This is the mail archive of the
binutils@sources.redhat.com
mailing list for the binutils project.
RE: Cygwin tester? Was: [rfa] Add bfd-in-memory io vector
- From: "Dave Korn" <dk at artimi dot com>
- To: "'Andrew Cagney'" <cagney at gnu dot org>,"'Alan Modra'" <amodra at bigpond dot net dot au>
- Cc: <binutils at sources dot redhat dot com>
- Date: Tue, 4 May 2004 18:38:01 +0100
- Subject: RE: Cygwin tester? Was: [rfa] Add bfd-in-memory io vector
> -----Original Message-----
> From: binutils-owner On Behalf Of Dave Korn
> Sent: 04 May 2004 10:59
> > I don't have access to cygwin so I'll need someones help
> > testing this.
> > Anyone?
> >
> > Andrew
>
>
> Hi Andrew,
>
> I see someone's already doing some mingw testing for you; I
> could run
> cygwin tests if you like. I'll have to do it as a background
> job but I
> should have some results later today if you still need them.
As promised. First, the relevant details:
dk@mace /usr/build/obj> runtest --version
WARNING: Couldn't find the global config file.
Expect version is 5.26
Tcl version is 8.4
Framework version is 1.4.4
dk@mace /usr/build/obj> uname -a
CYGWIN_NT-5.1 mace 1.5.7(0.109/3/2) 2004-01-30 19:32 i686 unknown unknown
Cygwin
Also, I'm configuring with the following options in both cases:
dk@mace /usr/build/obj-patched> ../src/configure -v
--prefix=/usr/build/install --disable-nls 2>&1 | tee conf.log
And I used the slightly revised version of the patch from
http://sources.redhat.com/ml/binutils/2004-05/msg00023.html which avoids a
potential bug in peicode.h, but I didn't bother to fix the spacing as Alan
requested in his reply :-O hope it doesn't affect the results!
Perhaps worth mentioning that I'm using a slightly old version of the
cygwin dll, but the two sets of results should at least be commensurate with
each other. First, CVS head:
------------------->begin testsuite results<-------------------
Test Run By dk on Tue May 4 15:17:37 2004
Native configuration is i686-pc-cygwin
=== binutils tests ===
Schedule of variations:
unix
XFAIL: simple objcopy of executable
=== binutils Summary ===
# of expected passes 68
# of expected failures 1
runtest completed at Tue May 4 15:17:52 2004
Test Run By dk on Tue May 4 15:17:55 2004
Native configuration is i686-pc-cygwin
=== gas tests ===
Schedule of variations:
unix
FAIL: i386 pcrel reloc
=== gas Summary ===
# of expected passes 47
# of unexpected failures 1
../as-new 20040504
runtest completed at Tue May 4 15:18:08 2004
Test Run By dk on Tue May 4 15:19:18 2004
Native configuration is i686-pc-cygwin
=== ld tests ===
Schedule of variations:
unix
FAIL: linking shared lib
FAIL: bootstrap
FAIL: bootstrap with strip
FAIL: bootstrap with --static
FAIL: bootstrap with --traditional-format
FAIL: bootstrap with --no-keep-memory
FAIL: bootstrap with --relax
FAIL: cdtest with -Ur
FAIL: NOCROSSREFS 2
FAIL: ld-scripts/provide-1
XFAIL: ld-scripts/provide-3
XFAIL: S-records
XFAIL: S-records with constructors
=== ld Summary ===
# of expected passes 22
# of unexpected failures 10
# of expected failures 3
/usr/build/obj/ld/ld-new 20040504
runtest completed at Tue May 4 15:19:47 2004
Test Run By dk on Tue May 4 15:20:56 2004
Native configuration is i686-pc-cygwin
=== winsup tests ===
setting trap for SIGTERM to terminated
setting trap for SIGINT to interrupted by user
setting trap for SIGQUIT to interrupted by user
setting trap for SIGSEGV to segmentation violation
dirlist is /artimi/swtools/windows/share/dejagnu/baseboards
pushing config for build, name is mace
dirlist is /artimi/swtools/windows/share/dejagnu/baseboards
pushing config for host, name is mace
Schedule of variations:
unix
FAIL: msgtest.c (execute)
FAIL: semtest.c (execute)
FAIL: shmtest.c (execute)
XFAIL: ltp/dup03.c (execute)
XFAIL: ltp/dup05.c (execute)
XFAIL: ltp/fcntl05.c (execute)
XFAIL: ltp/fcntl07B.c (execute)
XFAIL: ltp/lseek04.c (execute)
XFAIL: ltp/select03.c (execute)
XFAIL: ltp/setgroups01.c (execute)
XFAIL: ltp/setuid02.c (execute)
XFAIL: ltp/ulimit01.c (compile)
XFAIL: ltp/unlink06.c (execute)
XFAIL: ltp/unlink08.c (execute)
XFAIL: samples/sample-fail.c (execute)
XFAIL: samples/sample-miscompile.c (compile)
=== winsup Summary ===
# of expected passes 265
# of unexpected failures 3
# of expected failures 13
Binary file
/usr/build/obj/i686-pc-cygwin/winsup/testsuite/../cygwin/cygwin0.dll matches
CYGWIN=ntsec smbntsec notty
runtest completed at Tue May 4 15:34:07 2004
------------------->end testsuite results<-------------------
Which you'll notice agrees exactly with the results posted earlier today
at http://sources.redhat.com/ml/binutils/2004-05/msg00070.html despite not
even using quite the same cygwin version.
And now CVS HEAD plus the bfd_in_memory patch: (verified I correctly
applied it by running cvs diff -pu and then diffing the results from that
with the diffs.txt attachment from the original post: only lines with
datestamps were changed, and the ChangeLog entry wasn't there of course)
------------------->begin testsuite results<-------------------
Test Run By dk on Tue May 4 17:18:56 2004
Native configuration is i686-pc-cygwin
=== binutils tests ===
Schedule of variations:
unix
Running target unix
XFAIL: simple objcopy of executable
=== binutils Summary ===
# of expected passes 68
# of expected failures 1
runtest completed at Tue May 4 17:19:10 2004
Test Run By dk on Tue May 4 17:19:13 2004
Native configuration is i686-pc-cygwin
=== gas tests ===
Schedule of variations:
unix
Running target unix
FAIL: i386 pcrel reloc
=== gas Summary ===
# of expected passes 47
# of unexpected failures 1
../as-new 20040504
runtest completed at Tue May 4 17:19:24 2004
Test Run By dk on Tue May 4 17:19:56 2004
Native configuration is i686-pc-cygwin
=== ld tests ===
Schedule of variations:
unix
Running target unix
FAIL: linking shared lib
FAIL: bootstrap
FAIL: bootstrap with strip
FAIL: bootstrap with --static
FAIL: bootstrap with --traditional-format
FAIL: bootstrap with --no-keep-memory
FAIL: bootstrap with --relax
FAIL: cdtest with -Ur
FAIL: NOCROSSREFS 2
FAIL: ld-scripts/provide-1
XFAIL: ld-scripts/provide-3
XFAIL: S-records
XFAIL: S-records with constructors
=== ld Summary ===
# of expected passes 22
# of unexpected failures 10
# of expected failures 3
/usr/build/obj-patched/ld/ld-new 20040504
runtest completed at Tue May 4 17:20:17 2004
Test Run By dk on Tue May 4 17:21:04 2004
Native configuration is i686-pc-cygwin
=== winsup tests ===
setting trap for SIGTERM to terminated
setting trap for SIGINT to interrupted by user
setting trap for SIGQUIT to interrupted by user
setting trap for SIGSEGV to segmentation violation
dirlist is /artimi/swtools/windows/share/dejagnu/baseboards
pushing config for build, name is mace
dirlist is /artimi/swtools/windows/share/dejagnu/baseboards
pushing config for host, name is mace
Schedule of variations:
unix
target is unix
FAIL: msgtest.c (execute)
FAIL: semtest.c (execute)
FAIL: shmtest.c (execute)
XFAIL: ltp/dup03.c (execute)
XFAIL: ltp/dup05.c (execute)
XFAIL: ltp/fcntl05.c (execute)
XFAIL: ltp/fcntl07B.c (execute)
XFAIL: ltp/lseek04.c (execute)
XFAIL: ltp/select03.c (execute)
XFAIL: ltp/setgroups01.c (execute)
XFAIL: ltp/setuid02.c (execute)
XFAIL: ltp/ulimit01.c (compile)
XFAIL: ltp/unlink06.c (execute)
XFAIL: ltp/unlink08.c (execute)
XFAIL: samples/sample-fail.c (execute)
XFAIL: samples/sample-miscompile.c (compile)
=== winsup Summary ===
# of expected passes 265
# of unexpected failures 3
# of expected failures 13
Binary file
/usr/build/obj-patched/i686-pc-cygwin/winsup/testsuite/../cygwin/cygwin0.dll
matches
CYGWIN=ntsec smbntsec notty
runtest completed at Tue May 4 17:33:20 2004
------------------->end testsuite results<-------------------
Hey! Looks like no regressions to me. Just to be completely sure I've
installed the new patched binutils and rerun just the winsup build-and-test
using them:
=== winsup Summary ===
# of expected passes 265
# of unexpected failures 3
# of expected failures 13
Binary file
/usr/build/obj-patched/i686-pc-cygwin/winsup/testsuite/../cygwin/cygwin0.dll
matches
CYGWIN=ntsec smbntsec notty
[ To be completely certain the new as and ld were being used, I copied and
pasted one of the build commands from the test output, adding the -v flag;
then I did the same with the resulting collect2 command, and saw it was
correctly invoking /usr/build/install/bin/ld ]
cheers,
DaveK
--
Can't think of a witty .sigline today....