support/testing: test_file_capabilities: bump to kernel 4.19.204
PATCH 1/16 support/testing: test_file_capabilities: bump to kernel 4.19.204
The kernel 4.11.3 curently used by the test doesn't build with host gcc >= 10 due the gcc default -fno-common. See GCC 10 porting guide [1].
/usr/bin/ld: scripts/dtc/dtc-parser.tab.o:(.bss+0x20): multiple definition of `yylloc'; scripts/dtc/dtc-lexer.lex.o:(.bss+0x0): first defined here
Bump to the next LTS release.
[1] https://gcc.gnu.org/gcc-10/porting_to.html
--- support/testing/tests/core/test_file_capabilities.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
support/testing/tests/core/test_file_capabilities.py +1 -1
@@ -13,7 +13,7 @@ class TestFileCapabilities(infra.basetest.BRTest): BR2_TARGET_GENERIC_GETTY_PORT="ttyAMA0" BR2_LINUX_KERNEL=y BR2_LINUX_KERNEL_CUSTOM_VERSION=y - BR2_LINUX_KERNEL_CUSTOM_VERSION_VALUE="4.11.3" + BR2_LINUX_KERNEL_CUSTOM_VERSION_VALUE="4.19.204" BR2_LINUX_KERNEL_DEFCONFIG="vexpress" BR2_LINUX_KERNEL_CONFIG_FRAGMENT_FILES="{}" BR2_LINUX_KERNEL_DTS_SUPPORT=y
Signature · 6 lines
2.31.1 _______________________________________________ buildroot mailing list [email protected] https://lists.buildroot.org/mailman/listinfo/buildroot
PATCH 2/16 support/testing: test_systemd: bump to kernel 4.19.204
The kernel 4.11.3 curently used by the test doesn't build with host gcc >= 10 due the gcc default -fno-common. See GCC 10 porting guide [1].
/usr/bin/ld: scripts/dtc/dtc-parser.tab.o:(.bss+0x20): multiple definition of `yylloc'; scripts/dtc/dtc-lexer.lex.o:(.bss+0x0): first defined here
Bump to the next LTS release.
[1] https://gcc.gnu.org/gcc-10/porting_to.html
--- support/testing/tests/init/test_systemd.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
support/testing/tests/init/test_systemd.py +1 -1
@@ -13,7 +13,7 @@ class InitSystemSystemdBase(InitSystemBase): BR2_TARGET_GENERIC_GETTY_PORT="ttyAMA0" BR2_LINUX_KERNEL=y BR2_LINUX_KERNEL_CUSTOM_VERSION=y - BR2_LINUX_KERNEL_CUSTOM_VERSION_VALUE="4.11.3" + BR2_LINUX_KERNEL_CUSTOM_VERSION_VALUE="4.19.204" BR2_LINUX_KERNEL_DEFCONFIG="vexpress" BR2_LINUX_KERNEL_CONFIG_FRAGMENT_FILES="{}" BR2_LINUX_KERNEL_DTS_SUPPORT=y
Signature · 6 lines
2.31.1 _______________________________________________ buildroot mailing list [email protected] https://lists.buildroot.org/mailman/listinfo/buildroot
Re: [Buildroot] [PATCH 02/16] support/testing: test_systemd: bump to kernel 4.19.204
Quoted 5 lines
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
"Romain" == Romain Naour <[email protected]> writes:
The kernel 4.11.3 curently used by the test doesn't build with host gcc >= 10 due the gcc default -fno-common. See GCC 10 porting guide [1].
Quoted 7 lines
/usr/bin/ld: scripts/dtc/dtc-parser.tab.o:(.bss+0x20): multiple definition of `yylloc'; scripts/dtc/dtc-lexer.lex.o:(.bss+0x0): first defined hereBump to the next LTS release.
[1] https://gcc.gnu.org/gcc-10/porting_to.html
Signed-off-by Romain Naour <[email protected]>
Committed to 2021.02.x, 2021.05.x and 2021.08.x, thanks.
Signature · 5 lines
Bye, Peter Korsgaard _______________________________________________ buildroot mailing list [email protected] https://lists.buildroot.org/mailman/listinfo/buildroot
PATCH 3/16 support/testing: test_f2fs: bump to kernel 4.19.204
In reply to Romain NaourThe kernel 4.16.7 curently used by the test doesn't build with host gcc >= 10 due the gcc default -fno-common. See GCC 10 porting guide [1].
/usr/bin/ld: scripts/dtc/dtc-parser.tab.o:(.bss+0x20): multiple definition of `yylloc'; scripts/dtc/dtc-lexer.lex.o:(.bss+0x0): first defined here
Bump to the next LTS release.
[1] https://gcc.gnu.org/gcc-10/porting_to.html
--- support/testing/tests/fs/test_f2fs.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
support/testing/tests/fs/test_f2fs.py +1 -1
@@ -20,7 +20,7 @@ class TestF2FS(infra.basetest.BRTest): # BR2_TARGET_ROOTFS_TAR is not set BR2_LINUX_KERNEL=y BR2_LINUX_KERNEL_CUSTOM_VERSION=y - BR2_LINUX_KERNEL_CUSTOM_VERSION_VALUE="4.16.7" + BR2_LINUX_KERNEL_CUSTOM_VERSION_VALUE="4.19.204" BR2_LINUX_KERNEL_USE_DEFCONFIG=y BR2_LINUX_KERNEL_DEFCONFIG="vexpress" BR2_LINUX_KERNEL_CONFIG_FRAGMENT_FILES="{}"
Signature · 6 lines
2.31.1 _______________________________________________ buildroot mailing list [email protected] https://lists.buildroot.org/mailman/listinfo/buildroot
Quoted 5 lines
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
"Romain" == Romain Naour <[email protected]> writes:
The kernel 4.11.3 curently used by the test doesn't build with host gcc >= 10 due the gcc default -fno-common. See GCC 10 porting guide [1].
Quoted 7 lines
/usr/bin/ld: scripts/dtc/dtc-parser.tab.o:(.bss+0x20): multiple definition of `yylloc'; scripts/dtc/dtc-lexer.lex.o:(.bss+0x0): first defined hereBump to the next LTS release.
[1] https://gcc.gnu.org/gcc-10/porting_to.html
Signed-off-by Romain Naour <[email protected]>
Committed to 2021.02.x, 2021.05.x and 2021.08.x, thanks.
Signature · 5 lines
Bye, Peter Korsgaard _______________________________________________ buildroot mailing list [email protected] https://lists.buildroot.org/mailman/listinfo/buildroot
PATCH 3/16 support/testing: test_f2fs: bump to kernel 4.19.204
The kernel 4.16.7 curently used by the test doesn't build with host gcc >= 10 due the gcc default -fno-common. See GCC 10 porting guide [1].
/usr/bin/ld: scripts/dtc/dtc-parser.tab.o:(.bss+0x20): multiple definition of `yylloc'; scripts/dtc/dtc-lexer.lex.o:(.bss+0x0): first defined here
Bump to the next LTS release.
[1] https://gcc.gnu.org/gcc-10/porting_to.html
--- support/testing/tests/fs/test_f2fs.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
support/testing/tests/fs/test_f2fs.py +1 -1
@@ -20,7 +20,7 @@ class TestF2FS(infra.basetest.BRTest): # BR2_TARGET_ROOTFS_TAR is not set BR2_LINUX_KERNEL=y BR2_LINUX_KERNEL_CUSTOM_VERSION=y - BR2_LINUX_KERNEL_CUSTOM_VERSION_VALUE="4.16.7" + BR2_LINUX_KERNEL_CUSTOM_VERSION_VALUE="4.19.204" BR2_LINUX_KERNEL_USE_DEFCONFIG=y BR2_LINUX_KERNEL_DEFCONFIG="vexpress" BR2_LINUX_KERNEL_CONFIG_FRAGMENT_FILES="{}"
Signature · 6 lines
2.31.1 _______________________________________________ buildroot mailing list [email protected] https://lists.buildroot.org/mailman/listinfo/buildroot
PATCH 4/16 support/testing: test_glxinfo: bump to kernel 4.19.204
The kernel 4.16.7 curently used by the test doesn't build with host gcc >= 10 due the gcc default -fno-common. See GCC 10 porting guide [1].
/usr/bin/ld: scripts/dtc/dtc-parser.tab.o:(.bss+0x20): multiple definition of `yylloc'; scripts/dtc/dtc-lexer.lex.o:(.bss+0x0): first defined here
Bump to the next LTS release.
[1] https://gcc.gnu.org/gcc-10/porting_to.html
--- support/testing/tests/package/test_glxinfo.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
support/testing/tests/package/test_glxinfo.py +1 -1
@@ -20,7 +20,7 @@ class TestGlxinfo(infra.basetest.BRTest): BR2_TOOLCHAIN_EXTERNAL_CUSTOM_GLIBC=y BR2_LINUX_KERNEL=y BR2_LINUX_KERNEL_CUSTOM_VERSION=y - BR2_LINUX_KERNEL_CUSTOM_VERSION_VALUE="4.16.7" + BR2_LINUX_KERNEL_CUSTOM_VERSION_VALUE="4.19.204" BR2_LINUX_KERNEL_USE_CUSTOM_CONFIG=y BR2_LINUX_KERNEL_CUSTOM_CONFIG_FILE="board/qemu/x86/linux.config" BR2_PACKAGE_MESA3D_DEMOS=y
Signature · 6 lines
2.31.1 _______________________________________________ buildroot mailing list [email protected] https://lists.buildroot.org/mailman/listinfo/buildroot
Re: [Buildroot] [PATCH 04/16] support/testing: test_glxinfo: bump to kernel 4.19.204
Quoted 5 lines
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
"Romain" == Romain Naour <[email protected]> writes:
The kernel 4.16.7 curently used by the test doesn't build with host gcc >= 10 due the gcc default -fno-common. See GCC 10 porting guide [1].
Quoted 7 lines
/usr/bin/ld: scripts/dtc/dtc-parser.tab.o:(.bss+0x20): multiple definition of `yylloc'; scripts/dtc/dtc-lexer.lex.o:(.bss+0x0): first defined hereBump to the next LTS release.
[1] https://gcc.gnu.org/gcc-10/porting_to.html
Signed-off-by Romain Naour <[email protected]>
Committed to 2021.02.x, 2021.05.x and 2021.08.x, thanks.
Signature · 5 lines
Bye, Peter Korsgaard _______________________________________________ buildroot mailing list [email protected] https://lists.buildroot.org/mailman/listinfo/buildroot
PATCH 5/16 support/testing: test_openrc: switch to ARM arm toolchain and builtin kernel
In reply to Romain NaourThe kernel 5.5.7 curently used by the test doesn't build with host gcc >= 10 due the gcc default -fno-common. See GCC 10 porting guide [1].
/usr/bin/ld: scripts/dtc/dtc-parser.tab.o:(.bss+0x20): multiple definition of `yylloc'; scripts/dtc/dtc-lexer.lex.o:(.bss+0x0): first defined here
But we can't just update this test to the next linux kernel LTS 5.10.y since the minimum gcc version has been updated to gcc 4.9 since 5.8 kernel [2] and the Sourcery CodeBench ARM 2014.05 is used (gcc 4.8 based).
Enable arm cortex A9 and VFP support to switch to the ARM arm prebuilt toolchain (the Bootlin toolchain could be used).
While at it use the prebuilt buildin kernel for the vexpress target recently updated to 5.10.7.
Fixes:
https://gitlab.com/kubu93/buildroot/-/jobs/1564202094
[1] https://gcc.gnu.org/gcc-10/porting_to.html
[2] https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=6ec4476ac82512f09c94aff5972654b70f3772b2
[3] 3cf2782906d5cc05a3966c25518d2ce01f11afc8
--- support/testing/tests/init/test_openrc.py | 12 ++++-------- 1 file changed, 4 insertions(+), 8 deletions(-)
support/testing/tests/init/test_openrc.py +4 -8
@@ -5,15 +5,11 @@ class InitSystemOpenrcBase(InitSystemBase): config = \ """ BR2_arm=y + BR2_cortex_a9=y + BR2_ARM_ENABLE_VFP=y BR2_TOOLCHAIN_EXTERNAL=y BR2_INIT_OPENRC=y BR2_TARGET_GENERIC_GETTY_PORT="ttyAMA0" - BR2_LINUX_KERNEL=y - BR2_LINUX_KERNEL_CUSTOM_VERSION=y - BR2_LINUX_KERNEL_CUSTOM_VERSION_VALUE="5.5.7" - BR2_LINUX_KERNEL_DEFCONFIG="vexpress" - BR2_LINUX_KERNEL_DTS_SUPPORT=y - BR2_LINUX_KERNEL_INTREE_DTS_NAME="vexpress-v2p-ca9" # BR2_TARGET_ROOTFS_TAR is not set """
@@ -34,7 +30,7 @@ class TestInitSystemOpenrcRoFull(InitSystemOpenrcBase): """ def test_run(self): - self.start_emulator("squashfs", "zImage", "vexpress-v2p-ca9") + self.start_emulator("squashfs") self.check_init()
@@ -46,5 +42,5 @@ class TestInitSystemOpenrcRwFull(InitSystemOpenrcBase): """ def test_run(self): - self.start_emulator("ext2", "zImage", "vexpress-v2p-ca9") + self.start_emulator("ext2") self.check_init()
Signature · 6 lines
2.31.1 _______________________________________________ buildroot mailing list [email protected] https://lists.buildroot.org/mailman/listinfo/buildroot
Re: [Buildroot] [PATCH 05/16] support/testing: test_openrc: switch to ARM arm toolchain and builtin kernel
Quoted 16 lines
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
"Romain" == Romain Naour <[email protected]> writes:
The kernel 5.5.7 curently used by the test doesn't build with host gcc >= 10 due the gcc default -fno-common. See GCC 10 porting guide [1].
/usr/bin/ld: scripts/dtc/dtc-parser.tab.o:(.bss+0x20): multiple definition of `yylloc'; scripts/dtc/dtc-lexer.lex.o:(.bss+0x0): first defined hereBut we can't just update this test to the next linux kernel LTS 5.10.y since the minimum gcc version has been updated to gcc 4.9 since 5.8 kernel [2] and the Sourcery CodeBench ARM 2014.05 is used (gcc 4.8 based).
Enable arm cortex A9 and VFP support to switch to the ARM arm prebuilt toolchain (the Bootlin toolchain could be used).
While at it use the prebuilt buildin kernel for the vexpress target recently updated to 5.10.7.
Quoted 8 lines
Fixes:
https://gitlab.com/kubu93/buildroot/-/jobs/1564202094[1] https://gcc.gnu.org/gcc-10/porting_to.html
[2] https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=6ec4476ac82512f09c94aff5972654b70f3772b2
[3]3cf2782906d5cc05a3966c25518d2ce01f11afc8Signed-off-by Romain Naour <[email protected]>
Committed to 2021.02.x, 2021.05.x and 2021.08.x, thanks.
Signature · 5 lines
Bye, Peter Korsgaard _______________________________________________ buildroot mailing list [email protected] https://lists.buildroot.org/mailman/listinfo/buildroot
PATCH 6/16 support/testing: test_iso9660: bump to kernel 4.19.204
In reply to Romain NaourThe kernel 4.11 curently used by the test doesn't build with host gcc >= 10 due the gcc default -fno-common. See GCC 10 porting guide [1].
/usr/bin/ld: scripts/dtc/dtc-parser.tab.o:(.bss+0x20): multiple definition of `yylloc'; scripts/dtc/dtc-lexer.lex.o:(.bss+0x0): first defined here
Update to the latest 4.19 LTS version but doing so requires to disable CONFIG_RETPOLINE since the toolchain is not retpoline capable [2].
Fixes:
https://gitlab.com/kubu93/buildroot/-/jobs/1564202078
[1] https://gcc.gnu.org/gcc-10/porting_to.html
[2] https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=8c4ad5d39144776c2987e81609204e1766ed4190
--- support/testing/conf/minimal-x86-qemu-kernel.config | 1 + support/testing/tests/fs/test_iso9660.py | 2 +- 2 files changed, 2 insertions(+), 1 deletion(-)
support/testing/conf/minimal-x86-qemu-kernel.config +1 -0
@@ -2,6 +2,7 @@ CONFIG_SYSVIPC=y CONFIG_MODULES=y CONFIG_MODULE_UNLOAD=y CONFIG_SMP=y +# CONFIG_RETPOLINE is not set CONFIG_HYPERVISOR_GUEST=y CONFIG_PARAVIRT=y CONFIG_NET=y
support/testing/tests/fs/test_iso9660.py +1 -1
@@ -18,7 +18,7 @@ BASIC_CONFIG = \ BR2_TARGET_GENERIC_GETTY_BAUDRATE_115200=y BR2_LINUX_KERNEL=y BR2_LINUX_KERNEL_CUSTOM_VERSION=y - BR2_LINUX_KERNEL_CUSTOM_VERSION_VALUE="4.11" + BR2_LINUX_KERNEL_CUSTOM_VERSION_VALUE="4.19.204" BR2_LINUX_KERNEL_USE_CUSTOM_CONFIG=y BR2_LINUX_KERNEL_CUSTOM_CONFIG_FILE="{}" # BR2_TARGET_ROOTFS_TAR is not set
Signature · 6 lines
2.31.1 _______________________________________________ buildroot mailing list [email protected] https://lists.buildroot.org/mailman/listinfo/buildroot
Re: [Buildroot] [PATCH 06/16] support/testing: test_iso9660: bump to kernel 4.19.204
Quoted 11 lines
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
"Romain" == Romain Naour <[email protected]> writes:
The kernel 4.11 curently used by the test doesn't build with host gcc >= 10 due the gcc default -fno-common. See GCC 10 porting guide [1].
/usr/bin/ld: scripts/dtc/dtc-parser.tab.o:(.bss+0x20): multiple definition of `yylloc'; scripts/dtc/dtc-lexer.lex.o:(.bss+0x0): first defined hereUpdate to the latest 4.19 LTS version but doing so requires to disable
CONFIG_RETPOLINEsince the toolchain is not retpoline capable [2].
Quoted 7 lines
Fixes:
https://gitlab.com/kubu93/buildroot/-/jobs/1564202078[1] https://gcc.gnu.org/gcc-10/porting_to.html
[2] https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=8c4ad5d39144776c2987e81609204e1766ed4190Signed-off-by Romain Naour <[email protected]>
Committed to 2021.02.x, 2021.05.x and 2021.08.x, thanks.
Signature · 5 lines
Bye, Peter Korsgaard _______________________________________________ buildroot mailing list [email protected] https://lists.buildroot.org/mailman/listinfo/buildroot
PATCH 7/16 support/testing: test_lxc bump to kernel 4.19.204
In reply to Romain NaourThe kernel 4.19.79 curently used by the test doesn't build with host gcc >= 10 due the gcc default -fno-common. See GCC 10 porting guide [1].
/usr/bin/ld: scripts/dtc/dtc-parser.tab.o:(.bss+0x20): multiple definition of `yylloc'; scripts/dtc/dtc-lexer.lex.o:(.bss+0x0): first defined here
The issue was fixed in 4.19.114 [2]
Bump to the latest 4.19.x version.
[1] https://gcc.gnu.org/gcc-10/porting_to.html
[2] http://git.kernel.org/cgit/linux/kernel/git/stable/linux-stable.git/commit/?id=621f2ded601546119fabccd1651b1ae29d26cd38
--- support/testing/tests/package/test_lxc.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
support/testing/tests/package/test_lxc.py +1 -1
@@ -12,7 +12,7 @@ class TestLxc(infra.basetest.BRTest): BR2_TOOLCHAIN_EXTERNAL=y BR2_LINUX_KERNEL=y BR2_LINUX_KERNEL_CUSTOM_VERSION=y - BR2_LINUX_KERNEL_CUSTOM_VERSION_VALUE="4.19.79" + BR2_LINUX_KERNEL_CUSTOM_VERSION_VALUE="4.19.204" BR2_LINUX_KERNEL_DEFCONFIG="vexpress" BR2_LINUX_KERNEL_DTS_SUPPORT=y BR2_LINUX_KERNEL_INTREE_DTS_NAME="vexpress-v2p-ca9"
Signature · 6 lines
2.31.1 _______________________________________________ buildroot mailing list [email protected] https://lists.buildroot.org/mailman/listinfo/buildroot
Re: [Buildroot] [PATCH 07/16] support/testing: test_lxc bump to kernel 4.19.204
Quoted 7 lines
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
"Romain" == Romain Naour <[email protected]> writes:
The kernel 4.19.79 curently used by the test doesn't build with host gcc >= 10 due the gcc default -fno-common. See GCC 10 porting guide [1].
/usr/bin/ld: scripts/dtc/dtc-parser.tab.o:(.bss+0x20): multiple definition of `yylloc'; scripts/dtc/dtc-lexer.lex.o:(.bss+0x0): first defined here
Quoted 7 lines
The issue was fixed in 4.19.114 [2]
Bump to the latest 4.19.x version.[1] https://gcc.gnu.org/gcc-10/porting_to.html
[2] http://git.kernel.org/cgit/linux/kernel/git/stable/linux-stable.git/commit/?id=621f2ded601546119fabccd1651b1ae29d26cd38Signed-off-by Romain Naour <[email protected]>
Committed to 2021.02.x, 2021.05.x and 2021.08.x, thanks.
Signature · 5 lines
Bye, Peter Korsgaard _______________________________________________ buildroot mailing list [email protected] https://lists.buildroot.org/mailman/listinfo/buildroot
PATCH 8/16 support/testing: test_rust: remove kernel build
In reply to Romain NaourThis test already use builtin kernel provided by the testsuite infra:
self.emulator.boot(arch="armv7",
kernel="builtin",
options=["-initrd", img])But a second kernel is build from the its defconfig. This second kernel is not used by the test.
The TestRust (using BR2_PACKAGE_HOST_RUST=y) is really long to build, same some cpu time by removing the kernel build.
This unused kernel (based on 4.11.3 release) don't even build with host gcc >= 10.
--- support/testing/tests/package/test_rust.py | 12 ------------ 1 file changed, 12 deletions(-)
support/testing/tests/package/test_rust.py +0 -12
@@ -65,12 +65,6 @@ class TestRustBin(TestRustBase): BR2_TOOLCHAIN_EXTERNAL=y BR2_TARGET_GENERIC_GETTY_PORT="ttyAMA0" BR2_SYSTEM_DHCP="eth0" - BR2_LINUX_KERNEL=y - BR2_LINUX_KERNEL_CUSTOM_VERSION=y - BR2_LINUX_KERNEL_CUSTOM_VERSION_VALUE="4.11.3" - BR2_LINUX_KERNEL_DEFCONFIG="vexpress" - BR2_LINUX_KERNEL_DTS_SUPPORT=y - BR2_LINUX_KERNEL_INTREE_DTS_NAME="vexpress-v2p-ca9" BR2_TARGET_ROOTFS_CPIO=y # BR2_TARGET_ROOTFS_TAR is not set BR2_PACKAGE_HOST_RUSTC=y
@@ -92,12 +86,6 @@ class TestRust(TestRustBase): BR2_TOOLCHAIN_EXTERNAL=y BR2_TARGET_GENERIC_GETTY_PORT="ttyAMA0" BR2_SYSTEM_DHCP="eth0" - BR2_LINUX_KERNEL=y - BR2_LINUX_KERNEL_CUSTOM_VERSION=y - BR2_LINUX_KERNEL_CUSTOM_VERSION_VALUE="4.11.3" - BR2_LINUX_KERNEL_DEFCONFIG="vexpress" - BR2_LINUX_KERNEL_DTS_SUPPORT=y - BR2_LINUX_KERNEL_INTREE_DTS_NAME="vexpress-v2p-ca9" BR2_TARGET_ROOTFS_CPIO=y # BR2_TARGET_ROOTFS_TAR is not set BR2_PACKAGE_HOST_RUSTC=y
Signature · 6 lines
2.31.1 _______________________________________________ buildroot mailing list [email protected] https://lists.buildroot.org/mailman/listinfo/buildroot
Re: [Buildroot] [PATCH 08/16] support/testing: test_rust: remove kernel build
Quoted 10 lines
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
"Romain" == Romain Naour <[email protected]> writes:
This test already use builtin kernel provided by the testsuite infra:
self.emulator.boot(arch="armv7", kernel="builtin", options=["-initrd", img])But a second kernel is build from the its defconfig. This second kernel is not used by the test.
Quoted 7 lines
The TestRust (using BR2_PACKAGE_HOST_RUST=y) is really long to build, same some cpu time by removing the kernel build.
This unused kernel (based on 4.11.3 release) don't even build with host gcc >= 10.
Signed-off-by Romain Naour <[email protected]>
Committed to 2021.02.x, 2021.05.x and 2021.08.x, thanks.
Signature · 5 lines
Bye, Peter Korsgaard _______________________________________________ buildroot mailing list [email protected] https://lists.buildroot.org/mailman/listinfo/buildroot
PATCH 9/16 support/testing: test_openssh: use a prebuilt toolchain
In reply to Romain NaourSwitch from the Buildroot internal toolchain for armv5 to the prebuilt Bootlin external toolchain.
The test doesn't require to build a toolchain, there was no prebuilt glibc toolchain recent enough at the time this test has been introduced.
--- support/testing/tests/package/test_openssh.py | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-)
support/testing/tests/package/test_openssh.py +2 -3
@@ -47,9 +47,8 @@ class TestOpenSshGlibc(TestOpensshBase): TestOpensshBase.opensshconfig + \ """ BR2_arm=y - BR2_TOOLCHAIN_BUILDROOT_GLIBC=y - BR2_KERNEL_HEADERS_4_19=y - BR2_TOOLCHAIN_BUILDROOT_CXX=y + BR2_TOOLCHAIN_EXTERNAL=y + BR2_TOOLCHAIN_EXTERNAL_BOOTLIN=y BR2_PACKAGE_RNG_TOOLS=y BR2_TARGET_ROOTFS_CPIO=y """
Signature · 6 lines
2.31.1 _______________________________________________ buildroot mailing list [email protected] https://lists.buildroot.org/mailman/listinfo/buildroot
Re: [Buildroot] [PATCH 09/16] support/testing: test_openssh: use a prebuilt toolchain
Quoted 10 lines
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
"Romain" == Romain Naour <[email protected]> writes:
Switch from the Buildroot internal toolchain for armv5 to the prebuilt Bootlin external toolchain.
The test doesn't require to build a toolchain, there was no prebuilt glibc toolchain recent enough at the time this test has been introduced.
Signed-off-by Romain Naour <[email protected]>
Committed to 2021.02.x, 2021.05.x and 2021.08.x, thanks.
Signature · 5 lines
Bye, Peter Korsgaard _______________________________________________ buildroot mailing list [email protected] https://lists.buildroot.org/mailman/listinfo/buildroot
Re: [Buildroot] [PATCH 09/16] support/testing: test_openssh: use a prebuilt toolchain
Hello Peter,
Le 04/10/2021 à 22:06, Peter Korsgaard a écrit :
Quoted 5 lines
Quoted 4 lines
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
"Romain" == Romain Naour <[email protected]> writes:
Switch from the Buildroot internal toolchain for armv5 to the prebuilt Bootlin external toolchain.
Quoted 8 lines
Quoted 5 lines
The test doesn't require to build a toolchain, there was no prebuilt glibc toolchain recent enough at the time this test has been introduced.
Signed-off-by Romain Naour <[email protected]>Committed to 2021.02.x, 2021.05.x and 2021.08.x, thanks.
For 2021.02.x the Bootlin toolchain bleeding-edge use kernel-headers too new (5.4) for the 4.19 kernel used by this test.
The kernel doesn't boot with "kernel too old" message.
Best regards,
Romain
_______________________________________________
buildroot mailing list
[email protected]
https://lists.buildroot.org/mailman/listinfo/buildroot
PATCH 10/16 support/testing: switch to prebuilt toolchain, core2duo to Nehalem
In reply to Romain NaourTo avoid spending some time to build the x86_64 toolchain (~20min), switch to corei7 cpu (Nahalem) and use the prebuilt Bootlin toolchain.
We have to use the "stable" Bootlin toolchain to use the same kernel version for the toolchain kernel headers and the running kernel.
With the "bleeding-edge" toolchain we have the "kernel too old" issue (running kernel 4.19 vs kernel headers 5.4)
Runtime tested locally.
