This is the mail archive of the
gdb-testers@sources.redhat.com
mailing list for the GDB project.
Sunday Project test results, gdb, 2002-02-26
- From: Michael Elizabeth Chastain <mec at shout dot net>
- To: gdb-testers at sources dot redhat dot com
- Date: Wed, 27 Feb 2002 01:35:14 -0600
- Subject: Sunday Project test results, gdb, 2002-02-26
Here is an extra edition of the Sunday Project report. This is to give
more visibility into the health of gdb HEAD in time for the branch.
In particular, check out:
http://www.shout.net/~mec/sunday/2002-02-26/Counts.html
Counts
http://www.shout.net/~mec/sunday/2002-02-26/Differences-compare-by-gdb.html
Difference Tables: Compare by Gdb Version
Here is a difference grid:
gdb improved
gdb.base/constvars.exp +1 +2 +3 +4 +5 +6 +7 +8
gdb.base/interrupt.exp +1 +2 +3 +4
gdb.base/return.exp +1 +2 +3 +4 +5 +6 +7 +8
gdb.base/return2.exp +1 +2 +3 +4 +5 +6 +7 +8
gdb.base/scope.exp +6 +8
gdb.base/volatile.exp +1 +3 +5 +6 +7 +8
gdb.c++/cplusfuncs.exp +1 +2 +3 +4 +5 +6 +7 +8
gdb.c++/demangle.exp +1 +2 +3 +4 +5 +6 +7 +8
gdb.c++/derivation.exp +1 +2 +3 +4 +5 +6 +7 +8
gdb.c++/inherit.exp +1 +2 +3 +4 +5 +6 +7 +8
gdb.c++/namespace.exp +1 +2 +3 +4 +5 +6 +7 +8
gdb.c++/ovldbreak.exp +6 +8
gdb.c++/virtfunc.exp +1 +2 +3 +4 +5 +6 +7 +8
gdb.mi/mi-stack.exp +4 +6 +8
gdb.mi/mi-var-child.exp +1 +3 +5 +7
gdb.mi/mi-var-display.exp +1 +2 +3 +4 +5 +6 +7 +8
gdb.mi/mi0-stack.exp +4 +6 +8
gdb.mi/mi0-var-child.exp +1 +3 +5 +7
gdb.mi/mi0-var-display.exp +1 +2 +3 +4 +5 +6 +7 +8
gdb regressed
gdb.base/call-ar-st.exp -1 -2
gdb.base/display.exp ,1 -2 ,3 ,4 ,5 ,6 ,7 ,8
gdb.base/ending-run.exp -1 -2 ,3 ,4 ,5 ,6 ,7 ,8
gdb.base/funcargs.exp -2 -5 +6 -7 +8
gdb.base/list.exp -2
gdb.base/ptype.exp *1 -2 *3 *4 *5 *6 *7 *8
gdb.base/shlib-call.exp -2
gdb.base/step-line.exp -2
gdb.base/step-test.exp -2
gdb.c++/classes.exp -1 -2 +3 _4 +5 _6 +7 _8
gdb.c++/local.exp -1
gdb.c++/method.exp +3 +4 +5 -6 +7 -8
gdb.c++/overload.exp +1 -2 +3 +4 +5 +6 +7 +8
gdb.c++/templates.exp ,1 +2 +3 +4 +5 -6 +7 -8
gdb.mi/mi-disassemble.exp -6 -8
gdb.mi/mi-simplerun.exp -2
gdb.mi/mi-var-cmd.exp -1 +2 -3 +4 -5 +6 -7 +8
gdb.mi/mi0-disassemble.exp -6 -8
gdb.mi/mi0-simplerun.exp -2
gdb.mi/mi0-var-cmd.exp -1 +2 -3 +4 -5 +6 -7 +8
new test with good result
gdb.base/asm-source.exp *1 *2 *3 *4 *5 *6 *7 *8
gdb.base/break.exp *1 *2 *3 *4 *5 *6 *7 *8
gdb.base/code-expr.exp *1 *2 *3 *4 *5 *6 *7 *8
gdb.base/cvexpr.exp *1 *2 *3 *4 *5 *6 *7 *8
gdb.base/huge.exp *1 *2 *3 *4 *5 *6 *7 *8
gdb.base/info-proc.exp *1 *2 *3 *4 *5 *6 *7 *8
gdb.base/maint.exp *1 *2 *3 *4 *5 *6 *7 *8
gdb.base/sizeof.exp *1 *2 *3 *4 *5 *6 *7 *8
gdb.c++/userdef.exp *1 *2 *3 *4 *5 *6 *7 *8
gdb.threads/gcore-thread.exp *1 *2 *3 *4 *5 *6 *7 *8
new test with bad result
gdb.base/callfuncs.exp _1 _2 +3 *4 +5 *6 +7 *8
gdb.base/gcore.exp _1 _2 _3 _4 _5 _6 _7 _8
gdb.base/reread.exp *1 _2 _3 _4 _5 _6 _7 _8
noise: result changed but gdb behavior did not
gdb.base/annota1.exp ,1 ,2 ,3 ,4 ,5 ,6 ,7 ,8
gdb.base/call-rt-st.exp ,1 ,2 ,3 ,4 ,5 ,6 ,7 ,8
gdb.base/callfwmall.exp ,1 ,2 ,3 ,4 ,5 ,6 ,7 ,8
gdb.base/completion.exp ,1 ,2 ,3 ,4 ,5 ,6 ,7 ,8
gdb.base/corefile.exp ,1 ,2 ,3 ,4 ,5 ,6 ,7 ,8
gdb.base/restore.exp ,1 ,2 ,3 ,4 ,5 ,6 ,7 ,8
gdb.java/jmisc.exp ,1 ,2 ,3 ,4 ,5 ,6 ,7 ,8
gdb.stabs/weird.exp ,1 ,2 ,3 ,4 ,5 ,6 ,7 ,8
gdb.threads/linux-dp.exp ,5 ,6 .7 ,8
gdb.threads/pthreads.exp ,5 ,6 ,7 ,8
key
+ gdb improved
- gdb regressed
* new test with good result
_ new test with bad result
, noise: result changed but gdb behavior did not
Note that I test languages C and C++. No Java, no FORTRAN, no Pascal,
no Objective-C, no Chill.
