This is the mail archive of the
ecos-bugs@sources.redhat.com
mailing list for the eCos project.
[Bug 19308] New: SPARClite-sleb Hardware - clock - Fail
- From: bugzilla-daemon at ecoscentric dot com
- To: ecos-bugs at sources dot redhat dot com
- Date: Thu, 24 Apr 2003 14:47:26 +0100 (BST)
- Subject: [Bug 19308] New: SPARClite-sleb Hardware - clock - Fail
http://bugs.ecos.sourceware.org/show_bug.cgi?id=19308
Summary: SPARClite-sleb Hardware - clock - Fail
Product: eCos
Version: 1.2
Platform: sleb (Fujitsu MB86800-MA01 board)
OS/Version: SPARC
Status: UNCONFIRMED
Severity: normal
Priority: normal
Component: Other
AssignedTo: jifl at ecoscentric dot com
ReportedBy: support at ecoscentric dot com
Test output was as follows:
----------------------------------------------------------------------------
----------------------------------------------------------------------------
----------------------------------------------------------------------------
---------
1999-02-25 13:17:41 ginga:5003 SPARClite-sleb Hardware
clock Fail 23k/553k Elapsed: D=28.7/49.0 Total=30.0 E=0.3/30.0
"thundercat:/tmp/test7112/sparclite/sleb/H/combo1/install/tests/libc/time/cl
ock"
----------------------------------------------------------------------------
----------------------------------------------------------------------------
----------------------------------------------------------------------------
---------
GNU gdb 4.17-ecos-98r1p4
Copyright 1998 Free Software Foundation, Inc.
GDB is free software, covered by the GNU General Public License, and you are
welcome to change it and/or distribute copies of it under certain
conditions.
Type "show copying" to see the conditions. This version of GDB is supported
for customers of Cygnus Solutions. Type "show warranty" for details.
This GDB was configured as "--host=i386-cygwin32 --target=sparclite-elf"...
(gdb) set height 0
(gdb) set remotedebug 0
(gdb) set remotebaud 19200
(gdb) target remote com2
Remote debugging using com2
+0x8118 in ?? ()
(gdb) load
Loading section .rom_vectors, size 0x4c8 lma 0x4010000
Loading section .text, size 0x42ac lma 0x40104c8
Loading section .rodata, size 0x139f lma 0x4014778
Loading section .data, size 0x290 lma 0x4015b18
Start address 0x4010000 , load size 23971
Transfer rate: 6612 bits/sec.
(gdb) break cyg_test_exit
Breakpoint 1 at 0x40108d0: file
/tmp/v1_2_19990225_38/ecos-1.2/packages/infra/v1_2/src/tcdiag.cxx, line 90.
(gdb) break cyg_assert_fail
Breakpoint 2 at 0x4013434: file
/tmp/v1_2_19990225_38/ecos-1.2/packages/infra/v1_2/src/buffer.cxx, line 710.
(gdb) cont
Continuing.
INFO:<Starting tests from testcase
/tmp/v1_2_19990225_38/ecos-1.2/packages/language/c/libc/v1_2/tests/time/cloc
k.c for C library clock() function>
INFO:<Starting tests from testcase
/tmp/v1_2_19990225_38/ecos-1.2/packages/language/c/libc/v1_2/tests/time/cloc
k.c for C library clock() function>
FAIL:<clock() stability> Line: 182, File:
/tmp/v1_2_19990225_38/ecos-1.2/packages/language/c/libc/v1_2/tests/time/cloc
k.c
EXIT:<Finished tests from testcase
/tmp/v1_2_19990225_38/ecos-1.2/packages/language/c/libc/v1_2/tests/time/cloc
k.c for C library clock() function>
+
How-To-Repeat:
The following permutation(s) have been used:
combo1
Build was v1_2_19990225_38
Originator:
eCos Testing Group
Organization:
Cygnus
Audit-Trail:
Responsible-Changed-From-To: alexs->hmt
Responsible-Changed-By: alexs
Responsible-Changed-When: Thursday February 25 16:34:03 GMT 1999
Responsible-Changed-Why:
SPARClite h/w specific issue
Responsible-Changed-From-To: hmt->nickg
Responsible-Changed-By: alexs
Responsible-Changed-When: Tue Mar 16 03:14:01 PST 1999
Responsible-Changed-Why:
SPARClite h/w clock instability.
3/3 instability failures. See also 19410, 19528.
Responsible-Changed-From-To: nickg->jlarmour
Responsible-Changed-By: alexs
Responsible-Changed-When: Tue Mar 16 05:15:49 PST 1999
Responsible-Changed-Why:
The clock test (yours?) fails with instability error on TX39, SPARClite and PPC
hardware under various combos. They all look related, implying either a fault
with the kernel or with the test itself. Could you please investigate.
The other PRs are 19410 and 19528.
This is 2/3.
State-Changed-From-To: open-closed
State-Changed-By: jlarmour
State-Changed-When: Wed Mar 24 05:41:15 PST 1999
State-Changed-Why:
I have failed to reproduce this.
I have used sources checked out at the _exact_ same time, and
used the _exact_ same combo, run on Huge's sleb board.
I ran it over a dozen times and it still worked every time.
One thing I haven't been able to change is the CygMon on the board,
so it is possible that there has been a changein Cygmon which has
fixed this problem since then.
So in the absence of anything more likely right now, I'm closing
this PR.