--- support/testing/tests/package/test_docker_compose.py | 9 ++++----- 1 file changed, 4 insertions(+), 5 deletions(-)
support/testing/tests/package/test_docker_compose.py +4 -5
@@ -7,10 +7,9 @@ class TestDockerCompose(infra.basetest.BRTest): config = \ """ BR2_x86_64=y - BR2_x86_core2=y - BR2_TOOLCHAIN_BUILDROOT_GLIBC=y - BR2_KERNEL_HEADERS_4_19=y - BR2_TOOLCHAIN_BUILDROOT_CXX=y + BR2_x86_corei7=y + BR2_TOOLCHAIN_EXTERNAL=y + BR2_TOOLCHAIN_EXTERNAL_BOOTLIN_X86_64_CORE_I7_GLIBC_STABLE=y BR2_SYSTEM_DHCP="eth0" BR2_ROOTFS_POST_BUILD_SCRIPT="{}" BR2_ROOTFS_POST_SCRIPT_ARGS="{}"
@@ -53,7 +52,7 @@ class TestDockerCompose(infra.basetest.BRTest): self.emulator.boot(arch="x86_64", kernel=kernel, kernel_cmdline=["root=/dev/vda", "console=ttyS0"], - options=["-cpu", "core2duo", + options=["-cpu", "Nehalem", "-m", "512M", "-device", "virtio-rng-pci", "-drive", "file={},format=raw,if=virtio".format(rootfs),
Signature · 6 lines
2.31.1 _______________________________________________ buildroot mailing list [email protected] https://lists.buildroot.org/mailman/listinfo/buildroot
Re: [Buildroot] [PATCH 10/16] support/testing: switch to prebuilt toolchain, core2duo to Nehalem
Quoted 8 lines
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
"Romain" == Romain Naour <[email protected]> writes:
To avoid spending some time to build the
x86_64toolchain (~20min), switch to corei7 cpu (Nahalem) and use the prebuilt Bootlin toolchain.We have to use the "stable" Bootlin toolchain to use the same kernel version for the toolchain kernel headers and the running kernel.
Quoted 7 lines
With the "bleeding-edge" toolchain we have the "kernel too old" issue (running kernel 4.19 vs kernel headers 5.4)
Runtime tested locally.
Signed-off-by Romain Naour <[email protected]>Cc Peter Korsgaard <[email protected]>
Committed to 2021.02.x, 2021.05.x and 2021.08.x, thanks.
Signature · 5 lines
Bye, Peter Korsgaard _______________________________________________ buildroot mailing list [email protected] https://lists.buildroot.org/mailman/listinfo/buildroot
PATCH 11/16 support/docker: bump docker image for the gitlab-ci to Debian bullseye
In reply to Romain NaourStarting with Qemu 6.1.0, gcc 7.5 is needed to build.
Since we build host-qemu package for qemu defconfig, we have to upgrade to (at least) Debian buster that provide gcc 8 as host compiler.
While testing this upgrate, the test_edk2 failed since it actually requires Qemu >= 4.1.0 to support arm SBSA reference machine [1]. Debian Buster only provide Qemu 3.1.
Finally, upgrade to Debian bullseye but it requires some linux kernel version bump in several defconfigs since host gcc is based on gcc-10 [2].
Add explicitely python package in the Docker file to install python2 on the host.
[1] https://git.qemu.org/?p=qemu.git;a=commit;h=64580903c2b3aee08d74d64e6248a313b246cb69
[2] http://git.kernel.org/cgit/linux/kernel/git/stable/linux-stable.git/commit/?id=621f2ded601546119fabccd1651b1ae29d26cd38
---
The docker image must be rebuild and pushed to https://hub.docker.com/r/buildroot/base/
--- support/docker/Dockerfile | 4 +++- support/docker/apt-sources.list | 8 ++++---- 2 files changed, 7 insertions(+), 5 deletions(-)
support/docker/Dockerfile +3 -1
@@ -6,7 +6,7 @@ # We use a specific tag for the base image *and* the corresponding date # for the repository., so do not forget to update the apt-sources.list # file that is shipped next to this Dockerfile. -FROM debian:stretch-20171210 +FROM debian:bullseye-20210902 LABEL maintainer="Buildroot mailing list <[email protected]>" \ vendor="Buildroot" \
@@ -36,6 +36,8 @@ RUN apt-get install -y --no-install-recommends \ libncurses5-dev \ locales \ mercurial \ + python \ + python-six \ python3 \ python3-flake8 \ python3-nose2 \
support/docker/apt-sources.list +4 -4
@@ -1,4 +1,4 @@ -# Latest just before 20171210T000000Z: -deb [check-valid-until=no] http://snapshot.debian.org/archive/debian/20171209T220346Z/ stretch main -deb [check-valid-until=no] http://snapshot.debian.org/archive/debian/20171209T220346Z/ stretch-updates main -deb [check-valid-until=no] http://snapshot.debian.org/archive/debian-security/20171209T224618Z/ stretch/updates main +# Latest just before 20210817T000000Z: +deb [check-valid-until=no] http://snapshot.debian.org/archive/debian/20210902T205533Z/ bullseye main +deb [check-valid-until=no] http://snapshot.debian.org/archive/debian-security/20210902T193650Z/ bullseye-security main +deb [check-valid-until=no] http://snapshot.debian.org/archive/debian/20210902T205533Z/ bullseye-updates main
Signature · 6 lines
2.31.1 _______________________________________________ buildroot mailing list [email protected] https://lists.buildroot.org/mailman/listinfo/buildroot
Arnout Vandecappelle
AppliedRe: [Buildroot] [PATCH 11/16] support/docker: bump docker image for the gitlab-ci to Debian bullseye
I've applied patches 1-10 to master, thanks.
Quoted 8 lines · Romain Naour
Starting with Qemu 6.1.0, gcc 7.5 is needed to build.
Since we build host-qemu package for qemu defconfig, we have to upgrade to (at least) Debian buster that provide gcc 8 as host compiler.While testing this upgrate, the
test_edk2failed since it actually requires Qemu >= 4.1.0 to support arm SBSA reference machine [1]. Debian Buster only provide Qemu 3.1.
Quoted 6 lines
Finally, upgrade to Debian bullseye but it requires some linux kernel version bump in several defconfigs since host gcc is based on gcc-10 [2].
Add explicitely python package in the Docker file to install python2 on the host.
Once we convert all our support infra to python3 [1], this should no longer be
needed, right?
I propose to just apply it without python2, and see if things keep on running.
If not, we can still either add python2, or (better) fix the remaining issues.
Quoted 7 lines
[1] https://git.qemu.org/?p=qemu.git;a=commit;h=64580903c2b3aee08d74d64e6248a313b246cb69
[2] http://git.kernel.org/cgit/linux/kernel/git/stable/linux-stable.git/commit/?id=621f2ded601546119fabccd1651b1ae29d26cd38Signed-off-by Romain Naour <[email protected]>---
The docker image must be rebuild and pushed to https://hub.docker.com/r/buildroot/base/
And after that, gitlab-ci.yml has to be updated.
Regards,
Arnout
[1] https://patchwork.ozlabs.org/project/buildroot/patch/[email protected]/
Quoted 33 lines
--- support/docker/Dockerfile | 4 +++- support/docker/apt-sources.list | 8 ++++---- 2 files changed, 7 insertions(+), 5 deletions(-)support/docker/Dockerfile +3
-1@@ -6,7 +6,7 @@ # We use a specific tag for the base image *and* the corresponding date # for the repository., so do not forget to update the apt-sources.list # file that is shipped next to this Dockerfile. -FROM debian:stretch-20171210 +FROM debian:bullseye-20210902 LABEL maintainer="Buildroot mailing list <[email protected]>" \ vendor="Buildroot" \@@ -36,6 +36,8 @@ RUN apt-get install -y --no-install-recommends \ libncurses5-dev \ locales \ mercurial \ + python \ + python-six \ python3 \ python3-flake8 \ python3-nose2 \support/docker/apt-sources.list +0
-0
support/docker/apt-sources.list line 4
Quoted 8 lines
support/docker/apt-sources.list +4
-4@@ -1,4 +1,4 @@ -# Latest just before 20171210T000000Z: -deb [check-valid-until=no] http://snapshot.debian.org/archive/debian/20171209T220346Z/ stretch main -deb [check-valid-until=no] http://snapshot.debian.org/archive/debian/20171209T220346Z/ stretch-updates main -deb [check-valid-until=no] http://snapshot.debian.org/archive/debian-security/20171209T224618Z/ stretch/updates main +# Latest just before 20210817T000000Z: +deb [check-valid-until=no] http://snapshot.debian.org/archive/debian/20210902T205533Z/ bullseye main +deb [check-valid-until=no] http://snapshot.debian.org/archive/debian-security/20210902T193650Z/ bullseye-security main +deb [check-valid-until=no] http://snapshot.debian.org/archive/debian/20210902T205533Z/ bullseye-updates main
_______________________________________________
buildroot mailing list
[email protected]
https://lists.buildroot.org/mailman/listinfo/buildroot
PATCH 1/1 .gitlab-ci.yml: update docker image to 20210922.2200
In reply to Romain NaourCommit 71b83227128be62c169313e85bb8f339f977ad67 updated the Dockerfile to be used in CI tests. In order to actually use this new docker image, update .gitlab-ci.yml to point to the docker image that was created with the updated Dockerfile.
--- .gitlab-ci.yml | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
.gitlab-ci.yml +1 -1
@@ -1,7 +1,7 @@ # Configuration for Gitlab-CI. # Builds appear on https://gitlab.com/buildroot.org/buildroot/pipelines -image: buildroot/base:20200814.2228 +image: buildroot/base:20210922.2200 stages: - generate-gitlab-ci
Signature · 6 lines
2.31.1 _______________________________________________ buildroot mailing list [email protected] https://lists.buildroot.org/mailman/listinfo/buildroot
Re: [Buildroot] [PATCH] .gitlab-ci.yml: update docker image to 20210922.2200
Hi Arnout,
Le 22/09/2021 à 22:10, Arnout Vandecappelle (Essensium/Mind) a écrit :
Quoted 6 lines
Commit
71b83227128be62c169313e85bb8f339f977ad67updated the Dockerfile to be used in CI tests. In order to actually use this new docker image, update .gitlab-ci.yml to point to the docker image that was created with the updated Dockerfile.
There are some new issues since last time I tested:
Without python installed, syslinux doesn't build.
TestSysLinuxX86EFI/build/syslinux-6.03/com32/cmenu/menugen.py [...] make[6]: python: No such file or directory
https://gitlab.com/kubu93/buildroot/-/jobs/1614446766
There is a new issue with python-attrs test when python2 is not installed:
https://gitlab.com/kubu93/buildroot/-/jobs/1614446943
https://gitlab.com/kubu93/buildroot/-/jobs/1614446939
Strange python issue with refpolicy-2.20210908 build:
https://gitlab.com/kubu93/buildroot/-/jobs/1614446826/
But the issue I had with TestCheckPackage is fixed with the recent switch to python3.
https://gitlab.com/kubu93/buildroot/-/jobs/1614447236
With all issues fixed:
Best regards,
Romain
Quoted 10 lines
--- .gitlab-ci.yml | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-).gitlab-ci.yml +0
-0
.gitlab-ci.yml line 7
Quoted 8 lines
.gitlab-ci.yml +1
-1@@ -1,7 +1,7 @@ # Configuration for Gitlab-CI. # Builds appear on https://gitlab.com/buildroot.org/buildroot/pipelines -image: buildroot/base:20200814.2228 +image: buildroot/base:20210922.2200 stages: - generate-gitlab-ci
_______________________________________________
buildroot mailing list
[email protected]
https://lists.buildroot.org/mailman/listinfo/buildroot
Re: [Buildroot] [PATCH] .gitlab-ci.yml: update docker image to 20210922.2200
Quoted 10 lines · Romain Naour
Hi Arnout,
Le 22/09/2021 à 22:10, Arnout Vandecappelle (Essensium/Mind) a écrit :
Quoted 6 lines
Commit
71b83227128be62c169313e85bb8f339f977ad67updated the Dockerfile to be used in CI tests. In order to actually use this new docker image, update .gitlab-ci.yml to point to the docker image that was created with the updated Dockerfile.
Quoted 8 lines
There are some new issues since last time I tested:
Without python installed, syslinux doesn't build.
TestSysLinuxX86EFI/build/syslinux-6.03/com32/cmenu/menugen.py[...] make[6]: python: No such file or directory
For that, we just need the python-is-python3 package (or whatever it's
called). Or, better yet, let syslinux depend on host-python3.
Quoted 3 lines
There is a new issue with python-attrs test when python2 is not installed:
https://gitlab.com/kubu93/buildroot/-/jobs/1614446943
https://gitlab.com/kubu93/buildroot/-/jobs/1614446939
I suspect that that's just because python-attrs no longer supports python2. It
happens on the autobuilders as well:
http://autobuild.buildroot.net/results/175/175c2dd82dd305ed88b4d1780bd8c52e4f651de5/
And I have the same failure when I build on my laptop, which does have python2
installed (though /usr/bin/python is python3).
Quoted 2 lines
Strange python issue with refpolicy-2.20210908 build:
https://gitlab.com/kubu93/buildroot/-/jobs/1614446826/
Are you sure it is related to python? I also get the same error when I build
it locally on my laptop.
Regards,
Arnout
Quoted 26 lines
But the issue I had with TestCheckPackage is fixed with the recent switch to python3.
https://gitlab.com/kubu93/buildroot/-/jobs/1614447236
With all issues fixed:
Acked-by Romain Naour <[email protected]>Tested-by Romain Naour <[email protected]>Best regards,
RomainQuoted 12 lines
--- .gitlab-ci.yml | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-).gitlab-ci.yml +0
-0@@ -1,7 +1,7 @@ # Configuration for Gitlab-CI. # Builds appear on https://gitlab.com/buildroot.org/buildroot/pipelines
Quoted 7 lines
.gitlab-ci.yml line 7
Quoted 6 lines
.gitlab-ci.yml +1
-1-image: buildroot/base:20200814.2228 +image: buildroot/base:20210922.2200 stages: - generate-gitlab-ci
_______________________________________________
buildroot mailing list
[email protected]
https://lists.buildroot.org/mailman/listinfo/buildroot
Re: [Buildroot] [PATCH] .gitlab-ci.yml: update docker image to 20210922.2200
Hello Arnout,
Quoted 17 lines · Arnout Vandecappelle
Quoted 14 lines · Romain Naour
Hi Arnout,
Le 22/09/2021 à 22:10, Arnout Vandecappelle (Essensium/Mind) a écrit :
Quoted 6 lines
Commit
71b83227128be62c169313e85bb8f339f977ad67updated the Dockerfile to be used in CI tests. In order to actually use this new docker image, update .gitlab-ci.yml to point to the docker image that was created with the updated Dockerfile.There are some new issues since last time I tested:
Without python installed, syslinux doesn't build.
Quoted 7 lines
Quoted 4 lines
TestSysLinuxX86EFI/build/syslinux-6.03/com32/cmenu/menugen.py[...] make[6]: python: No such file or directoryFor that, we just need the python-is-python3 package (or whatever it's called). Or, better yet, let syslinux depend on host-python3.
Actually python is not a mandatory dependency of Buildroot.
With python-is-python3 installed it doesn't fix the dependency issue for good.
It seems we can override the python interpreter:
https://repo.or.cz/syslinux.git/commitdiff/4dec62ce9c2c0d170f21b3ae2d7c618eb7a30c05
So we can force syslinux using the python3 interpreter provided by host-python3.
