Buildroot defconfigs about to be removed
Hello,
This is another reminder about the below defconfigs, which are failing to build in Buildroot. If they are not fixed soon, they will be removed before the 2022.08 release, expected to be released at the end of August.
Please let us know if you are no longer interested in maintaining some of those defconfigs.
Thanks for your contribution!
Thomas
rockpro64_defconfig -> arm-trusted-firmware fails to build
https://gitlab.com/buildroot.org/buildroot/-/jobs/2812054016
rock_pi_ni10_defconfig -> arm-trusted-firmware fails to build
https://gitlab.com/buildroot.org/buildroot/-/jobs/2812054009
rock_pi_4_defconfig -> arm-trusted-firmware fails to build
https://gitlab.com/buildroot.org/buildroot/-/jobs/2812054006
roc_pc_rk3399_defconfig -> arm-trusted-firmware fails to build
https://gitlab.com/buildroot.org/buildroot/-/jobs/2812053999
raspberrypi3_qt5we_defconfig -> build failure in qt5webengine
https://gitlab.com/buildroot.org/buildroot/-/jobs/2812053961
qemu_xtensa_lx60_nommu_defconfig -> elf2flt segfaults
https://gitlab.com/buildroot.org/buildroot/-/jobs/2812053940
qemu_m68k_mcf5208_defconfig -> elf2flt segfaults
https://gitlab.com/buildroot.org/buildroot/-/jobs/2812053827
pine64_sopine_defconfig -> arm-trusted-firmware fails to build
https://gitlab.com/buildroot.org/buildroot/-/jobs/2812053812
pine64_defconfig -> arm-trusted-firmware fails to build
https://gitlab.com/buildroot.org/buildroot/-/jobs/2812053810
orangepi_win_defconfig -> arm-trusted-firmware fails to build
https://gitlab.com/buildroot.org/buildroot/-/jobs/2812053792
orangepi_rk3399_defconfig -> arm-trusted-firmware fails to build
https://gitlab.com/buildroot.org/buildroot/-/jobs/2812053790
orangepi_prime_defconfig -> arm-trusted-firmware fails to build
https://gitlab.com/buildroot.org/buildroot/-/jobs/2812053786
orangepi_plus_defconfig -> U-Boot needs Python 2.x
https://gitlab.com/buildroot.org/buildroot/-/jobs/2812053784
orangepi_pc2_defconfig -> arm-trusted-firmware fails to build
https://gitlab.com/buildroot.org/buildroot/-/jobs/2812053776
orangepi_one_plus_defconfig -> Device Tree syntax error
https://gitlab.com/buildroot.org/buildroot/-/jobs/2812053771
kontron_bl_imx8mm_defconfig -> -lgnutls -luuid missing when building U-Boot. Strange thing is that host-gnutls is definitely built before U-Boot, but it doesn't find it. for libuuid, BR2_TARGET_UBOOT_NEEDS_UTIL_LINUX=y is missing
https://gitlab.com/buildroot.org/buildroot/-/jobs/2812053608
friendlyarm_nanopi_r1_defconfig -> U-Boot needs Python 2.x
https://gitlab.com/buildroot.org/buildroot/-/jobs/2812053547
friendlyarm_nanopi_neo_plus2_defconfig -> arm-trusted-firmware fails to build
https://gitlab.com/buildroot.org/buildroot/-/jobs/2812053545
friendlyarm_nanopi_neo_defconfig -> U-Boot needs Python 2.x
https://gitlab.com/buildroot.org/buildroot/-/jobs/2812053540
friendlyarm_nanopi_neo4_defconfig -> arm-trusted-firmware fails to build
https://gitlab.com/buildroot.org/buildroot/-/jobs/2812053537
friendlyarm_nanopi_neo2_defconfig -> arm-trusted-firmware fails to build
https://gitlab.com/buildroot.org/buildroot/-/jobs/2812053533
friendlyarm_nanopi_m4_defconfig -> arm-trusted-firmware fails to build
https://gitlab.com/buildroot.org/buildroot/-/jobs/2812053529
friendlyarm_nanopi_m1_plus_defconfig -> U-Boot needs Python 2.x
https://gitlab.com/buildroot.org/buildroot/-/jobs/2812053526
friendlyarm_nanopi_m1_defconfig -> U-Boot needs Python 2.x
