qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
* [Qemu-devel] (TEST) memtest86+ & memtest86 & qemu 1.60
@ 2004-07-29 15:11 Leonardo Marques de Souza
  2004-07-29 16:31 ` André Braga
  0 siblings, 1 reply; 4+ messages in thread
From: Leonardo Marques de Souza @ 2004-07-29 15:11 UTC (permalink / raw)
  To: qemu-devel

Hello,

i tested this programs:

- Memtest86+ v1.20 (www.memtest.org)
- Memtest86    v3.1a (www.memtest86.com)

Well, booting the 'iso' version of this programs all goes fine:

- qemu -cdrom memtest86+1.20.iso (OK)
- qemu -cdrom memtest86_3.1a.iso    (OK)

Humm... interesting thing... the L1 cache is more slow than L2 cache.
This "PentiumPro" cpu need to switch this caches. hehe  :P

But using floppy version i got some "EIP" errors.
   (if you wait 10 seconds this show more errors)
- qemu -fda memtest86+-1.20.bin
- qemu -fda memtest86_3.1a.bin

using -isa or -stdvga i got "EIP" erros more "fast"


i tested other version of memtest (normal and plus) and the same error
happens.

^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: [Qemu-devel] (TEST) memtest86+ & memtest86 & qemu 1.60
  2004-07-29 15:11 [Qemu-devel] (TEST) memtest86+ & memtest86 & qemu 1.60 Leonardo Marques de Souza
@ 2004-07-29 16:31 ` André Braga
  2004-07-29 19:48   ` SPAM[RBL] " Leonardo Marques de Souza
  0 siblings, 1 reply; 4+ messages in thread
From: André Braga @ 2004-07-29 16:31 UTC (permalink / raw)
  To: qemu-devel

There's a bug in the floppy emulation code. And this patch (by Jani
Monoses) is supposed to fix it (I guess... Unless the bug in memtest
is triggered by a different event than failing to read the last
sectors, but that's highly unlikely since it works OK from the ISO)


- - - - - - - - - - - - - - - - - - - cut: fbblockpatch.diff - - - - -
- - - - - - - - - - - - - - - - -

Index: block.c
===================================================================
RCS file: /cvsroot/qemu/qemu/block.c,v
retrieving revision 1.10
diff -u -r1.10 block.c
--- block.c	14 Jul 2004 17:20:30 -0000	1.10
+++ block.c	23 Jul 2004 15:12:21 -0000
@@ -311,6 +311,7 @@
 {
     int ret, n, fd;
     int64_t offset;
+    int bytes = 512;
     
     if (!bs->inserted)
         return -1;
@@ -320,7 +321,7 @@
             fd = bs->cow_fd;
             offset = bs->cow_sectors_offset;
         } else if (sector_num == 0 && bs->boot_sector_enabled) {
-            memcpy(buf, bs->boot_sector_data, 512);
+            memcpy(buf, bs->boot_sector_data, bytes);
             n = 1;
             goto next;
         } else {
@@ -335,8 +336,11 @@
             offset += sector_num * 512;
             lseek64(fd, offset, SEEK_SET);
             ret = read(fd, buf, n * 512);
+	    
             if (ret != n * 512) {
-                return -1;
+		if (ret == -1)    
+	                return -1;
+		bytes = ret;
             }
         }
     next:

- - - - - - - - - - - - - - - - - - - cut: fbblockpatch.diff - - - - -
- - - - - - - - - - - - - - - - -


On Thu, 29 Jul 2004 12:11:06 -0300, Leonardo Marques de Souza
<leonardo@ambientebrasil.com.br> wrote:
> Hello,
> 
> i tested this programs:
> 
> - Memtest86+ v1.20 (www.memtest.org)
> - Memtest86    v3.1a (www.memtest86.com)
> 
> Well, booting the 'iso' version of this programs all goes fine:
> 
> - qemu -cdrom memtest86+1.20.iso (OK)
> - qemu -cdrom memtest86_3.1a.iso    (OK)
> 
> Humm... interesting thing... the L1 cache is more slow than L2 cache.
> This "PentiumPro" cpu need to switch this caches. hehe  :P
> 
> But using floppy version i got some "EIP" errors.
>    (if you wait 10 seconds this show more errors)
> - qemu -fda memtest86+-1.20.bin
> - qemu -fda memtest86_3.1a.bin
> 
> using -isa or -stdvga i got "EIP" erros more "fast"
> 
> i tested other version of memtest (normal and plus) and the same error
> happens.
> 
> _______________________________________________
> Qemu-devel mailing list
> Qemu-devel@nongnu.org
> http://lists.nongnu.org/mailman/listinfo/qemu-devel
> 


-- 
"No, no, you're not thinking; you're just being logical"
Niels Bohr

^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: SPAM[RBL] Re: [Qemu-devel] (TEST) memtest86+ & memtest86 & qemu 1.60
  2004-07-29 16:31 ` André Braga
@ 2004-07-29 19:48   ` Leonardo Marques de Souza
  2004-07-29 19:56     ` André Braga
  0 siblings, 1 reply; 4+ messages in thread
From: Leonardo Marques de Souza @ 2004-07-29 19:48 UTC (permalink / raw)
  To: qemu-devel

Great!

 Thanks André (and Jani ), it fix.




André Braga wrote:

>There's a bug in the floppy emulation code. And this patch (by Jani
>Monoses) is supposed to fix it (I guess... Unless the bug in memtest
>is triggered by a different event than failing to read the last
>sectors, but that's highly unlikely since it works OK from the ISO)
>
>
>- - - - - - - - - - - - - - - - - - - cut: fbblockpatch.diff - - - - -
>- - - - - - - - - - - - - - - - -
>
>  
>

^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: Re: SPAM[RBL] Re: [Qemu-devel] (TEST) memtest86+ & memtest86 & qemu 1.60
  2004-07-29 19:48   ` SPAM[RBL] " Leonardo Marques de Souza
@ 2004-07-29 19:56     ` André Braga
  0 siblings, 0 replies; 4+ messages in thread
From: André Braga @ 2004-07-29 19:56 UTC (permalink / raw)
  To: qemu-devel

Cool :)

Keep on testing, Léo.

On Thu, 29 Jul 2004 16:48:35 -0300, Leonardo Marques de Souza
<leonardo@ambientebrasil.com.br> wrote:
> Great!
> 
> Thanks André (and Jani ), it fix.
> 

-- 
"I invented the term Object-Oriented, and I can tell you I did not
have C++ in mind"
Alan Kay

^ permalink raw reply	[flat|nested] 4+ messages in thread

end of thread, other threads:[~2004-07-29 20:00 UTC | newest]

Thread overview: 4+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2004-07-29 15:11 [Qemu-devel] (TEST) memtest86+ & memtest86 & qemu 1.60 Leonardo Marques de Souza
2004-07-29 16:31 ` André Braga
2004-07-29 19:48   ` SPAM[RBL] " Leonardo Marques de Souza
2004-07-29 19:56     ` André Braga

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).