolchain/helpers.mk
+++ b/toolchain/helpers.mk
@@ -169,11 +169,9 @@ copy_toolchain_sysroot = \
# $2: sysroot directory
# $3: kernel version string, in the form: X.Y
# $4: test to do for the latest kernel version, 'strict' or 'loose'
-# always 'strict' if this is not the latest version.
#
check_kernel_headers_version = \
- if ! support/scripts/check-kernel-headers.sh $(1) $(2) $(3) \
- $(if $(BR2_TOOLCHAIN_HEADERS_LATEST),$(4),strict); \
+ if ! support/scripts/check-kernel-headers.sh $(1) $(2) $(3) $(4) ; \
then \
exit 1; \
fi
diff --git a/toolchain/toolchain-buildroot/Config.in b/toolchain/toolchain-buildroot/Config.in
index 0bccdc817c..592d88c68e 100644
--- a/toolchain/toolchain-buildroot/Config.in
+++ b/toolchain/toolchain-buildroot/Config.in
@@ -108,8 +108,6 @@ config BR2_TOOLCHAIN_BUILDROOT_LIBC
default "glibc" if BR2_TOOLCHAIN_BUILDROOT_GLIBC
default "musl" if BR2_TOOLCHAIN_BUILDROOT_MUSL
-source "package/linux-headers/Config.in.host"
-source "package/linux-headers/Config.in"
source "package/musl/Config.in"
source "package/uclibc/Config.in"
source "package/glibc/Config.in"
diff --git a/toolchain/toolchain-external/Config.in b/toolchain/toolchain-external/Config.in
index 831387bc4e..7ea71b5625 100644
--- a/toolchain/toolchain-external/Config.in
+++ b/toolchain/toolchain-external/Config.in
@@ -125,4 +125,16 @@ config BR2_TOOLCHAIN_EXTERNAL_GDB_SERVER_COPY
Copy the gdbserver provided by the external toolchain to the
target.
+config BR2_TOOLCHAIN_EXTERNAL_UPDATE_LINUX_HEADERS
+ bool "Update kernel headers"
+ # Restrict this option to glibc, as uClibc and musl should probably
+ # be rebuilt when updating kernel headers.
+ depends on BR2_TOOLCHAIN_EXTERNAL_GLIBC
+ select BR2_PACKAGE_LINUX_HEADERS
+ help
+ Select this option to update the kernel headers in the
+ external toolchain with headers from the kernel being built.
+ This is useful when the external toolchain has older kernel
+ headers than the kernel being built.
+
endif # BR2_TOOLCHAIN_EXTERNAL
diff --git a/toolchain/toolchain-external/pkg-toolchain-external.mk b/toolchain/toolchain-external/pkg-toolchain-external.mk
index 270cf814ed..80d8b0f017 100644
--- a/toolchain/toolchain-external/pkg-toolchain-external.mk
+++ b/toolchain/toolchain-external/pkg-toolchain-external.mk
@@ -542,6 +542,18 @@ define TOOLCHAIN_EXTERNAL_INSTALL_TARGET_LDD
fi
endef
+define TOOLCHAIN_EXTERNAL_UPDATE_LINUX_HEADERS
+ echo "Updating kernel headers in the external toolchain..." ; \
+ SYSROOT_DIR="$(call toolchain_find_sysroot,$(TOOLCHAIN_EXTERNAL_CC))" ; \
+ (cd $(LINUX_HEADERS_DIR); $(TARGET_MAKE_ENV) $(BR2_MAKE) \
+ ARCH=$(KERNEL_ARCH) \
+ HOSTCC="$(HOSTCC)" \
+ HOSTCFLAGS="$(HOSTCFLAGS)" \
+ HOSTCXX="$(HOSTCXX)" \
+ INSTALL_HDR_PATH=$${SYSROOT_DIR}/usr \
+ headers_install)
+endef
+
################################################################################
# inner-toolchain-external-package -- defines the generic installation rules
# for external toolchain packages
@@ -573,6 +585,11 @@ $(2)_POST_EXTRACT_HOOKS += \
TOOLCHAIN_EXTERNAL_MOVE
endif
+ifeq ($$(BR2_TOOLCHAIN_EXTERNAL_UPDATE_LINUX_HEADERS),y)
+$(2)_DEPENDENCIES += linux-headers
+$(2)_PRE_CONFIGURE_HOOKS += TOOLCHAIN_EXTERNAL_UPDATE_LINUX_HEADERS
+endif
+
# Checks for an already installed toolchain: check the toolchain
# location, check that it is usable, and then verify that it
# matches the configuration provided in Buildroot: ABI, C++ support,
@@ -585,7 +602,7 @@ define $(2)_CONFIGURE_CMDS
$$(BUILD_DIR),\
$$(call toolchain_find_sysroot,$$(TOOLCHAIN_EXTERNAL_CC)),\
$$(call qstrip,$$(BR2_TOOLCHAIN_HEADERS_AT_LEAST)),\
- $$(if $$(BR2_TOOLCHAIN_EXTERNAL_CUSTOM),loose,strict)); \