Quoted 11 lines
Quoted 5 lines
There is a new issue with python-attrs test when python2 is not installed:
https://gitlab.com/kubu93/buildroot/-/jobs/1614446943
https://gitlab.com/kubu93/buildroot/-/jobs/1614446939I suspect that that's just because python-attrs no longer supports python2. It happens on the autobuilders as well:
http://autobuild.buildroot.net/results/175/175c2dd82dd305ed88b4d1780bd8c52e4f651de5/
It seems a fix is now available:
http://patchwork.ozlabs.org/project/buildroot/patch/[email protected]/
Quoted 10 lines
And I have the same failure when I build on my laptop, which does have python2 installed (though
/usr/bin/pythonis python3).Quoted 2 lines
Strange python issue with refpolicy-2.20210908 build:
https://gitlab.com/kubu93/buildroot/-/jobs/1614446826/Are you sure it is related to python? I also get the same error when I build it locally on my laptop.
No, I reverted the last version bump of refpolicy and the test succeed. It seems a refpolicy issue.
Best regards,
Romain
Quoted 36 lines
Regards,
ArnoutQuoted 31 lines
But the issue I had with TestCheckPackage is fixed with the recent switch to python3.
https://gitlab.com/kubu93/buildroot/-/jobs/1614447236
With all issues fixed:
Acked-by Romain Naour <[email protected]>Tested-by Romain Naour <[email protected]>Best regards,
RomainQuoted 16 lines
---
.gitlab-ci.yml | 2 +-1 file changed, 1 insertion(+), 1 deletion(-).gitlab-ci.yml +0
-0@@ -1,7 +1,7 @@ # Configuration for Gitlab-CI. # Builds appear on https://gitlab.com/buildroot.org/buildroot/pipelines-image: buildroot/base:20200814.2228
+image: buildroot/base:20210922.2200
stages:
- generate-gitlab-ci
_______________________________________________
buildroot mailing list
[email protected]
https://lists.buildroot.org/mailman/listinfo/buildroot
Re: [Buildroot] [PATCH] .gitlab-ci.yml: update docker image to 20210922.2200
Hello Arnout,
Le 23/09/2021 à 22:52, Romain Naour a écrit :
Quoted 28 lines
Hello Arnout,
Quoted 17 lines · Arnout Vandecappelle
Quoted 14 lines · Romain Naour
Hi Arnout,
Le 22/09/2021 à 22:10, Arnout Vandecappelle (Essensium/Mind) a écrit :
Quoted 6 lines
Commit
71b83227128be62c169313e85bb8f339f977ad67updated the Dockerfile to be used in CI tests. In order to actually use this new docker image, update .gitlab-ci.yml to point to the docker image that was created with the updated Dockerfile.There are some new issues since last time I tested:
Without python installed, syslinux doesn't build.
Quoted 7 lines
Quoted 4 lines
TestSysLinuxX86EFI/build/syslinux-6.03/com32/cmenu/menugen.py[...] make[6]: python: No such file or directoryFor that, we just need the python-is-python3 package (or whatever it's called). Or, better yet, let syslinux depend on host-python3.
Quoted 7 lines
Actually python is not a mandatory dependency of Buildroot.
With python-is-python3 installed it doesn't fix the dependency issue for good.It seems we can override the python interpreter:
https://repo.or.cz/syslinux.git/commitdiff/4dec62ce9c2c0d170f21b3ae2d7c618eb7a30c05So we can force syslinux using the python3 interpreter provided by host-python3.
Fixed by:
https://git.buildroot.net/buildroot/commit/?id=6ccfd40711e122c720145044eb4d145a8bacd9db
Quoted 7 lines
Quoted 7 lines
Quoted 6 lines
There is a new issue with python-attrs test when python2 is not installed:
https://gitlab.com/kubu93/buildroot/-/jobs/1614446943
https://gitlab.com/kubu93/buildroot/-/jobs/1614446939
Quoted 8 lines
Quoted 5 lines
I suspect that that's just because python-attrs no longer supports python2. It happens on the autobuilders as well:
http://autobuild.buildroot.net/results/175/175c2dd82dd305ed88b4d1780bd8c52e4f651de5/
It seems a fix is now available:
http://patchwork.ozlabs.org/project/buildroot/patch/[email protected]/
Fixed by:
https://git.buildroot.net/buildroot/commit/?id=ef5c8be25ab373866b5a92d3d9f31bf36b7a8abf
Quoted 6 lines
Quoted 6 lines
And I have the same failure when I build on my laptop, which does have python2 installed (though
/usr/bin/pythonis python3).
Quoted 8 lines
Quoted 5 lines
Quoted 2 lines
Strange python issue with refpolicy-2.20210908 build:
https://gitlab.com/kubu93/buildroot/-/jobs/1614446826/Are you sure it is related to python? I also get the same error when I build it locally on my laptop.
No, I reverted the last version bump of refpolicy and the test succeed. It seems a refpolicy issue.
Fixed by:
https://git.buildroot.net/buildroot/commit/?id=60f991e68cac30c3e9912e73258f7804e7021d5e
So all the runtest issues has been fixed.
Before merging this patch we can merge the csky toolchain removal:
http://lists.busybox.net/pipermail/buildroot/2021-October/624866.html
Best regards,
Romain
Quoted 42 lines
Best regards,
RomainQuoted 36 lines
Regards,
ArnoutQuoted 31 lines
But the issue I had with TestCheckPackage is fixed with the recent switch to python3.
https://gitlab.com/kubu93/buildroot/-/jobs/1614447236
With all issues fixed:
Acked-by Romain Naour <[email protected]>Tested-by Romain Naour <[email protected]>Best regards,
RomainQuoted 16 lines
---
.gitlab-ci.yml | 2 +-1 file changed, 1 insertion(+), 1 deletion(-).gitlab-ci.yml +0
-0@@ -1,7 +1,7 @@ # Configuration for Gitlab-CI. # Builds appear on https://gitlab.com/buildroot.org/buildroot/pipelines-image: buildroot/base:20200814.2228
+image: buildroot/base:20210922.2200
stages:
- generate-gitlab-ci
_______________________________________________
buildroot mailing list
[email protected]
https://lists.buildroot.org/mailman/listinfo/buildroot
Yann E. MORIN
AppliedRe: [Buildroot] [PATCH] .gitlab-ci.yml: update docker image to 20210922.2200
In reply to Arnout Vandecappelle (Essensium/Mind)Arnout, All,
On 2021-09-22 22:10 +0200, Arnout Vandecappelle (Essensium/Mind) spake thusly:
Quoted 6 lines
Commit
71b83227128be62c169313e85bb8f339f977ad67updated the Dockerfile to be used in CI tests. In order to actually use this new docker image, update .gitlab-ci.yml to point to the docker image that was created with the updated Dockerfile.
Applied to master, thanks.
Regards,
Yann E. MORIN.
Quoted 24 lines
--- .gitlab-ci.yml | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-).gitlab-ci.yml +1
-1@@ -1,7 +1,7 @@ # Configuration for Gitlab-CI. # Builds appear on https://gitlab.com/buildroot.org/buildroot/pipelines -image: buildroot/base:20200814.2228 +image: buildroot/base:20210922.2200 stages: - generate-gitlab-ciSignature · 6 lines
2.31.1 _______________________________________________ buildroot mailing list [email protected] https://lists.buildroot.org/mailman/listinfo/buildroot
Signature · 10 lines
.-----------------.--------------------.------------------.--------------------. | Yann E. MORIN | Real-Time Embedded | /"\ ASCII RIBBON | Erics' conspiracy: | | +33 662 376 056 | Software Designer | \ / CAMPAIGN | ___ | | +33 561 099 427 `------------.-------: X AGAINST | \e/ There is no | | http://ymorin.is-a-geek.org/ | _/*\_ | / \ HTML MAIL | v conspiracy. | '------------------------------^-------^------------------^--------------------' _______________________________________________ buildroot mailing list [email protected] https://lists.buildroot.org/mailman/listinfo/buildroot
Re: [Buildroot] [PATCH 11/16] support/docker: bump docker image for the gitlab-ci to Debian bullseye
In reply to Romain NaourQuoted 40 lines · Romain Naour
Starting with Qemu 6.1.0, gcc 7.5 is needed to build.
Since we build host-qemu package for qemu defconfig, we have to upgrade to (at least) Debian buster that provide gcc 8 as host compiler.While testing this upgrate, the
test_edk2failed since it actually requires Qemu >= 4.1.0 to support arm SBSA reference machine [1]. Debian Buster only provide Qemu 3.1.Finally, upgrade to Debian bullseye but it requires some linux kernel version bump in several defconfigs since host gcc is based on gcc-10 [2].
Add explicitely python package in the Docker file to install python2 on the host.
[1] https://git.qemu.org/?p=qemu.git;a=commit;h=64580903c2b3aee08d74d64e6248a313b246cb69
[2] http://git.kernel.org/cgit/linux/kernel/git/stable/linux-stable.git/commit/?id=621f2ded601546119fabccd1651b1ae29d26cd38Signed-off-by Romain Naour <[email protected]>---
The docker image must be rebuild and pushed to https://hub.docker.com/r/buildroot/base/--- support/docker/Dockerfile | 4 +++- support/docker/apt-sources.list | 8 ++++---- 2 files changed, 7 insertions(+), 5 deletions(-)support/docker/Dockerfile +1
-1@@ -6,7 +6,7 @@ # We use a specific tag for the base image *and* the corresponding date # for the repository., so do not forget to update the apt-sources.list # file that is shipped next to this Dockerfile. -FROM debian:stretch-20171210 +FROM debian:bullseye-20210902 LABEL maintainer="Buildroot mailing list <[email protected]>" \ vendor="Buildroot" \
support/docker/Dockerfile RUN apt-get install -y --no-install-recommends \ · line 40
Quoted 5 lines
support/docker/Dockerfile +2
-0@@ -36,6 +36,8 @@ RUN apt-get install -y --no-install-recommends \ libncurses5-dev \ locales \ mercurial \ + python \ + python-six \
As discussed, I applied to master without these python changes, pushed a new
docker image to dockerhub, and sent a patch that updates .gitlab-ci.yml.
I believe the rest of the series can only be applied after that update is
committed, right?
Regards,
Arnout
Quoted 9 lines
support/docker/Dockerfile +0
-0python3 \ python3-flake8 \ python3-nose2 \support/docker/apt-sources.list +0
-0
support/docker/apt-sources.list line 4
Quoted 8 lines
support/docker/apt-sources.list +4
-4@@ -1,4 +1,4 @@ -# Latest just before 20171210T000000Z: -deb [check-valid-until=no] http://snapshot.debian.org/archive/debian/20171209T220346Z/ stretch main -deb [check-valid-until=no] http://snapshot.debian.org/archive/debian/20171209T220346Z/ stretch-updates main -deb [check-valid-until=no] http://snapshot.debian.org/archive/debian-security/20171209T224618Z/ stretch/updates main +# Latest just before 20210817T000000Z: +deb [check-valid-until=no] http://snapshot.debian.org/archive/debian/20210902T205533Z/ bullseye main +deb [check-valid-until=no] http://snapshot.debian.org/archive/debian-security/20210902T193650Z/ bullseye-security main +deb [check-valid-until=no] http://snapshot.debian.org/archive/debian/20210902T205533Z/ bullseye-updates main
_______________________________________________
buildroot mailing list
[email protected]
https://lists.buildroot.org/mailman/listinfo/buildroot
Re: [Buildroot] [PATCH 11/16] support/docker: bump docker image for the gitlab-ci to Debian bullseye
Hello Arnout,
Le 22/09/2021 à 22:13, Arnout Vandecappelle a écrit :
Quoted 47 lines · Romain Naour
Quoted 44 lines
Starting with Qemu 6.1.0, gcc 7.5 is needed to build.
Since we build host-qemu package for qemu defconfig, we have to upgrade to (at least) Debian buster that provide gcc 8 as host compiler.While testing this upgrate, the
test_edk2failed since it actually requires Qemu >= 4.1.0 to support arm SBSA reference machine [1]. Debian Buster only provide Qemu 3.1.Finally, upgrade to Debian bullseye but it requires some linux kernel version bump in several defconfigs since host gcc is based on gcc-10 [2].
Add explicitely python package in the Docker file to install python2 on the host.
[1]
https://git.qemu.org/?p=qemu.git;a=commit;h=64580903c2b3aee08d74d64e6248a313b246cb69Signed-off-by Romain Naour <[email protected]>---
The docker image must be rebuild and pushed to
https://hub.docker.com/r/buildroot/base/
---
support/docker/Dockerfile | 4 +++-
support/docker/apt-sources.list | 8 ++++----2 files changed, 7 insertions(+), 5 deletions(-)support/docker/Dockerfile +1
-1@@ -6,7 +6,7 @@ # We use a specific tag for the base image *and* the corresponding date # for the repository., so do not forget to update the apt-sources.list # file that is shipped next to this Dockerfile. -FROM debian:stretch-20171210 +FROM debian:bullseye-20210902 LABEL maintainer="Buildroot mailing list <[email protected]>" \ vendor="Buildroot" \@@ -36,6 +36,8 @@ RUN apt-get install -y --no-install-recommends \
Quoted 8 lines
support/docker/Dockerfile RUN apt-get install -y --no-install-recommends \ · line 40
Quoted 5 lines
support/docker/Dockerfile +2
-0libncurses5-dev \ locales \ mercurial \ + python \ + python-six \As discussed, I applied to master without these python changes, pushed a new docker image to dockerhub, and sent a patch that updates .gitlab-ci.yml.
Ok, great!
Quoted 2 lines
I believe the rest of the series can only be applied after that update is committed, right?
Yes, we need to switch to this new Docker image before merging qemu update. I tested the series using image stored in my dockerhub account. But we can't avoid the ubifs regression in the testsuite...
Best regards,
Romain
Quoted 21 lines
Regards,
ArnoutQuoted 17 lines
support/docker/Dockerfile +0
-0python3 \ python3-flake8 \ python3-nose2 \support/docker/apt-sources.list +0
-2@@ -1,4 +1,4 @@ -# Latest just before 20171210T000000Z: -deb [check-valid-until=no]http://snapshot.debian.org/archive/debian/20171209T220346Z/ stretch main -deb [check-valid-until=no]
http://snapshot.debian.org/archive/debian/20171209T220346Z/ stretch-updates main -deb [check-valid-until=no]
http://snapshot.debian.org/archive/debian-security/20171209T224618Z/
stretch/updates main
+# Latest just before 20210817T000000Z:
Quoted 8 lines
Quoted 8 lines
+deb [check-valid-until=no]
http://snapshot.debian.org/archive/debian/20210902T205533Z/ bullseye main +deb [check-valid-until=no]
http://snapshot.debian.org/archive/debian-security/20210902T193650Z/
bullseye-security main
+deb [check-valid-until=no]
http://snapshot.debian.org/archive/debian/20210902T205533Z/ bullseye-updates main
_______________________________________________
buildroot mailing list
[email protected]
https://lists.buildroot.org/mailman/listinfo/buildroot
PATCH 12/16 package/qemu: bump to version 6.1.0
In reply to Romain NaourRemove upstream patches.
