Comments (14)
glaubitz
commented on July 18, 2024
1
On Jun 25, 2024, at 6:09 PM, Dominique Dumont ***@***.***> wrote:
Unfortunately, I cannot reproduce this problem on Debian ppc64 machines, so I cannot strace it.
Did you remember to set the nice value accordingly?
from libuv.
bnoordhuis
commented on July 18, 2024
The test makes the (perhaps somewhat naive) assumption that the main thread is created with a priority of 0. Do you happen to know what scheduling policy and priority were used when you ran the test?
The fix is probably quite simple:
diff --git a/test/test-thread-priority.c b/test/test-thread-priority.c
index 0aaf2977..769b485e 100644
--- a/test/test-thread-priority.c
+++ b/test/test-thread-priority.c
@@ -88,8 +88,6 @@ TEST_IMPL(thread_priority) {
* test set nice value for the calling thread with default schedule policy
*/
#ifdef __linux__
- ASSERT_OK(uv_thread_getpriority(pthread_self(), &priority));
- ASSERT_EQ(priority, 0);
ASSERT_OK(uv_thread_setpriority(pthread_self(), UV_THREAD_PRIORITY_LOWEST));
ASSERT_OK(uv_thread_getpriority(pthread_self(), &priority));
ASSERT_EQ(priority, (0 - UV_THREAD_PRIORITY_LOWEST * 2));from libuv.
nikneym
commented on July 18, 2024
The test makes the (perhaps somewhat naive) assumption that the main thread is created with a priority of 0. Do you happen to know what scheduling policy and priority were used when you ran the test?
I don't know how to do this. Where should I look while tests are running?
I've applied the diff and it fails with the following:
not ok 355 - thread_priority
# exit code 134
# Output from process `thread_priority`:
# Assertion failed in test/test-thread-priority.c on line 93: `uv_thread_setpriority(pthread_self(), UV_THREAD_PRIORITY_LOWEST) == 0` (-13 == 0)
from libuv.
bnoordhuis
commented on July 18, 2024
That -13 error suggests it's a permission error. Can you run strace -o trace.log -f build/uv_run_tests_a thread_priority thread_priority and post the contents of trace.log?
from libuv.
nikneym
commented on July 18, 2024
Hi, sorry for the late reply. Directory named build has not created actually, I'm using make to build and test things. Should I switch to cmake?
from libuv.
bnoordhuis
commented on July 18, 2024
No, that's okay. As long as you can strace the test.
from libuv.
glaubitz
commented on July 18, 2024
The test makes the (perhaps somewhat naive) assumption that the main thread is created with a priority of 0. Do you happen to know what scheduling policy and priority were used when you ran the test?
The fix is probably quite simple:
diff --git a/test/test-thread-priority.c b/test/test-thread-priority.c index 0aaf2977..769b485e 100644 --- a/test/test-thread-priority.c +++ b/test/test-thread-priority.c @@ -88,8 +88,6 @@ TEST_IMPL(thread_priority) { * test set nice value for the calling thread with default schedule policy */ #ifdef __linux__ - ASSERT_OK(uv_thread_getpriority(pthread_self(), &priority)); - ASSERT_EQ(priority, 0); ASSERT_OK(uv_thread_setpriority(pthread_self(), UV_THREAD_PRIORITY_LOWEST)); ASSERT_OK(uv_thread_getpriority(pthread_self(), &priority)); ASSERT_EQ(priority, (0 - UV_THREAD_PRIORITY_LOWEST * 2));
This affects Debian as well, see: https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=1071194#10
I agree that the assumption about nice == 0 should just be removed here.
from libuv.
glaubitz
commented on July 18, 2024
Gentoo is also patching this line out: https://gitweb.gentoo.org/repo/gentoo.git/tree/dev-libs/libuv/files/libuv-1.48.0-test-thread-priority-portage.patch
So, I would suggest removing these two asserts as making assumptions about the build environment is not really good idea.
from libuv.
dod38fr
commented on July 18, 2024
ok, I'll patch this out from Debian package.
Thanks for the help
from libuv.
bnoordhuis
commented on July 18, 2024
If the ASSERT_EQ(priority, (0 - UV_THREAD_PRIORITY_LOWEST * 2)); assert trips up, that suggests the uv_thread_setpriority call isn't working, or isn't working properly. That's somewhat concerning and should be investigated.
from libuv.
dod38fr
commented on July 18, 2024
We're not there yet.
I've patched out the lines:
ASSERT_OK(uv_thread_getpriority(pthread_self(), &priority));
ASSERT_EQ(priority, 0);
Test on hppa, loong64, powerpc, ppc64, sparc64 and x32 fail on the line after:
not ok 354 - thread_priority
# exit code 6
# Output from process `thread_priority`:
# Assertion failed in ./test/test-thread-priority.c on line 91: `uv_thread_setpriority(pthread_self(), UV_THREAD_PRIORITY_LOWEST) == 0` (-13 == 0)
See detailed logs in https://buildd.debian.org/status/package.php?p=libuv1
from libuv.
bnoordhuis
commented on July 18, 2024
-13 is EACCES and suggests the setpriority(2) system call refuses to change the process's nice value.
My guess is debian buildbots run thing at low niceness (would make sense for a batch system) and that's what's tripping up the test. If you can strace the test, it should be very easy confirm; you can trace the test with:
strace <flags> ./build/uv_run_tests_a thread_priority thread_priority
I suppose libuv has a bug in that it sets niceness from -4 to +4 but the full range is -20 to +19.
from libuv.
dod38fr
commented on July 18, 2024
Unfortunately, I cannot reproduce this problem on Debian ppc64 machines, so I cannot strace it.
from libuv.
dod38fr
commented on July 18, 2024
Oops, I did not.
Here the strace output:
(sid_ppc64-dchroot)dod@perotto:~/libuv1-1.48.0$ nice strace ./obj-powerpc64-linux-gnu/uv_run_tests thread_priority
execve("./obj-powerpc64-linux-gnu/uv_run_tests", ["./obj-powerpc64-linux-gnu/uv_run"..., "thread_priority"], 0x7fffd5169808 /* 19 vars */) = 0
brk(NULL) = 0x10007280000
access("/etc/ld.so.preload", R_OK) = -1 ENOENT (No such file or directory)
openat(AT_FDCWD, "/home/dod/libuv1-1.48.0/obj-powerpc64-linux-gnu/libuv.so.1", O_RDONLY|O_CLOEXEC) = 3
read(3, "\177ELF\2\2\1\0\0\0\0\0\0\0\0\0\0\3\0\25\0\0\0\1\0\0\0\0\0\0\0\0"..., 832) = 832
newfstatat(3, "", {st_mode=S_IFREG|0755, st_size=1099664, ...}, AT_EMPTY_PATH) = 0
mmap(NULL, 394752, PROT_READ|PROT_EXEC, MAP_PRIVATE|MAP_DENYWRITE, 3, 0) = 0x7fffae520000
mmap(0x7fffae570000, 131072, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_FIXED|MAP_DENYWRITE, 3, 0x40000) = 0x7fffae570000
close(3) = 0
openat(AT_FDCWD, "/home/dod/libuv1-1.48.0/obj-powerpc64-linux-gnu/libc.so.6", O_RDONLY|O_CLOEXEC) = -1 ENOENT (No such file or directory)
openat(AT_FDCWD, "/etc/ld.so.cache", O_RDONLY|O_CLOEXEC) = 3
newfstatat(3, "", {st_mode=S_IFREG|0644, st_size=10991, ...}, AT_EMPTY_PATH) = 0
mmap(NULL, 10991, PROT_READ, MAP_PRIVATE, 3, 0) = 0x7fffae510000
close(3) = 0
openat(AT_FDCWD, "/lib/powerpc64-linux-gnu/libc.so.6", O_RDONLY|O_CLOEXEC) = 3
read(3, "\177ELF\2\2\1\3\0\0\0\0\0\0\0\0\0\3\0\25\0\0\0\1\0\0\0\0\0!\235\320"..., 832) = 832
newfstatat(3, "", {st_mode=S_IFREG|0755, st_size=2306480, ...}, AT_EMPTY_PATH) = 0
mmap(NULL, 2353712, PROT_READ|PROT_EXEC, MAP_PRIVATE|MAP_DENYWRITE, 3, 0) = 0x7fffae2d0000
mmap(0x7fffae4e0000, 196608, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_FIXED|MAP_DENYWRITE, 3, 0x210000) = 0x7fffae4e0000
close(3) = 0
set_tid_address(0x7fffae623950) = 2462874
set_robust_list(0x7fffae623960, 24) = 0
rseq(0x7fffae623fa0, 0x20, 0, 0xfe5000b) = 0
mprotect(0x7fffae4e0000, 131072, PROT_READ) = 0
mprotect(0x7fffae570000, 65536, PROT_READ) = 0
mprotect(0x125700000, 65536, PROT_READ) = 0
mprotect(0x7fffae610000, 65536, PROT_READ) = 0
prlimit64(0, RLIMIT_STACK, NULL, {rlim_cur=8192*1024, rlim_max=RLIM64_INFINITY}) = 0
munmap(0x7fffae510000, 10991) = 0
geteuid() = 3143
rt_sigaction(SIGPIPE, {sa_handler=SIG_IGN, sa_mask=[PIPE], sa_flags=SA_RESTART}, {sa_handler=SIG_DFL, sa_mask=[], sa_flags=0}, 8) = 0
getrandom("\x51\x38\x11\x19\x81\x82\x74\xcd", 8, GRND_NONBLOCK) = 8
brk(NULL) = 0x10007280000
brk(0x100072b0000) = 0x100072b0000
unlink("/tmp/uv-test-sock") = -1 ENOENT (No such file or directory)
unlink("/tmp/uv-test-sock2") = -1 ENOENT (No such file or directory)
unlink("/tmp/uv-test-sock3") = -1 ENOENT (No such file or directory)
openat(AT_FDCWD, "/tmp", O_RDWR|O_EXCL|O_TMPFILE, 0600) = 3
fcntl(3, F_GETFL) = 0x414002 (flags O_RDWR|O_LARGEFILE|O_TMPFILE)
clone(child_stack=NULL, flags=CLONE_CHILD_CLEARTID|CLONE_CHILD_SETTID|SIGCHLD, child_tidptr=0x7fffae623950) = 2462875
pipe2([4, 5], 0) = 0
rt_sigaction(SIGRT_1, {sa_handler=0x7fffae4efd10, sa_mask=[], sa_flags=SA_ONSTACK|SA_RESTART|SA_SIGINFO}, NULL, 8) = 0
rt_sigprocmask(SIG_UNBLOCK, [RTMIN RT_1], NULL, 8) = 0
mmap(NULL, 327680, PROT_NONE, MAP_PRIVATE|MAP_ANONYMOUS|MAP_STACK, -1, 0) = 0x7fffae280000
mprotect(0x7fffae290000, 262144, PROT_READ|PROT_WRITE) = 0
rt_sigprocmask(SIG_BLOCK, ~[], [], 8) = 0
clone3({flags=CLONE_VM|CLONE_FS|CLONE_FILES|CLONE_SIGHAND|CLONE_THREAD|CLONE_SYSVSEM|CLONE_SETTLS|CLONE_PARENT_SETTID|CLONE_CHILD_CLEARTID, child_tid=0x7fffae2cf1f0, parent_tid=0x7fffae2cf1f0, exit_signal=0, stack=0x7fffae280000, stack_size=0x4e970, tls=0x7fffae2d68e0} => {parent_tid=[2462876]}, 88) = 2462876
rt_sigprocmask(SIG_SETMASK, [], NULL, 8) = 0
pselect6(5, [4], NULL, NULL, {tv_sec=5, tv_nsec=0}, NULL) = ? ERESTARTNOHAND (To be restarted if no handler)
--- SIGCHLD {si_signo=SIGCHLD, si_code=CLD_KILLED, si_pid=2462875, si_uid=3143, si_status=SIGABRT, si_utime=0, si_stime=0} ---
pselect6(5, [4], NULL, NULL, {tv_sec=4, tv_nsec=998629413}, NULL) = 1 (in [4], left {tv_sec=4, tv_nsec=998626653})
close(4) = 0
close(5) = 0
write(1, "not ok 1 - thread_priority\n", 27not ok 1 - thread_priority
) = 27
write(1, "# exit code 6\n", 14# exit code 6
) = 14
write(1, "# ", 2# ) = 2
newfstatat(3, "", {st_mode=S_IFREG|0600, st_size=145, ...}, AT_EMPTY_PATH) = 0
write(1, "Output from process `thread_prio"..., 39Output from process `thread_priority`:
) = 39
newfstatat(3, "", {st_mode=S_IFREG|0600, st_size=145, ...}, AT_EMPTY_PATH) = 0
_llseek(3, 0, [0], SEEK_SET) = 0
read(3, "Assertion failed in ./test/test-"..., 8192) = 145
read(3, "", 8192) = 0
write(1, "# ", 2# ) = 2
write(1, "Assertion failed in ./test/test-"..., 144Assertion failed in ./test/test-thread-priority.c on line 91: `uv_thread_setpriority(pthread_self(), UV_THREAD_PRIORITY_LOWEST) == 0` (-13 == 0)) = 144
write(1, "\n", 1
) = 1
close(3) = 0
exit_group(6) = ?
+++ exited with 6 +++
from libuv.
Related Issues (20)
- System freeze after suspending two instances of Neovim HOT 5
- windows uv_poll_init
- doc: improve example docs/code/progress/main.c HOT 1
- win: use WinDNS family of functions for DNS resolving HOT 4
- Improper Validation of Certificate Trust in certifi and dependent packages HOT 4
- 测试在Windows XP系统上运行时遇到RegGetValueW无法定位于ADVAPI32.dll的问题 HOT 4
- uv_tty_reset_mode doesn't work as advertised if original handle is closed HOT 5
- Provide an OS independent way to set/get UV_THREADPOOL_SIZE HOT 8
- Build currently failing on AIX and OS/400 (IBM i) HOT 7
- about test udp_multicast_join | udp_multicast_join6 not ok | not ok 180 - metrics_pool_events HOT 3
- uv_tcp_close_reset() needs to handle some additional errno other than EINVAL on NetBSD HOT 2
- setuid() does not drop all privileges due to io_uring (CVE-2024-22017) HOT 3
- android and Debian: uv_fs_event stops working when file is deleted HOT 3
- unit test watcher_cross_stop fail in wsl2 HOT 1
- build fail HOT 6
- If you attempt to send a UDP packet to an IPv4 address using an IPv6 socket, sendto() will report the WSAEFAULT error. HOT 9
- Missing symbols in the library file compared to the headers HOT 4
- Issue with GetSystemTimePreciseAsFileTime on Windows Server 2008 in libuv 1.48.0 and Suggested Fix HOT 1
- CMake build fails with FetchContent HOT 2
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from libuv.