+ $$(if $$(or $$(BR2_TOOLCHAIN_EXTERNAL_CUSTOM),$$(BR2_TOOLCHAIN_EXTERNAL_UPDATE_LINUX_HEADERS)),loose,strict)); \
$$(call check_gcc_version,$$(TOOLCHAIN_EXTERNAL_CC),\
$$(call qstrip,$$(BR2_TOOLCHAIN_GCC_AT_LEAST))); \
if test "$$(BR2_arm)" = "y" ; then \
--
2.39.5
_______________________________________________
buildroot mailing list
buildroot@buildroot.org
https://lists.buildroot.org/mailman/listinfo/buildroot
^ permalink raw reply related [flat|nested] 6+ messages in thread
* Re: [Buildroot] [PATCH v2 1/1] toolchain/toolchain-external: introduce BR2_TOOLCHAIN_EXTERNAL_UPDATE_LINUX_HEADERS
2025-12-27 21:14 [Buildroot] [PATCH v2 1/1] toolchain/toolchain-external: introduce BR2_TOOLCHAIN_EXTERNAL_UPDATE_LINUX_HEADERS Rustam Abdullaev
@ 2025-12-27 22:24 ` Thomas Petazzoni via buildroot
2025-12-28 16:10 ` Rustam
2026-02-02 13:41 ` John Ernberg via buildroot
0 siblings, 2 replies; 6+ messages in thread
From: Thomas Petazzoni via buildroot @ 2025-12-27 22:24 UTC (permalink / raw)
To: Rustam Abdullaev; +Cc: buildroot, Andreas Ziegler, Giulio Benetti, Romain Naour
Hello Rustam,
Thanks for your patch!
On Sat, 27 Dec 2025 21:14:15 +0000
Rustam Abdullaev <rustamabd@gmail.com> wrote:
> It is important to have matching kernel headers when building packages
> that make use of kernel features such as seccomp, as they depend on
> system call definitions in linux/unistd.h. A typical example of this is
> openssh - building it with the official ARM GNU toolchain and kernel 6.x
> results in a non-working sshd.
Do you have more details on this? Which version of the ARM GNU
toolchain? Which kernel header version do they use? Which kernel
version are you running?
Indeed, there are two cases:
- The kernel headers in the toolchain are newer than the kernel you're
running => you are doing it wrong, and updating the headers in the
toolchain are not going to fix the fact that your C library is built
with headers that are too recent for your kernel (even though I admit
glibc has some tolerance on this)
- The kernel headers in the toolchain are older than the kernel you're
running and you're seeing breakage => this is a kernel bug as it
shouldn't happen.
In both cases, your proposed patch should therefore not be needed.
Could you clarify which situation you're in, so that we can
understand better whether the use-case/issue you're having warrants
this additional complexity?
> Currently restricted to glibc-based toolchains as glibc is backward- and
> forward-compatible with different kernel versions while uClibc and musl
> are not.
Not sure why this matters here though. Could you clarify?
Thanks!
Thomas
--
Thomas Petazzoni, co-owner and CEO, Bootlin
Embedded Linux and Kernel engineering and training
https://bootlin.com
_______________________________________________
buildroot mailing list
buildroot@buildroot.org
https://lists.buildroot.org/mailman/listinfo/buildroot
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [Buildroot] [PATCH v2 1/1] toolchain/toolchain-external: introduce BR2_TOOLCHAIN_EXTERNAL_UPDATE_LINUX_HEADERS
2025-12-27 22:24 ` Thomas Petazzoni via buildroot
@ 2025-12-28 16:10 ` Rustam
2026-03-14 7:27 ` Andreas Ziegler
2026-02-02 13:41 ` John Ernberg via buildroot
1 sibling, 1 reply; 6+ messages in thread
From: Rustam @ 2025-12-28 16:10 UTC (permalink / raw)
To: Thomas Petazzoni; +Cc: buildroot, Andreas Ziegler, Giulio Benetti, Romain Naour
[-- Attachment #1.1: Type: text/plain, Size: 4033 bytes --]
On Sat, 27 Dec 2025 at 23:24, Thomas Petazzoni <thomas.petazzoni@bootlin.com>
wrote:
> Hello Rustam,
>
Hi Thomas,
>
> Thanks for your patch!