https://gitlab.com/buildroot.org/buildroot/-/jobs/2812053522
friendlyarm_nanopi_a64_defconfig -> arm-trusted-firmware fails to build
https://gitlab.com/buildroot.org/buildroot/-/jobs/2812053520
friendlyarm_nanopc_t4_defconfig -> arm-trusted-firmware fails to build
https://gitlab.com/buildroot.org/buildroot/-/jobs/2812053518
bananapi_m64_defconfig -> arm-trusted-firmware fails to build
https://gitlab.com/buildroot.org/buildroot/-/jobs/2812053420
bananapi_m2_plus_defconfig -> U-Boot needs Python 2.x
https://gitlab.com/buildroot.org/buildroot/-/jobs/2812053415
bananapi_m1_plus_defconfig -> U-Boot needs Python 2.x
https://gitlab.com/buildroot.org/buildroot/-/jobs/2812053412
bananapi_m1_defconfig -> Device Tree syntax error
https://gitlab.com/buildroot.org/buildroot/-/jobs/2812053409
arm_juno_defconfig -> arm-trusted-firmware fails to build
https://gitlab.com/buildroot.org/buildroot/-/jobs/2812053341
amarula_a64_relic_defconfig -> arm-trusted-firmware fails to build
https://gitlab.com/buildroot.org/buildroot/-/jobs/2812053333
Signature · 7 lines
Thomas Petazzoni, co-owner and CEO, Bootlin Embedded Linux and Kernel engineering and training https://bootlin.com _______________________________________________ buildroot mailing list [email protected] https://lists.buildroot.org/mailman/listinfo/buildroot
Hi Thomas,
[snip]
Quoted 5 lines
kontron_bl_imx8mm_defconfig-> -lgnutls -luuid missing when building U-Boot. Strange thing is that host-gnutls is definitely built before U-Boot, but it doesn't find it. for libuuid, BR2_TARGET_UBOOT_NEEDS_UTIL_LINUX=y is missing
https://gitlab.com/buildroot.org/buildroot/-/jobs/2812053608
This is the first time I have been informed about this build failure. I am surprised that I did not get any mail about this.
I must confess that I could have guessed that this would be the same error as with the kontron_pitx_imx8m_defconfig. My fault.
Nonetheless I will fix that and send a patch.
Signature · 1 line
Heiko
Quoted 45 lines
friendlyarm_nanopi_r1_defconfig-> U-Boot needs Python 2.x
https://gitlab.com/buildroot.org/buildroot/-/jobs/2812053547
friendlyarm_nanopi_neo_plus2_defconfig-> arm-trusted-firmware fails to build
https://gitlab.com/buildroot.org/buildroot/-/jobs/2812053545
friendlyarm_nanopi_neo_defconfig-> U-Boot needs Python 2.x
https://gitlab.com/buildroot.org/buildroot/-/jobs/2812053540
friendlyarm_nanopi_neo4_defconfig-> arm-trusted-firmware fails to build
https://gitlab.com/buildroot.org/buildroot/-/jobs/2812053537
friendlyarm_nanopi_neo2_defconfig-> arm-trusted-firmware fails to build
https://gitlab.com/buildroot.org/buildroot/-/jobs/2812053533
friendlyarm_nanopi_m4_defconfig-> arm-trusted-firmware fails to build
https://gitlab.com/buildroot.org/buildroot/-/jobs/2812053529
friendlyarm_nanopi_m1_plus_defconfig-> U-Boot needs Python 2.x
https://gitlab.com/buildroot.org/buildroot/-/jobs/2812053526
friendlyarm_nanopi_m1_defconfig-> U-Boot needs Python 2.x
https://gitlab.com/buildroot.org/buildroot/-/jobs/2812053522
friendlyarm_nanopi_a64_defconfig-> arm-trusted-firmware fails to build
https://gitlab.com/buildroot.org/buildroot/-/jobs/2812053520
friendlyarm_nanopc_t4_defconfig-> arm-trusted-firmware fails to build
https://gitlab.com/buildroot.org/buildroot/-/jobs/2812053518
bananapi_m64_defconfig-> arm-trusted-firmware fails to build
https://gitlab.com/buildroot.org/buildroot/-/jobs/2812053420
bananapi_m2_plus_defconfig-> U-Boot needs Python 2.x
