if available use a Child's pidfd for kill/wait #117957

the8472 · 2023-11-16T01:07:32Z

This should get us closer to stabilization of pidfds since they now do something useful. And they're CLOEXEC now.

$ strace -ffe clone,sendmsg,recvmsg,execve,kill,pidfd_open,pidfd_send_signal,waitpid,waitid ./x test std --no-doc -- pidfd

[...]
running 1 tests
strace: Process 816007 attached
[pid 816007] pidfd_open(816006, 0)      = 3
[pid 816007] clone(child_stack=NULL, flags=CLONE_CHILD_CLEARTID|CLONE_CHILD_SETTID|SIGCHLD, child_tidptr=0x7f0c6b787990) = 816008
strace: Process 816008 attached
[pid 816007] recvmsg(3,  <unfinished ...>
[pid 816008] pidfd_open(816008, 0)      = 3
[pid 816008] sendmsg(4, {msg_name=NULL, msg_namelen=0, msg_iov=[{iov_base="", iov_len=0}], msg_iovlen=1, msg_control=[{cmsg_len=20, cmsg_level=SOL_SOCKET, cmsg_type=SCM_RIGHTS, cmsg_data=[3]}], msg_controllen=24, msg_flags=0}, 0) = 0
[pid 816007] <... recvmsg resumed>{msg_name=NULL, msg_namelen=0, msg_iov=[{iov_base="", iov_len=0}], msg_iovlen=1, msg_control=[{cmsg_len=20, cmsg_level=SOL_SOCKET, cmsg_type=SCM_RIGHTS, cmsg_data=[4]}], msg_controllen=24, msg_flags=MSG_CMSG_CLOEXEC}, MSG_CMSG_CLOEXEC) = 0
[pid 816008] execve("/usr/bin/false", ["false"], 0x7ffcf2100048 /* 105 vars */) = 0
[pid 816007] waitid(P_PIDFD, 4,  <unfinished ...>
[pid 816008] +++ exited with 1 +++
[pid 816007] <... waitid resumed>{si_signo=SIGCHLD, si_code=CLD_EXITED, si_pid=816008, si_uid=1001, si_status=1, si_utime=0, si_stime=0}, WEXITED, NULL) = 0
[pid 816007] --- SIGCHLD {si_signo=SIGCHLD, si_code=CLD_EXITED, si_pid=816008, si_uid=1001, si_status=1, si_utime=0, si_stime=0} ---
[pid 816007] clone(child_stack=NULL, flags=CLONE_CHILD_CLEARTID|CLONE_CHILD_SETTID|SIGCHLDstrace: Process 816009 attached
, child_tidptr=0x7f0c6b787990) = 816009
[pid 816007] recvmsg(3,  <unfinished ...>
[pid 816009] pidfd_open(816009, 0)      = 3
[pid 816009] sendmsg(5, {msg_name=NULL, msg_namelen=0, msg_iov=[{iov_base="", iov_len=0}], msg_iovlen=1, msg_control=[{cmsg_len=20, cmsg_level=SOL_SOCKET, cmsg_type=SCM_RIGHTS, cmsg_data=[3]}], msg_controllen=24, msg_flags=0}, 0) = 0
[pid 816007] <... recvmsg resumed>{msg_name=NULL, msg_namelen=0, msg_iov=[{iov_base="", iov_len=0}], msg_iovlen=1, msg_control=[{cmsg_len=20, cmsg_level=SOL_SOCKET, cmsg_type=SCM_RIGHTS, cmsg_data=[5]}], msg_controllen=24, msg_flags=MSG_CMSG_CLOEXEC}, MSG_CMSG_CLOEXEC) = 0
[pid 816009] execve("/usr/bin/sleep", ["sleep", "1000"], 0x7ffcf2100048 /* 105 vars */) = 0
[pid 816007] waitid(P_PIDFD, 5, {}, WNOHANG|WEXITED, NULL) = 0
[pid 816007] pidfd_send_signal(5, SIGKILL, NULL, 0) = 0
[pid 816007] waitid(P_PIDFD, 5,  <unfinished ...>
[pid 816009] +++ killed by SIGKILL +++
[pid 816007] <... waitid resumed>{si_signo=SIGCHLD, si_code=CLD_KILLED, si_pid=816009, si_uid=1001, si_status=SIGKILL, si_utime=0, si_stime=0}, WEXITED, NULL) = 0
[pid 816007] --- SIGCHLD {si_signo=SIGCHLD, si_code=CLD_KILLED, si_pid=816009, si_uid=1001, si_status=SIGKILL, si_utime=0, si_stime=0} ---
[pid 816007] +++ exited with 0 +++

rustbot · 2023-11-16T01:07:41Z

r? @Mark-Simulacrum

(rustbot has picked a reviewer for you, use r? to override)

Mark-Simulacrum · 2023-11-18T17:15:14Z

@bors r+

bors · 2023-11-18T17:15:16Z

📌 Commit 12efa53 has been approved by Mark-Simulacrum

It is now in the queue for this repository.

bors · 2023-11-19T04:57:40Z

⌛ Testing commit 12efa53 with merge e71a414...

if available use a Child's pidfd for kill/wait This should get us closer to stabilization of pidfds since they now do something useful. And they're `CLOEXEC` now. ``` $ strace -ffe clone,sendmsg,recvmsg,execve,kill,pidfd_open,pidfd_send_signal,waitpid,waitid ./x test std --no-doc -- pidfd [...] running 1 tests strace: Process 816007 attached [pid 816007] pidfd_open(816006, 0) = 3 [pid 816007] clone(child_stack=NULL, flags=CLONE_CHILD_CLEARTID|CLONE_CHILD_SETTID|SIGCHLD, child_tidptr=0x7f0c6b787990) = 816008 strace: Process 816008 attached [pid 816007] recvmsg(3, <unfinished ...> [pid 816008] pidfd_open(816008, 0) = 3 [pid 816008] sendmsg(4, {msg_name=NULL, msg_namelen=0, msg_iov=[{iov_base="", iov_len=0}], msg_iovlen=1, msg_control=[{cmsg_len=20, cmsg_level=SOL_SOCKET, cmsg_type=SCM_RIGHTS, cmsg_data=[3]}], msg_controllen=24, msg_flags=0}, 0) = 0 [pid 816007] <... recvmsg resumed>{msg_name=NULL, msg_namelen=0, msg_iov=[{iov_base="", iov_len=0}], msg_iovlen=1, msg_control=[{cmsg_len=20, cmsg_level=SOL_SOCKET, cmsg_type=SCM_RIGHTS, cmsg_data=[4]}], msg_controllen=24, msg_flags=MSG_CMSG_CLOEXEC}, MSG_CMSG_CLOEXEC) = 0 [pid 816008] execve("/usr/bin/false", ["false"], 0x7ffcf2100048 /* 105 vars */) = 0 [pid 816007] waitid(P_PIDFD, 4, <unfinished ...> [pid 816008] +++ exited with 1 +++ [pid 816007] <... waitid resumed>{si_signo=SIGCHLD, si_code=CLD_EXITED, si_pid=816008, si_uid=1001, si_status=1, si_utime=0, si_stime=0}, WEXITED, NULL) = 0 [pid 816007] --- SIGCHLD {si_signo=SIGCHLD, si_code=CLD_EXITED, si_pid=816008, si_uid=1001, si_status=1, si_utime=0, si_stime=0} --- [pid 816007] clone(child_stack=NULL, flags=CLONE_CHILD_CLEARTID|CLONE_CHILD_SETTID|SIGCHLDstrace: Process 816009 attached , child_tidptr=0x7f0c6b787990) = 816009 [pid 816007] recvmsg(3, <unfinished ...> [pid 816009] pidfd_open(816009, 0) = 3 [pid 816009] sendmsg(5, {msg_name=NULL, msg_namelen=0, msg_iov=[{iov_base="", iov_len=0}], msg_iovlen=1, msg_control=[{cmsg_len=20, cmsg_level=SOL_SOCKET, cmsg_type=SCM_RIGHTS, cmsg_data=[3]}], msg_controllen=24, msg_flags=0}, 0) = 0 [pid 816007] <... recvmsg resumed>{msg_name=NULL, msg_namelen=0, msg_iov=[{iov_base="", iov_len=0}], msg_iovlen=1, msg_control=[{cmsg_len=20, cmsg_level=SOL_SOCKET, cmsg_type=SCM_RIGHTS, cmsg_data=[5]}], msg_controllen=24, msg_flags=MSG_CMSG_CLOEXEC}, MSG_CMSG_CLOEXEC) = 0 [pid 816009] execve("/usr/bin/sleep", ["sleep", "1000"], 0x7ffcf2100048 /* 105 vars */) = 0 [pid 816007] waitid(P_PIDFD, 5, {}, WNOHANG|WEXITED, NULL) = 0 [pid 816007] pidfd_send_signal(5, SIGKILL, NULL, 0) = 0 [pid 816007] waitid(P_PIDFD, 5, <unfinished ...> [pid 816009] +++ killed by SIGKILL +++ [pid 816007] <... waitid resumed>{si_signo=SIGCHLD, si_code=CLD_KILLED, si_pid=816009, si_uid=1001, si_status=SIGKILL, si_utime=0, si_stime=0}, WEXITED, NULL) = 0 [pid 816007] --- SIGCHLD {si_signo=SIGCHLD, si_code=CLD_KILLED, si_pid=816009, si_uid=1001, si_status=SIGKILL, si_utime=0, si_stime=0} --- [pid 816007] +++ exited with 0 +++ ```

