https://bugzilla.novell.com/show_bug.cgi?id=359490 Christoph Thiel changed: What |Removed |Added ---------------------------------------------------------------------------- AssignedTo|cthiel@novell.com |duwe@novell.com -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=359490 User asvetter@cip.physik.uni-wuerzburg.de added comment https://bugzilla.novell.com/show_bug.cgi?id=359490#c2 --- Comment #2 from Andreas Vetter 2008-02-14 15:27:52 MST --- probably a dupl of bug 355044 -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=359490 User duwe@novell.com added comment https://bugzilla.novell.com/show_bug.cgi?id=359490#c4 Torsten Duwe changed: What |Removed |Added ---------------------------------------------------------------------------- CC| |casualprogrammer@yahoo.com --- Comment #4 from Torsten Duwe 2008-03-31 03:21:12 MST --- *** Bug 355044 has been marked as a duplicate of this bug. *** https://bugzilla.novell.com/show_bug.cgi?id=355044 -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=359490 User novell.com-pnt@ladisch.de added comment https://bugzilla.novell.com/show_bug.cgi?id=359490#c6 Julian Ladisch changed: What |Removed |Added ---------------------------------------------------------------------------- CC| |novell.com-pnt@ladisch.de Version|Alpha 2 |Beta 1 --- Comment #6 from Julian Ladisch 2008-04-20 17:51:25 MST --- I tried openSUSE-11.0-Beta1-KDE4-LiveCD-i386.iso. The memtest v1.70 incorrectly reports memory errors. It reports the same (!) errors on a laptop and a pc. When I hit space shortly after starting of memtest the display won't scroll off. These are the first 10 errors reported: Tst Pass Failing Address Good Bad Err-Bits Count Chan --- ---- ----------------------- -------- -------- -------- ----- ---- 2 0 000000ffffc - 0.9MB ffffffff 35fc0035 ca03ffca 1 2 0 000000ffff8 - 0.9MB ffffffff 302f3230 cfd0cdcf 1 2 0 000000ffff4 - 0.9MB ffffffff 2f3930f0 d0c6cf0f 1 2 0 000000ffff0 - 0.9MB ffffffff 00e05bea ff1fa415 1 2 0 000000fffec - 0.9MB ffffffff 00000000 ffffffff 1 2 0 000000fffe8 - 0.9MB ffffffff 00000000 ffffffff 1 2 0 000000fffe4 - 0.9MB ffffffff 00ab00e9 ff54ff16 1 2 0 000000fffe0 - 0.9MB ffffffff 000000cb ffffff34 1 2 0 000000fffdc - 0.9MB ffffffff fe93e800 016c17ff 1 2 0 000000fffd8 - 0.9MB ffffffff 00000000 ffffffff 1 memtest v1.70 from OpenSUSE 10.3 DVD does not report any error. -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=359490 User kernel01@hailfinger.org added comment https://bugzilla.novell.com/show_bug.cgi?id=359490#c8 Carl-Daniel Hailfinger changed: What |Removed |Added ---------------------------------------------------------------------------- CC| |kernel01@hailfinger.org --- Comment #8 from Carl-Daniel Hailfinger 2008-04-22 06:44:47 MST --- The failing memory area is decoded to ROM space on most (all?) x86-compatible systems since 8086. Memtest should skip it by default and the area is marked "reserved" in the BIOS e820 memory map. -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=359490 User lmichnovic@novell.com added comment https://bugzilla.novell.com/show_bug.cgi?id=359490#c10 --- Comment #10 from Ladislav Michnovič 2008-04-23 02:45:08 MST --- (In reply to comment #9 from Harald Koenig)

I've compared the source rpms -- they have identical contents (source, diffs):

-rw-r--r-- 1 harald users 158966 Sep 22 2007 memtest86+-1.70-40.src.rpm -rw-r--r-- 1 harald users 156612 Apr 12 22:53 memtest86+-1.70-68.src.rpm

You can't compare RPMs. The newer one uses different compression algorithm than the older one. (RPMPayload has changed from bzip2 to lzma in Alpha 3+.) -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=359490 User coolo@novell.com added comment https://bugzilla.novell.com/show_bug.cgi?id=359490#c12 --- Comment #12 from Stephan Kulow 2008-04-23 09:47:35 MST --- @#10 - only the payload of binary rpms has changed, the src.rpm used gzip then and now. Check yourself:

rpm -qp --nodigest --nosignature --qf "%{PAYLOADCOMPRESSOR}" memtest86+.spm gzip

-- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=359490 User casualprogrammer@yahoo.com added comment https://bugzilla.novell.com/show_bug.cgi?id=359490#c14 --- Comment #14 from Casual J. Programmer 2008-04-27 02:04:31 MST --- The Beta1 DVD still contains the broken memtest, not sure where the problem lies. This is getting ridiculous. -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=359490 User casualprogrammer@yahoo.com added comment https://bugzilla.novell.com/show_bug.cgi?id=359490#c16 --- Comment #16 from Casual J. Programmer 2008-04-27 03:25:43 MST --- Thanks, Dave, I pointed this out way back on Bug 355044 Comment #1 From Casual J. Programmer 2008-01-21 02:32:12 MST :-( -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=359490 User koenig@linux.de added comment https://bugzilla.novell.com/show_bug.cgi?id=359490#c18 --- Comment #18 from Harald Koenig 2008-04-28 07:22:00 MST --- (In reply to comment #17 from Carl-Daniel Hailfinger)

(In reply to comment #13 from Dave Plater)

...

