Bug#1070853: strace: test failures on 32bit with 64bit time_t
Hi, On Thu, Jun 6, 2024 at 6:12 PM Emanuele Rocca wrote: > > Hello! > > On 2024-06-06 12:48, Bo YU wrote: > > hmm, here I use libdw-dev instead of libunwind-dev as upstream > > suggested. I just have uploaded it to mentor because my mentor did not > > respond to me for a while. Could you help me to try to build it on > > arm64 or armhf again? I have tested it on qemu-user, but it failed. > > Unfortunately the following tests still fail on arm64 with libdw-dev: > > FAIL: strace-k-demangle.test > FAIL: strace-kk-p.test > FAIL: strace-kk.test > FAIL: strace-k-p.test > FAIL: strace-k.test > FAIL: strace-k-with-depth-limit.test > > You can find the full build logs here: > https://people.debian.org/~ema/strace_6.9-1_arm64.build Okay, thanks. I think we would better to upload 6.8-2 first and next I will battle with 6.9 release. ``` To access further information about this package, please visit the following URL: https://mentors.debian.net/package/strace/ Alternatively, you can download the package with 'dget' using this command: dget -x https://mentors.debian.net/debian/pool/main/s/strace/strace_6.8-2.dsc Changes since the last upload: strace (6.8-2) unstable; urgency=medium . * Team upload. * Only use libunwind on amd64 (Closes: #1070853) Thanks to Emanuele Rocca ``` Could you help me to sponsor it or I just send another RFS? Thanks again. BR, Bo
Bug#1070853: strace: test failures on 32bit with 64bit time_t
Hello! On 2024-06-06 12:48, Bo YU wrote: > hmm, here I use libdw-dev instead of libunwind-dev as upstream > suggested. I just have uploaded it to mentor because my mentor did not > respond to me for a while. Could you help me to try to build it on > arm64 or armhf again? I have tested it on qemu-user, but it failed. Unfortunately the following tests still fail on arm64 with libdw-dev: FAIL: strace-k-demangle.test FAIL: strace-kk-p.test FAIL: strace-kk.test FAIL: strace-k-p.test FAIL: strace-k.test FAIL: strace-k-with-depth-limit.test You can find the full build logs here: https://people.debian.org/~ema/strace_6.9-1_arm64.build
Bug#1070853: strace: test failures on 32bit with 64bit time_t
Hi, On Wed, Jun 5, 2024 at 9:38 PM Emanuele Rocca wrote: > > Control: retitle -1 strace: test failures on multiple architectures when > using libunwind > > On 2024-05-11 12:21, Adrian Bunk wrote: > > The tests also fail on i386 > > And on arm64 as well. > > The problem has to do with libunwind as mentioned on the upstream issue, > and it was introduced with: > https://salsa.debian.org/debian/strace/-/commit/711a9970c553b8fb9a5b9eb3a4b80600eb8112c4 > > With the attached patch strace builds fine on arm64 and armhf. Yeah. but before applying the revert, I was trying to update it with 6.9. hmm, here I use libdw-dev instead of libunwind-dev as upstream suggested. I just have uploaded it to mentor because my mentor did not respond to me for a while. Could you help me to try to build it on arm64 or armhf again? I have tested it on qemu-user, but it failed. ``` To access further information about this package, please visit the following URL: https://mentors.debian.net/package/strace/ Alternatively, you can download the package with 'dget' using this command: dget -x https://mentors.debian.net/debian/pool/main/s/strace/strace_6.9-1.dsc Changes since the last upload: strace (6.9-1) experimental; urgency=medium . * Team upload. * New upstream version 6.9. * Drop 03-fix-arm-time_t-ftbfs.patch which upstream has been applied. * Use libdw-dev instead of libunwind-dev. (Closes: #1070853) ``` Thanks for working on this. BR, Bo > > Cheers, > Emanuele