bors · 2023-11-19T05:34:01Z

💔 Test failed - checks-actions

the8472 · 2023-11-19T14:28:09Z

Looks like create_pidfd(true) was broken on older kernels and the test didn't catch that previously. sendmsg returned EINVAL when the cmsg fields point to an empty cmsg list.

Mark-Simulacrum · 2023-11-19T20:21:12Z

@bors r+

bors · 2023-11-19T20:21:15Z

📌 Commit f34e7f4 has been approved by Mark-Simulacrum

It is now in the queue for this repository.

if available use a Child's pidfd for kill/wait This should get us closer to stabilization of pidfds since they now do something useful. And they're `CLOEXEC` now. ``` $ strace -ffe clone,sendmsg,recvmsg,execve,kill,pidfd_open,pidfd_send_signal,waitpid,waitid ./x test std --no-doc -- pidfd [...] running 1 tests strace: Process 816007 attached [pid 816007] pidfd_open(816006, 0) = 3 [pid 816007] clone(child_stack=NULL, flags=CLONE_CHILD_CLEARTID|CLONE_CHILD_SETTID|SIGCHLD, child_tidptr=0x7f0c6b787990) = 816008 strace: Process 816008 attached [pid 816007] recvmsg(3, <unfinished ...> [pid 816008] pidfd_open(816008, 0) = 3 [pid 816008] sendmsg(4, {msg_name=NULL, msg_namelen=0, msg_iov=[{iov_base="", iov_len=0}], msg_iovlen=1, msg_control=[{cmsg_len=20, cmsg_level=SOL_SOCKET, cmsg_type=SCM_RIGHTS, cmsg_data=[3]}], msg_controllen=24, msg_flags=0}, 0) = 0 [pid 816007] <... recvmsg resumed>{msg_name=NULL, msg_namelen=0, msg_iov=[{iov_base="", iov_len=0}], msg_iovlen=1, msg_control=[{cmsg_len=20, cmsg_level=SOL_SOCKET, cmsg_type=SCM_RIGHTS, cmsg_data=[4]}], msg_controllen=24, msg_flags=MSG_CMSG_CLOEXEC}, MSG_CMSG_CLOEXEC) = 0 [pid 816008] execve("/usr/bin/false", ["false"], 0x7ffcf2100048 /* 105 vars */) = 0 [pid 816007] waitid(P_PIDFD, 4, <unfinished ...> [pid 816008] +++ exited with 1 +++ [pid 816007] <... waitid resumed>{si_signo=SIGCHLD, si_code=CLD_EXITED, si_pid=816008, si_uid=1001, si_status=1, si_utime=0, si_stime=0}, WEXITED, NULL) = 0 [pid 816007] --- SIGCHLD {si_signo=SIGCHLD, si_code=CLD_EXITED, si_pid=816008, si_uid=1001, si_status=1, si_utime=0, si_stime=0} --- [pid 816007] clone(child_stack=NULL, flags=CLONE_CHILD_CLEARTID|CLONE_CHILD_SETTID|SIGCHLDstrace: Process 816009 attached , child_tidptr=0x7f0c6b787990) = 816009 [pid 816007] recvmsg(3, <unfinished ...> [pid 816009] pidfd_open(816009, 0) = 3 [pid 816009] sendmsg(5, {msg_name=NULL, msg_namelen=0, msg_iov=[{iov_base="", iov_len=0}], msg_iovlen=1, msg_control=[{cmsg_len=20, cmsg_level=SOL_SOCKET, cmsg_type=SCM_RIGHTS, cmsg_data=[3]}], msg_controllen=24, msg_flags=0}, 0) = 0 [pid 816007] <... recvmsg resumed>{msg_name=NULL, msg_namelen=0, msg_iov=[{iov_base="", iov_len=0}], msg_iovlen=1, msg_control=[{cmsg_len=20, cmsg_level=SOL_SOCKET, cmsg_type=SCM_RIGHTS, cmsg_data=[5]}], msg_controllen=24, msg_flags=MSG_CMSG_CLOEXEC}, MSG_CMSG_CLOEXEC) = 0 [pid 816009] execve("/usr/bin/sleep", ["sleep", "1000"], 0x7ffcf2100048 /* 105 vars */) = 0 [pid 816007] waitid(P_PIDFD, 5, {}, WNOHANG|WEXITED, NULL) = 0 [pid 816007] pidfd_send_signal(5, SIGKILL, NULL, 0) = 0 [pid 816007] waitid(P_PIDFD, 5, <unfinished ...> [pid 816009] +++ killed by SIGKILL +++ [pid 816007] <... waitid resumed>{si_signo=SIGCHLD, si_code=CLD_KILLED, si_pid=816009, si_uid=1001, si_status=SIGKILL, si_utime=0, si_stime=0}, WEXITED, NULL) = 0 [pid 816007] --- SIGCHLD {si_signo=SIGCHLD, si_code=CLD_KILLED, si_pid=816009, si_uid=1001, si_status=SIGKILL, si_utime=0, si_stime=0} --- [pid 816007] +++ exited with 0 +++ ```