>
> On Sat, 27 Dec 2025 21:14:15 +0000
> Rustam Abdullaev <rustamabd@gmail.com> wrote:
>
> > It is important to have matching kernel headers when building packages
> > that make use of kernel features such as seccomp, as they depend on
> > system call definitions in linux/unistd.h. A typical example of this is
> > openssh - building it with the official ARM GNU toolchain and kernel 6.x
> > results in a non-working sshd.
>
> Do you have more details on this? Which version of the ARM GNU
> toolchain? Which kernel header version do they use? Which kernel
> version are you running?
>
This applies to the official ARM GNU toolchain up to and including 14.3:
https://developer.arm.com/downloads/-/arm-gnu-toolchain-downloads
They come with kernel headers 4.20.
I'm running kernel 6.12. Though that doesn't matter, what matters is that
it's virtually guaranteed that the actual kernel is going to be newer than
4.20.
>
> Indeed, there are two cases:
>
> - The kernel headers in the toolchain are newer than the kernel you're
> running => you are doing it wrong, and updating the headers in the
> toolchain are not going to fix the fact that your C library is built
> with headers that are too recent for your kernel (even though I admit
> glibc has some tolerance on this)
>
Yes, this wouldn't work at all, we're not planning to downgrade linux
headers, only upgrade.
>
> - The kernel headers in the toolchain are older than the kernel you're
> running and you're seeing breakage => this is a kernel bug as it
> shouldn't happen.
>
I think you're referring to the fact that linux kernel is
backward-compatible. That is only true in theory. In practice not matching
the actual kernel version can lead to various issues, among others,
security issues with seccomp.
In both cases, your proposed patch should therefore not be needed.
>
> Could you clarify which situation you're in, so that we can
> understand better whether the use-case/issue you're having warrants
> this additional complexity?
>
To give a concrete example, 64-bit time system calls that were added in 5.1
are automatically used by glibc (even if it was built against 4.20 kernel
which didn't have these). Applications that set up a seccomp sandbox (e.g.
sshd) won't be able to whitelist the needed system calls correctly because
they don't exist in linux/unistd.h.
Here's a minimal repro:
cat <<EOM >>configs/qemu_arm_vexpress_defconfig
BR2_TOOLCHAIN_EXTERNAL=y
BR2_PACKAGE_OPENSSH=y
EOM
make distclean
make qemu_arm_vexpress_defconfig
make -j$(nproc)
output/images/start-qemu.sh
. . .
Welcome to Buildroot
buildroot login: root
#
# ssh localhost
Connection reset by 127.0.0.1 port 22
#
^ notice sshd not working at all.
This is just an example, we had more issues elsewhere in our codebase.
Another downside is performance as application are unable to use the latest
kernel features for better efficiency and security (cgroups v2, new bpf
features, etc).
So it's quite important to match the kernel headers to the actual kernel
being used.
Regarding the complexity, I would say it's quite low - it's a fairly small
addition that re-uses most of the existing mechanisms and simply installs
the kernel headers into the external toolchain.
> Currently restricted to glibc-based toolchains as glibc is backward- and
> > forward-compatible with different kernel versions while uClibc and musl
> > are not.
>
> Not sure why this matters here though. Could you clarify?
>
This is just a safeguard to prevent issues. From personal experience,
uClibc and musl sometimes run into issues unless they are compiled for the
specific linux version on which they run.
So this feature is only usable in combination with glibc.
> Thanks!
>
> Thomas
> --
> Thomas Petazzoni, co-owner and CEO, Bootlin
> Embedded Linux and Kernel engineering and training
> https://bootlin.com
Hope this clarifies,
Regards,
Rustam
[-- Attachment #1.2: Type: text/html, Size: 5957 bytes --]
[-- Attachment #2: Type: text/plain, Size: 150 bytes --]
_______________________________________________
buildroot mailing list
buildroot@buildroot.org
https://lists.buildroot.org/mailman/listinfo/buildroot
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [Buildroot] [PATCH v2 1/1] toolchain/toolchain-external: introduce BR2_TOOLCHAIN_EXTERNAL_UPDATE_LINUX_HEADERS
2025-12-27 22:24 ` Thomas Petazzoni via buildroot
2025-12-28 16:10 ` Rustam
@ 2026-02-02 13:41 ` John Ernberg via buildroot
2026-03-13 23:03 ` Romain Naour via buildroot
1 sibling, 1 reply; 6+ messages in thread
From: John Ernberg via buildroot @ 2026-02-02 13:41 UTC (permalink / raw)
To: Thomas Petazzoni, Rustam Abdullaev
Cc: buildroot@buildroot.org, Andreas Ziegler, Giulio Benetti,
Romain Naour, Mårten Svanfeldt
Hi Thomas, Rustam,
On 12/27/25 11:24 PM, Thomas Petazzoni via buildroot wrote:
> Hello Rustam,
>
> Thanks for your patch!