Bug#1070853: strace: test failures on 32bit with 64bit time_t
Control: retitle -1 strace: test failures on multiple architectures when using libunwind On 2024-05-11 12:21, Adrian Bunk wrote: > The tests also fail on i386 And on arm64 as well. The problem has to do with libunwind as mentioned on the upstream issue, and it was introduced with: https://salsa.debian.org/debian/strace/-/commit/711a9970c553b8fb9a5b9eb3a4b80600eb8112c4 With the attached patch strace builds fine on arm64 and armhf. Cheers, Emanuele diff -Nru strace-6.8/debian/changelog strace-6.8/debian/changelog --- strace-6.8/debian/changelog 2024-05-10 06:53:53.0 + +++ strace-6.8/debian/changelog 2024-06-05 13:09:41.0 + @@ -1,3 +1,9 @@ +strace (6.8-1.1) unstable; urgency=medium + + * Only use libunwind on amd64 (Closes: #1070853) + + -- Emanuele Rocca Wed, 05 Jun 2024 13:09:41 + + strace (6.8-1) unstable; urgency=medium * Team upload. diff -Nru strace-6.8/debian/control strace-6.8/debian/control --- strace-6.8/debian/control 2024-05-10 06:53:53.0 + +++ strace-6.8/debian/control 2024-06-05 13:09:41.0 + @@ -2,7 +2,7 @@ Maintainer: Steve McIntyre <93...@debian.org> Section: utils Priority: optional -Build-Depends: libc6-dev (>= 2.2.2) [!alpha !ia64], libc6.1-dev (>= 2.2.2) [alpha ia64], gcc-multilib [amd64 i386 powerpc ppc64 s390 sparc sparc64 x32], debhelper (>= 10.0.0), gawk, libunwind-dev [!alpha !loong64 !m68k !sparc64 !x32], libiberty-dev, libbluetooth-dev +Build-Depends: libc6-dev (>= 2.2.2) [!alpha !ia64], libc6.1-dev (>= 2.2.2) [alpha ia64], gcc-multilib [amd64 i386 powerpc ppc64 s390 sparc sparc64 x32], debhelper (>= 10.0.0), gawk, libunwind-dev [amd64], libiberty-dev, libbluetooth-dev Standards-Version: 4.1.3 Homepage: https://strace.io Vcs-Git: https://salsa.debian.org/debian/strace.git
Bug#1070853: strace: test failures on 32bit with 64bit time_t
Hi, On Sat, May 11, 2024 at 12:21:05AM +0300, Adrian Bunk wrote: Control: retitle -1 strace: test failures on 32bit The tests also fail on i386, so do not seem to be related to time_t. Thanks for working on this. I will report this to upstream. -- Regards, -- Bo YU signature.asc Description: PGP signature
Bug#1070853: strace: test failures on 32bit with 64bit time_t
Control: retitle -1 strace: test failures on 32bit The tests also fail on i386, so do not seem to be related to time_t. cu Adrian
Bug#1070853: strace: test failures on 32bit with 64bit time_t
Source: strace
Version: 6.8-1
Severity: serious
Tags: ftbfs
https://buildd.debian.org/status/fetch.php?pkg=strace&arch=armhf&ver=6.8-1&stamp=1715348223&raw=0
https://buildd.debian.org/status/fetch.php?pkg=strace&arch=armel&ver=6.8-1&stamp=1715346678&raw=0
https://buildd.debian.org/status/fetch.php?pkg=strace&arch=powerpc&ver=6.8-1&stamp=1715349316&raw=0
...
FAIL: strace-k.test
PASS: umovestr_cached.test
FAIL: strace-k-with-depth-limit.test
FAIL: strace-k-demangle.test
PASS: strace-DD.test
FAIL: strace-k-p.test
PASS: strace--tips-full.test
PASS: qual_fault-syscall.test
PASS: qual_fault.test
PASS: filtering_syscall-syntax.test
PASS: looping_threads.test
==
strace 6.8: tests/test-suite.log
==
# TOTAL: 1350
# PASS: 1057
# SKIP: 289
# XFAIL: 0
# FAIL: 4
# XPASS: 0
# ERROR: 0
...