State-Changed-From-To: closed-open
State-Changed-By: jlarmour
State-Changed-When: Wed Mar 24 13:16:27 PST 1999
State-Changed-Why:
The bug is still around
State-Changed-From-To: open-closed
State-Changed-By: jlarmour
State-Changed-When: Thu Mar 25 12:00:34 PST 1999
State-Changed-Why:
I still cannot reproduce this problem, but there is a good chance
that Hugo's recent changes to the clock fix this problem.
I simply can't do anything else - I've run this test over 30 times
now, and it hasn't failed yet :-(.
State-Changed-From-To: closed-open
State-Changed-By: testuser
State-Changed-When: Friday March 26 22:01:58 GMT 1999
State-Changed-Why:
Failed again tonight.
--------------------------------------------------------------------------------
1999-03-26 20:20:22 bhuna:5003 SPARClite-sleb clock.exe Fail 61k/967k Elapsed: D=1.2/263.0 Total=4.9 E=2.0/30.0 "thundercat:c:tmptest1018sparcliteslebHcombo7installtestslibctimeclock.exe"
--------------------------------------------------------------------------------
Resetting target using ginga:5000 ab -b 19200
BOOTP got 194.130.39.55
<0.0/0.0> GNU gdb 4.17-ecos-98r1p5
<0.0/0.0> Copyright 1998 Free Software Foundation, Inc.
<0.0/0.0> GDB is free software, covered by the GNU General Public License, and you are
<0.0/0.0> welcome to change it and/or distribute copies of it under certain conditions.
<0.0/0.0> Type "show copying" to see the conditions. This version of GDB is supported
<0.0/0.0> for customers of Cygnus Solutions. Type "show warranty" for details.
<0.0/0.0> This GDB was configured as "--host=i386-cygwin32 --target=sparclite-elf"...
<0.6/0.0> (gdb) set height 0
<0.6/0.0> (gdb) set remotedebug 0
<0.6/0.0> (gdb) set watchdog 0
<0.6/0.0> (gdb) target remote 194.130.39.55:1000
<0.6/0.0> Remote debugging using 194.130.39.55:1000
<0.6/0.0> 0x9c30 in ?? ()
<0.6/0.0> (gdb) load
<0.6/0.0> Loading section .rom_vectors, size 0x65c lma 0x4010000
<0.7/0.1> Loading section .text, size 0xca20 lma 0x401065c
<1.9/1.2> Loading section .rodata, size 0x1ed0 lma 0x401d080
<2.2/0.3> Loading section .data, size 0x764 lma 0x401ef50
<2.2/0.1> Start address 0x4010000 , load size 63152
<2.2/0.0> Transfer rate: 505216 bits/sec.
<2.2/0.0> (gdb) break cyg_test_exit
<2.3/0.0> Breakpoint 1 at 0x401311c: file //c/tmp/v1_2_19990326_1604/ecos-1.2/packages/infra/v1_2/src/tcdiag.cxx, line 99.
<2.3/0.0> (gdb) break cyg_assert_fail
<2.3/0.0> Breakpoint 2 at 0x401925c: file //c/tmp/v1_2_19990326_1604/ecos-1.2/packages/infra/v1_2/src/buffer.cxx, line 709.
<2.3/0.0> (gdb) cont
<2.3/0.0> Continuing.
<2.8/0.5> INFO:<Starting tests from testcase //c/tmp/v1_2_19990326_1604/ecos-1.2/packages/language/c/libc/v1_2/tests/time/clock.c for C library clock() function>
<2.9/0.0> FAIL:<clock() stability> Line: 171, File: //c/tmp/v1_2_19990326_1604/ecos-1.2/packages/language/c/libc/v1_2/tests/time/clock.c
<2.9/0.0> EXIT:<Finished tests from testcase //c/tmp/v1_2_19990326_1604/ecos-1.2/packages/language/c/libc/v1_2/tests/time/clock.c for C library clock() function>
<4.9/2.0>
<4.9/2.0> Breakpoint 1, cyg_test_exit ()
<4.9/2.0> at //c/tmp/v1_2_19990326_1604/ecos-1.2/packages/infra/v1_2/src/tcdiag.cxx:99
<4.9/2.0> 99 for(;;)
<4.9/2.0> Current language: auto; currently c++
<4.9/2.0> (gdb) q
No flames please.
State-Changed-From-To: open-closed
State-Changed-By: jlarmour
State-Changed-When: Tue Mar 30 14:42:20 PST 1999
State-Changed-Why:
Definitely fixed now :-).
State-Changed-From-To: closed-open
State-Changed-By: testuser
State-Changed-When: Sunday April 18 14:45:25 BST 1999
State-Changed-Why:
Definitely failed again last night :-).
Responsible-Changed-From-To: jlarmour->jskov
Responsible-Changed-By: jskov
Responsible-Changed-When: Mon Apr 19 09:23:08 PDT 1999
Responsible-Changed-Why:
State-Changed-From-To: open-open
State-Changed-By: jskov
State-Changed-When: Thu Apr 22 08:48:13 PDT 1999
State-Changed-Why:
gonna fix it properly.
Responsible-Changed-From-To: jskov->jlarmour
Responsible-Changed-By: jskov
Responsible-Changed-When: Thu Apr 22 08:48:13 PDT 1999
Responsible-Changed-Why:
------- Additional Comments From alexs at ecoscentric dot com 2003-24-04 14:47 BST -------
Proper assignment
------- You are receiving this mail because: -------
You are the assignee for the bug, or are watching the assignee.