public inbox for buildbot@sourceware.org
 help / color / mirror / Atom feed
* [PATCH 1/2] gcc_full_build_factory: pass '-k' to make check to maximise results
@ 2023-01-31  1:56 Sam James
  2023-01-31  1:56 ` [PATCH 2/2] gcc_full_build_factory: call contrib/test_summary after make check Sam James
  2023-01-31  9:27 ` [PATCH 1/2] gcc_full_build_factory: pass '-k' to make check to maximise results Mark Wielaard
  0 siblings, 2 replies; 5+ messages in thread
From: Sam James @ 2023-01-31  1:56 UTC (permalink / raw)
  To: buildbot; +Cc: Sam James

We want to keep going even if we hit failures for a maximally useful
log and set of results.
---
 builder/master.cfg | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/builder/master.cfg b/builder/master.cfg
index 84d457e..71efbe9 100644
--- a/builder/master.cfg
+++ b/builder/master.cfg
@@ -3254,7 +3254,7 @@ def gcc_full_build_factory_gen(extra_configure_arg=None):
         # Even if that means bunsen gets some parallel/duplicate log files
         gcc_full_build_factory.addStep(steps.Test(
                 workdir='gcc-build',
-                command=addOutputSync.withArgs(['make', 'check',
+                command=addOutputSync.withArgs(['make', 'check', '-k',
                     util.Interpolate('-j%(prop:maxcpus)s')]),
                 name='make check',
                 timeout=3600,
-- 
2.39.1


^ permalink raw reply	[flat|nested] 5+ messages in thread

* [PATCH 2/2] gcc_full_build_factory: call contrib/test_summary after make check
  2023-01-31  1:56 [PATCH 1/2] gcc_full_build_factory: pass '-k' to make check to maximise results Sam James
@ 2023-01-31  1:56 ` Sam James
  2023-01-31  9:31   ` Mark Wielaard
  2023-01-31  9:27 ` [PATCH 1/2] gcc_full_build_factory: pass '-k' to make check to maximise results Mark Wielaard
  1 sibling, 1 reply; 5+ messages in thread
From: Sam James @ 2023-01-31  1:56 UTC (permalink / raw)
  To: buildbot; +Cc: Sam James

This gives us a nice summary of the various dejagnu results.
---
 builder/master.cfg | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/builder/master.cfg b/builder/master.cfg
index 71efbe9..562e794 100644
--- a/builder/master.cfg
+++ b/builder/master.cfg
@@ -3259,6 +3259,13 @@ def gcc_full_build_factory_gen(extra_configure_arg=None):
                 name='make check',
                 timeout=3600,
                 haltOnFailure=False, flunkOnFailure=True))
+        # Get a useful summary (repeats the various dejagnu summaries) at
+        # the end. From https://gcc.gnu.org/install/test.html.
+        gcc_full_build_factory.addStep(steps.Test(
+                workdir='gcc-build',
+                command=['../gcc/contrib/test_summary'],
+                name='contrib/test_summary',
+                haltOnFailure=False, flunkOnFailure=True))
         gcc_full_build_factory.addSteps(bunsen_logfile_upload_cpio_steps(
                 ["*.log", "*.sum"],
                 workdir='gcc-build'))
-- 
2.39.1


^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: [PATCH 1/2] gcc_full_build_factory: pass '-k' to make check to maximise results
  2023-01-31  1:56 [PATCH 1/2] gcc_full_build_factory: pass '-k' to make check to maximise results Sam James
  2023-01-31  1:56 ` [PATCH 2/2] gcc_full_build_factory: call contrib/test_summary after make check Sam James
@ 2023-01-31  9:27 ` Mark Wielaard
  1 sibling, 0 replies; 5+ messages in thread
From: Mark Wielaard @ 2023-01-31  9:27 UTC (permalink / raw)
  To: Sam James; +Cc: buildbot

Hi Sam,

On Tue, Jan 31, 2023 at 01:56:10AM +0000, Sam James wrote:
> We want to keep going even if we hit failures for a maximally useful
> log and set of results.

Yes, that makes sense for the full build.

Pushed,

Mark

^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: [PATCH 2/2] gcc_full_build_factory: call contrib/test_summary after make check
  2023-01-31  1:56 ` [PATCH 2/2] gcc_full_build_factory: call contrib/test_summary after make check Sam James
@ 2023-01-31  9:31   ` Mark Wielaard
  2023-01-31 17:34     ` Sam James
  0 siblings, 1 reply; 5+ messages in thread
From: Mark Wielaard @ 2023-01-31  9:31 UTC (permalink / raw)
  To: Sam James; +Cc: buildbot

On Tue, Jan 31, 2023 at 01:56:11AM +0000, Sam James wrote:
> This gives us a nice summary of the various dejagnu results.

Looks like a nice idea in general.

But it also renames the *.sum files to .sum.sent, which means
they next bunsen step will fail to fetch them. You need to add
-t: prevents logs from being renamed

Also the "wrapper" text about emailing the sums is a little
odd imho. But it looks like there is no way to suppress that.

Cheers,

Mark

> ---
>  builder/master.cfg | 7 +++++++
>  1 file changed, 7 insertions(+)
> 
> diff --git a/builder/master.cfg b/builder/master.cfg
> index 71efbe9..562e794 100644
> --- a/builder/master.cfg
> +++ b/builder/master.cfg
> @@ -3259,6 +3259,13 @@ def gcc_full_build_factory_gen(extra_configure_arg=None):
>                  name='make check',
>                  timeout=3600,
>                  haltOnFailure=False, flunkOnFailure=True))
> +        # Get a useful summary (repeats the various dejagnu summaries) at
> +        # the end. From https://gcc.gnu.org/install/test.html.
> +        gcc_full_build_factory.addStep(steps.Test(
> +                workdir='gcc-build',
> +                command=['../gcc/contrib/test_summary'],
> +                name='contrib/test_summary',
> +                haltOnFailure=False, flunkOnFailure=True))
>          gcc_full_build_factory.addSteps(bunsen_logfile_upload_cpio_steps(
>                  ["*.log", "*.sum"],
>                  workdir='gcc-build'))
> -- 
> 2.39.1
> 

^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: [PATCH 2/2] gcc_full_build_factory: call contrib/test_summary after make check
  2023-01-31  9:31   ` Mark Wielaard
@ 2023-01-31 17:34     ` Sam James
  0 siblings, 0 replies; 5+ messages in thread
From: Sam James @ 2023-01-31 17:34 UTC (permalink / raw)
  To: Mark Wielaard; +Cc: buildbot

[-- Attachment #1: Type: text/plain, Size: 707 bytes --]



> On 31 Jan 2023, at 09:31, Mark Wielaard <mark@klomp.org> wrote:
> 
> On Tue, Jan 31, 2023 at 01:56:11AM +0000, Sam James wrote:
>> This gives us a nice summary of the various dejagnu results.
> 
> Looks like a nice idea in general.
> 
> But it also renames the *.sum files to .sum.sent, which means
> they next bunsen step will fail to fetch them. You need to add
> -t: prevents logs from being renamed
> 

Oh, good spot! Let me fix.

> Also the "wrapper" text about emailing the sums is a little
> odd imho. But it looks like there is no way to suppress that.
> 
I want to add an option for that as it's a bit confusing when we run
It in our Gentoo packaging too. We'll circle back to that.

Best,
sam

[-- Attachment #2: Message signed with OpenPGP --]
[-- Type: application/pgp-signature, Size: 358 bytes --]

^ permalink raw reply	[flat|nested] 5+ messages in thread

end of thread, other threads:[~2023-01-31 17:45 UTC | newest]

Thread overview: 5+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2023-01-31  1:56 [PATCH 1/2] gcc_full_build_factory: pass '-k' to make check to maximise results Sam James
2023-01-31  1:56 ` [PATCH 2/2] gcc_full_build_factory: call contrib/test_summary after make check Sam James
2023-01-31  9:31   ` Mark Wielaard
2023-01-31 17:34     ` Sam James
2023-01-31  9:27 ` [PATCH 1/2] gcc_full_build_factory: pass '-k' to make check to maximise results Mark Wielaard

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).