* [PATCH] Fix `testrun.log' output of manual `testrun's
@ 2007-01-13 9:08 Jan Kratochvil
2007-01-13 13:47 ` Mark Kettenis
2007-01-13 16:01 ` Daniel Jacobowitz
0 siblings, 2 replies; 7+ messages in thread
From: Jan Kratochvil @ 2007-01-13 9:08 UTC (permalink / raw)
To: gdb-patches
[-- Attachment #1: Type: text/plain, Size: 228 bytes --]
Hi,
if you run by hand:
cd gdb/testsuite; runtest gdb.base/complex.exp
(any test using get_compiler_info())
"testrun.log" will get empty (except its header).
This problem does not occur during `make check'.
Regards,
Jan
[-- Attachment #2: gdb-6.5-testsuite-log.patch --]
[-- Type: text/plain, Size: 787 bytes --]
2007-01-13 Jan Kratochvil <jan.kratochvil@redhat.com>
* lib/gdb.exp (get_compiler_info): Cope with nondefault output logs.
diff -u -rupN gdb-6.5-ppc/gdb/testsuite/lib/gdb.exp gdb-6.5/gdb/testsuite/lib/gdb.exp
--- gdb-6.5-ppc/gdb/testsuite/lib/gdb.exp 2007-01-12 14:33:26.000000000 -0500
+++ gdb-6.5/gdb/testsuite/lib/gdb.exp 2007-01-12 21:15:02.000000000 -0500
@@ -1351,9 +1351,10 @@ proc get_compiler_info {binfile args} {
# Run $ifile through the right preprocessor.
# Toggle gdb.log to keep the compiler output out of the log.
+ set log_old [log_file -info]
log_file
set cppout [ gdb_compile "${ifile}" "" preprocess [list "$args" quiet] ]
- log_file -a "$outdir/$tool.log"
+ eval "log_file $log_old"
# Eval the output.
set unknown 0
^ permalink raw reply [flat|nested] 7+ messages in thread* Re: [PATCH] Fix `testrun.log' output of manual `testrun's
2007-01-13 9:08 [PATCH] Fix `testrun.log' output of manual `testrun's Jan Kratochvil
@ 2007-01-13 13:47 ` Mark Kettenis
2007-01-13 16:01 ` Daniel Jacobowitz
1 sibling, 0 replies; 7+ messages in thread
From: Mark Kettenis @ 2007-01-13 13:47 UTC (permalink / raw)
To: jan.kratochvil; +Cc: gdb-patches
> Date: Sat, 13 Jan 2007 10:08:37 +0100
> From: Jan Kratochvil <jan.kratochvil@redhat.com>
>
> Hi,
>
> if you run by hand:
> cd gdb/testsuite; runtest gdb.base/complex.exp
> (any test using get_compiler_info())
>
> "testrun.log" will get empty (except its header).
>
> This problem does not occur during `make check'.
I always do
$ make RUNTESTFLAGS=gdb.base/complex.exp check
That works in both the gdb directory and the gdb/testsuite directory
and puts the output in gdb.log and gdb.sum.
Cheers,
Mark
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [PATCH] Fix `testrun.log' output of manual `testrun's
2007-01-13 9:08 [PATCH] Fix `testrun.log' output of manual `testrun's Jan Kratochvil
2007-01-13 13:47 ` Mark Kettenis
@ 2007-01-13 16:01 ` Daniel Jacobowitz
2007-01-14 15:34 ` Jan Kratochvil
1 sibling, 1 reply; 7+ messages in thread
From: Daniel Jacobowitz @ 2007-01-13 16:01 UTC (permalink / raw)
To: Jan Kratochvil; +Cc: gdb-patches
On Sat, Jan 13, 2007 at 10:08:37AM +0100, Jan Kratochvil wrote:
> Hi,
>
> if you run by hand:
> cd gdb/testsuite; runtest gdb.base/complex.exp
> (any test using get_compiler_info())
>
> "testrun.log" will get empty (except its header).
>
> This problem does not occur during `make check'.
Does site.exp exist before this happens? It has never happened to me;
site.exp sets tool to gdb.
I get a testrun.log if I run runtest before make site.exp, but then
nothing works.
--
Daniel Jacobowitz
CodeSourcery
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [PATCH] Fix `testrun.log' output of manual `testrun's
2007-01-13 16:01 ` Daniel Jacobowitz
@ 2007-01-14 15:34 ` Jan Kratochvil
2007-01-20 19:59 ` Daniel Jacobowitz
0 siblings, 1 reply; 7+ messages in thread
From: Jan Kratochvil @ 2007-01-14 15:34 UTC (permalink / raw)
To: gdb-patches; +Cc: Daniel Jacobowitz
[-- Attachment #1: Type: text/plain, Size: 633 bytes --]
On Sat, 13 Jan 2007 17:00:55 +0100, Daniel Jacobowitz wrote:
...
> Does site.exp exist before this happens? It has never happened to me;
> site.exp sets tool to gdb.
It does not exists as gdb has been built in a non-srcdir directory.
> I get a testrun.log if I run runtest before make site.exp, but then
> nothing works.
You are right. While I believe the former patch makes the testsuite more
consistent I now see it makes no practical difference.
Rather proposing a new patch as only the warning message
WARNING: No tool specified
was proven as not stopping the testsuite users in such local configurations.
Regards,
Jan
[-- Attachment #2: gdb-site-require.patch --]
[-- Type: text/plain, Size: 670 bytes --]
2007-01-14 Jan Kratochvil <jan.kratochvil@redhat.com>
* lib/gdb.exp: Abort on missing `site.exp'.
diff -u -rup gdb-6.5-orig/gdb/testsuite/lib/gdb.exp gdb-6.5/gdb/testsuite/lib/gdb.exp
--- gdb-6.5-orig/gdb/testsuite/lib/gdb.exp 2006-12-17 11:59:42.000000000 +0100
+++ gdb-6.5/gdb/testsuite/lib/gdb.exp 2007-01-14 16:27:03.000000000 +0100
@@ -22,6 +22,12 @@
# need to be modified for any target, it can be done with a variable
# or by passing arguments.
+if {$tool == ""} {
+ # Tests would fail, logs on get_compiler_info() would be missing.
+ send_error "`site.exp' not found, run `make site.exp'!\n"
+ exit 2
+}
+
load_lib libgloss.exp
global GDB
^ permalink raw reply [flat|nested] 7+ messages in thread* Re: [PATCH] Fix `testrun.log' output of manual `testrun's
2007-01-14 15:34 ` Jan Kratochvil
@ 2007-01-20 19:59 ` Daniel Jacobowitz
2007-01-21 22:28 ` Jan Kratochvil
0 siblings, 1 reply; 7+ messages in thread
From: Daniel Jacobowitz @ 2007-01-20 19:59 UTC (permalink / raw)
To: Jan Kratochvil; +Cc: gdb-patches
On Sun, Jan 14, 2007 at 04:34:38PM +0100, Jan Kratochvil wrote:
> Rather proposing a new patch as only the warning message
> WARNING: No tool specified
> was proven as not stopping the testsuite users in such local configurations.
Yeah, I've made this mistake several times.
> 2007-01-14 Jan Kratochvil <jan.kratochvil@redhat.com>
>
> * lib/gdb.exp: Abort on missing `site.exp'.
OK. You have write access now, so feel free to commit this.
--
Daniel Jacobowitz
CodeSourcery
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [PATCH] Fix `testrun.log' output of manual `testrun's
2007-01-20 19:59 ` Daniel Jacobowitz
@ 2007-01-21 22:28 ` Jan Kratochvil
2007-01-21 22:51 ` Daniel Jacobowitz
0 siblings, 1 reply; 7+ messages in thread
From: Jan Kratochvil @ 2007-01-21 22:28 UTC (permalink / raw)
To: gdb-patches; +Cc: Daniel Jacobowitz
On Sat, 20 Jan 2007 20:59:21 +0100, Daniel Jacobowitz wrote:
> On Sun, Jan 14, 2007 at 04:34:38PM +0100, Jan Kratochvil wrote:
...
> > 2007-01-14 Jan Kratochvil <jan.kratochvil@redhat.com>
> >
> > * lib/gdb.exp: Abort on missing `site.exp'.
>
> OK. You have write access now, so feel free to commit this.
Checked in.
I assume one does not need to post here about an unchanged approved commit.
Regards,
Jan
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [PATCH] Fix `testrun.log' output of manual `testrun's
2007-01-21 22:28 ` Jan Kratochvil
@ 2007-01-21 22:51 ` Daniel Jacobowitz
0 siblings, 0 replies; 7+ messages in thread
From: Daniel Jacobowitz @ 2007-01-21 22:51 UTC (permalink / raw)
To: Jan Kratochvil; +Cc: gdb-patches
On Sun, Jan 21, 2007 at 11:28:18PM +0100, Jan Kratochvil wrote:
> Checked in.
>
> I assume one does not need to post here about an unchanged approved commit.
Thanks.
It's not a big deal to me either way. I don't know what other
maintainers think; I need to clean up the associated discussion from my
inbox, but it's not hard to do that from the gdb-cvs messages.
--
Daniel Jacobowitz
CodeSourcery
^ permalink raw reply [flat|nested] 7+ messages in thread
end of thread, other threads:[~2007-01-21 22:51 UTC | newest]
Thread overview: 7+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2007-01-13 9:08 [PATCH] Fix `testrun.log' output of manual `testrun's Jan Kratochvil
2007-01-13 13:47 ` Mark Kettenis
2007-01-13 16:01 ` Daniel Jacobowitz
2007-01-14 15:34 ` Jan Kratochvil
2007-01-20 19:59 ` Daniel Jacobowitz
2007-01-21 22:28 ` Jan Kratochvil
2007-01-21 22:51 ` Daniel Jacobowitz
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox