--------------7356AE7C25C804A11BBC1088
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
-- Linux - it isn't just for breakfast anymore...--------------7356AE7C25C804A11BBC1088 Content-Type: message/rfc822 Content-Transfer-Encoding: 7bit Content-Disposition: inline
Message-ID: <35846868.412EF2FC@ibm.net> Date: Sun, 14 Jun 1998 17:18:48 -0700 From: "Ya'akov N. Miles" <ynmiles@ibm.net> Organization: Linux Kernel Hackers X-Mailer: Mozilla 3.01 (X11; I; Linux 2.0.30 i586) MIME-Version: 1.0 To: linux@ubc.ee.ca Subject: I typed in your program on genuine intel P233/MMX and got segmentation fault Content-Type: multipart/mixed; boundary="------------580E4F094BE86BF266DDC1B9"
This is a multi-part message in MIME format.
--------------580E4F094BE86BF266DDC1B9 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit
-- Linux - it isn't just for breakfast anymore...--------------580E4F094BE86BF266DDC1B9 Content-Type: message/rfc822 Content-Transfer-Encoding: 7bit Content-Disposition: inline
Received: from fs3.ee.ubc.ca [137.82.52.241] by in2.ibm.net id 897859696.23566-1 ; Sun, 14 Jun 1998 21:28:16 +0000 Received: from alphamb2.phys.vt.edu (alphamb2.phys.vt.edu [128.173.176.26]) by fs3.ee.ubc.ca (8.9.0/8.9.0) with ESMTP id OAA02231 for <linux@ee.ubc.ca>; Sun, 14 Jun 1998 14:22:54 -0700 (PDT) Received: from localhost (siglercm@localhost) by alphamb2.phys.vt.edu (8.8.5/8.8.5) with SMTP id RAA00900; Sun, 14 Jun 1998 17:26:14 -0400 (EDT) Date: Sun, 14 Jun 1998 17:26:14 -0400 (EDT) From: Clemmitt Sigler <siglercm@alphamb2.phys.vt.edu> To: Mike Montour <mmontour@iname.com> cc: "Vancouver (UBC) Linux Users' Group" <linux@ee.ubc.ca> Subject: Re: WARNING: New K6 bug In-Reply-To: <3580C07C.8C603115@iname.com> Message-ID: <Pine.OSF.3.92.980614171904.871A-100000@alphamb2.phys.vt.edu> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII
On Fri, 12 Jun 1998, Mike Montour wrote: > Clemmitt Sigler wrote: > Please try running the attached program from a regular user (non-root) > account. <snip> > > Here is how to reproduce it : > > > > $ cat a.s > > .text > > .align 4096 /* r1 */ > > .globl _start > > _start: > > movl _start, %edi /* S1 */ > > cmpb 0x80000000(%edi),%dl /* r2, S2 */ > > je nowhere /* r3 */ > > ret > > $ as -o a.o a.s > > $ ld -defsym nowhere=0xc0000000 a.o > > $ ./a.out
This causes an immediate and total lock-up of my CPU, so it does have this CPU bug in it. It's a brand new AMD K6-266, so I guess they didn't implement a new stepping on these 0.25 micron CPUs :^( They said it would be fixed in the next stepping.
> > [...] > > > > Also, you might > > > > try running the program cpuspd3e that you can download from the AMD > > > > web site (http://www.amd.com/K6/k6docs/cpuspd3e.exe) on your chip and > > > > posting the results. > > CPUSPD3.EXE says : > > AMD-K6/PR2-233 Processor > > So, the labelling appears to be correct. Not as many technical details as I > had hoped for, however.
Mine just said "Unknown speed detected." It knew it was an AMD CPU, but it couldn't figure out its speed. The little assembler source has a data table in it that only goes up to 233MHz, so it couldn't figure out my K6-266 running at 300MHz :^(
> Last night, I got another clue. I was doing something that caused a 'sleeping' > hard drive to spin up, and the system locked right at that moment. I have > not been able to reproduce the problem, but I turned off hard-drive spindown > and ran some 'crashme's overnight (2.1.105) and the system survived. I will > abuse the system some more today
OK, I would still be really surprised if this CPU bug, which apparently has existed since the original K6's, is causing your RH5.1 kernel crashes. Please let us know what the problem turns out to be when you diagnose it.
Clemmitt Sigler Va. Tech Physics Dept.
--------------580E4F094BE86BF266DDC1B9 Content-Type: text/plain; charset=us-ascii; name="crashme.s" Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="crashme.s"
.text .align 4096 /* r1 */ .globl _start _start: movl _start,%edi/* S1 */ cmpb 0x80000000(%edi),%dl/* r2,S2 */ je nowhere/* r3 */ ret
--------------580E4F094BE86BF266DDC1B9--
--------------7356AE7C25C804A11BBC1088--