Upgraded to factory memtest86 1.70-69 i586 running Athlon 64 X2 4000 @2110Mhz and can confirm that memtest is definitely sick. It starts to fail and then goes haywire and it is impossible to exit except a hard reset.

Very interesting. So can you confirm that the problem not only affects the memtest on the CD, but also the installed version? And does a memtest from opensuse 10.3 work if you install it inside your 11.0beta1?

correct, same here on an IBM T60 (T7200 CPU). booting from the same grub from hard disk, the new memtest shows errors starting with test #2, while the memtest.bin from 10.3 is happy. details: errors from: harald > rpm -qf /boot/memtest.bin memtest86+-1.70-70 harald > md5sum /boot/memtest.bin 60f4e98a0fb2882334518c5b6d587745 /boot/memtest.bin working (no errors): harald > rpm -r /10.3 -qf /boot/memtest.bin memtest86+-1.70-40 harald > md5sum /10.3/boot/memtest.bin cdcba28768ebd7a6262f89165cf9f4f2 /10.3/boot/memtest.bin both binary are same size but different: harald > ll /boot/memtest.bin /10.3/boot/memtest.bin -rw-r--r-- 1 root root 103100 Apr 25 22:54 /boot/memtest.bin -rw-r--r-- 1 root root 103100 Sep 21 2007 /10.3/boot/memtest.bin harald > cmp -l /boot/memtest.bin /10.3/boot/memtest.bin | wc 68989 206967 1034835 please remember my comment #9 above! the sources/diffs in memtest *.src.rpm did not change between those versions, so it looks like a problem in the tool chain, isn't it ?!? -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=359490 User kernel01@hailfinger.org added comment https://bugzilla.novell.com/show_bug.cgi?id=359490#c20 --- Comment #20 from Carl-Daniel Hailfinger 2008-04-28 08:38:59 MST --- Thanks! That means it is definitely a toolchain problem. The most prominent change that could affect some memtest routines is that vanilla gcc 4.3.0 for x86 assumes DF is cleared when execution starts. About comment #9: The decimal/hexadecimal address conversion problem is known and fixed in later upstream versions. -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=359490 User coolo@novell.com added comment https://bugzilla.novell.com/show_bug.cgi?id=359490#c22 --- Comment #22 from Stephan Kulow 2008-04-30 04:53:35 MST --- we looked into this and it's the actual test routine - also in 2.01 -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=359490 User coolo@novell.com added comment https://bugzilla.novell.com/show_bug.cgi?id=359490#c24 Stephan Kulow changed: What |Removed |Added ---------------------------------------------------------------------------- Status|ASSIGNED |RESOLVED Resolution| |FIXED --- Comment #24 from Stephan Kulow 2008-05-01 00:40:05 MST --- submitted fixed package -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=359490 User casualprogrammer@yahoo.com added comment https://bugzilla.novell.com/show_bug.cgi?id=359490#c26 Casual J. Programmer changed: What |Removed |Added ---------------------------------------------------------------------------- Status|RESOLVED |REOPENED Resolution|FIXED | --- Comment #26 from Casual J. Programmer 2008-05-02 01:11:07 MST --- It is still broken on http://download.opensuse.org/distribution/SL-OSS-factory/inst-source/boot/bo... dated 01-May-2008 19:27. Not sure why it would not be possible to include a working binary until compiler issues are sorted out for good. -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=359490 User casualprogrammer@yahoo.com added comment https://bugzilla.novell.com/show_bug.cgi?id=359490#c28 --- Comment #28 from Casual J. Programmer 2008-05-02 02:14:06 MST --- As I stated in various places now: Closing bugs does not make them go away. As far as I see it, the broken memtest is now part of Beta2. People using it will probably spend a lot of time ( as we have already ) to find out that it's a bug and not broken memory. As I also pointed out on various occasions, debugging would be a lot less time- and resource consuming if we were able to not propagate known problems over and over. Feel free to rebuke people spending time and effort to help you. -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=359490 User koenig@linux.de added comment https://bugzilla.novell.com/show_bug.cgi?id=359490#c30 --- Comment #30 from Harald Koenig 2008-05-05 13:58:26 MST --- (In reply to comment #29 from Stephan Kulow)

...

As far as I see it, the broken memtest is now part of Beta2.

But you can't see beta2, only factory's boot.iso

ACK: memtest86+-2.01-2 works for me (tm). -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=359490 User coolo@novell.com added comment https://bugzilla.novell.com/show_bug.cgi?id=359490#c32 --- Comment #32 from Stephan Kulow 2008-05-07 11:13:36 MST --- different problem, please file a new bug. -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.