>
> On Sat, 27 Dec 2025 21:14:15 +0000
> Rustam Abdullaev <rustamabd@gmail.com> wrote:
>
>> It is important to have matching kernel headers when building packages
>> that make use of kernel features such as seccomp, as they depend on
>> system call definitions in linux/unistd.h. A typical example of this is
>> openssh - building it with the official ARM GNU toolchain and kernel 6.x
>> results in a non-working sshd.
>
> Do you have more details on this? Which version of the ARM GNU
> toolchain? Which kernel header version do they use? Which kernel
> version are you running?
>
> Indeed, there are two cases:
>
> - The kernel headers in the toolchain are newer than the kernel you're
> running => you are doing it wrong, and updating the headers in the
> toolchain are not going to fix the fact that your C library is built
> with headers that are too recent for your kernel (even though I admit
> glibc has some tolerance on this)
>
> - The kernel headers in the toolchain are older than the kernel you're
> running and you're seeing breakage => this is a kernel bug as it
> shouldn't happen.
>
> In both cases, your proposed patch should therefore not be needed.
>
> Could you clarify which situation you're in, so that we can
> understand better whether the use-case/issue you're having warrants
> this additional complexity?
>
>> Currently restricted to glibc-based toolchains as glibc is backward- and
>> forward-compatible with different kernel versions while uClibc and musl
>> are not.
>
> Not sure why this matters here though. Could you clarify?
We ran into the same issue with the pre-compiled toolchain from ARM.
We needed the CAN J1939 socket support in python, and python enables
this based on tests done on the installed kernel headers.
Since the J1939 support landed after Linux 4.20 we ended up spending
some hours figuring out how python enabled the CAN J1939 socket support.
Having the ability to use kernel headers from the selected kernel would
alleviate such problems where the project depends on the installed
kernel headers to enable / disable features.
Unless the python package is broken instead and should bundle the
necessary headers rather than depending on them being provided by the
toolchain?
Best regards // John Ernberg
_______________________________________________
buildroot mailing list
buildroot@buildroot.org
https://lists.buildroot.org/mailman/listinfo/buildroot
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [Buildroot] [PATCH v2 1/1] toolchain/toolchain-external: introduce BR2_TOOLCHAIN_EXTERNAL_UPDATE_LINUX_HEADERS
2026-02-02 13:41 ` John Ernberg via buildroot
@ 2026-03-13 23:03 ` Romain Naour via buildroot
0 siblings, 0 replies; 6+ messages in thread
From: Romain Naour via buildroot @ 2026-03-13 23:03 UTC (permalink / raw)
To: John Ernberg, Thomas Petazzoni, Rustam Abdullaev
Cc: buildroot@buildroot.org, Andreas Ziegler, Giulio Benetti,
Romain Naour, Mårten Svanfeldt
Hello John, Rustam, All,
Le 02/02/2026 à 14:41, John Ernberg via buildroot a écrit :
> Hi Thomas, Rustam,
>
> On 12/27/25 11:24 PM, Thomas Petazzoni via buildroot wrote:
>> Hello Rustam,
>>
>> Thanks for your patch!
>>
>> On Sat, 27 Dec 2025 21:14:15 +0000
>> Rustam Abdullaev <rustamabd@gmail.com> wrote:
>>
>>> It is important to have matching kernel headers when building packages
>>> that make use of kernel features such as seccomp, as they depend on
>>> system call definitions in linux/unistd.h. A typical example of this is
>>> openssh - building it with the official ARM GNU toolchain and kernel 6.x
>>> results in a non-working sshd.
>>
>> Do you have more details on this? Which version of the ARM GNU
>> toolchain? Which kernel header version do they use? Which kernel
>> version are you running?
>>
>> Indeed, there are two cases:
>>
>> - The kernel headers in the toolchain are newer than the kernel you're
>> running => you are doing it wrong, and updating the headers in the
>> toolchain are not going to fix the fact that your C library is built
>> with headers that are too recent for your kernel (even though I admit
>> glibc has some tolerance on this)
>>
>> - The kernel headers in the toolchain are older than the kernel you're
>> running and you're seeing breakage => this is a kernel bug as it
>> shouldn't happen.
>>
>> In both cases, your proposed patch should therefore not be needed.