FAIL: strace-k
==
+ ../src/strace -V
+ TIMEOUT=timeout -k 5 -s XCPU 1500
+ timeout -k 5 -s XCPU 1500 true
+ [ 1 -eq 0 ]
+ exec timeout -k 5 -s XCPU 1500 ../../tests/strace-k.test
+ : 0
+ : -k
+ : --stack-trace
+ :
+ [ -f /proc/self/maps ]
+ check_prog grep
+ type grep
+ check_prog readlink
+ type readlink
+ check_prog sed
+ type sed
+ check_prog tr
+ type tr
+ command -v sed
+ path_to_sed=/usr/bin/sed
+ [ -x /usr/bin/sed ]
+ readlink -ev -- /usr/bin/sed
+ path_to_sed=/usr/bin/sed
+ /usr/bin/sed -n s/^[^/]\+[[:space:]]\(\/.*\)$/\1/p /proc/self/maps
+ grep -F -x -e /usr/bin/sed
+ run_prog ../stack-fcall
+ [ 1 -eq 0 ]
+ args=../stack-fcall
+ ../stack-fcall
+ [ x0 = x1 ]
+ run_strace -e chdir -k ../stack-fcall
+
+ args=-e chdir -k ../stack-fcall
+ ../../src/strace -o log -e chdir -k ../stack-fcall
+ expected=../../../tests/strace-k.expected
+ awk_script_common=
/^[^ ]/ {
if (out != "")
print out
syscall = gensub(/^([[:alnum:]_]+)\(.*/, "\\1", 1)
signal = gensub(/^--- ([A-Z]+) .*/, "\\1", 1)
if (syscall != $0) {
out = syscall
stop = 0
} else if (signal != $0) {
out = signal
stop = 0
} else {
out = ""
}
}
/^ > too many stack frames/ {
out = out " _too_many_stack_frames_"
stop = 1
}
+ awk_script_symbol=
/^ >[^(]+\(([^+]+)\+0x[a-f0-9]+\) / && !stop {
sym = gensub(/^ >[^(]+\(([^+]+)\+0x[a-f0-9]+\) .*$/, "\\1", 1)
out = out " " sym
if (sym == "main")
stop = 1
}
+ awk_script_source=
/^ >[^(]+\(([^+]+)\+0x[a-f0-9]+\) \[0x[a-f0-9]+\] ([^:]+):([0-9]+)$/ && !stop {
sym = gensub(/^ >[^(]+\(([^+]+)\+0x[a-f0-9]+\) .*$/, "\\1", 1)
if (sym == "main" || sym ~ /f[0-9]/) {
file = gensub(/^ >[^(]+\(([^+]+)\+0x[a-f0-9]+\) \[0x[a-f0-9]+\]
([^:]+):([0-9]+)$/, "\\2", 1)
line = gensub(/^ >[^(]+\(([^+]+)\+0x[a-f0-9]+\) \[0x[a-f0-9]+\]
([^:]+):([0-9]+)$/, "\\3", 1)
sub(".*/", "", file)
out = out " " sym "<" file ":" line ">"
if (sym == "main")
stop = 1
}
}
+ [ --stack-trace = --stack-trace ]
+ awk_script=
/^[^ ]/ {
if (out != "")
print out
syscall = gensub(/^([[:alnum:]_]+)\(.*/, "\\1", 1)
signal = gensub(/^--- ([A-Z]+) .*/, "\\1", 1)
if (syscall != $0) {
out = syscall
stop = 0
} else if (signal != $0) {
out = signal
stop = 0
} else {
out = ""
}
}
/^ > too many stack frames/ {
out = out " _too_many_stack_frames_"
stop = 1
}
/^ >[^(]+\(([^+]+)\+0x[a-f0-9]+\) / && !stop {
sym = gensub(/^ >[^(]+\(([^+]+)\+0x[a-f0-9]+\) .*$/, "\\1", 1)
out = out " " sym
if (sym == "main")
stop = 1
}
+ awk
/^[^ ]/ {
if (out != "")
print out
syscall = gensub(/^([[:alnum:]_]+)\(.*/, "\\1", 1)
signal = gensub(/^--- ([A-Z]+) .*/, "\\1", 1)
if (syscall != $0) {
out = syscall
stop = 0
} else if (signal != $0) {
out = signal
stop = 0
} else {
out = ""
}
}
/^ > too many stack frames/ {
out = out " _too_many_stack_frames_"
stop = 1
}
/^ >[^(]+\(([^+]+)\+0x[a-f0-9]+\) / && !stop {
sym = gensub(/^ >[^(]+\(([^+]+)\+0x[a-f0-9]+\) .*$/, "\\1", 1)
out = out " " sym
if (sym == "main")
stop = 1
}
log
+ LC_ALL=C grep -E -x -f ../../../tests/strace-k.expected
+ cat ../../../tests/strace-k.expected
+ cat out
+ cat
Failed pattern of expected output:
^chdir (__kernel_vsyscall )?(__)?chdir f3 f2 f1 f0 main
^SIGURG (__kernel_vsyscall )?(__)?kill f3 f2 f1 f0 main
Actual output:
chdir chdir f3
SIGURG kill f3
+ pattern=
+ pattern=No DWARF information found
+ [ -n No DWARF information found ]
+ LC_ALL=C grep -x > No DWARF information found
+ dump_log_and_fai