https://gitlab.com/buildroot.org/buildroot/-/jobs/2812053415
bananapi_m1_plus_defconfig-> U-Boot needs Python 2.x
https://gitlab.com/buildroot.org/buildroot/-/jobs/2812053412
bananapi_m1_defconfig-> Device Tree syntax error
https://gitlab.com/buildroot.org/buildroot/-/jobs/2812053409
arm_juno_defconfig-> arm-trusted-firmware fails to build
https://gitlab.com/buildroot.org/buildroot/-/jobs/2812053341
Quoted 8 lines
amarula_a64_relic_defconfig-> arm-trusted-firmware fails to build
https://gitlab.com/buildroot.org/buildroot/-/jobs/2812053333--
Thomas Petazzoni, co-owner and CEO, Bootlin Embedded Linux and Kernel engineering and training
https://bootlin.com
_______________________________________________
buildroot mailing list
[email protected]
https://lists.buildroot.org/mailman/listinfo/buildroot
Hello Heiko,
Quoted 2 lines · On Sun, 7 Aug 2022 00:28:46 +0200 Heiko Thiery <[email protected]> wrote:
This is the first time I have been informed about this build failure. I am surprised that I did not get any mail about this.
Sorry, your issue is potentially new. It's mainly the other issues that I have already reported in the past, and which did not get any feedback, on defconfigs that are not properly maintained (all those friendlyarm/bananapi/orangepi boards).
Quoted 1 line
Nonetheless I will fix that and send a patch.
Thanks a lot!
Thomas
Signature · 7 lines
Thomas Petazzoni, co-owner and CEO, Bootlin Embedded Linux and Kernel engineering and training https://bootlin.com _______________________________________________ buildroot mailing list [email protected] https://lists.buildroot.org/mailman/listinfo/buildroot
Quoted 5 lines · Thomas Petazzoni
Hello Heiko,
On Sun, 7 Aug 2022 00:28:46 +0200
Heiko Thiery <[email protected]> wrote:
Quoted 8 lines
Quoted 2 lines
This is the first time I have been informed about this build failure. I am surprised that I did not get any mail about this.
Sorry, your issue is potentially new. It's mainly the other issues that I have already reported in the past, and which did not get any feedback, on defconfigs that are not properly maintained (all those friendlyarm/bananapi/orangepi boards).
Hi Thomas,
I will try to fix the defconfigs related to friendlyarm/bananapi/orangepi boards ASAP.
Thanks
Suniel
Quoted 11 lines
Quoted 2 lines
Nonetheless I will fix that and send a patch.
Thanks a lot!
Thomas
--
Thomas Petazzoni, co-owner and CEO, Bootlin Embedded Linux and Kernel engineering and training
https://bootlin.com
_______________________________________________
buildroot mailing list
[email protected]
https://lists.buildroot.org/mailman/listinfo/buildroot
Am Sat, 6 Aug 2022 22:43:38 +0200
schrieb Thomas Petazzoni <[email protected]>:
Quoted 9 lines
Hello,
This is another reminder about the below defconfigs, which are failing to build in Buildroot. If they are not fixed soon, they will be removed before the 2022.08 release, expected to be released at the end of August.
Please let us know if you are no longer interested in maintaining some of those defconfigs.
[...]
Quoted 5 lines
qemu_xtensa_lx60_nommu_defconfig-> elf2flt segfaults
https://gitlab.com/buildroot.org/buildroot/-/jobs/2812053940
qemu_m68k_mcf5208_defconfig-> elf2flt segfaults
https://gitlab.com/buildroot.org/buildroot/-/jobs/2812053827
I'm interested in keeping the qemu defconfigs alive, please don't remove them yet. I'll try to figure out what's going wrong here.
