|
From: <js...@ac...> - 2006-03-28 02:55:30
|
Nightly build on g5 ( YDL 4.0, ppc970 ) started at 2006-03-28 04:40:01 CEST Results differ from 24 hours ago Checking out valgrind source tree ... done Configuring valgrind ... done Building valgrind ... done Running regression tests ... failed Regression test results follow == 201 tests, 7 stderr failures, 2 stdout failures ================= memcheck/tests/leak-cycle (stderr) memcheck/tests/leak-tree (stderr) memcheck/tests/leakotron (stdout) memcheck/tests/pointer-trace (stderr) memcheck/tests/sh-mem (stderr) none/tests/faultstatus (stderr) none/tests/fdleak_fcntl (stderr) none/tests/mremap (stderr) none/tests/ppc32/mftocrf (stdout) ================================================= == Results from 24 hours ago == ================================================= Checking out valgrind source tree ... done Configuring valgrind ... done Building valgrind ... done Running regression tests ... failed Regression test results follow == 199 tests, 6 stderr failures, 2 stdout failures ================= memcheck/tests/leak-cycle (stderr) memcheck/tests/leak-tree (stderr) memcheck/tests/leakotron (stdout) memcheck/tests/pointer-trace (stderr) none/tests/faultstatus (stderr) none/tests/fdleak_fcntl (stderr) none/tests/mremap (stderr) none/tests/ppc32/mftocrf (stdout) ================================================= == Difference between 24 hours ago and now == ================================================= *** old.short Tue Mar 28 04:47:50 2006 --- new.short Tue Mar 28 04:55:24 2006 *************** *** 8,10 **** ! == 199 tests, 6 stderr failures, 2 stdout failures ================= memcheck/tests/leak-cycle (stderr) --- 8,10 ---- ! == 201 tests, 7 stderr failures, 2 stdout failures ================= memcheck/tests/leak-cycle (stderr) *************** *** 13,14 **** --- 13,15 ---- memcheck/tests/pointer-trace (stderr) + memcheck/tests/sh-mem (stderr) none/tests/faultstatus (stderr) |
|
From: Nicholas N. <nj...@cs...> - 2006-03-28 03:09:55
|
On Tue, 28 Mar 2006 js...@ac... wrote: > Nightly build on g5 ( YDL 4.0, ppc970 ) started at 2006-03-28 04:40:01 CEST > Results differ from 24 hours ago > > ================================================= > == Difference between 24 hours ago and now == > ================================================= > > *************** > *** 13,14 **** > --- 13,15 ---- > memcheck/tests/pointer-trace (stderr) > + memcheck/tests/sh-mem (stderr) > none/tests/faultstatus (stderr) What went wrong, Julian? Can you post the diff? Nick |
|
From: Julian S. <js...@ac...> - 2006-03-28 03:31:22
|
> What went wrong, Julian? Can you post the diff? *** sh-mem.stderr.exp Tue Mar 28 04:47:54 2006 --- sh-mem.stderr.out Tue Mar 28 04:53:29 2006 *************** *** 12,33 **** ! h = 0 (checking 0..60) 0...32...64...96...128...160...192...224... ! h = 1 (checking 1..61) 0...32...64...96...128...160...192...224... ! h = 2 (checking 2..62) 0...32...64...96...128...160...192...224... ! h = 3 (checking 3..63) 0...32...64...96...128...160...192...224... ! -- NNN: 8 U8 U8 ------------------------ ! h = 0 (checking 0..56) 0...32...64...96...128...160...192...224... ! h = 1 (checking 1..57) 0...32...64...96...128...160...192...224... ! h = 2 (checking 2..58) 0...32...64...96...128...160...192...224... ! h = 3 (checking 3..59) 