The next step is to probe further into the regressions and then file
PR's to cover time. But I am out of time for tonight.
The worst regression appears to be the line number bug + step bug which
appeared between 2002-02-16 and 2002-02-22. See pr gdb/379, pr gdb/380,
pr gdb/381.
gdb.c++ has some regressions that are likely to turn out "this is
actually as good as 5.1.1 but the test suite does not know it".
Michael C
===
. Summary
. Test Matrix
. Matrix
target: native
host: i686-pc-linux-gnulibc2.2%rh-7.2
gdb: 5.1.1, HEAD%20020226
gcc: 2.95.3, 3.0.4, gcc-3_1-branch%20020226, HEAD%20020226
glibc: vendor
goption: -gdwarf-2, -gstabs+
count: 16=1*1*2*4*1*2
. Notes
target and host names are gnu triples with extra information after
a '%' delimiter.
versions which start with a number are official releases or snapshots.
versions named "vendor" are the vendor-supplied facility.
versions named "HEAD" are the cvs HEAD (mainline).
versions with any other name are cvs branches.
cvs versions show the pull date after a '%' delimiter.
gcc releases and snapshots are built with binutils 2.11.2
gcc cvs versions are built with binutils HEAD from the same pull date.
. libiberty
all tests passed
. gdb
. Tables
http://www.shout.net/~mec/sunday/2002-02-26/index.html
. Overview
gdb 5.1.1: 0 build failures, 0 test run failures, 634 attention lines
gdb HEAD: 0 build failures, 0 test run failures, 341 attention lines
an attention line is:
error warning fail kfail unresolved untested unsupported
an attention line is not:
pass xpass xfail
. Old bugs fixed
None.
. New bugs detected
None.
. Test protocol changes since last report
I dropped gcc 3.0.3, gcc 3.0.4-20020215, and gcc gcc-3_0-branch.
I added gcc gcc-3_1-branch.
I upgraded my baseline gcc from 3.0.3 to 3.0.4. The baseline gcc is
what I use to build gdb and gcc and so on; it is different from the
list of tested gcc's, which is what I use to build the test programs
for gdb. Changing the baseline gcc should not affect the test results.
I ran before-and-after test runs (not shown) and found no difference.
I upgraded the baseline bison from 1.30 to 1.33. I ran before-and-after
test runs (not shown) and found no difference.
I fixed my implementation of goption. This fixes some spurious FAILs
I had in gdb.base/nodebug.exp. Props to Daniel Jacobowitz.
. Baseline software
. host i686-pc-linux-gnu%rh-7.2
make 3.79.1
binutils 2.11.2
gcc 3.0.4
flex 2.5.4
bison 1.33
tcl 8.3.4
expect 5.33.0
dejagnu 1.4.2
. Analysis
. libiberty
. results
. target native
. host i686-pc-linux-gnu%rh-7.2
binutils HEAD All 648 tests passed
gcc gcc-3_1-branch All 648 tests passed
gcc HEAD All 648 tests passed
gdb HEAD All 648 tests passed
. gdb
The last report was 2002-02-22.
. 5.1.1
No significant change.
. HEAD
. Difference Grid
gdb.base/nodebug.exp ,7 ,8 ,9 ,10 ,11 ,12
gdb.c++/templates.exp ,8 ,10 ,12
gdb.threads/gcore-thread.exp ,11 ,12
gdb.threads/linux-dp.exp ,11 ,12
gdb.threads/pthreads.exp ,11 ,12
Column numbers are configuration numbers in:
http://www.shout.net/~mec/sunday/2002-02-26/Differences-compare-by-date-two.html
+ gdb improved
- gdb regressed
* new test with good result
_ new test with bad result
, noise: result changed but gdb behavior did not
. Noise
All the noise changes this week were reductions in spurious
attention results, which is good.
. gdb.base/nodebug.exp
This used to FAIL a lot due to a bug in my test harness
regarding goption handling.
. gdb.c++/templates.exp
An old setup_xfail is gone because all configurations are PASS
or XPASS now.
. gdb.threads/gcore.exp
gdb.threads/linux-dp.exp
gdb.threads/pthreads.exp
The test programs did not compile with gcc HEAD. It does now.