GNU bug report logs - #10038
[PATCH] tests: avoid rare fp failure in du/move-dir-while-traversing

Previous Next

Package: coreutils;

Reported by: Nix <nix <at> esperi.org.uk>

Date: Sun, 13 Nov 2011 18:16:01 UTC

Severity: normal

Tags: patch

Done: Jim Meyering <jim <at> meyering.net>

Bug is archived. No further changes may be made.

To add a comment to this bug, you must first unarchive it, by sending
a message to control AT debbugs.gnu.org, with unarchive 10038 in the body.
You can then email your comments to 10038 AT debbugs.gnu.org in the normal way.

Toggle the display of automated, internal messages from the tracker.

View this report as an mbox folder, status mbox, maintainer mbox


Report forwarded to bug-coreutils <at> gnu.org:
bug#10038; Package coreutils. (Sun, 13 Nov 2011 18:16:01 GMT) Full text and rfc822 format available.

Acknowledgement sent to Nix <nix <at> esperi.org.uk>:
New bug report received and forwarded. Copy sent to bug-coreutils <at> gnu.org. (Sun, 13 Nov 2011 18:16:01 GMT) Full text and rfc822 format available.

Message #5 received at submit <at> debbugs.gnu.org (full text, mbox):

From: Nix <nix <at> esperi.org.uk>
To: bug-coreutils <at> gnu.org
Subject: [PATCH] tests: avoid rare fp failure in du/move-dir-while-traversing
Date: Sun, 13 Nov 2011 18:15:10 +0000
du/move-dir-while-traversing: boost the mkdir iteration count yet again.
---
 tests/du/move-dir-while-traversing |    4 ++--
 1 files changed, 2 insertions(+), 2 deletions(-)

I just had an FP in du/move-dir-while-traversing, due to the traversal
terminating too soon. My autobuilder is running on a 2.2GHz Nehalem,
with the source tree located on a loopback ext4 filesystem on a tmpfs,
so particularly rapid test execution is understandable.

Patch obvious (I ran 200 trials of this just to avoid having to change
the comment!):

diff --git a/tests/du/move-dir-while-traversing b/tests/du/move-dir-while-traversing
index 322a9a5..d0969fe 100755
--- a/tests/du/move-dir-while-traversing
+++ b/tests/du/move-dir-while-traversing
@@ -66,8 +66,8 @@ long=d/e/f/g/h/i/j/k/l/m/n/o/p/q/r/s/t/u/v/w/x/y/z
 # Five iterations was not enough in 2 of 7 "make -j20 check" runs on a
 # 6/12-core system.  However, using "10", I saw no failure in 20 trials.
 # Using 10 iterations was not enough, either.
-# Using 30, I saw no failure in 200 trials.
-for i in $(seq 30); do
+# Using 50, I saw no failure in 200 trials.
+for i in $(seq 50); do
   mkdir -p $t/3/a/b/c/$i/$long || framework_failure_
 done
 
-- 
1.7.7.2.144.g3624f




Reply sent to Jim Meyering <jim <at> meyering.net>:
You have taken responsibility. (Sun, 13 Nov 2011 19:07:01 GMT) Full text and rfc822 format available.

Notification sent to Nix <nix <at> esperi.org.uk>:
bug acknowledged by developer. (Sun, 13 Nov 2011 19:07:01 GMT) Full text and rfc822 format available.

Message #10 received at 10038-done <at> debbugs.gnu.org (full text, mbox):

From: Jim Meyering <jim <at> meyering.net>
To: Nix <nix <at> esperi.org.uk>
Cc: 10038-done <at> debbugs.gnu.org
Subject: Re: bug#10038: [PATCH] tests: avoid rare fp failure in
	du/move-dir-while-traversing
Date: Sun, 13 Nov 2011 20:06:06 +0100
Nix wrote:
> du/move-dir-while-traversing: boost the mkdir iteration count yet again.
> ---
>  tests/du/move-dir-while-traversing |    4 ++--
>  1 files changed, 2 insertions(+), 2 deletions(-)
>
> I just had an FP in du/move-dir-while-traversing, due to the traversal
> terminating too soon. My autobuilder is running on a 2.2GHz Nehalem,
> with the source tree located on a loopback ext4 filesystem on a tmpfs,
> so particularly rapid test execution is understandable.
>
> Patch obvious (I ran 200 trials of this just to avoid having to change
> the comment!):
>
> diff --git a/tests/du/move-dir-while-traversing b/tests/du/move-dir-while-traversing
> index 322a9a5..d0969fe 100755
> --- a/tests/du/move-dir-while-traversing
> +++ b/tests/du/move-dir-while-traversing
> @@ -66,8 +66,8 @@ long=d/e/f/g/h/i/j/k/l/m/n/o/p/q/r/s/t/u/v/w/x/y/z
>  # Five iterations was not enough in 2 of 7 "make -j20 check" runs on a
>  # 6/12-core system.  However, using "10", I saw no failure in 20 trials.
>  # Using 10 iterations was not enough, either.
> -# Using 30, I saw no failure in 200 trials.
> -for i in $(seq 30); do
> +# Using 50, I saw no failure in 200 trials.
> +for i in $(seq 50); do
>    mkdir -p $t/3/a/b/c/$i/$long || framework_failure_
>  done

Thanks.
I've applied that and pushed it along with a tiny admin patch.
(also tweaked your log message to conform)

From 47052ddbfa04b16d649ae0ecbaacfd3ce1f03c53 Mon Sep 17 00:00:00 2001
From: Jim Meyering <meyering <at> redhat.com>
Date: Sun, 13 Nov 2011 20:00:08 +0100
Subject: [PATCH 1/2] maint: adjust git hook to allow "copy: ..." on line 1 of
 commit message

* scripts/git-hooks/commit-msg: Add "copy" to the list of valid
command-name-like summary line prefixes.
---
 scripts/git-hooks/commit-msg |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)

diff --git a/scripts/git-hooks/commit-msg b/scripts/git-hooks/commit-msg
index 96d5947..76b829d 100755
--- a/scripts/git-hooks/commit-msg
+++ b/scripts/git-hooks/commit-msg
@@ -42,7 +42,7 @@ check_msg()
       test|timeout|touch|tr|true|truncate|tsort|tty|uname|unexpand) ;;
       uniq|unlink|uptime|users|vdir|wc|who|whoami|yes) ;;
       # other tags
-      gnulib|tests|maint|doc|build|scripts) ;;
+      copy|gnulib|tests|maint|doc|build|scripts) ;;
       *) echo "invalid first word of summary line: $w"; fail=1;;
     esac
   done
--
1.7.8.rc2


From b8b6dd27a61dce58d6537dfbee65c7428f4ae3cd Mon Sep 17 00:00:00 2001
From: Nix <nix <at> esperi.org.uk>
Date: Sun, 13 Nov 2011 18:15:10 +0000
Subject: [PATCH 2/2] tests: avoid rare fp failure in
 du/move-dir-while-traversing

* tests/du/move-dir-while-traversing: Boost the mkdir iteration count
yet again.
---
 tests/du/move-dir-while-traversing |    4 ++--
 1 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/tests/du/move-dir-while-traversing b/tests/du/move-dir-while-traversing
index 322a9a5..d0969fe 100755
--- a/tests/du/move-dir-while-traversing
+++ b/tests/du/move-dir-while-traversing
@@ -66,8 +66,8 @@ long=d/e/f/g/h/i/j/k/l/m/n/o/p/q/r/s/t/u/v/w/x/y/z
 # Five iterations was not enough in 2 of 7 "make -j20 check" runs on a
 # 6/12-core system.  However, using "10", I saw no failure in 20 trials.
 # Using 10 iterations was not enough, either.
-# Using 30, I saw no failure in 200 trials.
-for i in $(seq 30); do
+# Using 50, I saw no failure in 200 trials.
+for i in $(seq 50); do
   mkdir -p $t/3/a/b/c/$i/$long || framework_failure_
 done

--
1.7.8.rc2




bug archived. Request was from Debbugs Internal Request <help-debbugs <at> gnu.org> to internal_control <at> debbugs.gnu.org. (Mon, 12 Dec 2011 12:24:02 GMT) Full text and rfc822 format available.

This bug report was last modified 13 years and 193 days ago.

Previous Next


GNU bug tracking system
Copyright (C) 1999 Darren O. Benham, 1997,2003 nCipher Corporation Ltd, 1994-97 Ian Jackson.