freebsd-amd64 Digest, Vol 171, Issue 4
freebsd-am...@freebsd.org
freebs...@freebsd.org
To subscribe or unsubscribe via the World Wide Web, visit
http://lists.freebsd.org/mailman/listinfo/freebsd-amd64
or, via email, send a message with subject or body 'help' to
freebsd-am...@freebsd.org
You can reach the person managing the list at
freebsd-a...@freebsd.org
When replying, please edit your Subject line so it is more specific
than "Re: Contents of freebsd-amd64 digest..."
Today's Topics:
1. testing RAID perf. & choosing a motherboard (Michael Bernstein)
2. hardware donation ? (was Re: SMP system not running SMP)
(Pete French)
3. Re: amd64/95167: driver for SuperMicro H8DAR-T (Adaptec
AIC-8130: (Marvell 88SX6041) 4-port Hercules-2 (Anthony Lau)
4. Re: amd64/95056: Nvidia Nforce ETH Driver -> Timeout Errors
(Remko Lodder)
5. Re: keyboard LEDs CAPS and NUM problem (Zhu Sha Zang)
6. Re: amd64/102996: powerpc cross-build fails on amd64 host,
works fine on i386 (Ruslan Ermilov)
7. Re: amd64/102996: powerpc cross-build fails on amd64 host,
works fine on i386 (Ruslan Ermilov)
8. [SOLVED IN -CURRENT] Interrupt storm on ASUS M2N-E
(Angelo Turetta)
----------------------------------------------------------------------
Message: 1
Date: Wed, 13 Sep 2006 05:19:50 -0700 (PDT)
From: Michael Bernstein <mb_...@yahoo.com>
Subject: testing RAID perf. & choosing a motherboard
To: freebs...@freebsd.org
Message-ID: <2006091312195...@web33413.mail.mud.yahoo.com>
Content-Type: text/plain; charset=iso-8859-1
Hi, I know there's been quite some talk on RAID performance. Can someone here please tell me how you specifically test disk perf?
Also, I want to buy another motherboard for use for AMD 64 chip that I have - Athlon 64. Currently, I have an ATI MB, it's not the best. Do you think I should get a Tyan - again, this is for an Athlon 64 2GHz.
Thanks in advance.
Michael
CCIE Security 16395
---------------------------------
How low will we go? Check out Yahoo! Messenger’s low PC-to-Phone call rates.
------------------------------
Message: 2
Date: Wed, 13 Sep 2006 14:46:32 +0100
From: Pete French <petef...@ticketswitch.com>
Subject: hardware donation ? (was Re: SMP system not running SMP)
To: freebs...@FreeBSD.org
Message-ID: <E1GNV4S-...@dilbert.firstcallgroup.co.uk>
I;m resurrecting this thread because I got to the poitn where I
decided to replace the motherboard on which I was having the issue.
So as of this weekend I will have a spare motherboard which exhibits
the problem. If anyone is interested in working on the SMP problem then
I would be happy to donate the board to the cause. If anyone here
wants to work on this, or can suggest anyone in the SMP camp
whoc might be interested, then please let me know.
thanks,
-pete french.
------------------------------
Message: 3
Date: Wed, 13 Sep 2006 17:10:21 GMT
From: "Anthony Lau" <to...@asacomputers.com>
Subject: Re: amd64/95167: driver for SuperMicro H8DAR-T (Adaptec
AIC-8130: (Marvell 88SX6041) 4-port Hercules-2
To: freebs...@FreeBSD.org
Message-ID: <200609131710....@freefall.freebsd.org>
The following reply was made to PR amd64/95167; it has been noted by GNATS.
From: "Anthony Lau" <to...@asacomputers.com>
To: <bug-fo...@FreeBSD.org>, "Anthony Lau" <to...@asacomputers.com>
Cc:
Subject: Re: amd64/95167: driver for SuperMicro H8DAR-T (Adaptec AIC-8130: (Marvell 88SX6041) 4-port Hercules-2
Date: Wed, 13 Sep 2006 10:01:08 -0700
Hi all,
Unfortunately, the machine still crash with this driver. When this driver
was developed, BSD6.1 was under Beta testing and I haven't get a chance to
test it after 6.1 is officially release.
Thanks~
Anthony Lau
------------------------------
Message: 4
Date: Wed, 13 Sep 2006 18:50:06 GMT
From: Remko Lodder <re...@FreeBSD.org>
Subject: Re: amd64/95056: Nvidia Nforce ETH Driver -> Timeout Errors
To: linu...@gmail.com, re...@FreeBSD.org, freebs...@FreeBSD.org
Message-ID: <200609131850....@freefall.freebsd.org>
Synopsis: Nvidia Nforce ETH Driver -> Timeout Errors
State-Changed-From-To: feedback->closed
State-Changed-By: remko
State-Changed-When: Wed Sep 13 18:50:06 UTC 2006
State-Changed-Why:
The submitter moved on to Gentoo Linux and cannot test the presented
patch, close the PR since we cannot determine if it will work. If
someone has feedback, please contact me so that we can look into it. To
the submitter, we hope to see you back someday! Thanks for using FreeBSD
http://www.freebsd.org/cgi/query-pr.cgi?pr=95056
------------------------------
Message: 5
Date: Wed, 13 Sep 2006 16:10:33 -0300
From: Zhu Sha Zang <zhush...@yahoo.com.br>
Subject: Re: keyboard LEDs CAPS and NUM problem
To: freebs...@freebsd.org
Message-ID: <ee9l31$kp2$1...@sea.gmane.org>
Content-Type: text/plain; charset="iso-8859-1"
Gobbledegeek wrote:
> Hi
> My 6.1 has a keyboard driver problem - CAPS and NUM LEDs only
> display if I press them alternately twice or thrice in a row. They do
> not work otherwise... is this a known issue?
>
>
>
Hey, i have this problem too. But think this is problem on my keyboard.
SUCH A PITY!!!!
------------------------------
Message: 6
Date: Thu, 14 Sep 2006 07:50:27 GMT
From: Ruslan Ermilov <r...@FreeBSD.org>
Subject: Re: amd64/102996: powerpc cross-build fails on amd64 host,
works fine on i386
To: freebs...@FreeBSD.org
Message-ID: <200609140750....@freefall.freebsd.org>
The following reply was made to PR amd64/102996; it has been noted by GNATS.
From: Ruslan Ermilov <r...@FreeBSD.org>
To: bug-fo...@FreeBSD.org
Cc: John Baldwin <j...@FreeBSD.org>, Peter Grehan <gre...@FreeBSD.org>
Subject: Re: amd64/102996: powerpc cross-build fails on amd64 host, works fine on i386
Date: Thu, 14 Sep 2006 11:49:37 +0400
On Fri, Sep 08, 2006 at 11:10:52AM -0400, John Baldwin wrote:
> On Thursday 07 September 2006 17:51, Peter Grehan wrote:
> > %%%
> > Index: Makefile
> > ===================================================================
> > RCS file: /home/ncvs/src/gnu/usr.bin/binutils/libbfd/Makefile,v
> > retrieving revision 1.33
> > diff -u -r1.33 Makefile
> > --- Makefile 24 Oct 2004 15:32:25 -0000 1.33
> > +++ Makefile 26 Aug 2006 00:48:26 -0000
> > @@ -48,6 +48,12 @@
> > CFLAGS+= -DSELECT_VECS="${SELVEC}"
> > CFLAGS+= -DDEFAULT_VECTOR=${DEFAULT_VECTOR}
> >
> > +.if ${TARGET_ARCH:M*64}
> > +CFLAGS+= -DTARGET_64BIT=1
> > +.else
> > +CFLAGS+= -DTARGET_64BIT=0
> > +.endif
> > +
>
> This doesn't match alpha (though that only matters if you MFC this). Seems
> that a more general purpose solution would be to test TARGET_ARCH and use
> that to set TARGET properties
>
I just quickly prototyped the patch back then. How's this instead?
