devel

devel@uclibc-ng.org

1 participants
1164 discussions

Re: [uclibc-ng-devel] CONFIG_UCLIBC_SUSV2_LEGACY triggers odd SIGSEGV in busybox's ash on MIPS
by Joshua Kinard 23 Oct '16

23 Oct '16

On 10/12/2016 04:27, Waldemar Brodkorb wrote: > hi, > > i am visiting elce, be back on monday. > can you try 1.0.17 please. > 1.0.18 introduced some interesting regressions i habe not covered in my tests. > i have solved most of them, but not all is pushed, yet. > > looks related to the patch i sent out to lance last week on the list. > best regards > Waldemar > > Von meinem iPhone gesendet > >> Am 12.10.2016 um 09:59 schrieb Joshua Kinard <kumba(a)gentoo.org>: >> >>> On 10/12/2016 03:53, Joshua Kinard wrote: >>> Hello, >>> >>> I think I've run into a rather odd bug on a big-endian MIPS platform while >>> trying to hand-assemble a MIPS-II ISA netboot image built from a uclibc-ng >>> chroot. >> >> [snip] >> >> PS, I forgot to add, this is using uclibc-ng-1.0.18 and busybox-1.24.2. >> (Resending to the actual list, sorry Waldemar!) Unfortunately, my base root for building the netboot is built on 1.0.18 at the moment. It'd take about two days to do a full rebuild. That said, I think I might have an idea. The bit of code cited in Bug #3919 for old uclibc only defines and uses null_not_ptr in __uClibc_main.c, but it looks like the code in jmp-unwind.c does not. So I am going to try moving the null_not_ptr definition to a header somewhere, mark it non-static (maybe inline?), then try using it on the __pthread_cleanup_upto test and see if that might resolve the issue. Sound sane? -- Joshua Kinard Gentoo/MIPS kumba(a)gentoo.org 6144R/F5C6C943 2015-04-27 177C 1972 1FB8 F254 BAD0 3E72 5C63 F4E3 F5C6 C943 "The past tempts us, the present confuses us, the future frightens us. And our lives slip away, moment by moment, lost in that vast, terrible in-between." --Emperor Turhan, Centauri Republic

2 10

pthread.h build issue on Blackfin, Microblaze, m68k
by Thomas Petazzoni 23 Oct '16

23 Oct '16

Hello, I was investigating the recent build failures of the libbluray package in Buildroot, which occur on the Microblaze, Blackfin and m68k architecture (at least), using the latest uClibc-ng release. The failure looks like this: http://autobuild.buildroot.net/results/941/941c06e29f6542e258f20799dc6f3f94… Relevant part: In file included from /home/buildroot/autobuild/run/instance-3/output/host/usr/microblazeel-buildroot-linux-uclibc/sysroot/usr/include/pthread.h:686:0, from src/util/mutex.c:32: /home/buildroot/autobuild/run/instance-3/output/host/usr/microblazeel-buildroot-linux-uclibc/sysroot/usr/include/bits/sigthread.h:31:14: error: unknown type name '__sigset_t' const __sigset_t *__restrict __newmask, ^ /home/buildroot/autobuild/run/instance-3/output/host/usr/microblazeel-buildroot-linux-uclibc/sysroot/usr/include/bits/sigthread.h:32:8: error: unknown type name '__sigset_t' __sigset_t *__restrict __oldmask)__THROW; And indeed, one can reproduce this problem with a very simple test case: ==== #include <pthread.h> int main(void) { return 0; } ==== If you build it with the Microblaze/uClibc compiler with no special flags, it works just fine: $ ./output/host/usr/bin/microblazeel-linux-gcc -o test test.c $ However, as soon as you start adding the -std=c99 flag, the build fails: $ ./output/host/usr/bin/microblazeel-linux-gcc -std=c99 -o test test.c In file included from /home/thomas/projets/buildroot/output/host/usr/microblazeel-buildroot-linux-uclibc/sysroot/usr/include/pthread.h:686:0, from pouet.c:1: /home/thomas/projets/buildroot/output/host/usr/microblazeel-buildroot-linux-uclibc/sysroot/usr/include/bits/sigthread.h:31:14: error: unknown type name ‘__sigset_t’ const __sigset_t *__restrict __newmask, ^ /home/thomas/projets/buildroot/output/host/usr/microblazeel-buildroot-linux-uclibc/sysroot/usr/include/bits/sigthread.h:32:8: error: unknown type name ‘__sigset_t’ __sigset_t *__restrict __oldmask)__THROW; ^ Is this problem already known? Best regards, Thomas Petazzoni -- Thomas Petazzoni, CTO, Free Electrons Embedded Linux and Kernel engineering http://free-electrons.com

2 2

new release 1.0.19
by Waldemar Brodkorb 23 Oct '16

23 Oct '16

Hi Embedded-Linux-Hackers, this release is mostly bug fixing and regression fixes from the single-libc patch. Next release will separate the testsuite to its own project. Thanks for testing and bug reports, Waldemar Changes: John Ata (1): Fix nftw when called with FTW_CHDIR and FTW_DEPTH Vineet Gupta (4): ARC: nptl: cancellable wrappers were broken ARC: nptl: cancellable wrappers were broken #2 ARC: update .note.ABI-tag for ABIv4 ARC: build: don't force usage of llock and swape instructions Waldemar Brodkorb (14): locale: remove building for the host, as it breaks ubacktrace/uargp: remove unneeded and false linker scripts remove linux kernel 2.4 modules support cleanup ppc port test: fix tests including non-wrapper libgcc exception handling code cleanup libc.a ppc: do not include copysgnl.c if UCLIBC_HAS_LONG_DOUBLE_MATH enabled fix mips/mips64 build for old compilers cleanup and fix static linking issues linuxthreads: allow to choose on all supported architectures create empty static files conditionally test: add nftw test case test: add a simple check, that -std=c99 working bump for release

1 0

uClibc-ng - small C library for embedded systems annotated tag v1.0.19 created. v1.0.19
by wbx＠helium.openadk.org 23 Oct '16

23 Oct '16

This is an automated email from the git hooks/post-receive script. It was generated because a ref change was pushed to the repository containing the project "uClibc-ng - small C library for embedded systems". The annotated tag, v1.0.19 has been created at 20c83f773d9b1dd7739dcbfdb9be92245b216ad1 (tag) tagging e87c79f7983d995b1a69151e3dd43e9e7466b333 (commit) replaces v1.0.18 tagged by Waldemar Brodkorb on Sun Oct 23 09:19:06 2016 +0200 - Log ----------------------------------------------------------------- release 1.0.19 - Trapistes Rochefort 6 -----BEGIN PGP SIGNATURE----- Comment: GPGTools - http://gpgtools.org iQIcBAABCAAGBQJYDGRsAAoJEMEz+QKmmH5i4iUQALKz7YezIqW1aIB4eX9U5enz ja+ZkHhVa8iS3MUYNQ16zA67IAMYikRKDnPQ2gLjv4OXXSP+mUe7zvCCDpW0I9S7 3AdyvZye3u/KvZ0xyEMJMITT2FSC/ksw0jJk9olCYPHSjHnjrHW6Yi/yeluotVJj 9gpxL3KsKDqKJGPtdo5v7Q0jee3OWPxHYJUTkfXI4OM/4bXlqv32diMpdRLw8rMP M/vPk0NAYPVVY2sFCkvdPQ3y7Yj4CicTzSG+CyjJ+as2PMUuYxMLSFwovl1v/YtB 7twyhCACr1+LiDG4K/v+i/RG8VYcDCaKfCtEhZ12yKOVKmcrY2b9mx1TGCY/sNmQ AwRTzD3NihKJnywIZ3Vz3cbA6g2P/oUc7sjkSque/s2iUHNvhUWZzMvCPtIwYmm1 0WWfPUk6LKjnvQm4I5R3NrOZVw8o3V90tTxdejIIiO0LxrhBvrZop9iPiXt0qqyE q3PLhUzCebmweyzJI+WzrFbAWFVoJuc5NqAL+Slj3Cfz3xGl4aFz8zlYeqAuCGtX T7h52DePA5jL0VVP9p2VhrWzOHF2175KKrCYGFKk77LpFXSsVAGGmJcHZJXxNDQM anPc9mC932G9mRABErGFB+qcbGxPUC9E+ZqQi3InhtobhkQ3oXeY3u9F550D7zaH FGCe4Jc+EE8jikQ0ibnW =RXBs -----END PGP SIGNATURE----- John Ata (1): Fix nftw when called with FTW_CHDIR and FTW_DEPTH Vineet Gupta (4): ARC: nptl: cancellable wrappers were broken ARC: nptl: cancellable wrappers were broken #2 ARC: update .note.ABI-tag for ABIv4 ARC: build: don't force usage of llock and swape instructions Waldemar Brodkorb (14): locale: remove building for the host, as it breaks ubacktrace/uargp: remove unneeded and false linker scripts remove linux kernel 2.4 modules support cleanup ppc port test: fix tests including non-wrapper libgcc exception handling code cleanup libc.a ppc: do not include copysgnl.c if UCLIBC_HAS_LONG_DOUBLE_MATH enabled fix mips/mips64 build for old compilers cleanup and fix static linking issues linuxthreads: allow to choose on all supported architectures create empty static files conditionally test: add nftw test case test: add a simple check, that -std=c99 working bump for release ----------------------------------------------------------------------- hooks/post-receive -- uClibc-ng - small C library for embedded systems

1 0

uClibc-ng - small C library for embedded systems branch master updated. v1.0.18-19-ge87c79f
by wbx＠helium.openadk.org 23 Oct '16

23 Oct '16

1 0

[PATCH] ARC: update .note.ABI-tag for ABIv4
by Vineet Gupta 22 Oct '16

22 Oct '16

Cc: Anton Kolesov <akolesov(a)synopsys.com> Cc: linux-snps-arc(a)lists.infradead.org Signed-off-by: Vineet Gupta <vgupta(a)synopsys.com> --- libc/sysdeps/linux/arc/crt1.S | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) diff --git a/libc/sysdeps/linux/arc/crt1.S b/libc/sysdeps/linux/arc/crt1.S index ae863e7acf5f..70a06e058d3b 100644 --- a/libc/sysdeps/linux/arc/crt1.S +++ b/libc/sysdeps/linux/arc/crt1.S @@ -72,5 +72,9 @@ __start: 0: .asciz "GNU" /* Vendor name */ 1: .align 4 2: .long 0 /* Note data: Linux executable */ - .long 3,9,0 /* Earliest compatible kernel */ +#if (__GNUC__ > 4) + .long 4,8,0 /* Earliest compatible kernel for ABI v4 */ +#else + .long 3,9,0 /* Earliest compatible kernel for ABI v3 */ +#endif 3: .align 4 /* Pad out section */ -- 2.7.4

2 1

Re: [uclibc-ng-devel] PATCH] Fix ntfw when called with FTW_CHDIR and FTW_DEPTH to change directory back to the parent before processing the directory (after the contents have already been processed).
by Waldemar Brodkorb 22 Oct '16

22 Oct '16

Hi, Ata, John (US) wrote, > Hi all, > > > > In using ftw/nftw, I’ve discovered a bug when using FTW_CHDIR and FTW_DEPTH. > After changing the working directory to a subdirectory, nftw would process the > contents fine. However, it would then try and process the directory itself > while still in that directory and then switch back to the parent. This would > result in an error (ENOENT). This patch will change working directories to the > parent before processing the directory itself. Thanks, test case and patch applied, best regards Waldemar

1 0

uClibc-ng - small C library for embedded systems branch master updated. v1.0.18-16-ga6cdfdf
by wbx＠helium.openadk.org 22 Oct '16

22 Oct '16

This is an automated email from the git hooks/post-receive script. It was generated because a ref change was pushed to the repository containing the project "uClibc-ng - small C library for embedded systems". The branch, master has been updated via a6cdfdfff38584cadec40b6d6f2df2651e0b34b8 (commit) via 545beca17885560298580b51619979529a309360 (commit) via 688563159f328726f958163fcdab8088df66e977 (commit) via d87cc4d9a6945daf48848022c93d0d9a0489a87e (commit) from 6630ac965279ed2a76394fbad8fd861f275ed24c (commit) Those revisions listed above that are new to this repository have not appeared on any other notification email; so we list those revisions in full, below. - Log ----------------------------------------------------------------- commit a6cdfdfff38584cadec40b6d6f2df2651e0b34b8 Author: John Ata <john.ata(a)baesystems.com> Date: Mon Oct 17 14:47:02 2016 -0400 Fix nftw when called with FTW_CHDIR and FTW_DEPTH Change directory back to the parent before processing the directory (after the contents have already been processed). Signed-off-by: John Ata <john.ata(a)baesystems.com> commit 545beca17885560298580b51619979529a309360 Author: Waldemar Brodkorb <wbx(a)uclibc-ng.org> Date: Thu Oct 20 21:18:05 2016 +0200 test: add nftw test case commit 688563159f328726f958163fcdab8088df66e977 Author: Vineet Gupta <Vineet.Gupta1(a)synopsys.com> Date: Thu Oct 13 10:46:14 2016 -0700 ARC: update .note.ABI-tag for ABIv4 Cc: Anton Kolesov <akolesov(a)synopsys.com> Cc: linux-snps-arc(a)lists.infradead.org Signed-off-by: Vineet Gupta <vgupta(a)synopsys.com> commit d87cc4d9a6945daf48848022c93d0d9a0489a87e Author: Waldemar Brodkorb <wbx(a)uclibc-ng.org> Date: Wed Oct 19 19:16:37 2016 +0200 create empty static files conditionally ----------------------------------------------------------------------- Summary of changes: Makefile.in | 26 +++++++++++++++++++- libc/Makefile.in | 12 ++++----- libc/misc/ftw/ftw.c | 13 +++++----- libc/sysdeps/linux/arc/crt1.S | 6 ++++- test/misc/tst-nftw.c | 57 +++++++++++++++++++++++++++++++++++++++++++ 5 files changed, 99 insertions(+), 15 deletions(-) create mode 100644 test/misc/tst-nftw.c diff --git a/Makefile.in b/Makefile.in index c289ed3..877a0b6 100644 --- a/Makefile.in +++ b/Makefile.in @@ -363,7 +363,31 @@ else $(Q)cd $(PREFIX)$(DEVEL_PREFIX)include && $(RM) -f wchar-stub.h endif -EMPTY_LIB_NAMES = m crypt nsl resolv util dl rt pthread intl +EMPTY_LIB_NAMES = dl +ifeq ($(DO_C99_MATH),y) +EMPTY_LIB_NAMES += m +endif +ifeq ($(UCLIBC_HAS_THREADS),y) +EMPTY_LIB_NAMES += pthread +endif +ifeq ($(UCLIBC_HAS_REALTIME),y) +EMPTY_LIB_NAMES += rt +endif +ifeq ($(UCLIBC_HAS_CRYPT),y) +EMPTY_LIB_NAMES += crypt +endif +ifeq ($(UCLIBC_HAS_LIBUTIL),y) +EMPTY_LIB_NAMES += util +endif +ifeq ($(UCLIBC_HAS_LIBRESOLV_STUB),y) +EMPTY_LIB_NAMES += resolv +endif +ifeq ($(UCLIBC_HAS_LIBNSL_STUB),y) +EMPTY_LIB_NAMES += nsl +endif +ifeq ($(UCLIBC_HAS_GETTEXT_AWARENESS),y) +EMPTY_LIB_NAMES += intl +endif EMPTY_LIBS = $(EMPTY_LIB_NAMES:%=lib/lib%.a) $(EMPTY_LIBS): diff --git a/libc/Makefile.in b/libc/Makefile.in index 85986a7..714a66e 100644 --- a/libc/Makefile.in +++ b/libc/Makefile.in @@ -42,9 +42,9 @@ else libc-a-y = $(libc-y) $(libc-static-y) endif -libc-a-y += $(libnsl-a-y) -libc-a-y += $(libresolv-a-y) -libc-a-y += $(libm-a-y) +libc-a-$(DO_C99_MATH) += $(libm-a-y) +libc-a-$(UCLIBC_HAS_LIBNSL_STUB) += $(libnsl-a-y) +libc-a-$(UCLIBC_HAS_LIBRESOLV_STUB) += $(libresolv-a-y) libc-a-$(UCLIBC_HAS_LIBUTIL) += $(libutil-a-y) libc-a-$(UCLIBC_HAS_CRYPT) += $(libcrypt-a-y) libc-a-$(UCLIBC_HAS_ARGP) += $(libuargp-a-y) @@ -55,10 +55,10 @@ libc-a-$(UCLIBC_HAS_BACKTRACE) += $(libubacktrace-a-y) libc-so-y = $(libc-y:.o=.os) $(libc-shared-y) -libc-so-y += $(libnsl-so-y) -libc-so-y += $(libresolv-so-y) libc-so-y += $(libdl-so-y) -libc-so-y += $(libm-so-y) +libc-so-$(DO_C99_MATH) += $(libm-so-y) +libc-so-$(UCLIBC_HAS_LIBNSL_STUB) += $(libnsl-so-y) +libc-so-$(UCLIBC_HAS_LIBRESOLV_STUB) += $(libresolv-so-y) libc-so-$(UCLIBC_HAS_LIBUTIL) += $(libutil-so-y) libc-so-$(UCLIBC_HAS_CRYPT) += $(libcrypt-so-y) libc-so-$(UCLIBC_HAS_ARGP) += $(libuargp-so-y) diff --git a/libc/misc/ftw/ftw.c b/libc/misc/ftw/ftw.c index 05ea0fa..c6241ed 100644 --- a/libc/misc/ftw/ftw.c +++ b/libc/misc/ftw/ftw.c @@ -558,19 +558,14 @@ fail: --data->ftw.level; data->ftw.base = previous_base; - /* Finally, if we process depth-first report the directory. */ - if (result == 0 && (data->flags & FTW_DEPTH)) - result = (*data->func) (data->dirbuf, st, FTW_DP, &data->ftw); - - if (old_dir - && (data->flags & FTW_CHDIR) + if ((data->flags & FTW_CHDIR) && (result == 0 || ((data->flags & FTW_ACTIONRETVAL) && (result != -1 && result != FTW_STOP)))) { /* Change back to the parent directory. */ int done = 0; - if (old_dir->stream != NULL) + if (old_dir && old_dir->stream != NULL) if (__fchdir (dirfd (old_dir->stream)) == 0) done = 1; @@ -587,6 +582,10 @@ fail: } } + /* Finally, if we process depth-first report the directory. */ + if (result == 0 && (data->flags & FTW_DEPTH)) + result = (*data->func) (data->dirbuf, st, FTW_DP, &data->ftw); + return result; } diff --git a/libc/sysdeps/linux/arc/crt1.S b/libc/sysdeps/linux/arc/crt1.S index ae863e7..70a06e0 100644 --- a/libc/sysdeps/linux/arc/crt1.S +++ b/libc/sysdeps/linux/arc/crt1.S @@ -72,5 +72,9 @@ __start: 0: .asciz "GNU" /* Vendor name */ 1: .align 4 2: .long 0 /* Note data: Linux executable */ - .long 3,9,0 /* Earliest compatible kernel */ +#if (__GNUC__ > 4) + .long 4,8,0 /* Earliest compatible kernel for ABI v4 */ +#else + .long 3,9,0 /* Earliest compatible kernel for ABI v3 */ +#endif 3: .align 4 /* Pad out section */ diff --git a/test/misc/tst-nftw.c b/test/misc/tst-nftw.c new file mode 100644 index 0000000..76d11eb --- /dev/null +++ b/test/misc/tst-nftw.c @@ -0,0 +1,57 @@ +#define _XOPEN_SOURCE 500 +#define _GNU_SOURCE +#include <ftw.h> +#include <stdio.h> +#include <sys/types.h> +#include <sys/stat.h> +#include <fcntl.h> +#include <unistd.h> + +int result = 0; + +static int process_one_entry(const char *fpath, const struct stat *sb, + int typeflag, struct FTW *ftwbuf) +{ + + struct stat buf; + const char *rel_path = fpath+ftwbuf->base; + + printf("Processing %s in working dir %s\n", + rel_path, get_current_dir_name()); + if (stat(rel_path, &buf) < 0) { + perror("Oops...relative path does not exist in current directory"); + result = 1; + } +} + +static int +do_test(void) +{ + char *path = "/tmp/stest_dir"; + char *subpath = "/tmp/stest_dir/d1"; + char *filepath = "/tmp/stest_dir/f1"; + char *filesubpath = "/tmp/stest_dir/d1/f2"; + + if ((mkdir(path, 0700)) < 0) + perror("Creating path"); + if ((mkdir(subpath, 0700)) < 0) + perror("Creating subpath"); + if ((open(filepath, O_CREAT)) < 0) + perror("Opening filepath"); + if ((open(filesubpath, O_CREAT)) < 0) + perror("Opening filesubpath"); + + if (nftw(path, process_one_entry, 100, (FTW_CHDIR|FTW_DEPTH|FTW_PHYS)) < 0) + perror("ntfw"); + + unlink(filesubpath); + unlink(filepath); + rmdir(subpath); + rmdir(path); + + return result; +} + +#define TIMEOUT 5 +#define TEST_FUNCTION do_test () +#include "../test-skeleton.c" hooks/post-receive -- uClibc-ng - small C library for embedded systems

