* Re: Memory performance on Serverworks GC-LE based system poor?
@ 2002-11-12 18:42 Martin Knoblauch
0 siblings, 0 replies; 9+ messages in thread
From: Martin Knoblauch @ 2002-11-12 18:42 UTC (permalink / raw)
To: alan; +Cc: linux-kernel
[-- Attachment #1: Type: text/plain, Size: 1106 bytes --]
> Re: Memory performance on Serverworks GC-LE based system poor?
>
>>
> On Mon, 2002-11-11 at 00:51, Dave Jones wrote:
> > On Mon, Nov 11, 2002 at 01:30:13AM +0100, Martin Knoblauch wrote:
> >
> > > I have experienced extreme low STREAMS numbers (about 600 MB/sec for Triad)
> > > on two dual CPU systems based on the ServerWorks GC-LE chipset (SuperMicro
> > > P4DLR+ mainboard). Both systems had 2x2.4 GHz XEONs, 4GB of DDR memory and
> > > were running kernel 2.4.18. I would usually expect STREAMS numbers of about
> > > 2000 MB/sec for this kind of systems.
> > >
> > > Does this ring any bells?
> >
> > ISTR serverworks LE errata with MTRRs and write-combining.
> > Whether this is biting you or not I can't say.
>
> Write combining would really only bite graphics cards. The only other
> performance errata I know about affects the CIOB20 earlier revisions
> (vendor serverworks id 0x0006)
>
does anybody know what the following ServerWorks device ids are? I
could not find them in the include file:
0012
0000
0101
Maybe their nonrecognition by the kernel is related to my problem?
Martin
[-- Attachment #2: lspci.marge1 --]
[-- Type: text/plain, Size: 8406 bytes --]
marge1$ lspci
00:00.0 Host bridge: ServerWorks: Unknown device 0012 (rev 13)
00:00.1 Host bridge: ServerWorks: Unknown device 0012
00:00.2 Host bridge: ServerWorks: Unknown device 0000
00:02.0 VGA compatible controller: ATI Technologies Inc Rage XL (rev 27)
00:04.0 Ethernet controller: Intel Corp. 82557/8/9 [Ethernet Pro 100] (rev 0d)
00:05.0 Ethernet controller: Intel Corp. 82557/8/9 [Ethernet Pro 100] (rev 0d)
00:0f.0 ISA bridge: ServerWorks CSB5 South Bridge (rev 93)
00:0f.1 IDE interface: ServerWorks CSB5 IDE Controller (rev 93)
00:0f.2 USB Controller: ServerWorks OSB4/CSB5 USB Controller (rev 05)
00:0f.3 Host bridge: ServerWorks: Unknown device 0225
00:11.0 Host bridge: ServerWorks: Unknown device 0101 (rev 03)
00:11.2 Host bridge: ServerWorks: Unknown device 0101 (rev 03)
02:02.0 SCSI storage controller: Adaptec AIC-7899P U160/m (rev 01)
02:02.1 SCSI storage controller: Adaptec AIC-7899P U160/m (rev 01)
marge1$ lspci -v
00:00.0 Host bridge: ServerWorks: Unknown device 0012 (rev 13)
Flags: fast devsel
00:00.1 Host bridge: ServerWorks: Unknown device 0012
Flags: fast devsel
00:00.2 Host bridge: ServerWorks: Unknown device 0000
Flags: fast devsel
00:02.0 VGA compatible controller: ATI Technologies Inc Rage XL (rev 27) (prog-if 00 [VGA])
Subsystem: ATI Technologies Inc Rage XL
Flags: bus master, stepping, medium devsel, latency 64, IRQ 18
Memory at fd000000 (32-bit, non-prefetchable) [size=16M]
I/O ports at c800 [size=256]
Memory at fe7ff000 (32-bit, non-prefetchable) [size=4K]