0...32...64...96...128...160...192...224... ! h = 4 (checking 4..60) 0...32...64...96...128...160...192...224... ! h = 5 (checking 5..61) 0...32...64...96...128...160...192...224... ! h = 6 (checking 6..62) 0...32...64...96...128...160...192...224... ! h = 7 (checking 7..63) 0...32...64...96...128...160...192...224... ! -- NNN: 8 F8 U8 ------------------------ ! h = 0 (checking 0..56) 0...32...64...96...128...160...192...224... ! h = 1 (checking 1..57) 0...32...64...96...128...160...192...224... ! h = 2 (checking 2..58) 0...32...64...96...128...160...192...224... ! h = 3 (checking 3..59) 0...32...64...96...128...160...192...224... ! h = 4 (checking 4..60) 0...32...64...96...128...160...192...224... ! h = 5 (checking 5..61) 0...32...64...96...128...160...192...224... ! h = 6 (checking 6..62) 0...32...64...96...128...160...192...224... ! h = 7 (checking 7..63) 0...32...64...96...128...160...192...224... --- 12,15 ---- ! h = 0 (checking 0..60) 0... ! ! FAILURE: STOREVn, offset 0, byte 0 -- is 0x........, should be 0x........ ! Hmm. That's not very helpful. The actual results are: sewardj@g5:~/VgTRUNK/trunk$ file ./memcheck/tests/sh-mem ./memcheck/tests/sh-mem: ELF 64-bit MSB executable, cisco 7500, version 1 (SYSV), for GNU/Linux 2.4.19, dynamically linked (uses shared libs), not stripped and actual results are: -- NNN: 1 U1 U1 ------------------------ h = 0 (checking 0..63) 0...32...64...96...128...160...192...224... -- NNN: 2 U2 U2 ------------------------ h = 0 (checking 0..62) 0...32...64...96...128...160...192...224... h = 1 (checking 1..63) 0...32...64...96...128...160...192...224... -- NNN: 4 U4 U4 ------------------------ h = 0 (checking 0..60) 0...32...64...96...128...160...192...224... h = 1 (checking 1..61) 0...32...64...96...128...160...192...224... h = 2 (checking 2..62) 0...32...64...96...128...160...192...224... h = 3 (checking 3..63) 0...32...64...96...128...160...192...224... -- NNN: 4 F4 U4 ------------------------ h = 0 (checking 0..60) 0... FAILURE: STOREVn, offset 0, byte 0 -- is 0xff, should be 0x1 Um. Does that mean anything to you? J |
|
From: Nicholas N. <nj...@cs...> - 2006-03-28 03:46:46
|
On Tue, 28 Mar 2006, Julian Seward wrote: > -- NNN: 1 U1 U1 ------------------------ > h = 0 (checking 0..63) 0...32...64...96...128...160...192...224... > -- NNN: 2 U2 U2 ------------------------ > h = 0 (checking 0..62) 0...32...64...96...128...160...192...224... > h = 1 (checking 1..63) 0...32...64...96...128...160...192...224... > -- NNN: 4 U4 U4 ------------------------ > h = 0 (checking 0..60) 0...32...64...96...128...160...192...224... > h = 1 (checking 1..61) 0...32...64...96...128...160...192...224... > h = 2 (checking 2..62) 0...32...64...96...128...160...192...224... > h = 3 (checking 3..63) 0...32...64...96...128...160...192...224... > -- NNN: 4 F4 U4 ------------------------ > h = 0 (checking 0..60) 0... > > FAILURE: STOREVn, offset 0, byte 0 -- is 0xff, should be 0x1 > > Um. Does that mean anything to you? Something went wrong with 4 byte loads and stores of floating point values. Can you try commenting out PERF_FAST_LOADV and/or PERF_FAST_STOREV to see if that helps? Nick |
|
From: Julian S. <js...@ac...> - 2006-03-28 04:07:49
|
> Can you try commenting out PERF_FAST_LOADV and/or PERF_FAST_STOREV to see > if that helps? No change, with both commented out - same outcome. J |