>>
>> Could you clarify which situation you're in, so that we can
>> understand better whether the use-case/issue you're having warrants
>> this additional complexity?
>>
>>> Currently restricted to glibc-based toolchains as glibc is backward- and
>>> forward-compatible with different kernel versions while uClibc and musl
>>> are not.
>>
>> Not sure why this matters here though. Could you clarify?
>
> We ran into the same issue with the pre-compiled toolchain from ARM.
> We needed the CAN J1939 socket support in python, and python enables
> this based on tests done on the installed kernel headers.
>
> Since the J1939 support landed after Linux 4.20 we ended up spending
> some hours figuring out how python enabled the CAN J1939 socket support.
>
> Having the ability to use kernel headers from the selected kernel would
> alleviate such problems where the project depends on the installed
> kernel headers to enable / disable features.
>
> Unless the python package is broken instead and should bundle the
> necessary headers rather than depending on them being provided by the
> toolchain?
It reminds me the issue with openssh with the ARM toolchains:
"TestOpenSsh needs a toolchain >= 5.x due to a openssh issue
(Similar to: https://bugs.busybox.net/show_bug.cgi?id=13671)
Use the Bootlin toolchain bleeding-edge 2022.08-1 rhat provide
kernel headers 5.4"
https://gitlab.com/buildroot.org/buildroot/-/commit/f89f52168fcb667a3e8e43f6f44d5b1ca3961a8c
Can you try with this newer ARM toolchain that uses 6.6 linux-headers?
https://patchwork.ozlabs.org/project/buildroot/list/?series=495567
Best regards,
Romain
>
> Best regards // John Ernberg
> _______________________________________________
> buildroot mailing list
> buildroot@buildroot.org
> https://lists.buildroot.org/mailman/listinfo/buildroot
_______________________________________________
buildroot mailing list
buildroot@buildroot.org
https://lists.buildroot.org/mailman/listinfo/buildroot
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [Buildroot] [PATCH v2 1/1] toolchain/toolchain-external: introduce BR2_TOOLCHAIN_EXTERNAL_UPDATE_LINUX_HEADERS
2025-12-28 16:10 ` Rustam
@ 2026-03-14 7:27 ` Andreas Ziegler
0 siblings, 0 replies; 6+ messages in thread
From: Andreas Ziegler @ 2026-03-14 7:27 UTC (permalink / raw)
To: Rustam; +Cc: Thomas Petazzoni, buildroot, Giulio Benetti, Romain Naour
Hi ALL,
On 2025-12-28 16:10, Rustam wrote:
> On Sat, 27 Dec 2025 at 23:24, Thomas Petazzoni
> <thomas.petazzoni@bootlin.com>
> wrote:
>
>> Hello Rustam,
>>
> Hi Thomas,
>
[...]
>> Currently restricted to glibc-based toolchains as glibc is backward-
>> and
>> > forward-compatible with different kernel versions while uClibc and musl
>> > are not.
>>
>> Not sure why this matters here though. Could you clarify?
>>
> This is just a safeguard to prevent issues. From personal experience,
> uClibc and musl sometimes run into issues unless they are compiled for
> the
> specific linux version on which they run.
> So this feature is only usable in combination with glibc.
>
Actually this suggestion came from me. I have Buildroot build my
toolchains, using uClibc and musl. I occasionally had problems when
switching kernel versions and forgetting to rebuild everything. This
might be a dependency issue, not directly related to the toolchain, but
since Rustam only tested this using glibc, the restriction seems
appropriate.
>
>> Thanks!
>>
>> Thomas
>> --
>> Thomas Petazzoni, co-owner and CEO, Bootlin
>> Embedded Linux and Kernel engineering and training
>> https://bootlin.com
>
> Hope this clarifies,
> Regards,
> Rustam
Kind regards,
Andreas
_______________________________________________
buildroot mailing list
buildroot@buildroot.org
https://lists.buildroot.org/mailman/listinfo/buildroot
^ permalink raw reply [flat|nested] 6+ messages in thread
end of thread, other threads:[~2026-03-14 7:27 UTC | newest]
Thread overview: 6+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2025-12-27 21:14 [Buildroot] [PATCH v2 1/1] toolchain/toolchain-external: introduce BR2_TOOLCHAIN_EXTERNAL_UPDATE_LINUX_HEADERS Rustam Abdullaev
2025-12-27 22:24 ` Thomas Petazzoni via buildroot
2025-12-28 16:10 ` Rustam
2026-03-14 7:27 ` Andreas Ziegler
2026-02-02 13:41 ` John Ernberg via buildroot
2026-03-13 23:03 ` Romain Naour via buildroot
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox