Skip to contentSkip to conversation

ltp: Bump to 20190115

PATCH In review Daniel Díaz 2 participants 4 messages Feb 21, 2019 – Feb 25, 2019

lore
Share & tools
lore
#0 lore raw

PATCH 1/1 ltp: Bump to 20190115

This patch has been refreshed:
* 0008-Check-if-__GLIBC_PREREQ-is-defined-before-using-it.patch

These patches have been dropped as they have been merged:
* 0001-netns_helper.sh-use-ping-6-when-ping6-is-not-avaliab.patch
* 0001-setrlimit05-Use-another-method-to-get-bad-address.patch
* 0001-sigwaitinfo01-recent-glibc-calls-syscall-directly.patch
* 0001-statx-fix-compile-errors.patch
* 0001-syscalls-fcntl-make-OFD-command-use-fcntl64-syscall-.patch
* 0001-getcpu01-Rename-getcpu-to-avoid-conflict-with-glibc-.patch

This patch has been added:
* define-sigrtmin-and-sigrtmax-for-musl.patch

Added these rdependencies:
* file: for ld01, file01 and logrotate
* quota: for quota_remount_test01

---
 ...getcpu-to-avoid-conflict-with-glibc-.patch |  57 ------
 ...use-ping-6-when-ping6-is-not-avaliab.patch |  45 -----
 ...suite-mmap24-2-Relax-condition-a-bit.patch |   6 +-
 ...se-another-method-to-get-bad-address.patch |  61 ------
 ...-recent-glibc-calls-syscall-directly.patch |  75 -------
 .../ltp/0001-statx-fix-compile-errors.patch   |  35 ----
 ...ake-OFD-command-use-fcntl64-syscall-.patch | 184 -----------------
 ...option-to-select-libc-implementation.patch |   3 +-
 ...E-undeclared-when-building-with-musl.patch |   2 +-
 ...BC_PREREQ-is-defined-before-using-it.patch | 190 +++---------------
 .../ltp/0035-fix-test_proc_kill-hang.patch    |   2 +-
 ...1-Fix-for-test-in-deterministic-mode.patch |   2 +-
 ...efine-sigrtmin-and-sigrtmax-for-musl.patch | 114 +++++++++++
 .../ltp/{ltp_20180926.bb => ltp_20190115.bb}  |  11 +-
 14 files changed, 154 insertions(+), 633 deletions(-)
 delete mode 100644 meta/recipes-extended/ltp/ltp/0001-getcpu01-Rename-getcpu-to-avoid-conflict-with-glibc-.patch
 delete mode 100644 meta/recipes-extended/ltp/ltp/0001-netns_helper.sh-use-ping-6-when-ping6-is-not-avaliab.patch
 delete mode 100644 meta/recipes-extended/ltp/ltp/0001-setrlimit05-Use-another-method-to-get-bad-address.patch
 delete mode 100644 meta/recipes-extended/ltp/ltp/0001-sigwaitinfo01-recent-glibc-calls-syscall-directly.patch
 delete mode 100644 meta/recipes-extended/ltp/ltp/0001-statx-fix-compile-errors.patch
 delete mode 100644 meta/recipes-extended/ltp/ltp/0001-syscalls-fcntl-make-OFD-command-use-fcntl64-syscall-.patch
 create mode 100644 meta/recipes-extended/ltp/ltp/define-sigrtmin-and-sigrtmax-for-musl.patch
 rename meta/recipes-extended/ltp/{ltp_20180926.bb => ltp_20190115.bb} (91%)
meta/recipes-extended/ltp/ltp/0001-getcpu01-Rename-getcpu-to-avoid-conflict-with-glibc-.patch deleted +0 -57
deleted file mode 100644
index d0c847774d..0000000000
@@ -1,57 +0,0 @@
-From c9edb0eaaca83c18ac7b5ea3330faad8d47d3fd1 Mon Sep 17 00:00:00 2001
-From: Khem Raj <[email protected]>
-Date: Fri, 14 Dec 2018 06:44:02 -0800
-Subject: [PATCH] getcpu01: Rename getcpu to avoid conflict with glibc >= 2.29
-
-glibc has added getcpu() API starting with glibc 2.29 release, this
-makes the static function in conflict and signatures are not same as
-well.
-
-Renaming helps to keep ltp build going
-
-Upstream-Status: Submitted [http://lists.linux.it/pipermail/ltp/2018-December/010321.html]
-
-Signed-off-by: Khem Raj <[email protected]>
----
- testcases/kernel/syscalls/getcpu/getcpu01.c | 10 +++++-----
- 1 file changed, 5 insertions(+), 5 deletions(-)
-
-diff --git a/testcases/kernel/syscalls/getcpu/getcpu01.c b/testcases/kernel/syscalls/getcpu/getcpu01.c
-index c927512ca..e850fe844 100644
---- a/testcases/kernel/syscalls/getcpu/getcpu01.c
-+++ b/testcases/kernel/syscalls/getcpu/getcpu01.c
-@@ -82,7 +82,7 @@ int sys_support = 0;
- 
- void cleanup(void);
- void setup(void);
--static inline int getcpu(unsigned int *, unsigned int *, void *);
-+static inline int get_cpu(unsigned int *, unsigned int *, void *);
- unsigned int set_cpu_affinity(void);
- unsigned int get_nodeid(unsigned int);
- unsigned int max_cpuid(size_t, cpu_set_t *);
-@@ -124,7 +124,7 @@ int main(int ac, char **av)
- #ifdef __i386__
- 		node_set = get_nodeid(cpu_set);
- #endif
--		TEST(getcpu(&cpu_id, &node_id, NULL));
-+		TEST(get_cpu(&cpu_id, &node_id, NULL));
- 		if (TEST_RETURN == 0) {
- 			if (cpu_id != cpu_set) {
- 				tst_resm(TFAIL, "getcpu() returned wrong value"
-@@ -157,10 +157,10 @@ int main(int ac, char **av)
- }
- 
- /*
-- * getcpu() - calls the system call
-+ * get_cpu() - calls the system call
-  */
--static inline int getcpu(unsigned *cpu_id, unsigned *node_id,
--			 void *cache_struct)
-+static inline int get_cpu(unsigned *cpu_id, unsigned *node_id,
-+			  void *cache_struct)
- {
- #if defined(__i386__)
- 	return syscall(318, cpu_id, node_id, cache_struct);
--- 
-2.20.0
-
meta/recipes-extended/ltp/ltp/0001-netns_helper.sh-use-ping-6-when-ping6-is-not-avaliab.patch deleted +0 -45

45 lines not on this page · raw message

meta/recipes-extended/ltp/ltp/0001-open_posix_testsuite-mmap24-2-Relax-condition-a-bit.patch +3 -3
index 4d771c0b9c..84736053bf 100644
@@ -43,8 +43,8 @@ index de51d43..810e5c8 100644
 + * 3. Should get ENOMEM or EINVAL.
   */
  
- #define _XOPEN_SOURCE 600
-@@ -93,8 +93,8 @@ int main(void)
+ #include <stdio.h>
+@@ -92,8 +92,8 @@ int main(void)
  	       (unsigned long)len);
  	pa = mmap(addr, len, PROT_READ | PROT_WRITE, MAP_FIXED | MAP_SHARED, fd,
  		  0);
@@ -55,7 +55,7 @@ index de51d43..810e5c8 100644
  		exit(PTS_PASS);
  	}
  
-@@ -103,6 +103,6 @@ int main(void)
+@@ -102,6 +102,6 @@ int main(void)
  	else
  		munmap(pa, len);
  	close(fd);
meta/recipes-extended/ltp/ltp/0001-setrlimit05-Use-another-method-to-get-bad-address.patch deleted +0 -61

61 lines not on this page · raw message

meta/recipes-extended/ltp/ltp/0001-sigwaitinfo01-recent-glibc-calls-syscall-directly.patch deleted +0 -75

75 lines not on this page · raw message

meta/recipes-extended/ltp/ltp/0001-statx-fix-compile-errors.patch deleted +0 -35
deleted file mode 100644
index 140bfeb879..0000000000
@@ -1,35 +0,0 @@
-From a945304b6a2286d0665c70250cc5475eb07fd21e Mon Sep 17 00:00:00 2001
-From: Thadeu Lima de Souza Cascardo <[email protected]>
-Date: Wed, 26 Sep 2018 11:15:45 -0300
-Subject: [PATCH] statx: fix compile errors
-
-struct statx is only defined in sys/stat.h when _GNU_SOURCE is defined.
-The testcases do define _GNU_SOURCE, but not the m4 struct check.
-
-Upstream-Status: Backport
-Signed-off-by: Thadeu Lima de Souza Cascardo <[email protected]>
-Acked-by: Cyril Hrubis <[email protected]>
----
- m4/ltp-statx.m4 | 9 +++++++--
- 1 file changed, 7 insertions(+), 2 deletions(-)
-
-diff --git a/m4/ltp-statx.m4 b/m4/ltp-statx.m4
-index 44303d30c..5f79a94cd 100644
---- a/m4/ltp-statx.m4
-+++ b/m4/ltp-statx.m4
-@@ -23,6 +23,11 @@ dnl
- AC_DEFUN([LTP_CHECK_STATX],[
- AC_CHECK_FUNCS(statx,,,[[#include <sys/stat.h>]])
- AC_CHECK_HEADER(linux/fs.h,,,)
--AC_CHECK_TYPES([struct statx],,,[[#include <sys/stat.h>]])
--AC_CHECK_TYPES([struct statx_timestamp],,,[[#include <sys/stat.h>]])
-+AC_CHECK_TYPES([struct statx],,,[[
-+	#define _GNU_SOURCE
-+	#include <sys/stat.h>
-+]])
-+AC_CHECK_TYPES([struct statx_timestamp],,,[[
-+	#define _GNU_SOURCE
-+	#include <sys/stat.h>]])
- ])
---
-2.19.1
meta/recipes-extended/ltp/ltp/0001-syscalls-fcntl-make-OFD-command-use-fcntl64-syscall-.patch deleted +0 -184

184 lines not on this page · raw message

meta/recipes-extended/ltp/ltp/0004-build-Add-option-to-select-libc-implementation.patch +1 -2
index 2df385591b..3c66c9436f 100644
@@ -42,7 +42,7 @@ diff --git a/testcases/kernel/Makefile b/testcases/kernel/Makefile
 index 50a12fa..4f1987f 100644
 --- a/testcases/kernel/Makefile
 +++ b/testcases/kernel/Makefile
-@@ -48,13 +48,16 @@ SUBDIRS			+= connectors \
+@@ -49,12 +49,15 @@ SUBDIRS			+= connectors \
  			   logging \
  			   mem \
  			   numa \
@@ -51,7 +51,6 @@ index 50a12fa..4f1987f 100644
  			   security \
  			   timers \
  			   tracing \
- 			   module \
  
 +ifneq ($(LIBC),musl)
 +SUBDIRS			+= pty
meta/recipes-extended/ltp/ltp/0007-fix-__WORDSIZE-undeclared-when-building-with-musl.patch +1 -1
index 2f4ca63750..6e6dbf398b 100644
@@ -16,7 +16,7 @@ diff --git a/include/old/test.h b/include/old/test.h
 index b36764d83..cc6f1b551 100644
 --- a/include/old/test.h
 +++ b/include/old/test.h
-@@ -44,6 +44,9 @@
+@@ -17,6 +17,9 @@
  #include <string.h>
  #include <stdlib.h>
  #include <stdint.h>
meta/recipes-extended/ltp/ltp/0008-Check-if-__GLIBC_PREREQ-is-defined-before-using-it.patch +29 -161

264 lines not on this page · raw message

meta/recipes-extended/ltp/ltp/0035-fix-test_proc_kill-hang.patch +1 -1
index 71e32a5e51..665da3bef3 100644
@@ -19,7 +19,7 @@ diff --git a/testcases/kernel/controllers/memcg/functional/memcg_lib.sh b/testca
 index b785fe3..2918cc5 100755
 --- a/testcases/kernel/controllers/memcg/functional/memcg_lib.sh
 +++ b/testcases/kernel/controllers/memcg/functional/memcg_lib.sh
-@@ -283,6 +283,7 @@ test_proc_kill()
+@@ -291,6 +291,7 @@ test_proc_kill()
  	pid=$!
  	TST_CHECKPOINT_WAIT 0
  	echo $pid > tasks
meta/recipes-extended/ltp/ltp/0039-commands-ar01-Fix-for-test-in-deterministic-mode.patch +1 -1
index 9244a06b44..ec5e2ef95a 100644
@@ -32,7 +32,7 @@ index be105f6da..813a51d9c 100644
  TST_TESTFUNC=test
  TST_NEEDS_TMPDIR=1
  TST_NEEDS_CMDS="$AR"
-
+ 
  . tst_test.sh
  
 +setup()
meta/recipes-extended/ltp/ltp/define-sigrtmin-and-sigrtmax-for-musl.patch new file +114 -0

114 lines not on this page · raw message

meta/recipes-extended/ltp/ltp_20190115.bb renamed from meta/recipes-extended/ltp/ltp_20180926.bb +4 -7
similarity index 91%
rename from meta/recipes-extended/ltp/ltp_20180926.bb
rename to meta/recipes-extended/ltp/ltp_20190115.bb
index 99e7023b25..4bc9224a1d 100644
@@ -27,7 +27,7 @@ CFLAGS_append_x86-64 = " -fomit-frame-pointer"
 
 CFLAGS_append_powerpc64 = " -D__SANE_USERSPACE_TYPES__"
 CFLAGS_append_mipsarchn64 = " -D__SANE_USERSPACE_TYPES__"
-SRCREV = "f424769b1ad9fca477118763f88a6cec98ea2c0a"
+SRCREV = "a6a5caef13632e669dda27b0461726eba546a2f3"
 
 SRC_URI = "git://github.com/linux-test-project/ltp.git \
            file://0004-build-Add-option-to-select-libc-implementation.patch \
@@ -46,13 +46,8 @@ SRC_URI = "git://github.com/linux-test-project/ltp.git \
            file://0035-fix-test_proc_kill-hang.patch \
            file://0036-testcases-network-nfsv4-acl-acl1.c-Security-fix-on-s.patch \
            file://0039-commands-ar01-Fix-for-test-in-deterministic-mode.patch \
-           file://0001-syscalls-fcntl-make-OFD-command-use-fcntl64-syscall-.patch \
-           file://0001-sigwaitinfo01-recent-glibc-calls-syscall-directly.patch \
-           file://0001-netns_helper.sh-use-ping-6-when-ping6-is-not-avaliab.patch \
            file://0001-open_posix_testsuite-mmap24-2-Relax-condition-a-bit.patch \
-           file://0001-statx-fix-compile-errors.patch \
-           file://0001-setrlimit05-Use-another-method-to-get-bad-address.patch \
-           file://0001-getcpu01-Rename-getcpu-to-avoid-conflict-with-glibc-.patch \
+           file://define-sigrtmin-and-sigrtmax-for-musl.patch \
            "
 