%%%
Index: Makefile
===================================================================
RCS file: /home/ncvs/src/gnu/usr.bin/binutils/libbfd/Makefile,v
retrieving revision 1.33
diff -u -p -u -r1.33 Makefile
--- Makefile 24 Oct 2004 15:32:25 -0000 1.33
+++ Makefile 14 Sep 2006 07:45:19 -0000
@@ -48,6 +48,14 @@ SELVEC+= ,&${_v}
CFLAGS+= -DSELECT_VECS="${SELVEC}"
CFLAGS+= -DDEFAULT_VECTOR=${DEFAULT_VECTOR}
+.if ${DEFAULT_VECTOR:Mbfd_elf64_*}
+CFLAGS+= -DTARGET_64BIT=1
+.elif ${DEFAULT_VECTOR:Mbfd_elf32_*}
+CFLAGS+= -DTARGET_64BIT=0
+.else
+.error unknown DEFAULT_VECTOR=${DEFAULT_VECTOR}
+.endif
+
# XXX:DEO should grab BFD_VERSION_DATE from ${VERSION}...
bfdver.h: Makefile
echo '#define BFD_VERSION 215000000' > ${.TARGET}
Index: bfd.h
===================================================================
RCS file: /home/ncvs/src/gnu/usr.bin/binutils/libbfd/bfd.h,v
retrieving revision 1.3
diff -u -p -u -r1.3 bfd.h
--- bfd.h 8 Jul 2004 17:05:32 -0000 1.3
+++ bfd.h 14 Sep 2006 07:34:28 -0000
@@ -57,15 +57,18 @@ extern "C" {
#define BFD_ARCH_SIZE 64
-#if defined(__i386__) || defined(__powerpc__) || defined(__arm__)
/* The word size of the default bfd target. */
+#if TARGET_64BIT
+#define BFD_DEFAULT_TARGET_SIZE 64
+#else
#define BFD_DEFAULT_TARGET_SIZE 32
+#endif
+
+#if defined(__i386__) || defined(__powerpc__) || defined(__arm__)
#define BFD_HOST_64BIT_LONG 0
#define BFD_HOST_64_BIT long long
#define BFD_HOST_U_64_BIT unsigned long long
#elif defined(__alpha__) || defined(__sparc64__) || defined(__amd64__) || defined(__ia64__)
-/* The word size of the default bfd target. */
-#define BFD_DEFAULT_TARGET_SIZE 64
#define BFD_HOST_64BIT_LONG 1
#define BFD_HOST_64_BIT long
#define BFD_HOST_U_64_BIT unsigned long
%%%
> (i.e. the bug seems to be that it's setting
> TARGET properties based on the host arch, which is wrong).
>
Exactly!
Cheers,
--
Ruslan Ermilov
r...@FreeBSD.org
FreeBSD committer
------------------------------
Message: 7
Date: Thu, 14 Sep 2006 08:08:08 GMT
From: Ruslan Ermilov <r...@FreeBSD.org>
Subject: Re: amd64/102996: powerpc cross-build fails on amd64 host,
works fine on i386
To: r...@FreeBSD.org, freebs...@FreeBSD.org, obr...@FreeBSD.org
Message-ID: <200609140808....@freefall.freebsd.org>
Synopsis: powerpc cross-build fails on amd64 host, works fine on i386
Responsible-Changed-From-To: freebsd-amd64->obrien
Responsible-Changed-By: ru
Responsible-Changed-When: Thu Sep 14 08:07:01 UTC 2006
Responsible-Changed-Why:
Pass over to the BU maintainer.
http://www.freebsd.org/cgi/query-pr.cgi?pr=102996
------------------------------
Message: 8
Date: Thu, 14 Sep 2006 13:16:53 +0200
From: Angelo Turetta <atur...@commit.it>
Subject: [SOLVED IN -CURRENT] Interrupt storm on ASUS M2N-E
To: freebs...@freebsd.org
Cc: Frank Reppin <fr...@undermydesk.org>, freebsd...@freebsd.org
Message-ID: <45093A2...@commit.it>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Frank Reppin wrote:
> Angelo Turetta wrote:
>> Could you please post the result of 'vmstat -i' from -CURRENT ?
>
> interrupt total rate
> irq23: atapci1 242237559 218625 <= Arrr!
> cpu0: timer 2214521 1998
> cpu1: timer 2214030 1998
> Total 246693998 222648
>
I've not had the time to identify any committer to thank in particular,
but I inform you that a 7-CURRENT release CD home-built on 2006-09-11
does not show this interrupt storm anymore on my motherboard.
I'll try 6-STABLE this evening and will report back.
Thanks to all involved.
Angelo.
------------------------------
End of freebsd-amd64 Digest, Vol 171, Issue 4
*********************************************