Drop 0003-meson-add-tests-option.patch and disable the fp-bench build by default [1]
Qemu now requires at least gcc 7.x to build.
Disable optional dependency for eBPF RSS loader.
See:
https://wiki.qemu.org/ChangeLog/6.1#Build_Information
[1] https://lists.nongnu.org/archive/html/qemu-devel/2021-03/msg00947.html
---
This version bump will break test_ubi since it requires qemu 2.8.0 to pass (due to a qemu bug). The test will be updated/fixed in the follow-up commits.
---
.../0002-Fix-build-with-64-bits-time_t.patch | 98 -------------------
...fp-disable-fp-bench-build-by-default.patch | 28 ++++++
....c-fix-fuse-lseek-on-uclibc-or-musl.patch} | 0
.../qemu/0003-meson-add-tests-option.patch | 94 ------------------
.../0004-Makefile-ignore-long-options.patch | 41 --------
...5-block-export-fuse.c-fix-musl-build.patch | 46 ---------
package/qemu/Config.in | 6 +-
package/qemu/Config.in.host | 5 +-
package/qemu/qemu.hash | 2 +-
package/qemu/qemu.mk | 6 +-
10 files changed, 37 insertions(+), 289 deletions(-)
delete mode 100644 package/qemu/0002-Fix-build-with-64-bits-time_t.patch
create mode 100644 package/qemu/0002-tests-fp-disable-fp-bench-build-by-default.patch
rename package/qemu/{0006-block-export-fuse.c-fix-fuse-lseek-on-uclibc-or-musl.patch => 0003-block-export-fuse.c-fix-fuse-lseek-on-uclibc-or-musl.patch} (100%)
delete mode 100644 package/qemu/0003-meson-add-tests-option.patch
delete mode 100644 package/qemu/0004-Makefile-ignore-long-options.patch
delete mode 100644 package/qemu/0005-block-export-fuse.c-fix-musl-build.patchpackage/qemu/0002-Fix-build-with-64-bits-time_t.patch deleted +0 -98
@@ -1,98 +0,0 @@ -From 839e51aa452345b440f8d2d0df84ab58bdedfcd1 Mon Sep 17 00:00:00 2001 -From: Fabrice Fontaine <[email protected]> -Date: Sat, 14 Nov 2020 21:54:17 +0100 -Subject: [PATCH] Fix build with 64 bits time_t - -time element is deprecated on new input_event structure in kernel's -input.h [1] - -This will avoid the following build failure: - -hw/input/virtio-input-host.c: In function 'virtio_input_host_handle_status': -hw/input/virtio-input-host.c:198:28: error: 'struct input_event' has no member named 'time' - 198 | if (gettimeofday(&evdev.time, NULL)) { - | ^ - -Fixes: - - http://autobuild.buildroot.org/results/a538167e288c14208d557cd45446df86d3d599d5 - - http://autobuild.buildroot.org/results/efd4474fb4b6c0ce0ab3838ce130429c51e43bbb - -[1] https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit?id=152194fe9c3f - -Signed-off-by: Fabrice Fontaine <[email protected]> ---- - contrib/vhost-user-input/main.c | 10 +++++++++- - hw/input/virtio-input-host.c | 10 +++++++++- - 2 files changed, 18 insertions(+), 2 deletions(-) - -diff --git a/contrib/vhost-user-input/main.c b/contrib/vhost-user-input/main.c -index 6020c6f33a..e688c3e0a9 100644 ---- a/contrib/vhost-user-input/main.c -+++ b/contrib/vhost-user-input/main.c -@@ -15,6 +15,11 @@ - #include "standard-headers/linux/virtio_input.h" - #include "qapi/error.h" - -+#ifndef input_event_sec -+#define input_event_sec time.tv_sec -+#define input_event_usec time.tv_usec -+#endif -+ - enum { - VHOST_USER_INPUT_MAX_QUEUES = 2, - }; -@@ -113,13 +118,16 @@ vi_evdev_watch(VuDev *dev, int condition, void *data) - static void vi_handle_status(VuInput *vi, virtio_input_event *event) - { - struct input_event evdev; -+ struct timeval tval; - int rc; - -- if (gettimeofday(&evdev.time, NULL)) { -+ if (gettimeofday(&tval, NULL)) { - perror("vi_handle_status: gettimeofday"); - return; - } - -+ evdev.input_event_sec = tval.tv_sec; -+ evdev.input_event_usec = tval.tv_usec; - evdev.type = le16toh(event->type); - evdev.code = le16toh(event->code); - evdev.value = le32toh(event->value); -diff --git a/hw/input/virtio-input-host.c b/hw/input/virtio-input-host.c -index 85daf73f1a..2e261737e1 100644 ---- a/hw/input/virtio-input-host.c -+++ b/hw/input/virtio-input-host.c -@@ -16,6 +16,11 @@ - #include <sys/ioctl.h> - #include "standard-headers/linux/input.h" - -+#ifndef input_event_sec -+#define input_event_sec time.tv_sec -+#define input_event_usec time.tv_usec -+#endif -+ - /* ----------------------------------------------------------------- */ - - static struct virtio_input_config virtio_input_host_config[] = { -@@ -193,13 +198,16 @@ static void virtio_input_host_handle_status(VirtIOInput *vinput, - { - VirtIOInputHost *vih = VIRTIO_INPUT_HOST(vinput); - struct input_event evdev; -+ struct timeval tval; - int rc; - -- if (gettimeofday(&evdev.time, NULL)) { -+ if (gettimeofday(&tval, NULL)) { - perror("virtio_input_host_handle_status: gettimeofday"); - return; - } - -+ evdev.input_event_sec = tval.tv_sec; -+ evdev.input_event_usec = tval.tv_usec; - evdev.type = le16_to_cpu(event->type); - evdev.code = le16_to_cpu(event->code); - evdev.value = le32_to_cpu(event->value); --- -2.29.2 -
package/qemu/0002-tests-fp-disable-fp-bench-build-by-default.patch new file +28 -0
@@ -0,0 +1,28 @@ +From f82238299d3c4cd23ca60cfc0cf4271f5c860873 Mon Sep 17 00:00:00 2001 +From: Romain Naour <[email protected]> +Date: Wed, 25 Aug 2021 21:55:53 +0200 +Subject: [PATCH] tests/fp: disable fp-bench build by default + +Fixes: +https://lists.nongnu.org/archive/html/qemu-devel/2021-03/msg00947.html + +Signed-off-by: Romain Naour <[email protected]> +--- + tests/fp/meson.build | 1 + + 1 file changed, 1 insertion(+) + +diff --git a/tests/fp/meson.build b/tests/fp/meson.build +index 07e2cdc8d2..c96460b7f1 100644 +--- a/tests/fp/meson.build ++++ b/tests/fp/meson.build +@@ -631,6 +631,7 @@ fpbench = executable( + dependencies: [qemuutil], + include_directories: [sfinc, include_directories(tfdir)], + c_args: fpcflags, ++ build_by_default: false, + ) + + fptestlog2 = executable( +-- +2.31.1 +
package/qemu/0003-block-export-fuse.c-fix-fuse-lseek-on-uclibc-or-musl.patch renamed from package/qemu/0006-block-export-fuse.c-fix-fuse-lseek-on-uclibc-or-musl.patch +0 -0
package/qemu/0003-meson-add-tests-option.patch deleted +0 -94
94 lines not on this page · raw message
package/qemu/0004-Makefile-ignore-long-options.patch deleted +0 -41
41 lines not on this page · raw message
package/qemu/0005-block-export-fuse.c-fix-musl-build.patch deleted +0 -46
46 lines not on this page · raw message
package/qemu/Config.in +4 -2
@@ -5,14 +5,16 @@ config BR2_PACKAGE_QEMU_ARCH_SUPPORTS_TARGET || BR2_x86_64 || BR2_arm \ || BR2_powerpc64 || BR2_powerpc64le -comment "QEMU requires a toolchain with wchar, threads" +comment "QEMU requires a toolchain with wchar, threads, gcc >= 7" depends on BR2_PACKAGE_QEMU_ARCH_SUPPORTS_TARGET depends on BR2_USE_MMU - depends on !(BR2_TOOLCHAIN_HAS_THREADS && BR2_USE_WCHAR) + depends on !(BR2_TOOLCHAIN_HAS_THREADS && BR2_USE_WCHAR) || \ + !BR2_TOOLCHAIN_GCC_AT_LEAST_7 config BR2_PACKAGE_QEMU bool "QEMU" depends on BR2_PACKAGE_QEMU_ARCH_SUPPORTS_TARGET + depends on BR2_TOOLCHAIN_GCC_AT_LEAST_7 depends on BR2_TOOLCHAIN_HAS_THREADS depends on BR2_USE_WCHAR # gettext depends on BR2_USE_MMU # fork()
package/qemu/Config.in.host +1 -4
@@ -37,6 +37,7 @@ config BR2_PACKAGE_HOST_QEMU_USER_ARCH_SUPPORTS config BR2_PACKAGE_HOST_QEMU bool "host qemu" depends on BR2_PACKAGE_HOST_QEMU_SYSTEM_ARCH_SUPPORTS || BR2_PACKAGE_HOST_QEMU_USER_ARCH_SUPPORTS + depends on BR2_HOST_GCC_AT_LEAST_7 select BR2_PACKAGE_HOST_QEMU_LINUX_USER_MODE \ if !BR2_PACKAGE_HOST_QEMU_SYSTEM_MODE && BR2_PACKAGE_HOST_QEMU_USER_ARCH_SUPPORTS select BR2_PACKAGE_HOST_QEMU_SYSTEM_MODE \
@@ -93,11 +94,7 @@ config BR2_PACKAGE_HOST_QEMU_VIRTFS config BR2_PACKAGE_HOST_QEMU_USB bool "USB passthrough support" - depends on BR2_HOST_GCC_AT_LEAST_4_9 # host-libusb help Enables USB passthrough support from guest to host. -comment "USB passthrough support needs a toolchain w/ host gcc >= 4.9" - depends on !BR2_HOST_GCC_AT_LEAST_4_9 - endif
package/qemu/qemu.hash +1 -1
@@ -1,4 +1,4 @@ # Locally computed, tarball verified with GPG signature -sha256 87bc1a471ca24b97e7005711066007d443423d19aacda3d442558ae032fa30b9 qemu-6.0.0.tar.xz +sha256 eebc089db3414bbeedf1e464beda0a7515aad30f73261abc246c9b27503a3c96 qemu-6.1.0.tar.xz sha256 6f04ae8364d0079a192b14635f4b1da294ce18724c034c39a6a41d1b09df6100 COPYING sha256 dc626520dcd53a22f727af3ee42c770e56c97a64fe3adb063799d8ab032fe551 COPYING.LIB
package/qemu/qemu.mk +3 -3
@@ -4,7 +4,7 @@ # ################################################################################ -QEMU_VERSION = 6.0.0 +QEMU_VERSION = 6.1.0 QEMU_SOURCE = qemu-$(QEMU_VERSION).tar.xz QEMU_SITE = http://download.qemu.org QEMU_LICENSE = GPL-2.0, LGPL-2.1, MIT, BSD-3-Clause, BSD-2-Clause, Others/BSD-1c
@@ -186,6 +186,7 @@ define QEMU_CONFIGURE_CMDS --enable-kvm \ --enable-attr \ --enable-vhost-net \ + --disable-bpf \ --disable-bsd-user \ --disable-containers \ --disable-xen \
@@ -215,7 +216,6 @@ define QEMU_CONFIGURE_CMDS --disable-opengl \ --disable-vhost-user-blk-server \ --disable-virtiofsd \ - --disable-tests \ $(QEMU_OPTS) endef
@@ -351,6 +351,7 @@ define HOST_QEMU_CONFIGURE_CMDS --extra-ldflags="$(HOST_LDFLAGS)" \ --meson=$(HOST_DIR)/bin/meson \ --ninja=$(HOST_DIR)/bin/ninja \ + --disable-bpf \ --disable-bzip2 \ --disable-containers \ --disable-curl \
@@ -362,7 +363,6 @@ define HOST_QEMU_CONFIGURE_CMDS --disable-vnc-jpeg \ --disable-vnc-png \ --disable-vnc-sasl \ - --disable-tests \ $(HOST_QEMU_OPTS) endef
Signature · 6 lines
2.31.1 _______________________________________________ buildroot mailing list [email protected] https://lists.buildroot.org/mailman/listinfo/buildroot
Yann E. MORIN
AppliedRe: [Buildroot] [PATCH 12/16] package/qemu: bump to version 6.1.0
Romain, All,
On 2021-09-12 16:11 +0200, Romain Naour spake thusly:
Quoted 5 lines
Remove upstream patches.
Drop 0003-meson-add-tests-option.patch and disable the fp-bench build by default [1]Qemu now requires at least gcc 7.x to build.
Quoted 8 lines
Disable optional dependency for eBPF RSS loader.
See:
https://wiki.qemu.org/ChangeLog/6.1#Build_Information[1] https://lists.nongnu.org/archive/html/qemu-devel/2021-03/msg00947.html
Signed-off-by Romain Naour <[email protected]>
Applied to master, thanks.
Regards,
Yann E. MORIN.