…mpiler-errors Rollup of 8 pull requests Successful merges: - rust-lang#115526 (Add arm64e-apple-ios & arm64e-apple-darwin targets) - rust-lang#115691 (Add `$message_type` field to distinguish json diagnostic outputs) - rust-lang#117828 (Avoid iterating over hashmaps in astconv) - rust-lang#117832 (interpret: simplify handling of shifts by no longer trying to handle signed and unsigned shift amounts in the same branch) - rust-lang#117891 (Recover `dyn` and `impl` after `for<...>`) - rust-lang#117957 (if available use a Child's pidfd for kill/wait) - rust-lang#117994 (Ignore but do not assume region obligations from unifying headers in negative coherence) - rust-lang#118068 (subtree update cg_gcc 2023/11/17) r? `@ghost` `@rustbot` modify labels: rollup

…mpiler-errors Rollup of 8 pull requests Successful merges: - rust-lang#117828 (Avoid iterating over hashmaps in astconv) - rust-lang#117832 (interpret: simplify handling of shifts by no longer trying to handle signed and unsigned shift amounts in the same branch) - rust-lang#117891 (Recover `dyn` and `impl` after `for<...>`) - rust-lang#117957 (if available use a Child's pidfd for kill/wait) - rust-lang#117988 (Handle attempts to have multiple `cfg`d tail expressions) - rust-lang#117994 (Ignore but do not assume region obligations from unifying headers in negative coherence) - rust-lang#118000 (Make regionck care about placeholders in outlives components) - rust-lang#118068 (subtree update cg_gcc 2023/11/17) r? `@ghost` `@rustbot` modify labels: rollup

…tthiaskrgr Rollup of 8 pull requests Successful merges: - rust-lang#117828 (Avoid iterating over hashmaps in astconv) - rust-lang#117832 (interpret: simplify handling of shifts by no longer trying to handle signed and unsigned shift amounts in the same branch) - rust-lang#117891 (Recover `dyn` and `impl` after `for<...>`) - rust-lang#117957 (if available use a Child's pidfd for kill/wait) - rust-lang#117988 (Handle attempts to have multiple `cfg`d tail expressions) - rust-lang#117994 (Ignore but do not assume region obligations from unifying headers in negative coherence) - rust-lang#118000 (Make regionck care about placeholders in outlives components) - rust-lang#118068 (subtree update cg_gcc 2023/11/17) r? `@ghost` `@rustbot` modify labels: rollup

Rollup merge of rust-lang#117957 - the8472:pidfd-wait, r=Mark-Simulacrum if available use a Child's pidfd for kill/wait This should get us closer to stabilization of pidfds since they now do something useful. And they're `CLOEXEC` now. ``` $ strace -ffe clone,sendmsg,recvmsg,execve,kill,pidfd_open,pidfd_send_signal,waitpid,waitid ./x test std --no-doc -- pidfd [...] running 1 tests strace: Process 816007 attached [pid 816007] pidfd_open(816006, 0) = 3 [pid 816007] clone(child_stack=NULL, flags=CLONE_CHILD_CLEARTID|CLONE_CHILD_SETTID|SIGCHLD, child_tidptr=0x7f0c6b787990) = 816008 strace: Process 816008 attached [pid 816007] recvmsg(3, <unfinished ...> [pid 816008] pidfd_open(816008, 0) = 3 [pid 816008] sendmsg(4, {msg_name=NULL, msg_namelen=0, msg_iov=[{iov_base="", iov_len=0}], msg_iovlen=1, msg_control=[{cmsg_len=20, cmsg_level=SOL_SOCKET, cmsg_type=SCM_RIGHTS, cmsg_data=[3]}], msg_controllen=24, msg_flags=0}, 0) = 0 [pid 816007] <... recvmsg resumed>{msg_name=NULL, msg_namelen=0, msg_iov=[{iov_base="", iov_len=0}], msg_iovlen=1, msg_control=[{cmsg_len=20, cmsg_level=SOL_SOCKET, cmsg_type=SCM_RIGHTS, cmsg_data=[4]}], msg_controllen=24, msg_flags=MSG_CMSG_CLOEXEC}, MSG_CMSG_CLOEXEC) = 0 [pid 816008] execve("/usr/bin/false", ["false"], 0x7ffcf2100048 /* 105 vars */) = 0 [pid 816007] waitid(P_PIDFD, 4, <unfinished ...> [pid 816008] +++ exited with 1 +++ [pid 816007] <... waitid resumed>{si_signo=SIGCHLD, si_code=CLD_EXITED, si_pid=816008, si_uid=1001, si_status=1, si_utime=0, si_stime=0}, WEXITED, NULL) = 0 [pid 816007] --- SIGCHLD {si_signo=SIGCHLD, si_code=CLD_EXITED, si_pid=816008, si_uid=1001, si_status=1, si_utime=0, si_stime=0} --- [pid 816007] clone(child_stack=NULL, flags=CLONE_CHILD_CLEARTID|CLONE_CHILD_SETTID|SIGCHLDstrace: Process 816009 attached , child_tidptr=0x7f0c6b787990) = 816009 [pid 816007] recvmsg(3, <unfinished ...> [pid 816009] pidfd_open(816009, 0) = 3 [pid 816009] sendmsg(5, {msg_name=NULL, msg_namelen=0, msg_iov=[{iov_base="", iov_len=0}], msg_iovlen=1, msg_control=[{cmsg_len=20, cmsg_level=SOL_SOCKET, cmsg_type=SCM_RIGHTS, cmsg_data=[3]}], msg_controllen=24, msg_flags=0}, 0) = 0 [pid 816007] <... recvmsg resumed>{msg_name=NULL, msg_namelen=0, msg_iov=[{iov_base="", iov_len=0}], msg_iovlen=1, msg_control=[{cmsg_len=20, cmsg_level=SOL_SOCKET, cmsg_type=SCM_RIGHTS, cmsg_data=[5]}], msg_controllen=24, msg_flags=MSG_CMSG_CLOEXEC}, MSG_CMSG_CLOEXEC) = 0 [pid 816009] execve("/usr/bin/sleep", ["sleep", "1000"], 0x7ffcf2100048 /* 105 vars */) = 0 [pid 816007] waitid(P_PIDFD, 5, {}, WNOHANG|WEXITED, NULL) = 0 [pid 816007] pidfd_send_signal(5, SIGKILL, NULL, 0) = 0 [pid 816007] waitid(P_PIDFD, 5, <unfinished ...> [pid 816009] +++ killed by SIGKILL +++ [pid 816007] <... waitid resumed>{si_signo=SIGCHLD, si_code=CLD_KILLED, si_pid=816009, si_uid=1001, si_status=SIGKILL, si_utime=0, si_stime=0}, WEXITED, NULL) = 0 [pid 816007] --- SIGCHLD {si_signo=SIGCHLD, si_code=CLD_KILLED, si_pid=816009, si_uid=1001, si_status=SIGKILL, si_utime=0, si_stime=0} --- [pid 816007] +++ exited with 0 +++ ```

the8472 added 3 commits November 16, 2023 01:36

set CLOEXEC on pidfd received from child process

10127d9

update comment, we're currently using a different syscall

3ffbb48

if available use a Child's pidfd for kill/wait

12efa53

the8472 added the O-linux Operating system: Linux label Nov 16, 2023

rustbot assigned Mark-Simulacrum Nov 16, 2023

rustbot added O-unix Operating system: Unix-like S-waiting-on-review Status: Awaiting review from the assignee but also interested parties. T-libs Relevant to the library team, which will review and decide on the PR/issue. labels Nov 16, 2023

the8472 mentioned this pull request Nov 16, 2023

Tracking Issue for linux_pidfd #82971

Open

6 tasks

bors added S-waiting-on-bors Status: Waiting on bors to run and complete tests. Bors will change the label on completion. and removed S-waiting-on-review Status: Awaiting review from the assignee but also interested parties. labels Nov 18, 2023

This comment has been minimized.

Sign in to view

bors added S-waiting-on-review Status: Awaiting review from the assignee but also interested parties. and removed S-waiting-on-bors Status: Waiting on bors to run and complete tests. Bors will change the label on completion. labels Nov 19, 2023

Don't set cmsg fields in msghdr if we have no cmsg to send

f34e7f4

bors added S-waiting-on-bors Status: Waiting on bors to run and complete tests. Bors will change the label on completion. and removed S-waiting-on-review Status: Awaiting review from the assignee but also interested parties. labels Nov 19, 2023

compiler-errors mentioned this pull request Nov 20, 2023

Rollup of 8 pull requests #118078

Closed

compiler-errors mentioned this pull request Nov 20, 2023

Rollup of 8 pull requests #118082

Merged

bors merged commit 6d33e90 into rust-lang:master Nov 20, 2023
11 checks passed

rustbot added this to the 1.76.0 milestone Nov 20, 2023

qarmin mentioned this pull request Nov 22, 2023

undefined reference to `waitid' sunfishcode/c-ward#75

Closed

the8472 mentioned this pull request Nov 23, 2023

no method named si_pid found for struct siginfo_t for armv7-unknown-linux-uclibceabihf #118190

Open

the8472 mentioned this pull request Apr 17, 2024

Add PidFd::{kill, wait, try_wait} #124101

Open

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

if available use a Child's pidfd for kill/wait #117957

if available use a Child's pidfd for kill/wait #117957

the8472 commented Nov 16, 2023

rustbot commented Nov 16, 2023

Mark-Simulacrum commented Nov 18, 2023

bors commented Nov 18, 2023

bors commented Nov 19, 2023

This comment has been minimized.

bors commented Nov 19, 2023

the8472 commented Nov 19, 2023 •

edited

Mark-Simulacrum commented Nov 19, 2023

bors commented Nov 19, 2023

if available use a Child's pidfd for kill/wait #117957

if available use a Child's pidfd for kill/wait #117957

Conversation

the8472 commented Nov 16, 2023

rustbot commented Nov 16, 2023

Mark-Simulacrum commented Nov 18, 2023

bors commented Nov 18, 2023

bors commented Nov 19, 2023

This comment has been minimized.

bors commented Nov 19, 2023

the8472 commented Nov 19, 2023 • edited

Mark-Simulacrum commented Nov 19, 2023

bors commented Nov 19, 2023

the8472 commented Nov 19, 2023 •

edited