From: Jacob Moody <moody@posixcafe.org>
To: 9fans@9fans.net
Subject: Re: [9fans] syscall silently kill processes
Date: Sun, 19 Jun 2022 00:13:50 -0600 [thread overview]
Message-ID: <48e83724-8ea9-79b2-536a-b6d3c952c152@posixcafe.org> (raw)
In-Reply-To: <f75f7af1-7d33-5422-17b-3cccc92ce20@SDF.ORG>
On 6/18/22 23:54, adr wrote:
> On Sat, 18 Jun 2022, Jacob Moody wrote:
>> I've attempted to reproduce it, trying to remove the libthread/notify
>> factors. I've come up with this:
>>
>> #include <u.h>
>> #include <libc.h>
>>
>> static void
>> proc_udp(void*)
>> {
>> char resp[512];
>> char req[] = "request";
>> int fd;
>> int n;
>> int pid;
>>
>> fd = dial("udp!185.157.221.201!5678", nil, nil, nil);
>> if(fd < 0)
>> exits("can't dial");
>>
>> if(write(fd, req, strlen(req)) != strlen(req))
>> exits("can't write");
>>
>> pid = getpid();
>> fprint(1, "start %d\n", pid);
>> n = read(fd, resp, sizeof(resp)-1);
>> fprint(1, "end %d %d\n", pid, n);
>> exits(nil);
>> }
>>
>> void
>> main(int, char**)
>> {
>> int i;
>> Waitmsg *wm;
>>
>> for(i = 0; i < 10; i++){
>> switch(fork()){
>> case -1:
>> sysfatal("fork %r");
>> case 0:
>> proc_udp(nil);
>> sysfatal("ret");
>> default:
>> break;
>> }
>> }
>> for(i = 0; i < 10; i++){
>> wm = wait();
>> print("proc %d died with message %s\n", wm->pid, wm->msg);
>> }
>> exits(nil);
>> }
>>
>> This code makes it pretty obvious that we are losing some children;
>> on my machine this program never exits. I see some portion of the
>> readers correctly returning -1, and the parent is able to get their
>> Waitmsg but not all of them.
>
> Moody I think this old thread will interest you:
>
> https://marc.info/?t=112730920400001&r=1&w=2
>
> Russ Cox explained there:
> It appears that your program, at its core, it is doing this:
>
> void
> readproc(void *v)
> {
> int fd;
> char buf[100];
> fd = (int)v;
> read(fd, buf, sizeof buf);
> }
>
> void
> threadmain(int argc, char **argv)
> {
> int p[2];
> pipe(p);
> proccreate(readproc, (void*)p[0], 8192);
> proccreate(readproc, (void*)p[1], 8192);
> close(p[0]);
> /* and here you expect the first readproc to be done */
> close(p[1]);
> /* and here the second */
> }
>
> Each read call is holding up a reference to its channel
> inside the kernel, so that even though you've closed the fd
> and removed the ref from the fd table, there is still a reference
> to each side of the pipe in the form of the process blocked
> on the read.
>
> I've never been sure whether the implicit ref held during
> the system call is good behavior, but it's hard to change.
>
> In your case, writing 0 (or anything) makes the read
> finish, releasing the last ref to the underlying pipe when
> the system call finishes, and then everything cleans up
> as expected. So you've found your workaround, and now
> we understand why it works.
>
I was just making the wrong observation here.
I thought I had observed the child procs getting
murdered mid read, and the parent never getting
the Waitmsg. Testing again I see as Andrej had observed,
they are just blocking. I thought I was seeing a bug
related to just udp, nothing to do with notes/threads.
I apologize for the confusion, interesting thread
you linked regardless.
> ------------------------------------------
> 9fans: 9fans
> Permalink: https://9fans.topicbox.com/groups/9fans/Tfa6823048ad90a21-M6e48031f9e8673387c0b47b8
> Delivery options: https://9fans.topicbox.com/groups/9fans/subscription
------------------------------------------
9fans: 9fans
Permalink: https://9fans.topicbox.com/groups/9fans/Tfa6823048ad90a21-Md81beb48e514ad6a776fa41d
Delivery options: https://9fans.topicbox.com/groups/9fans/subscription
next prev parent reply other threads:[~2022-06-19 6:14 UTC|newest]
Thread overview: 50+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-06-17 9:37 andrey100100100
2022-06-17 13:46 ` Thaddeus Woskowiak
2022-06-17 14:11 ` Jacob Moody
2022-06-17 14:39 ` Thaddeus Woskowiak
2022-06-17 15:06 ` andrey100100100
2022-06-17 16:08 ` Skip Tavakkolian
2022-06-17 16:11 ` Skip Tavakkolian
2022-06-17 16:16 ` Skip Tavakkolian
2022-06-17 17:42 ` adr
2022-06-17 16:11 ` Jacob Moody
2022-06-17 18:48 ` andrey100100100
2022-06-17 19:28 ` Jacob Moody
2022-06-17 21:15 ` adr
2022-06-18 6:40 ` andrey100100100
2022-06-18 8:37 ` adr
2022-06-18 9:22 ` adr
2022-06-18 12:53 ` Jacob Moody
2022-06-18 22:03 ` andrey100100100
2022-06-19 5:54 ` adr
2022-06-19 6:13 ` Jacob Moody [this message]
2022-06-18 22:22 ` andrey100100100
2022-06-18 16:57 ` andrey100100100
2022-06-19 2:40 ` adr
2022-06-19 5:01 ` adr
2022-06-19 8:52 ` andrey100100100
2022-06-19 10:32 ` adr
2022-06-19 11:40 ` andrey100100100
2022-06-19 12:01 ` andrey100100100
2022-06-19 15:10 ` andrey100100100
2022-06-19 16:41 ` adr
2022-06-19 21:22 ` andrey100100100
2022-06-19 21:26 ` andrey100100100
2022-06-20 4:41 ` adr
2022-06-20 5:39 ` andrey100100100
2022-06-20 5:59 ` adr
2022-06-20 15:56 ` andrey100100100
2022-06-20 22:29 ` Skip Tavakkolian
2022-06-21 7:07 ` andrey100100100
2022-06-21 11:26 ` adr
2022-06-21 13:03 ` andrey100100100
2022-06-21 13:22 ` adr
2022-06-28 15:28 ` adr
2022-06-28 16:43 ` ori
2022-06-28 18:19 ` adr
2022-06-28 18:28 ` adr
2022-06-28 19:09 ` andrey100100100
2022-06-28 19:42 ` adr
2022-06-29 13:14 ` adr
2022-06-21 13:47 ` andrey100100100
2022-06-21 7:22 ` adr
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=48e83724-8ea9-79b2-536a-b6d3c952c152@posixcafe.org \
--to=moody@posixcafe.org \
--cc=9fans@9fans.net \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).