Quoted 472 lines
---
This version bump will breaktest_ubisince it requires qemu 2.8.0 to pass (due to a qemu bug). The test will be updated/fixed in the follow-up commits.--- .../0002-Fix-build-with-64-bits-time_t.patch | 98 ------------------- ...fp-disable-fp-bench-build-by-default.patch | 28 ++++++ ....c-fix-fuse-lseek-on-uclibc-or-musl.patch} | 0 .../qemu/0003-meson-add-tests-option.patch | 94 ------------------ .../0004-Makefile-ignore-long-options.patch | 41 -------- ...5-block-export-fuse.c-fix-musl-build.patch | 46 --------- package/qemu/Config.in | 6 +- package/qemu/Config.in.host | 5 +- package/qemu/qemu.hash | 2 +- package/qemu/qemu.mk | 6 +- 10 files changed, 37 insertions(+), 289 deletions(-) delete mode 100644 package/qemu/0002-Fix-build-with-64-bits-time_t.patch create mode 100644 package/qemu/0002-tests-fp-disable-fp-bench-build-by-default.patch rename package/qemu/{0006-block-export-fuse.c-fix-fuse-lseek-on-uclibc-or-musl.patch => 0003-block-export-fuse.c-fix-fuse-lseek-on-uclibc-or-musl.patch} (100%) delete mode 100644 package/qemu/0003-meson-add-tests-option.patch delete mode 100644 package/qemu/0004-Makefile-ignore-long-options.patch delete mode 100644 package/qemu/0005-block-export-fuse.c-fix-musl-build.patchpackage/qemu/0002-Fix-build-with-64-bits-time_t.patch deleted +0
-98@@ -1,98 +0,0 @@ -From 839e51aa452345b440f8d2d0df84ab58bdedfcd1 Mon Sep 17 00:00:00 2001 -From: Fabrice Fontaine <[email protected]> -Date: Sat, 14 Nov 2020 21:54:17 +0100 -Subject: [PATCH] Fix build with 64 bits time_t - -time element is deprecated on new input_event structure in kernel's -input.h [1] - -This will avoid the following build failure: - -hw/input/virtio-input-host.c: In function 'virtio_input_host_handle_status': -hw/input/virtio-input-host.c:198:28: error: 'struct input_event' has no member named 'time' - 198 | if (gettimeofday(&evdev.time, NULL)) { - | ^ - -Fixes: - - http://autobuild.buildroot.org/results/a538167e288c14208d557cd45446df86d3d599d5 - - http://autobuild.buildroot.org/results/efd4474fb4b6c0ce0ab3838ce130429c51e43bbb - -[1] https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit?id=152194fe9c3f - -Signed-off-by: Fabrice Fontaine <[email protected]> ---- - contrib/vhost-user-input/main.c | 10 +++++++++- - hw/input/virtio-input-host.c | 10 +++++++++- - 2 files changed, 18 insertions(+), 2 deletions(-) - -diff --git a/contrib/vhost-user-input/main.c b/contrib/vhost-user-input/main.c -index 6020c6f33a..e688c3e0a9 100644 ---- a/contrib/vhost-user-input/main.c -+++ b/contrib/vhost-user-input/main.c -@@ -15,6 +15,11 @@ - #include "standard-headers/linux/virtio_input.h" - #include "qapi/error.h" - -+#ifndef input_event_sec -+#define input_event_sec time.tv_sec -+#define input_event_usec time.tv_usec -+#endif -+ - enum { - VHOST_USER_INPUT_MAX_QUEUES = 2, - }; -@@ -113,13 +118,16 @@ vi_evdev_watch(VuDev *dev, int condition, void *data) - static void vi_handle_status(VuInput *vi, virtio_input_event *event) - { - struct input_event evdev; -+ struct timeval tval; - int rc; - -- if (gettimeofday(&evdev.time, NULL)) { -+ if (gettimeofday(&tval, NULL)) { - perror("vi_handle_status: gettimeofday"); - return; - } - -+ evdev.input_event_sec = tval.tv_sec; -+ evdev.input_event_usec = tval.tv_usec; - evdev.type = le16toh(event->type); - evdev.code = le16toh(event->code); - evdev.value = le32toh(event->value); -diff --git a/hw/input/virtio-input-host.c b/hw/input/virtio-input-host.c -index 85daf73f1a..2e261737e1 100644 ---- a/hw/input/virtio-input-host.c -+++ b/hw/input/virtio-input-host.c -@@ -16,6 +16,11 @@ - #include <sys/ioctl.h> - #include "standard-headers/linux/input.h" - -+#ifndef input_event_sec -+#define input_event_sec time.tv_sec -+#define input_event_usec time.tv_usec -+#endif -+ - /* ----------------------------------------------------------------- */ - - static struct virtio_input_config virtio_input_host_config[] = { -@@ -193,13 +198,16 @@ static void virtio_input_host_handle_status(VirtIOInput *vinput, - { - VirtIOInputHost *vih = VIRTIO_INPUT_HOST(vinput); - struct input_event evdev; -+ struct timeval tval; - int rc; - -- if (gettimeofday(&evdev.time, NULL)) { -+ if (gettimeofday(&tval, NULL)) { - perror("virtio_input_host_handle_status: gettimeofday"); - return; - } - -+ evdev.input_event_sec = tval.tv_sec; -+ evdev.input_event_usec = tval.tv_usec; - evdev.type = le16_to_cpu(event->type); - evdev.code = le16_to_cpu(event->code); - evdev.value = le32_to_cpu(event->value); --- -2.29.2 -package/qemu/0002-tests-fp-disable-fp-bench-build-by-default.patch new file +28
-0@@ -0,0 +1,28 @@ +From f82238299d3c4cd23ca60cfc0cf4271f5c860873 Mon Sep 17 00:00:00 2001 +From: Romain Naour <[email protected]> +Date: Wed, 25 Aug 2021 21:55:53 +0200 +Subject: [PATCH] tests/fp: disable fp-bench build by default + +Fixes: +https://lists.nongnu.org/archive/html/qemu-devel/2021-03/msg00947.html + +Signed-off-by: Romain Naour <[email protected]> +--- + tests/fp/meson.build | 1 + + 1 file changed, 1 insertion(+) + +diff --git a/tests/fp/meson.build b/tests/fp/meson.build +index 07e2cdc8d2..c96460b7f1 100644 +--- a/tests/fp/meson.build ++++ b/tests/fp/meson.build +@@ -631,6 +631,7 @@ fpbench = executable( + dependencies: [qemuutil], + include_directories: [sfinc, include_directories(tfdir)], + c_args: fpcflags, ++ build_by_default: false, + ) + + fptestlog2 = executable( +-- +2.31.1 +package/qemu/0003-block-export-fuse.c-fix-fuse-lseek-on-uclibc-or-musl.patch renamed from package/qemu/0006-block-export-fuse.c-fix-fuse-lseek-on-uclibc-or-musl.patch +0
-0package/qemu/0003-meson-add-tests-option.patch deleted +0
-94@@ -1,94 +0,0 @@ -From c24f6cc8f847a4f90cd8fe0f94604862c5ac4a5e Mon Sep 17 00:00:00 2001 -From: Romain Naour <[email protected]> -Date: Mon, 21 Dec 2020 22:12:14 +0100 -Subject: [PATCH] meson: add tests option - -tests/fp/fp-bench.c use fenv.h that is not always provided -by the libc (uClibc). - -To workaround this issue, add an new meson option to -disable tests while building Qemu. - -Fixes: -http://autobuild.buildroot.net/results/53f5d8baa994d599b9da013ee643b82353366ec3/build-end.log - -Signed-off-by: Romain Naour <[email protected]> -Signed-off-by: Adam Duskett <[email protected]> -Refresh for 6.0.0 ---- - configure | 7 +++++++ - meson.build | 5 ++++- - meson_options.txt | 2 ++ - 3 files changed, 13 insertions(+), 1 deletion(-) - -diff --git a/configure b/configure -index 18c26e0389..7cceae4418 100755 ---- a/configure -+++ b/configure -@@ -460,6 +460,7 @@ meson="" - ninja="" - skip_meson=no - gettext="auto" -+tests="auto" - fuse="auto" - fuse_lseek="auto" - multiprocess="auto" -@@ -1549,6 +1550,10 @@ for opt do - ;; - --disable-libdaxctl) libdaxctl=no - ;; -+ --disable-tests) tests="disabled" -+ ;; -+ --enable-tests) tests="enabled" -+ ;; - --enable-fuse) fuse="enabled" - ;; - --disable-fuse) fuse="disabled" -@@ -1915,6 +1920,7 @@ disabled with --disable-FEATURE, default is enabled if available - xkbcommon xkbcommon support - rng-none dummy RNG, avoid using /dev/(u)random and getrandom() - libdaxctl libdaxctl support -+ tests build tests - fuse FUSE block device export - fuse-lseek SEEK_HOLE/SEEK_DATA support for FUSE exports - multiprocess Out of process device emulation support -@@ -6422,6 +6428,7 @@ NINJA=$ninja $meson setup \ - -Dzstd=$zstd -Dseccomp=$seccomp -Dvirtfs=$virtfs -Dcap_ng=$cap_ng \ - -Dattr=$attr -Ddefault_devices=$default_devices \ - -Ddocs=$docs -Dsphinx_build=$sphinx_build -Dinstall_blobs=$blobs \ -+ -Dtests=$tests \ - -Dvhost_user_blk_server=$vhost_user_blk_server -Dmultiprocess=$multiprocess \ - -Dfuse=$fuse -Dfuse_lseek=$fuse_lseek -Dguest_agent_msi=$guest_agent_msi \ - $(if test "$default_features" = no; then echo "-Dauto_features=disabled"; fi) \ -diff --git a/meson.build b/meson.build -index e3386196ba..7f4efdc347 100644 ---- a/meson.build -+++ b/meson.build -@@ -2381,7 +2381,10 @@ subdir('scripts') - subdir('tools') - subdir('pc-bios') - subdir('docs') --subdir('tests') -+if get_option('tests').enabled() -+ subdir('tests') -+endif -+ - if gtk.found() - subdir('po') - endif -diff --git a/meson_options.txt b/meson_options.txt -index f6f64785fe..81bffbfe87 100644 ---- a/meson_options.txt -+++ b/meson_options.txt -@@ -111,6 +111,8 @@ option('fuse_lseek', type : 'feature', value : 'auto', - - option('vhost_user_blk_server', type: 'feature', value: 'auto', - description: 'build vhost-user-blk server') -+option('tests', type : 'feature', value : 'auto', -+ description: 'Tests build support') - option('virtfs', type: 'feature', value: 'auto', - description: 'virtio-9p support') - option('virtiofsd', type: 'feature', value: 'auto', --- -2.29.2 -package/qemu/0004-Makefile-ignore-long-options.patch deleted +0
-41@@ -1,41 +0,0 @@ -From 14833e24dea49303ebc2464813601054b6cdfcac Mon Sep 17 00:00:00 2001 -From: Alexey Neyman <[email protected]> -Date: Wed, 21 Jul 2021 19:08:46 -0700 -Subject: [PATCH] Makefile: ignore long options - -When searching for options like -n in MAKEFLAGS, current code may result -in a false positive match when make is invoked with long options like ---no-print-directory. This has been observed with certain versions of -host make (e.g. 3.82) while building the Qemu package in buildroot. - -Filter out such long options before searching for one-character options. - -Signed-off-by: Alexey Neyman <[email protected]> -Message-Id: <[email protected]> -Signed-off-by: Paolo Bonzini <[email protected]> ---- - Makefile | 8 +++++--- - 1 file changed, 5 insertions(+), 3 deletions(-) - -diff --git a/Makefile b/Makefile -index 6c36330eef..401c623a65 100644 ---- a/Makefile -+++ b/Makefile -@@ -129,9 +129,11 @@ endif - # 4. Rules to bridge to other makefiles - - ifneq ($(NINJA),) --MAKE.n = $(findstring n,$(firstword $(MAKEFLAGS))) --MAKE.k = $(findstring k,$(firstword $(MAKEFLAGS))) --MAKE.q = $(findstring q,$(firstword $(MAKEFLAGS))) -+# Filter out long options to avoid flags like --no-print-directory which -+# may result in false positive match for MAKE.n -+MAKE.n = $(findstring n,$(firstword $(filter-out --%,$(MAKEFLAGS)))) -+MAKE.k = $(findstring k,$(firstword $(filter-out --%,$(MAKEFLAGS)))) -+MAKE.q = $(findstring q,$(firstword $(filter-out --%,$(MAKEFLAGS)))) - MAKE.nq = $(if $(word 2, $(MAKE.n) $(MAKE.q)),nq) - NINJAFLAGS = $(if $V,-v) $(if $(MAKE.n), -n) $(if $(MAKE.k), -k0) \ - $(filter-out -j, $(lastword -j1 $(filter -l% -j%, $(MAKEFLAGS)))) \ --- -2.27.0 -package/qemu/0005-block-export-fuse.c-fix-musl-build.patch deleted +0
-46@@ -1,46 +0,0 @@ -From 4d5280c066eb46a31556044fb52c2618928b0019 Mon Sep 17 00:00:00 2001 -From: Fabrice Fontaine <[email protected]> -Date: Mon, 9 Aug 2021 10:04:07 +0200 -Subject: [PATCH] block/export/fuse.c: fix musl build - -Fix the following build failure on musl raised since version 6.0.0 and -https://gitlab.com/qemu-project/qemu/-/commit/4ca37a96a75aafe7a37ba51ab1912b09b7190a6b -because musl does not define FALLOC_FL_ZERO_RANGE: - -../block/export/fuse.c: In function 'fuse_fallocate': -../block/export/fuse.c:563:23: error: 'FALLOC_FL_ZERO_RANGE' undeclared (first use in this function) - 563 | } else if (mode & FALLOC_FL_ZERO_RANGE) { - | ^~~~~~~~~~~~~~~~~~~~ - -Fixes: - - http://autobuild.buildroot.org/results/b96e3d364fd1f8bbfb18904a742e73327d308f64 - -Signed-off-by: Fabrice Fontaine <[email protected]> -[Upstream status: sent to [email protected]] ---- - block/export/fuse.c | 2 ++ - 1 file changed, 2 insertions(+) - -diff --git a/block/export/fuse.c b/block/export/fuse.c -index ada9e263eb..07e31129a6 100644 ---- a/block/export/fuse.c -+++ b/block/export/fuse.c -@@ -635,6 +635,7 @@ static void fuse_fallocate(fuse_req_t req, fuse_ino_t inode, int mode, - offset += size; - length -= size; - } while (ret == 0 && length > 0); -+#ifdef FALLOC_FL_ZERO_RANGE - } else if (mode & FALLOC_FL_ZERO_RANGE) { - if (!(mode & FALLOC_FL_KEEP_SIZE) && offset + length > blk_len) { - /* No need for zeroes, we are going to write them ourselves */ -@@ -654,6 +655,7 @@ static void fuse_fallocate(fuse_req_t req, fuse_ino_t inode, int mode, - offset += size; - length -= size; - } while (ret == 0 && length > 0); -+#endif - } else if (!mode) { - /* We can only fallocate at the EOF with a truncate */ - if (offset < blk_len) { --- -2.30.2 -package/qemu/Config.in +4
-2@@ -5,14 +5,16 @@ config BR2_PACKAGE_QEMU_ARCH_SUPPORTS_TARGET || BR2_x86_64 || BR2_arm \ || BR2_powerpc64 || BR2_powerpc64le -comment "QEMU requires a toolchain with wchar, threads" +comment "QEMU requires a toolchain with wchar, threads, gcc >= 7" depends on BR2_PACKAGE_QEMU_ARCH_SUPPORTS_TARGET depends on BR2_USE_MMU - depends on !(BR2_TOOLCHAIN_HAS_THREADS && BR2_USE_WCHAR) + depends on !(BR2_TOOLCHAIN_HAS_THREADS && BR2_USE_WCHAR) || \ + !BR2_TOOLCHAIN_GCC_AT_LEAST_7 config BR2_PACKAGE_QEMU bool "QEMU" depends on BR2_PACKAGE_QEMU_ARCH_SUPPORTS_TARGET + depends on BR2_TOOLCHAIN_GCC_AT_LEAST_7 depends on BR2_TOOLCHAIN_HAS_THREADS depends on BR2_USE_WCHAR # gettext depends on BR2_USE_MMU # fork()package/qemu/Config.in.host +1
-4@@ -37,6 +37,7 @@ config BR2_PACKAGE_HOST_QEMU_USER_ARCH_SUPPORTS config BR2_PACKAGE_HOST_QEMU bool "host qemu" depends on BR2_PACKAGE_HOST_QEMU_SYSTEM_ARCH_SUPPORTS || BR2_PACKAGE_HOST_QEMU_USER_ARCH_SUPPORTS + depends on BR2_HOST_GCC_AT_LEAST_7 select BR2_PACKAGE_HOST_QEMU_LINUX_USER_MODE \ if !BR2_PACKAGE_HOST_QEMU_SYSTEM_MODE && BR2_PACKAGE_HOST_QEMU_USER_ARCH_SUPPORTS select BR2_PACKAGE_HOST_QEMU_SYSTEM_MODE \@@ -93,11 +94,7 @@ config BR2_PACKAGE_HOST_QEMU_VIRTFS config BR2_PACKAGE_HOST_QEMU_USB bool "USB passthrough support" - depends on BR2_HOST_GCC_AT_LEAST_4_9 # host-libusb help Enables USB passthrough support from guest to host. -comment "USB passthrough support needs a toolchain w/ host gcc >= 4.9" - depends on !BR2_HOST_GCC_AT_LEAST_4_9 - endifpackage/qemu/qemu.hash +1
-1@@ -1,4 +1,4 @@ # Locally computed, tarball verified with GPG signature -sha256 87bc1a471ca24b97e7005711066007d443423d19aacda3d442558ae032fa30b9 qemu-6.0.0.tar.xz +sha256 eebc089db3414bbeedf1e464beda0a7515aad30f73261abc246c9b27503a3c96 qemu-6.1.0.tar.xz sha256 6f04ae8364d0079a192b14635f4b1da294ce18724c034c39a6a41d1b09df6100 COPYING sha256 dc626520dcd53a22f727af3ee42c770e56c97a64fe3adb063799d8ab032fe551 COPYING.LIBpackage/qemu/qemu.mk +3
-3@@ -4,7 +4,7 @@ # ################################################################################ -QEMU_VERSION = 6.0.0 +QEMU_VERSION = 6.1.0 QEMU_SOURCE = qemu-$(QEMU_VERSION).tar.xz QEMU_SITE = http://download.qemu.org QEMU_LICENSE = GPL-2.0, LGPL-2.1, MIT, BSD-3-Clause, BSD-2-Clause, Others/BSD-1c@@ -186,6 +186,7 @@ define QEMU_CONFIGURE_CMDS --enable-kvm \ --enable-attr \ --enable-vhost-net \ + --disable-bpf \ --disable-bsd-user \ --disable-containers \ --disable-xen \@@ -215,7 +216,6 @@ define QEMU_CONFIGURE_CMDS --disable-opengl \ --disable-vhost-user-blk-server \ --disable-virtiofsd \ - --disable-tests \ $(QEMU_OPTS) endef@@ -351,6 +351,7 @@ define HOST_QEMU_CONFIGURE_CMDS --extra-ldflags="$(HOST_LDFLAGS)" \ --meson=$(HOST_DIR)/bin/meson \ --ninja=$(HOST_DIR)/bin/ninja \ + --disable-bpf \ --disable-bzip2 \ --disable-containers \ --disable-curl \@@ -362,7 +363,6 @@ define HOST_QEMU_CONFIGURE_CMDS --disable-vnc-jpeg \ --disable-vnc-png \ --disable-vnc-sasl \ - --disable-tests \ $(HOST_QEMU_OPTS) endefSignature · 6 lines
2.31.1 _______________________________________________ buildroot mailing list [email protected] https://lists.buildroot.org/mailman/listinfo/buildroot
Signature · 10 lines
.-----------------.--------------------.------------------.--------------------. | Yann E. MORIN | Real-Time Embedded | /"\ ASCII RIBBON | Erics' conspiracy: | | +33 662 376 056 | Software Designer | \ / CAMPAIGN | ___ | | +33 561 099 427 `------------.-------: X AGAINST | \e/ There is no | | http://ymorin.is-a-geek.org/ | _/*\_ | / \ HTML MAIL | v conspiracy. | '------------------------------^-------^------------------^--------------------' _______________________________________________ buildroot mailing list [email protected] https://lists.buildroot.org/mailman/listinfo/buildroot
PATCH 13/16 support/testing: test_ubi: reduce the rootfs.ubi size to 64M to match the Qemu emulated flash device
In reply to Romain NaourThe size of the cfi flash device emulated by Qemu is 64M not 128M [1]. Since Qemu >= 4.0, the size of the device must match the size of the block backend [2].
Fixes:
qemu-system-arm: device requires 67108864 bytes, block backend provides 134217728 bytes
[1] https://git.qemu.org/?p=qemu.git;a=blob;f=hw/arm/vexpress.c;h=58481c07629aedb09864dcc72757ff7947e733bb;hb=f9baca549e44791be0dd98de15add3d8452a8af0#l50
[2] https://git.qemu.org/?p=qemu.git;a=commitdiff;h=06f1521795207359a395996c253c306f4ab7586e
--- support/testing/tests/fs/test_ubi.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
support/testing/tests/fs/test_ubi.py +1 -1
@@ -25,7 +25,7 @@ class TestUbi(infra.basetest.BRTest): out = out.splitlines() self.assertIn("UBI image, version 1", out[0]) - subprocess.call(["truncate", "-s 128M", img]) + subprocess.call(["truncate", "-s 64M", img]) self.emulator.boot(arch="armv7", kernel="builtin",
Signature · 6 lines
2.31.1 _______________________________________________ buildroot mailing list [email protected] https://lists.buildroot.org/mailman/listinfo/buildroot
Yann E. MORIN
AppliedRe: [Buildroot] [PATCH 13/16] support/testing: test_ubi: reduce the rootfs.ubi size to 64M to match the Qemu emulated flash device
Romain, All,
On 2021-09-12 16:11 +0200, Romain Naour spake thusly:
Quoted 11 lines
The size of the cfi flash device emulated by Qemu is 64M not 128M [1]. Since Qemu >= 4.0, the size of the device must match the size of the block backend [2].
Fixes:
qemu-system-arm: device requires 67108864 bytes, block backend provides 134217728 bytes
[1] https://git.qemu.org/?p=qemu.git;a=blob;f=hw/arm/vexpress.c;h=58481c07629aedb09864dcc72757ff7947e733bb;hb=f9baca549e44791be0dd98de15add3d8452a8af0#l50
[2] https://git.qemu.org/?p=qemu.git;a=commitdiff;h=06f1521795207359a395996c253c306f4ab7586eSigned-off-by Romain Naour <[email protected]>
Applied to master, thanks.
Regards,
Yann E. MORIN.
Quoted 24 lines
--- support/testing/tests/fs/test_ubi.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)support/testing/tests/fs/test_ubi.py +1
-1@@ -25,7 +25,7 @@ class TestUbi(infra.basetest.BRTest): out = out.splitlines() self.assertIn("UBI image, version 1", out[0]) - subprocess.call(["truncate", "-s 128M", img]) + subprocess.call(["truncate", "-s 64M", img]) self.emulator.boot(arch="armv7", kernel="builtin",Signature · 6 lines
2.31.1 _______________________________________________ buildroot mailing list [email protected] https://lists.buildroot.org/mailman/listinfo/buildroot
Signature · 10 lines
.-----------------.--------------------.------------------.--------------------. | Yann E. MORIN | Real-Time Embedded | /"\ ASCII RIBBON | Erics' conspiracy: | | +33 662 376 056 | Software Designer | \ / CAMPAIGN | ___ | | +33 561 099 427 `------------.-------: X AGAINST | \e/ There is no | | http://ymorin.is-a-geek.org/ | _/*\_ | / \ HTML MAIL | v conspiracy. | '------------------------------^-------^------------------^--------------------' _______________________________________________ buildroot mailing list [email protected] https://lists.buildroot.org/mailman/listinfo/buildroot
PATCH 14/16 support/testing: test_ubi: add image format on the qemu command line
In reply to Romain NaourAdding the Image format on the Qemu command line avoid this warning:
"WARNING: Image format was not specified for 'output/TestUbi/images/rootfs.ubi' and probing guessed raw.
Automatically detecting the format is dangerous for raw images, write operations on block 0 will be restricted. Specify the 'raw' format explicitly to remove the restrictions."
--- support/testing/tests/fs/test_ubi.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
support/testing/tests/fs/test_ubi.py +1 -1
@@ -32,7 +32,7 @@ class TestUbi(infra.basetest.BRTest): kernel_cmdline=["root=ubi0:rootfs", "ubi.mtd=0", "rootfstype=ubifs"], - options=["-drive", "file={},if=pflash".format(img)]) + options=["-drive", "file={},if=pflash,format=raw".format(img)]) self.emulator.login() cmd = "mount | grep 'ubi0:rootfs on / type ubifs'" _, exit_code = self.emulator.run(cmd)
Signature · 6 lines
2.31.1 _______________________________________________ buildroot mailing list [email protected] https://lists.buildroot.org/mailman/listinfo/buildroot
Yann E. MORIN
AppliedRe: [Buildroot] [PATCH 14/16] support/testing: test_ubi: add image format on the qemu command line
Romain, All,
On 2021-09-12 16:11 +0200, Romain Naour spake thusly:
Quoted 7 lines
Adding the Image format on the Qemu command line avoid this warning:
"WARNING: Image format was not specified for 'output/TestUbi/images/rootfs.ubi' and probing guessed raw.
Automatically detecting the format is dangerous for raw images, write operations on block 0 will be restricted. Specify the 'raw' format explicitly to remove the restrictions."Signed-off-by Romain Naour <[email protected]>
Applied to master, thanks.
Regards,
Yann E. MORIN.
Quoted 24 lines
--- support/testing/tests/fs/test_ubi.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)support/testing/tests/fs/test_ubi.py +1
-1@@ -32,7 +32,7 @@ class TestUbi(infra.basetest.BRTest): kernel_cmdline=["root=ubi0:rootfs", "ubi.mtd=0", "rootfstype=ubifs"], - options=["-drive", "file={},if=pflash".format(img)]) + options=["-drive", "file={},if=pflash,format=raw".format(img)]) self.emulator.login() cmd = "mount | grep 'ubi0:rootfs on / type ubifs'" _, exit_code = self.emulator.run(cmd)Signature · 6 lines
2.31.1 _______________________________________________ buildroot mailing list [email protected] https://lists.buildroot.org/mailman/listinfo/buildroot
Signature · 10 lines
.-----------------.--------------------.------------------.--------------------. | Yann E. MORIN | Real-Time Embedded | /"\ ASCII RIBBON | Erics' conspiracy: | | +33 662 376 056 | Software Designer | \ / CAMPAIGN | ___ | | +33 561 099 427 `------------.-------: X AGAINST | \e/ There is no | | http://ymorin.is-a-geek.org/ | _/*\_ | / \ HTML MAIL | v conspiracy. | '------------------------------^-------^------------------^--------------------' _______________________________________________ buildroot mailing list [email protected] https://lists.buildroot.org/mailman/listinfo/buildroot
PATCH 15/16 support/testing: update logical eraseblock and physical eraseblock size for qemu >= 2.9
In reply to Romain NaourThe current ubi/ubifs test (test_ubi.py) rely on a Qemu bug present in 2.8.0 that was fixed in Qemu 2.9.0 [1]. The ubi/ubifs settings is updated to run with Qemu >= 2.9.0 using the new multiple chip handling.
If needed, the old behavior can be enabled using the pflash01 property "old-multiple-chip-handling" [2].
The issue was not detected until now since we are sill using an old qemu (2.8 from Debian stretch) for testing in gitlab (using the Buildroot Docker image used by gitlab-ci.yml).
First the logical eraseblock size (LEB) must be updated to the value 0x3ff80 reported by the kernel when using qemu >= 2.9.0.
UBIFS (ubi0:0): Mounting in unauthenticated mode
UBIFS error (ubi0:0 pid 1): ubifs_read_superblock: LEB size mismatch: 524160 in superblock, 262016 real UBIFS error (ubi0:0 pid 1): ubifs_read_superblock: bad superblock, error 1
But the system is still failing to boot:
UBIFS error (ubi0:0 pid 1): ubifs_scan: garbage
UBIFS error (ubi0:0 pid 1): ubifs_recover_master_node: failed to recover master node
ubifs is reading garbage since Qemu >= 2.9.0 report a sector length per device divided by the number of devices (see commit [1]).
The kernel detect two flash devices (dmesg):
Concatenating MTD devices:
(0): "40000000.flash"
(1): "40000000.flash"
into device "40000000.flash"
Divide the physical eraseblock (PEB) size by two.
Tested with qemu 2.9.0, 5.1.0.
Fixes:
https://gitlab.com/kubu93/buildroot/-/jobs/1543100932
[1] https://git.qemu.org/?p=qemu.git;a=commitdiff;h=feb0b1aa11f14ee71660aba46b46387d1f923c9e
[2] http://lists.busybox.net/pipermail/buildroot/2021-September/622069.html
--- support/testing/tests/fs/test_ubi.py | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
support/testing/tests/fs/test_ubi.py +2 -2
@@ -8,10 +8,10 @@ class TestUbi(infra.basetest.BRTest): config = infra.basetest.BASIC_TOOLCHAIN_CONFIG + \ """ BR2_TARGET_ROOTFS_UBIFS=y - BR2_TARGET_ROOTFS_UBIFS_LEBSIZE=0x7ff80 + BR2_TARGET_ROOTFS_UBIFS_LEBSIZE=0x3ff80 BR2_TARGET_ROOTFS_UBIFS_MINIOSIZE=0x1 BR2_TARGET_ROOTFS_UBI=y - BR2_TARGET_ROOTFS_UBI_PEBSIZE=0x80000 + BR2_TARGET_ROOTFS_UBI_PEBSIZE=0x40000 BR2_TARGET_ROOTFS_UBI_SUBSIZE=1 """
Signature · 6 lines
2.31.1 _______________________________________________ buildroot mailing list [email protected] https://lists.buildroot.org/mailman/listinfo/buildroot
Yann E. MORIN
AppliedRe: [Buildroot] [PATCH 15/16] support/testing: update logical eraseblock and physical eraseblock size for qemu >= 2.9
Romain, All,
On 2021-09-12 16:11 +0200, Romain Naour spake thusly:
Quoted 35 lines
The current ubi/ubifs test (
test_ubi.py) rely on a Qemu bug present in 2.8.0 that was fixed in Qemu 2.9.0 [1]. The ubi/ubifs settings is updated to run with Qemu >= 2.9.0 using the new multiple chip handling.If needed, the old behavior can be enabled using the pflash01 property "old-multiple-chip-handling" [2].
