* [Buildroot] [PATCH 0/3] Fix our defconfigs (branch yem/fix-defconfigs)
@ 2016-08-07 20:44 Yann E. MORIN
2016-08-07 20:44 ` [Buildroot] [PATCH 1/3] defconfigs/kb9202: fix headers version Yann E. MORIN
` (3 more replies)
0 siblings, 4 replies; 6+ messages in thread
From: Yann E. MORIN @ 2016-08-07 20:44 UTC (permalink / raw)
To: buildroot
Hello All!
This series fixes most build failures for our defconfigs:
- patch 1 fixes kb9202 after we changed the way we handled the kernel
headers selection to default to the built kernel;
- patch 2 fixes the build of uboot for the snps defconfigs, as it
needs dtc;
- patch 3 fixes all our thumb2-only altera and atmel defconfigs, by
reverting an upstream commit that broke the build.
There are a two other build failures for demo-like defconfigs. Those are
not yet fixed.
Series meant to fix failures seen with Travis:
https://travis-ci.org/buildroot/buildroot-defconfig-testing/builds/150333638
Regards,
Yann E. MORIN.
The following changes since commit 023bba7797132713c9e656cd45846596a2b16bf5
package/hidapi: depends on NPTL threads (2016-08-06 23:59:38 +0200)
are available in the git repository at:
https://gitlab.com/ymorin/buildroot.git
for you to fetch changes up to f410a42ac202f8266ff8f4ace32f97e5c68ac0fe
pcakage/uclibc: fix build on thumb2-only arm cpus (2016-08-07 22:39:30 +0200)
----------------------------------------------------------------
Yann E. MORIN (3):
defconfigs/kb9202: fix headers version
defconfigs/snps_axs_10*: uboot needs dtc
pcakage/uclibc: fix build on thumb2-only arm cpus
configs/kb9202_defconfig | 3 +-
configs/snps_axs101_defconfig | 1 +
configs/snps_axs103_defconfig | 1 +
...arm-cleanup-redundant-macros-for-syscalls.patch | 150 +++++++++++++++++++++
4 files changed, 154 insertions(+), 1 deletion(-)
create mode 100644 package/uclibc/0001-Revert-arm-cleanup-redundant-macros-for-syscalls.patch
--
.-----------------.--------------------.------------------.--------------------.
| Yann E. MORIN | Real-Time Embedded | /"\ ASCII RIBBON | Erics' conspiracy: |
| +33 662 376 056 | Software Designer | \ / CAMPAIGN | ___ |
| +33 223 225 172 `------------.-------: X AGAINST | \e/ There is no |
| http://ymorin.is-a-geek.org/ | _/*\_ | / \ HTML MAIL | v conspiracy. |
'------------------------------^-------^------------------^--------------------'
^ permalink raw reply [flat|nested] 6+ messages in thread* [Buildroot] [PATCH 1/3] defconfigs/kb9202: fix headers version 2016-08-07 20:44 [Buildroot] [PATCH 0/3] Fix our defconfigs (branch yem/fix-defconfigs) Yann E. MORIN @ 2016-08-07 20:44 ` Yann E. MORIN 2016-08-07 20:44 ` [Buildroot] [PATCH 2/3] defconfigs/snps_axs_10*: uboot needs dtc Yann E. MORIN ` (2 subsequent siblings) 3 siblings, 0 replies; 6+ messages in thread From: Yann E. MORIN @ 2016-08-07 20:44 UTC (permalink / raw) To: buildroot Fixes: https://travis-ci.org/buildroot/buildroot-defconfig-testing/jobs/150333697 Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr> --- configs/kb9202_defconfig | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/configs/kb9202_defconfig b/configs/kb9202_defconfig index c38adbf..0aea0eb 100644 --- a/configs/kb9202_defconfig +++ b/configs/kb9202_defconfig @@ -5,7 +5,8 @@ BR2_arm920t=y # Filesystem BR2_TARGET_ROOTFS_TAR=y -# Linux headers same as kernel, a 2.6 series (so, no option selected) +# Linux headers same as kernel, a 2.6 series +BR2_PACKAGE_HOST_LINUX_HEADERS_CUSTOM_REALLY_OLD=y # Kernel BR2_LINUX_KERNEL=y -- 2.7.4 ^ permalink raw reply related [flat|nested] 6+ messages in thread
* [Buildroot] [PATCH 2/3] defconfigs/snps_axs_10*: uboot needs dtc 2016-08-07 20:44 [Buildroot] [PATCH 0/3] Fix our defconfigs (branch yem/fix-defconfigs) Yann E. MORIN 2016-08-07 20:44 ` [Buildroot] [PATCH 1/3] defconfigs/kb9202: fix headers version Yann E. MORIN @ 2016-08-07 20:44 ` Yann E. MORIN 2016-08-07 20:44 ` [Buildroot] [PATCH 3/3] pcakage/uclibc: fix build on thumb2-only arm cpus Yann E. MORIN 2016-08-07 20:54 ` [Buildroot] [PATCH 0/3] Fix our defconfigs (branch yem/fix-defconfigs) Thomas Petazzoni 3 siblings, 0 replies; 6+ messages in thread From: Yann E. MORIN @ 2016-08-07 20:44 UTC (permalink / raw) To: buildroot Fixes: https://travis-ci.org/buildroot/buildroot-defconfig-testing/jobs/150333758 https://travis-ci.org/buildroot/buildroot-defconfig-testing/jobs/150333759 Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr> --- configs/snps_axs101_defconfig | 1 + configs/snps_axs103_defconfig | 1 + 2 files changed, 2 insertions(+) diff --git a/configs/snps_axs101_defconfig b/configs/snps_axs101_defconfig index f7bcf4d..f272853 100644 --- a/configs/snps_axs101_defconfig +++ b/configs/snps_axs101_defconfig @@ -23,3 +23,4 @@ BR2_TARGET_UBOOT_BUILD_SYSTEM_KCONFIG=y BR2_TARGET_UBOOT_CUSTOM_VERSION=y BR2_TARGET_UBOOT_CUSTOM_VERSION_VALUE="2016.07" BR2_TARGET_UBOOT_BOARD_DEFCONFIG="axs101" +BR2_TARGET_UBOOT_NEEDS_DTC=y diff --git a/configs/snps_axs103_defconfig b/configs/snps_axs103_defconfig index 840db74..75596c9 100644 --- a/configs/snps_axs103_defconfig +++ b/configs/snps_axs103_defconfig @@ -24,3 +24,4 @@ BR2_TARGET_UBOOT_BUILD_SYSTEM_KCONFIG=y BR2_TARGET_UBOOT_CUSTOM_VERSION=y BR2_TARGET_UBOOT_CUSTOM_VERSION_VALUE="2016.07" BR2_TARGET_UBOOT_BOARD_DEFCONFIG="axs103" +BR2_TARGET_UBOOT_NEEDS_DTC=y -- 2.7.4 ^ permalink raw reply related [flat|nested] 6+ messages in thread
* [Buildroot] [PATCH 3/3] pcakage/uclibc: fix build on thumb2-only arm cpus 2016-08-07 20:44 [Buildroot] [PATCH 0/3] Fix our defconfigs (branch yem/fix-defconfigs) Yann E. MORIN 2016-08-07 20:44 ` [Buildroot] [PATCH 1/3] defconfigs/kb9202: fix headers version Yann E. MORIN 2016-08-07 20:44 ` [Buildroot] [PATCH 2/3] defconfigs/snps_axs_10*: uboot needs dtc Yann E. MORIN @ 2016-08-07 20:44 ` Yann E. MORIN 2016-08-07 20:54 ` [Buildroot] [PATCH 0/3] Fix our defconfigs (branch yem/fix-defconfigs) Thomas Petazzoni 3 siblings, 0 replies; 6+ messages in thread From: Yann E. MORIN @ 2016-08-07 20:44 UTC (permalink / raw) To: buildroot Add a patch to revert an upstream commit that broke building for armv7 thumb2-only CPUs. Commit pointed out by Thomas. Fixes: https://travis-ci.org/buildroot/buildroot-defconfig-testing/jobs/150333657 https://travis-ci.org/buildroot/buildroot-defconfig-testing/jobs/150333658 ... and all the other atmel_sama5d* defconfigs https://travis-ci.org/buildroot/buildroot-defconfig-testing/jobs/150333643 https://travis-ci.org/buildroot/buildroot-defconfig-testing/jobs/150333644 Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr> Cc: Thomas Petazzoni <thomas.petazzoni@free-electrons.com> --- ...arm-cleanup-redundant-macros-for-syscalls.patch | 150 +++++++++++++++++++++ 1 file changed, 150 insertions(+) create mode 100644 package/uclibc/0001-Revert-arm-cleanup-redundant-macros-for-syscalls.patch diff --git a/package/uclibc/0001-Revert-arm-cleanup-redundant-macros-for-syscalls.patch b/package/uclibc/0001-Revert-arm-cleanup-redundant-macros-for-syscalls.patch new file mode 100644 index 0000000..184dc62 --- /dev/null +++ b/package/uclibc/0001-Revert-arm-cleanup-redundant-macros-for-syscalls.patch @@ -0,0 +1,150 @@ +From cdac8bef6b8e94518f78f2e9182cb3fa0f78a57b Mon Sep 17 00:00:00 2001 +From: "Yann E. MORIN" <yann.morin.1998@free.fr> +Date: Sun, 7 Aug 2016 21:52:46 +0200 +Subject: [PATCH] Revert "arm: cleanup redundant macros for syscalls" + +This reverts commit 0550ecce0e6580c5ad34e9a9a39ff18ccf8774f9. +--- + libc/sysdeps/linux/arm/bits/syscalls.h | 4 -- + libc/sysdeps/linux/arm/sysdep.h | 111 +++++++++++++++++++++++++++++++++ + 2 files changed, 111 insertions(+), 4 deletions(-) + +diff --git a/libc/sysdeps/linux/arm/bits/syscalls.h b/libc/sysdeps/linux/arm/bits/syscalls.h +index 5b30564..e447a7b 100644 +--- a/libc/sysdeps/linux/arm/bits/syscalls.h ++++ b/libc/sysdeps/linux/arm/bits/syscalls.h +@@ -83,10 +83,6 @@ + } \ + (int) __internal_sys_result; }) \ + ) +- +-#undef INTERNAL_SYSCALL_ARM +-#define INTERNAL_SYSCALL_ARM(name, err, nr, args...) \ +- INTERNAL_SYSCALL_NCS(__ARM_NR_##name, err, nr, args) + #endif + + #define INTERNAL_SYSCALL_ERROR_P(val, err) \ +diff --git a/libc/sysdeps/linux/arm/sysdep.h b/libc/sysdeps/linux/arm/sysdep.h +index 80bf9ec..65d9c2b 100644 +--- a/libc/sysdeps/linux/arm/sysdep.h ++++ b/libc/sysdeps/linux/arm/sysdep.h +@@ -254,5 +254,116 @@ __local_syscall_error: \ + #define UNDOARGS_6 ldmfd sp!, {r4, r5}; + #define UNDOARGS_7 ldmfd sp!, {r4, r5, r6}; + ++#else /* not __ASSEMBLER__ */ ++/* Define a macro which expands into the inline wrapper code for a system ++ call. */ ++#undef INLINE_SYSCALL ++#define INLINE_SYSCALL(name, nr, args...) \ ++ ({ unsigned int _inline_sys_result = INTERNAL_SYSCALL (name, , nr, args); \ ++ if (unlikely (INTERNAL_SYSCALL_ERROR_P (_inline_sys_result, ))) \ ++ { \ ++ __set_errno (INTERNAL_SYSCALL_ERRNO (_inline_sys_result, )); \ ++ _inline_sys_result = (unsigned int) -1; \ ++ } \ ++ (int) _inline_sys_result; }) ++ ++#undef INTERNAL_SYSCALL_DECL ++#define INTERNAL_SYSCALL_DECL(err) do { } while (0) ++ ++#undef INTERNAL_SYSCALL_RAW ++#if defined(__thumb__) ++/* Hide the use of r7 from the compiler, this would be a lot ++ * easier but for the fact that the syscalls can exceed 255. ++ * For the moment the LOAD_ARG_7 is sacrificed. ++ * We can't use push/pop inside the asm because that breaks ++ * unwinding (ie. thread cancellation). ++ */ ++#define INTERNAL_SYSCALL_RAW(name, err, nr, args...) \ ++ ({ unsigned int _internal_sys_result; \ ++ { \ ++ int _sys_buf[2]; \ ++ register int __a1 __asm__ ("a1"); \ ++ register int *_v3 __asm__ ("v3") = _sys_buf; \ ++ LOAD_ARGS_##nr (args) \ ++ *_v3 = (int) (name); \ ++ __asm__ __volatile__ ("str r7, [v3, #4]\n" \ ++ "\tldr r7, [v3]\n" \ ++ "\tswi 0 @ syscall " #name "\n" \ ++ "\tldr r7, [v3, #4]" \ ++ : "=r" (__a1) \ ++ : "r" (_v3) ASM_ARGS_##nr \ ++ : "memory"); \ ++ _internal_sys_result = __a1; \ ++ } \ ++ (int) _internal_sys_result; }) ++#elif defined(__ARM_EABI__) ++#define INTERNAL_SYSCALL_RAW(name, err, nr, args...) \ ++ ({unsigned int _internal_sys_result; \ ++ { \ ++ register int __a1 __asm__ ("r0"), _nr __asm__ ("r7"); \ ++ LOAD_ARGS_##nr (args) \ ++ _nr = name; \ ++ __asm__ __volatile__ ("swi 0x0 @ syscall " #name \ ++ : "=r" (__a1) \ ++ : "r" (_nr) ASM_ARGS_##nr \ ++ : "memory"); \ ++ _internal_sys_result = __a1; \ ++ } \ ++ (int) _internal_sys_result; }) ++#else /* !defined(__ARM_EABI__) */ ++#define INTERNAL_SYSCALL_RAW(name, err, nr, args...) \ ++ ({ unsigned int _internal_sys_result; \ ++ { \ ++ register int __a1 __asm__ ("a1"); \ ++ LOAD_ARGS_##nr (args) \ ++ __asm__ __volatile__ ("swi %1 @ syscall " #name \ ++ : "=r" (__a1) \ ++ : "i" (name) ASM_ARGS_##nr \ ++ : "memory"); \ ++ _internal_sys_result = __a1; \ ++ } \ ++ (int) _internal_sys_result; }) ++#endif ++ ++#undef INTERNAL_SYSCALL ++#define INTERNAL_SYSCALL(name, err, nr, args...) \ ++ INTERNAL_SYSCALL_RAW(SYS_ify(name), err, nr, args) ++ ++#undef INTERNAL_SYSCALL_ARM ++#define INTERNAL_SYSCALL_ARM(name, err, nr, args...) \ ++ INTERNAL_SYSCALL_RAW(__ARM_NR_##name, err, nr, args) ++ ++#undef INTERNAL_SYSCALL_ERROR_P ++#define INTERNAL_SYSCALL_ERROR_P(val, err) \ ++ ((unsigned int) (val) >= 0xfffff001u) ++ ++#undef INTERNAL_SYSCALL_ERRNO ++#define INTERNAL_SYSCALL_ERRNO(val, err) (-(val)) ++ ++#if defined(__ARM_EABI__) ++#undef INTERNAL_SYSCALL_NCS ++#define INTERNAL_SYSCALL_NCS(number, err, nr, args...) \ ++ INTERNAL_SYSCALL_RAW(number, err, nr, args) ++#else ++/* We can't implement non-constant syscalls directly since the syscall ++ number is normally encoded in the instruction. So use SYS_syscall. */ ++#undef INTERNAL_SYSCALL_NCS ++#define INTERNAL_SYSCALL_NCS(number, err, nr, args...) \ ++ INTERNAL_SYSCALL_NCS_##nr (number, err, args) ++ ++#define INTERNAL_SYSCALL_NCS_0(number, err, args...) \ ++ INTERNAL_SYSCALL (syscall, err, 1, number, args) ++#define INTERNAL_SYSCALL_NCS_1(number, err, args...) \ ++ INTERNAL_SYSCALL (syscall, err, 2, number, args) ++#define INTERNAL_SYSCALL_NCS_2(number, err, args...) \ ++ INTERNAL_SYSCALL (syscall, err, 3, number, args) ++#define INTERNAL_SYSCALL_NCS_3(number, err, args...) \ ++ INTERNAL_SYSCALL (syscall, err, 4, number, args) ++#define INTERNAL_SYSCALL_NCS_4(number, err, args...) \ ++ INTERNAL_SYSCALL (syscall, err, 5, number, args) ++#define INTERNAL_SYSCALL_NCS_5(number, err, args...) \ ++ INTERNAL_SYSCALL (syscall, err, 6, number, args) ++#endif ++ + #endif /* __ASSEMBLER__ */ + #endif /* linux/arm/sysdep.h */ +-- +2.7.4 + -- 2.7.4 ^ permalink raw reply related [flat|nested] 6+ messages in thread
* [Buildroot] [PATCH 0/3] Fix our defconfigs (branch yem/fix-defconfigs) 2016-08-07 20:44 [Buildroot] [PATCH 0/3] Fix our defconfigs (branch yem/fix-defconfigs) Yann E. MORIN ` (2 preceding siblings ...) 2016-08-07 20:44 ` [Buildroot] [PATCH 3/3] pcakage/uclibc: fix build on thumb2-only arm cpus Yann E. MORIN @ 2016-08-07 20:54 ` Thomas Petazzoni 2016-08-07 20:59 ` Yann E. MORIN 3 siblings, 1 reply; 6+ messages in thread From: Thomas Petazzoni @ 2016-08-07 20:54 UTC (permalink / raw) To: buildroot Hello, On Sun, 7 Aug 2016 22:44:33 +0200, Yann E. MORIN wrote: > defconfigs/kb9202: fix headers version > defconfigs/snps_axs_10*: uboot needs dtc > pcakage/uclibc: fix build on thumb2-only arm cpus All three applied. If not already done, please make sure to report on the uclibc mailing list the thumb2 problem. BTW, now that I re-read your commit log, it's wrong: the SAMA5 are not at all Thumb2 only CPUs. They are regular Cortex-A5 CPUs, i.e they support both Thumb-2 and ARM. Best regards, Thomas -- Thomas Petazzoni, CTO, Free Electrons Embedded Linux, Kernel and Android engineering http://free-electrons.com ^ permalink raw reply [flat|nested] 6+ messages in thread
* [Buildroot] [PATCH 0/3] Fix our defconfigs (branch yem/fix-defconfigs) 2016-08-07 20:54 ` [Buildroot] [PATCH 0/3] Fix our defconfigs (branch yem/fix-defconfigs) Thomas Petazzoni @ 2016-08-07 20:59 ` Yann E. MORIN 0 siblings, 0 replies; 6+ messages in thread From: Yann E. MORIN @ 2016-08-07 20:59 UTC (permalink / raw) To: buildroot Thomas, All, On 2016-08-07 22:54 +0200, Thomas Petazzoni spake thusly: > On Sun, 7 Aug 2016 22:44:33 +0200, Yann E. MORIN wrote: > > defconfigs/kb9202: fix headers version > > defconfigs/snps_axs_10*: uboot needs dtc > > pcakage/uclibc: fix build on thumb2-only arm cpus > All three applied. If not already done, please make sure to report on > the uclibc mailing list the thumb2 problem. Sure, I was about to do so... > BTW, now that I re-read your commit log, it's wrong: the SAMA5 are not > at all Thumb2 only CPUs. They are regular Cortex-A5 CPUs, i.e they > support both Thumb-2 and ARM. Right, what I meant to say was that it is fixing our thumb2 defconfigs. Regards, Yann E. MORIN. -- .-----------------.--------------------.------------------.--------------------. | Yann E. MORIN | Real-Time Embedded | /"\ ASCII RIBBON | Erics' conspiracy: | | +33 662 376 056 | Software Designer | \ / CAMPAIGN | ___ | | +33 223 225 172 `------------.-------: X AGAINST | \e/ There is no | | http://ymorin.is-a-geek.org/ | _/*\_ | / \ HTML MAIL | v conspiracy. | '------------------------------^-------^------------------^--------------------' ^ permalink raw reply [flat|nested] 6+ messages in thread
end of thread, other threads:[~2016-08-07 20:59 UTC | newest] Thread overview: 6+ messages (download: mbox.gz follow: Atom feed -- links below jump to the message on this page -- 2016-08-07 20:44 [Buildroot] [PATCH 0/3] Fix our defconfigs (branch yem/fix-defconfigs) Yann E. MORIN 2016-08-07 20:44 ` [Buildroot] [PATCH 1/3] defconfigs/kb9202: fix headers version Yann E. MORIN 2016-08-07 20:44 ` [Buildroot] [PATCH 2/3] defconfigs/snps_axs_10*: uboot needs dtc Yann E. MORIN 2016-08-07 20:44 ` [Buildroot] [PATCH 3/3] pcakage/uclibc: fix build on thumb2-only arm cpus Yann E. MORIN 2016-08-07 20:54 ` [Buildroot] [PATCH 0/3] Fix our defconfigs (branch yem/fix-defconfigs) Thomas Petazzoni 2016-08-07 20:59 ` Yann E. MORIN
This is a public inbox, see mirroring instructions for how to clone and mirror all data and code used for this inbox