 S = "${WORKDIR}/git"
@@ -96,6 +91,7 @@ RDEPENDS_${PN} = "\
     curl \
     e2fsprogs-mke2fs \
     expect \
+    file \
     gawk \
     gzip \
     iproute2 \
@@ -105,6 +101,7 @@ RDEPENDS_${PN} = "\
     perl \
     python-core \
     procps \
+    quota \
     unzip \
     util-linux \
     which \
Signature · 1 line
2.17.1
#1 lore raw
Quoted 11 lines · Daniel Díaz

This patch has been refreshed:
* 0008-Check-if-__GLIBC_PREREQ-is-defined-before-using-it.patch

These patches have been dropped as they have been merged:
* 0001-netns_helper.sh-use-ping-6-when-ping6-is-not-avaliab.patch
* 0001-setrlimit05-Use-another-method-to-get-bad-address.patch
* 0001-sigwaitinfo01-recent-glibc-calls-syscall-directly.patch
* 0001-statx-fix-compile-errors.patch
* 0001-syscalls-fcntl-make-OFD-command-use-fcntl64-syscall-.patch
* 0001-getcpu01-Rename-getcpu-to-avoid-conflict-with-glibc-.patch

Quoted 8 lines

This patch has been added:
* define-sigrtmin-and-sigrtmax-for-musl.patch

Added these rdependencies:
* file: for ld01, file01 and logrotate
* quota: for quota_remount_test01

I think there was already a previous version of this upgrade which also didn't work:

http://errors.yoctoproject.org/Errors/Latest/?filter=480efd388db578106a8a2129c0382b7e56a1ee36&type=commit&limit=150

From the -Wformat-security -Werror=format-security security flags.

Cheers,

Richard

#2 lore raw

Hello!

Quoted 20 lines · Richard Purdie
Quoted 18 lines · Daniel Díaz

This patch has been refreshed:
* 0008-Check-if-__GLIBC_PREREQ-is-defined-before-using-it.patch

These patches have been dropped as they have been merged:
* 0001-netns_helper.sh-use-ping-6-when-ping6-is-not-avaliab.patch
* 0001-setrlimit05-Use-another-method-to-get-bad-address.patch
* 0001-sigwaitinfo01-recent-glibc-calls-syscall-directly.patch
* 0001-statx-fix-compile-errors.patch
* 0001-syscalls-fcntl-make-OFD-command-use-fcntl64-syscall-.patch
* 0001-getcpu01-Rename-getcpu-to-avoid-conflict-with-glibc-.patch

This patch has been added:
* define-sigrtmin-and-sigrtmax-for-musl.patch

Added these rdependencies:
* file: for ld01, file01 and logrotate
* quota: for quota_remount_test01

Quoted 7 lines
Quoted 1 line

I think there was already a previous version of this upgrade which also didn't work:

http://errors.yoctoproject.org/Errors/Latest/?filter=480efd388db578106a8a2129c0382b7e56a1ee36&type=commit&limit=150
From the -Wformat-security -Werror=format-security security flags.

I thought the problems were with MUSL and Glibc 2.29 (for which I built). Let me look at these problems, then.

Thanks and greetings!

Daniel Díaz
[email protected]

#3 lore raw
Quoted 7 lines · Daniel Díaz

Hello!

Quoted 4 lines · Richard Purdie

I think there was already a previous version of this upgrade which also
didn't work:

Quoted 5 lines
Quoted 2 lines

http://errors.yoctoproject.org/Errors/Latest/?filter=480efd388db578106a8a2129c0382b7e56a1ee36&type=commit&limit=150
From the -Wformat-security -Werror=format-security security flags.

I thought the problems were with MUSL and Glibc 2.29 (for which I built). Let me look at these problems, then.

Yes, thanks for looking at those.

For now I tweaked the recipe to disable the string security flags for ltp as I wanted to get the upgrade in. As soon as we get a patchset which builds, we can remove that workaround. I'm hoping we can get something upstream for this too...

Cheers,

Richard