1 0

[PATCH] ARC: build: don't force usage of llock and swape instructions
by Vineet Gupta 21 Oct '16

21 Oct '16

This is because some old ARC750 cores lack these instructions. We now rely on the default for the toolchain driver: -mcpu=A7 won't enable those, while -mcpu=archs will as these instructions are baseline (and thus is not impacted with this change). If some arc700 based cpu does have them, it can override the driver defaults in it's one level up build system. Signed-off-by: Vineet Gupta <vgupta(a)synopsys.com> --- Rules.mak | 1 - 1 file changed, 1 deletion(-) diff --git a/Rules.mak b/Rules.mak index 8d0e66a49b72..cc1545c81f40 100644 --- a/Rules.mak +++ b/Rules.mak @@ -507,7 +507,6 @@ ifeq ($(TARGET_ARCH),c6x) endif ifeq ($(TARGET_ARCH),arc) - CPU_CFLAGS-y += -mlock -mswape CPU_CFLAGS-$(CONFIG_ARC_CPU_700) += -mA7 CPU_CFLAGS-$(CONFIG_ARC_CPU_HS) += -mcpu=archs CPU_LDFLAGS-y += $(CPU_CFLAGS) -marclinux -- 2.7.4

1 0

18 Oct '16

Hi John, Ata, John (US) wrote, > Hi all, > > > > In using ftw/nftw, I’ve discovered a bug when using FTW_CHDIR and FTW_DEPTH. > After changing the working directory to a subdirectory, nftw would process the > contents fine. However, it would then try and process the directory itself > while still in that directory and then switch back to the parent. This would > result in an error (ENOENT). This patch will change working directories to the > parent before processing the directory itself. Thanks for the report. Are you able to provide a small test-case showing the issue? best regards Waldemar

2 1

Jump to page:

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

devel