Results for 3.0 20010501 (prerelease) testsuite on sparc-sun-solaris2.8

[Phil Edwards]

On Thu, May 03, 2001 at 02:00:04PM -0700, Mark Mitchell wrote:
> I comitted the first, but not the latter.  I'll wait until I hear from
> you, or from someone else.  It's hard for me to see how this could
> cause the kind of problem you mention, but anything is possible.

Including this?


83% grep loop- BEFORE AFTER
BEFORE:XPASS: gcc.c-torture/execute/loop-2b.c execution,  -O0
BEFORE:XPASS: gcc.c-torture/execute/loop-2b.c execution,  -O1
BEFORE:XPASS: gcc.c-torture/execute/loop-2b.c execution,  -O2
BEFORE:XPASS: gcc.c-torture/execute/loop-2b.c execution,  -O3 -fomit-frame-pointer
BEFORE:XPASS: gcc.c-torture/execute/loop-2b.c execution,  -O3 -fomit-frame-pointer -funroll-loops
BEFORE:XPASS: gcc.c-torture/execute/loop-2b.c execution,  -O3 -fomit-frame-pointer -funroll-all-loops -finline-functions
BEFORE:XPASS: gcc.c-torture/execute/loop-2b.c execution,  -O3 -g
BEFORE:XPASS: gcc.c-torture/execute/loop-2b.c execution,  -Os
AFTER:XPASS: gcc.c-torture/execute/loop-2b.c execution,  -O0
AFTER:XPASS: gcc.c-torture/execute/loop-2b.c execution,  -O1
AFTER:XPASS: gcc.c-torture/execute/loop-2b.c execution,  -O2
AFTER:XPASS: gcc.c-torture/execute/loop-2b.c execution,  -O3 -fomit-frame-pointer
AFTER:XPASS: gcc.c-torture/execute/loop-2b.c execution,  -O3 -fomit-frame-pointer -funroll-loops
AFTER:XPASS: gcc.c-torture/execute/loop-2b.c execution,  -O3 -fomit-frame-pointer -funroll-all-loops -finline-functions
AFTER:XPASS: gcc.c-torture/execute/loop-2b.c execution,  -O3 -g
AFTER:XPASS: gcc.c-torture/execute/loop-2b.c execution,  -Os
84%
84%
84%
84% diff BEFORE AFTER
24a25
> FAIL: gcc.c-torture/execute/990604-1.c execution,  -Os
45,46c46,47
< # of expected passes          15061
< # of unexpected failures      18
---
> # of expected passes          15060
> # of unexpected failures      19
85%


BEFORE and AFTER are the mail messages generated by contrib/test_summary.
The sources were the 3.0 branch as of the time of my email Thursday evening;
the only difference was your patch.  (I've been sick as a dog between then
and now, so this report's a bit delayed.)

Note that loop-8.c, mentioned in your email, doesn't show up at all.
The other failure just came out of the blue, I think.  We really ought to get
independant confirmation of this before assigning any weight to it, however.


Phil

-- 
pedwards at disaster dot jaj dot com  |  pme at sources dot redhat dot com
devphil at several other less interesting addresses in various dot domains
The gods do not protect fools.  Fools are protected by more capable fools.