Thomas
_______________________________________________
buildroot mailing list
[email protected]
https://lists.buildroot.org/mailman/listinfo/buildroot
Hello Thomas,
Quoted 8 lines · On Mon, 8 Aug 2022 10:14:52 +0200 Thomas Huth <[email protected]> wrote:
Quoted 5 lines
qemu_xtensa_lx60_nommu_defconfig-> elf2flt segfaults
https://gitlab.com/buildroot.org/buildroot/-/jobs/2812053940
qemu_m68k_mcf5208_defconfig-> elf2flt segfaults
https://gitlab.com/buildroot.org/buildroot/-/jobs/2812053827I'm interested in keeping the qemu defconfigs alive, please don't remove them yet. I'll try to figure out what's going wrong here.
So these are "special" issues. The elf2flt segfault is caused by the changes in elf2flt to support RISC-V 64-bit noMMU, in commit 32f93b0ef2c602dfa1f5b243ba865d5b16e2fa33.
However, even with this removed, we have issues with elf2flt. See
https://lore.kernel.org/buildroot/20220725164425.7316b2e3@windsurf/ for more details, and also
https://lore.kernel.org/buildroot/[email protected]/.
Best regards,
Thomas
Signature · 7 lines
Thomas Petazzoni, co-owner and CEO, Bootlin Embedded Linux and Kernel engineering and training https://bootlin.com _______________________________________________ buildroot mailing list [email protected] https://lists.buildroot.org/mailman/listinfo/buildroot
Am Mon, 8 Aug 2022 10:26:53 +0200
schrieb Thomas Petazzoni <[email protected]>:
Quoted 14 lines
Hello Thomas,
Quoted 8 lines · On Mon, 8 Aug 2022 10:14:52 +0200 Thomas Huth <[email protected]> wrote:
Quoted 5 lines
qemu_xtensa_lx60_nommu_defconfig-> elf2flt segfaults
https://gitlab.com/buildroot.org/buildroot/-/jobs/2812053940
qemu_m68k_mcf5208_defconfig-> elf2flt segfaults
https://gitlab.com/buildroot.org/buildroot/-/jobs/2812053827I'm interested in keeping the qemu defconfigs alive, please don't remove them yet. I'll try to figure out what's going wrong here.
Quoted 8 lines
So these are "special" issues. The elf2flt segfault is caused by the changes in elf2flt to support RISC-V 64-bit noMMU, in commit
32f93b0ef2c602dfa1f5b243ba865d5b16e2fa33.However, even with this removed, we have issues with elf2flt. See
https://lore.kernel.org/buildroot/20220725164425.7316b2e3@windsurf/ for more details, and also
https://lore.kernel.org/buildroot/[email protected]/.
Thanks for the pointers! I had a look at the source for a while, and from what I can see, I came to the same conclusion that the modifications for ARM should really not have been done for m68k / binutils 2.32.
I can get the mcf5208 board back working by applying a patch like this on top of the current state in buildroot (fixes both, the segfault and the section overlapping):
build/host-elf2flt-2021.08/elf2flt.c.new renamed from build/host-elf2flt-2021.08/elf2flt.c +15 -14
@@ -346,6 +346,17 @@ } #endif +static int is_text_sec(asection *s) +{ +#if defined(TARGET_m68k) || defined(TARGET_xtensa) + return (s->flags & SEC_CODE); +#else + return ((s->flags & SEC_CODE) || + ((s->flags & (SEC_DATA | SEC_READONLY | SEC_RELOC)) == + (SEC_DATA | SEC_READONLY | SEC_RELOC))); +#endif +} + static uint32_t * output_relocs ( bfd *abs_bfd,
@@ -435,10 +446,7 @@ * Only relocate things in the writable data sections if we are PIC/GOT. * Otherwise do text (and read only data) as well. */ - if ((!pic_with_got || ALWAYS_RELOC_TEXT) && - ((a->flags & SEC_CODE) || - ((a->flags & (SEC_DATA | SEC_READONLY | SEC_RELOC)) == - (SEC_DATA | SEC_READONLY | SEC_RELOC)))) + if ((!pic_with_got || ALWAYS_RELOC_TEXT) && is_text_sec(a)) sectionp = text + (a->vma - text_vma); else if (a->flags & SEC_DATA) sectionp = data + (a->vma - data_vma);
@@ -1901,10 +1909,7 @@ bfd_size_type sec_size; bfd_vma sec_vma; - if ((s->flags & SEC_CODE) || - (((s->flags & (SEC_DATA | SEC_READONLY | SEC_RELOC)) == - (SEC_DATA | SEC_READONLY | SEC_RELOC)) && - !data_len)) { + if (is_text_sec(s) && !data_len) { vma = &text_vma; len = &text_len; } else if (s->flags & SEC_DATA) {
@@ -1941,9 +1946,7 @@ * Includes code sections, but also includes read-only relocation * data sections.*/ for (s = abs_bfd->sections; s != NULL; s = s->next) - if ((s->flags & SEC_CODE) || - ((s->flags & (SEC_DATA | SEC_READONLY | SEC_RELOC)) == - (SEC_DATA | SEC_READONLY | SEC_RELOC))) + if (is_text_sec(s)) if (!bfd_get_section_contents(abs_bfd, s, text + (s->vma - text_vma), 0, elf2flt_bfd_section_size(s)))
@@ -1971,9 +1974,7 @@ * Includes data sections, but not those read-only relocation * data sections already included in the text output section.*/ for (s = abs_bfd->sections; s != NULL; s = s->next) - if ((s->flags & SEC_DATA) && - ((s->flags & (SEC_READONLY | SEC_RELOC)) != - (SEC_READONLY | SEC_RELOC))) + if ((s->flags & SEC_DATA) && !is_text_sec(s)) if (!bfd_get_section_contents(abs_bfd, s, data + (s->vma - data_vma), 0, elf2flt_bfd_section_size(s)))
Does that look passable for buildroot? If so, I can send it as a proper patch for buildroot (i.e. a patch that adds a patch to the packages/elf2flt/ directory) and also try to get it upstream later. If not ... I'm also running out of ideas here ... then maybe it's really time to retire the old boards that depend on binutils 2.32 nowadays...
Thomas
_______________________________________________
buildroot mailing list
[email protected]
https://lists.buildroot.org/mailman/listinfo/buildroot
Hello Thomas,
Quoted 7 lines · On Mon, 8 Aug 2022 21:59:00 +0200 Thomas Huth <[email protected]> wrote:
Thanks for the pointers! I had a look at the source for a while, and from what I can see, I came to the same conclusion that the modifications for ARM should really not have been done for m68k / binutils 2.32.
I can get the mcf5208 board back working by applying a patch like this on top of the current state in buildroot (fixes both, the segfault and the section overlapping):
FYI, a patch was posted and merged:
https://git.buildroot.org/buildroot/commit/?id=95402089c314fbc03b934f5fcf9c29db043d758c
https://git.buildroot.org/buildroot/commit/?id=9dd179d43ffe9be98035f921a0db2bf051b632fb
Don't hesitate to give a test on your side!
Best regards,
Thomas
Signature · 7 lines
Thomas Petazzoni, co-owner and CEO, Bootlin Embedded Linux and Kernel engineering and training https://bootlin.com _______________________________________________ buildroot mailing list [email protected] https://lists.buildroot.org/mailman/listinfo/buildroot
Am Fri, 12 Aug 2022 15:52:33 +0200
schrieb Thomas Petazzoni <[email protected]>:
Quoted 13 lines
Hello Thomas,
Quoted 7 lines · On Mon, 8 Aug 2022 21:59:00 +0200 Thomas Huth <[email protected]> wrote:
Thanks for the pointers! I had a look at the source for a while, and from what I can see, I came to the same conclusion that the modifications for ARM should really not have been done for m68k / binutils 2.32.
I can get the mcf5208 board back working by applying a patch like this on top of the current state in buildroot (fixes both, the segfault and the section overlapping):
Quoted 6 lines
FYI, a patch was posted and merged:
https://git.buildroot.org/buildroot/commit/?id=95402089c314fbc03b934f5fcf9c29db043d758c
https://git.buildroot.org/buildroot/commit/?id=9dd179d43ffe9be98035f921a0db2bf051b632fbDon't hesitate to give a test on your side!
Thanks, I've checked it and it works fine for me, too!
Thomas H
_______________________________________________
buildroot mailing list
[email protected]
https://lists.buildroot.org/mailman/listinfo/buildroot