The issue was not detected until now since we are sill using an old qemu (2.8 from Debian stretch) for testing in gitlab (using the Buildroot Docker image used by gitlab-ci.yml).
First the logical eraseblock size (LEB) must be updated to the value
0x3ff80reported by the kernel when using qemu >= 2.9.0.UBIFS (ubi0:0): Mounting in unauthenticated mode
UBIFS error (ubi0:0 pid 1):ubifs_read_superblock: LEB size mismatch: 524160 in superblock, 262016 real UBIFS error (ubi0:0 pid 1):ubifs_read_superblock: bad superblock, error 1But the system is still failing to boot:
UBIFS error (ubi0:0 pid 1):
ubifs_scan: garbage
UBIFS error (ubi0:0 pid 1):ubifs_recover_master_node: failed to recover master nodeubifs is reading garbage since Qemu >= 2.9.0 report a sector length per device divided by the number of devices (see commit [1]).
The kernel detect two flash devices (dmesg):
Concatenating MTD devices:
(0): "40000000.flash"
(1): "40000000.flash"
into device "40000000.flash"Divide the physical eraseblock (PEB) size by two.
Quoted 7 lines
Tested with qemu 2.9.0, 5.1.0.
Fixes:
https://gitlab.com/kubu93/buildroot/-/jobs/1543100932[1] https://git.qemu.org/?p=qemu.git;a=commitdiff;h=feb0b1aa11f14ee71660aba46b46387d1f923c9e
[2] http://lists.busybox.net/pipermail/buildroot/2021-September/622069.html
I am not sure I understood that reference: the patch in [2] was never applied, it is no longer present in patchwork, but I could not find a discsuuion about it either... Did you withdraw that patch?
Quoted 2 lines
Signed-off-by Romain Naour <[email protected]>Cc Thomas Petazzoni <[email protected]>
Applied to master, thanks.
Regards,
Yann E. MORIN.
Quoted 28 lines
--- support/testing/tests/fs/test_ubi.py | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)support/testing/tests/fs/test_ubi.py +2
-2@@ -8,10 +8,10 @@ class TestUbi(infra.basetest.BRTest): config = infra.basetest.BASIC_TOOLCHAIN_CONFIG + \ """ BR2_TARGET_ROOTFS_UBIFS=y - BR2_TARGET_ROOTFS_UBIFS_LEBSIZE=0x7ff80 + BR2_TARGET_ROOTFS_UBIFS_LEBSIZE=0x3ff80 BR2_TARGET_ROOTFS_UBIFS_MINIOSIZE=0x1 BR2_TARGET_ROOTFS_UBI=y - BR2_TARGET_ROOTFS_UBI_PEBSIZE=0x80000 + BR2_TARGET_ROOTFS_UBI_PEBSIZE=0x40000 BR2_TARGET_ROOTFS_UBI_SUBSIZE=1 """Signature · 6 lines
2.31.1 _______________________________________________ buildroot mailing list [email protected] https://lists.buildroot.org/mailman/listinfo/buildroot
Signature · 10 lines
.-----------------.--------------------.------------------.--------------------. | Yann E. MORIN | Real-Time Embedded | /"\ ASCII RIBBON | Erics' conspiracy: | | +33 662 376 056 | Software Designer | \ / CAMPAIGN | ___ | | +33 561 099 427 `------------.-------: X AGAINST | \e/ There is no | | http://ymorin.is-a-geek.org/ | _/*\_ | / \ HTML MAIL | v conspiracy. | '------------------------------^-------^------------------^--------------------' _______________________________________________ buildroot mailing list [email protected] https://lists.buildroot.org/mailman/listinfo/buildroot
Re: [Buildroot] [PATCH 15/16] support/testing: update logical eraseblock and physical eraseblock size for qemu >= 2.9
Hello Yann,
Le 17/10/2021 à 21:51, Yann E. MORIN a écrit :
Quoted 43 lines
Romain, All,
On 2021-09-12 16:11 +0200, Romain Naour spake thusly:
Quoted 40 lines
The current ubi/ubifs test (
test_ubi.py) rely on a Qemu bug present in 2.8.0 that was fixed in Qemu 2.9.0 [1]. The ubi/ubifs settings is updated to run with Qemu >= 2.9.0 using the new multiple chip handling.If needed, the old behavior can be enabled using the pflash01 property "old-multiple-chip-handling" [2].
The issue was not detected until now since we are sill using an old qemu (2.8 from Debian stretch) for testing in gitlab (using the Buildroot Docker image used by gitlab-ci.yml).
First the logical eraseblock size (LEB) must be updated to the value
0x3ff80reported by the kernel when using qemu >= 2.9.0.UBIFS (ubi0:0): Mounting in unauthenticated mode
UBIFS error (ubi0:0 pid 1):ubifs_read_superblock: LEB size mismatch: 524160 in superblock, 262016 real UBIFS error (ubi0:0 pid 1):ubifs_read_superblock: bad superblock, error 1But the system is still failing to boot:
UBIFS error (ubi0:0 pid 1):
ubifs_scan: garbage
UBIFS error (ubi0:0 pid 1):ubifs_recover_master_node: failed to recover master nodeubifs is reading garbage since Qemu >= 2.9.0 report a sector length per device divided by the number of devices (see commit [1]).
The kernel detect two flash devices (dmesg):
Concatenating MTD devices:
(0): "40000000.flash"
(1): "40000000.flash"
into device "40000000.flash"Divide the physical eraseblock (PEB) size by two.
Tested with qemu 2.9.0, 5.1.0.
Quoted 6 lines
Quoted 2 lines
[1] https://git.qemu.org/?p=qemu.git;a=commitdiff;h=feb0b1aa11f14ee71660aba46b46387d1f923c9e
[2] http://lists.busybox.net/pipermail/buildroot/2021-September/622069.htmlI am not sure I understood that reference: the patch in [2] was never applied, it is no longer present in patchwork, but I could not find a discsuuion about it either... Did you withdraw that patch?
Yes, the patch [2] was a first attempt to fix the issue. Thanks to this patch, it's not needed anymore.
Best regards,
Romain
Quoted 30 lines
Quoted 3 lines
Signed-off-by Romain Naour <[email protected]>Cc Thomas Petazzoni <[email protected]>Applied to master, thanks.
Regards,
Yann E. MORIN.Quoted 21 lines
--- support/testing/tests/fs/test_ubi.py | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)support/testing/tests/fs/test_ubi.py +2
-2@@ -8,10 +8,10 @@ class TestUbi(infra.basetest.BRTest): config = infra.basetest.BASIC_TOOLCHAIN_CONFIG + \ """ BR2_TARGET_ROOTFS_UBIFS=y - BR2_TARGET_ROOTFS_UBIFS_LEBSIZE=0x7ff80 + BR2_TARGET_ROOTFS_UBIFS_LEBSIZE=0x3ff80 BR2_TARGET_ROOTFS_UBIFS_MINIOSIZE=0x1 BR2_TARGET_ROOTFS_UBI=y - BR2_TARGET_ROOTFS_UBI_PEBSIZE=0x80000 + BR2_TARGET_ROOTFS_UBI_PEBSIZE=0x40000 BR2_TARGET_ROOTFS_UBI_SUBSIZE=1 """
Quoted 8 lines
Quoted 7 lines
Signature · 6 lines
2.31.1 _______________________________________________ buildroot mailing list [email protected] https://lists.buildroot.org/mailman/listinfo/buildroot
_______________________________________________
buildroot mailing list
[email protected]
https://lists.buildroot.org/mailman/listinfo/buildroot
PATCH 16/16 support/testing: test_ubi: allow to boot several times using the same ubi image.
In reply to Romain NaourSince the ubi/ubifs test has been introduced, it's not possible to boot the same ubi image twice [1]:
"TODO: if you boot Qemu twice on the same UBI image, it fails to attach the image the second time, with "ubi0 error:ubi_read_volume_table: the layout volume was not found"."
For some reason, the kernel corrupt the ubi image if the ubifs rootfs is mounted with write access. Use a custom config file to mount the rootfs readonly (vol_type=static). Doing so requires to add the flash size (vol_size=64MiB).
At least it allows to boot several times the same ubi image.
[1] bf4a6490e4ee70f0a46e588602995ba34e6c872a
--- support/testing/tests/fs/test_ubi.py | 9 ++++----- .../tests/fs/test_ubi/ubinize_qemu_pflash_cfi01.cfg | 8 ++++++++ 2 files changed, 12 insertions(+), 5 deletions(-) create mode 100644 support/testing/tests/fs/test_ubi/ubinize_qemu_pflash_cfi01.cfg
support/testing/tests/fs/test_ubi.py +4 -5
@@ -13,12 +13,11 @@ class TestUbi(infra.basetest.BRTest): BR2_TARGET_ROOTFS_UBI=y BR2_TARGET_ROOTFS_UBI_PEBSIZE=0x40000 BR2_TARGET_ROOTFS_UBI_SUBSIZE=1 - """ + BR2_TARGET_ROOTFS_UBI_USE_CUSTOM_CONFIG=y + BR2_TARGET_ROOTFS_UBI_CUSTOM_CONFIG_FILE="{}" + """.format( + infra.filepath("tests/fs/test_ubi/ubinize_qemu_pflash_cfi01.cfg")) - # TODO: if you boot Qemu twice on the same UBI image, it fails to - # attach the image the second time, with "ubi0 error: - # ubi_read_volume_table: the layout volume was not found". - # To be investigated. def test_run(self): img = os.path.join(self.builddir, "images", "rootfs.ubi") out = infra.run_cmd_on_host(self.builddir, ["file", img])
support/testing/tests/fs/test_ubi/ubinize_qemu_pflash_cfi01.cfg new file +8 -0
@@ -0,0 +1,8 @@ +[ubifs] +mode=ubi +vol_id=0 +vol_type=static +vol_name=rootfs +vol_alignment=1 +vol_size=64MiB +image=BR2_ROOTFS_UBIFS_PATH
Signature · 6 lines
2.31.1 _______________________________________________ buildroot mailing list [email protected] https://lists.buildroot.org/mailman/listinfo/buildroot
Yann E. MORIN
AppliedRe: [Buildroot] [PATCH 16/16] support/testing: test_ubi: allow to boot several times using the same ubi image.
Romain, All,
On 2021-09-12 16:11 +0200, Romain Naour spake thusly:
Quoted 12 lines
Since the ubi/ubifs test has been introduced, it's not possible to boot the same ubi image twice [1]:
"TODO: if you boot Qemu twice on the same UBI image, it fails to attach the image the second time, with "ubi0 error:
ubi_read_volume_table: the layout volume was not found"."For some reason, the kernel corrupt the ubi image if the ubifs rootfs is mounted with write access. Use a custom config file to mount the rootfs readonly (vol_type=static). Doing so requires to add the flash size (vol_size=64MiB).
Quoted 6 lines
At least it allows to boot several times the same ubi image.
[1]
bf4a6490e4ee70f0a46e588602995ba34e6c872aSigned-off-by Romain Naour <[email protected]>Cc Thomas Petazzoni <[email protected]>
Applied to master, thanks.
Regards,
Yann E. MORIN.
Quoted 48 lines
--- support/testing/tests/fs/test_ubi.py | 9 ++++----- .../tests/fs/test_ubi/ubinize_qemu_pflash_cfi01.cfg | 8 ++++++++ 2 files changed, 12 insertions(+), 5 deletions(-) create mode 100644 support/testing/tests/fs/test_ubi/ubinize_qemu_pflash_cfi01.cfgsupport/testing/tests/fs/test_ubi.py +4
-5@@ -13,12 +13,11 @@ class TestUbi(infra.basetest.BRTest): BR2_TARGET_ROOTFS_UBI=y BR2_TARGET_ROOTFS_UBI_PEBSIZE=0x40000 BR2_TARGET_ROOTFS_UBI_SUBSIZE=1 - """ + BR2_TARGET_ROOTFS_UBI_USE_CUSTOM_CONFIG=y + BR2_TARGET_ROOTFS_UBI_CUSTOM_CONFIG_FILE="{}" + """.format( + infra.filepath("tests/fs/test_ubi/ubinize_qemu_pflash_cfi01.cfg")) - # TODO: if you boot Qemu twice on the same UBI image, it fails to - # attach the image the second time, with "ubi0 error: - # ubi_read_volume_table: the layout volume was not found". - # To be investigated. def test_run(self): img = os.path.join(self.builddir, "images", "rootfs.ubi") out = infra.run_cmd_on_host(self.builddir, ["file", img])support/testing/tests/fs/test_ubi/ubinize_qemu_pflash_cfi01.cfg new file +8
-0@@ -0,0 +1,8 @@ +[ubifs] +mode=ubi +vol_id=0 +vol_type=static +vol_name=rootfs +vol_alignment=1 +vol_size=64MiB +image=BR2_ROOTFS_UBIFS_PATHSignature · 6 lines
2.31.1 _______________________________________________ buildroot mailing list [email protected] https://lists.buildroot.org/mailman/listinfo/buildroot
Signature · 10 lines
.-----------------.--------------------.------------------.--------------------. | Yann E. MORIN | Real-Time Embedded | /"\ ASCII RIBBON | Erics' conspiracy: | | +33 662 376 056 | Software Designer | \ / CAMPAIGN | ___ | | +33 561 099 427 `------------.-------: X AGAINST | \e/ There is no | | http://ymorin.is-a-geek.org/ | _/*\_ | / \ HTML MAIL | v conspiracy. | '------------------------------^-------^------------------^--------------------' _______________________________________________ buildroot mailing list [email protected] https://lists.buildroot.org/mailman/listinfo/buildroot
Re: [Buildroot] [PATCH 03/16] support/testing: test_f2fs: bump to kernel 4.19.204
In reply to Romain NaourQuoted 5 lines
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
"Romain" == Romain Naour <[email protected]> writes:
The kernel 4.16.7 curently used by the test doesn't build with host gcc >= 10 due the gcc default -fno-common. See GCC 10 porting guide [1].
Quoted 7 lines
/usr/bin/ld: scripts/dtc/dtc-parser.tab.o:(.bss+0x20): multiple definition of `yylloc'; scripts/dtc/dtc-lexer.lex.o:(.bss+0x0): first defined hereBump to the next LTS release.
[1] https://gcc.gnu.org/gcc-10/porting_to.html
Signed-off-by Romain Naour <[email protected]>
Committed to 2021.02.x, 2021.05.x and 2021.08.x, thanks.
Signature · 5 lines
Bye, Peter Korsgaard _______________________________________________ buildroot mailing list [email protected] https://lists.buildroot.org/mailman/listinfo/buildroot