From: Eric Wong <e@80x24.org>
To: meta@public-inbox.org
Subject: [PATCH 02/11] t/check-www-inbox: replace IPC::Run with PublicInbox::Spawn
Date: Wed, 30 Jan 2019 04:44:21 +0000 [thread overview]
Message-ID: <20190130044430.28189-3-e@80x24.org> (raw)
In-Reply-To: <20190130044430.28189-1-e@80x24.org>
Because WWW::Mechanize uses truckload of memory, fork
needs to prepare all that memory for CoW, which ends up
bailing with ENOMEM.
---
t/check-www-inbox.perl | 39 ++++++++++++++++++++++++++++++++++-----
1 file changed, 34 insertions(+), 5 deletions(-)
diff --git a/t/check-www-inbox.perl b/t/check-www-inbox.perl
index 1e88e95..933362a 100644
--- a/t/check-www-inbox.perl
+++ b/t/check-www-inbox.perl
@@ -14,6 +14,12 @@ use POSIX qw(:sys_wait_h);
use Time::HiRes qw(gettimeofday tv_interval);
use WWW::Mechanize;
use Data::Dumper;
+
+# we want to use vfork+exec with spawn, WWW::Mechanize can use too much
+# memory and fork(2) fails
+use PublicInbox::Spawn qw(spawn which);
+$ENV{PERL_INLINE_DIRECTORY} or warn "PERL_INLINE_DIRECTORY unset, may OOM\n";
+
our $tmp_owner = $$;
my $nproc = 4;
my $slow = 0.5;
@@ -24,13 +30,35 @@ my %opts = (
GetOptions(%opts) or die "bad command-line args\n$usage";
my $root_url = shift or die $usage;
-chomp(my $xmlstarlet = `which xmlstarlet 2>/dev/null`);
+chomp(my $xmlstarlet = which('xmlstarlet'));
my $atom_check = eval {
- require IPC::Run;
my $cmd = [ qw(xmlstarlet val -e -) ];
sub {
my ($in, $out, $err) = @_;
- IPC::Run::run($cmd, $in, $out, $err);
+ use autodie;
+ open my $in_fh, '+>', undef;
+ open my $out_fh, '+>', undef;
+ open my $err_fh, '+>', undef;
+ print $in_fh $$in;
+ $in_fh->flush;
+ sysseek($in_fh, 0, 0);
+ my $rdr = {
+ 0 => fileno($in_fh),
+ 1 => fileno($out_fh),
+ 2 => fileno($err_fh),
+ };
+ my $pid = spawn($cmd, undef, $rdr);
+ defined $pid or die "spawn failure: $!";
+ while (waitpid($pid, 0) != $pid) {
+ next if $!{EINTR};
+ warn "waitpid(xmlstarlet, $pid) $!";
+ return $!;
+ }
+ sysseek($out_fh, 0, 0);
+ sysread($out_fh, $$out, -s $out_fh);
+ sysseek($err_fh, 0, 0);
+ sysread($err_fh, $$err, -s $err_fh);
+ $?
}
} if $xmlstarlet;
@@ -120,6 +148,7 @@ while (keys %workers) { # reacts to SIGCHLD
sub worker_loop {
my ($todo_rd, $done_wr) = @_;
+ $SIG{CHLD} = 'DEFAULT';
my $m = WWW::Mechanize->new(autocheck => 0);
my $cc = LWP::ConnCache->new;
$m->conn_cache($cc);
@@ -164,8 +193,8 @@ sub worker_loop {
if ($atom_check && $ct =~ m!\bapplication/atom\+xml\b!) {
my $raw = $r->decoded_content;
my ($out, $err) = ('', '');
- $atom_check->(\$raw, \$out, \$err) and
- warn "Atom ($?) - $u - <1:$out> <2:$err>\n";
+ my $fail = $atom_check->(\$raw, \$out, \$err);
+ warn "Atom ($fail) - $u - <1:$out> <2:$err>\n" if $fail;
}
next if $ct !~ m!\btext/html\b!;
--
EW
next prev parent reply other threads:[~2019-01-30 4:44 UTC|newest]
Thread overview: 12+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-01-30 4:44 [PATCH 00/11] viewvcs: more fixes Eric Wong
2019-01-30 4:44 ` [PATCH 01/11] hval: add src_escape for highlight post-processing Eric Wong
2019-01-30 4:44 ` Eric Wong [this message]
2019-01-30 4:44 ` [PATCH 03/11] t/check-www-inbox: don't follow mboxes Eric Wong
2019-01-30 4:44 ` [PATCH 04/11] t/check-www-inbox: disable history Eric Wong
2019-01-30 4:44 ` [PATCH 05/11] solvergit: do not solve blobs twice Eric Wong
2019-01-30 4:44 ` [PATCH 06/11] viewvcs: avoid segfault with highlight.pm at shutdown Eric Wong
2019-01-30 4:44 ` [PATCH 07/11] css/216dark: add comments and tweak highlight colors Eric Wong
2019-01-30 4:44 ` [PATCH 08/11] solvergit: do not show full path names to "git apply" Eric Wong
2019-01-30 4:44 ` [PATCH 09/11] solvergit: avoid "Wide character" warnings Eric Wong
2019-01-30 4:44 ` [PATCH 10/11] solvergit: extract mode from diff headers properly Eric Wong
2019-01-30 4:44 ` [PATCH 11/11] solvergit: deal with alternative diff prefixes Eric Wong
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
List information: https://public-inbox.org/README
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20190130044430.28189-3-e@80x24.org \
--to=e@80x24.org \
--cc=meta@public-inbox.org \
/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 read-only IMAP folder(s) and NNTP newsgroup(s).