Expansion ROM at fe7c0000 [disabled] [size=128K]
Capabilities: <available only to root>
00:04.0 Ethernet controller: Intel Corp. 82557/8/9 [Ethernet Pro 100] (rev 0d)
Subsystem: Intel Corp. EtherExpress PRO/100 S Server Adapter
Flags: bus master, medium devsel, latency 64, IRQ 17
Memory at fe7fc000 (32-bit, non-prefetchable) [size=4K]
I/O ports at ce80 [size=64]
Memory at fe780000 (32-bit, non-prefetchable) [size=128K]
Expansion ROM at fe770000 [disabled] [size=64K]
Capabilities: <available only to root>
00:05.0 Ethernet controller: Intel Corp. 82557/8/9 [Ethernet Pro 100] (rev 0d)
Subsystem: Intel Corp. EtherExpress PRO/100 S Server Adapter
Flags: bus master, medium devsel, latency 64, IRQ 19
Memory at fe7fd000 (32-bit, non-prefetchable) [size=4K]
I/O ports at cf00 [size=64]
Memory at fe7a0000 (32-bit, non-prefetchable) [size=128K]
Expansion ROM at fe7e0000 [disabled] [size=64K]
Capabilities: <available only to root>
00:0f.0 ISA bridge: ServerWorks CSB5 South Bridge (rev 93)
Subsystem: Unknown device d915:5539
Flags: bus master, medium devsel, latency 64
00:0f.1 IDE interface: ServerWorks CSB5 IDE Controller (rev 93) (prog-if 8a [Master SecP PriP])
Subsystem: ServerWorks CSB5 IDE Controller
Flags: bus master, medium devsel, latency 64
I/O ports at 01f0 [size=8]
I/O ports at 03f4
I/O ports at 0170 [size=8]
I/O ports at 0374
I/O ports at ffa0 [size=16]
00:0f.2 USB Controller: ServerWorks OSB4/CSB5 USB Controller (rev 05) (prog-if 10 [OHCI])
Subsystem: ServerWorks OSB4/CSB5 USB Controller
Flags: bus master, medium devsel, latency 64, IRQ 10
Memory at fe7fe000 (32-bit, non-prefetchable) [size=4K]
00:0f.3 Host bridge: ServerWorks: Unknown device 0225
Subsystem: Unknown device d915:5539
Flags: bus master, medium devsel, latency 0
00:11.0 Host bridge: ServerWorks: Unknown device 0101 (rev 03)
Flags: 66Mhz, medium devsel
Capabilities: <available only to root>
00:11.2 Host bridge: ServerWorks: Unknown device 0101 (rev 03)
Flags: 66Mhz, medium devsel
Capabilities: <available only to root>
02:02.0 SCSI storage controller: Adaptec AIC-7899P U160/m (rev 01)
Subsystem: Unknown device d915:5539
Flags: bus master, 66Mhz, medium devsel, latency 64, IRQ 30
BIST result: 00
I/O ports at e400 [disabled] [size=256]
Memory at febfe000 (64-bit, non-prefetchable) [size=4K]
Expansion ROM at feba0000 [disabled] [size=128K]
Capabilities: <available only to root>
02:02.1 SCSI storage controller: Adaptec AIC-7899P U160/m (rev 01)
Subsystem: Unknown device d915:5539
Flags: bus master, 66Mhz, medium devsel, latency 64, IRQ 31
BIST result: 00
I/O ports at e800 [disabled] [size=256]
Memory at febff000 (64-bit, non-prefetchable) [size=4K]
Expansion ROM at febc0000 [disabled] [size=128K]
Capabilities: <available only to root>
marge1$ lspci -vn
00:00.0 Class 0600: 1166:0012 (rev 13)
Flags: fast devsel
00:00.1 Class 0600: 1166:0012
Flags: fast devsel
00:00.2 Class 0600: 1166:0000
Flags: fast devsel
00:02.0 Class 0300: 1002:4752 (rev 27)
Subsystem: 1002:0008
Flags: bus master, stepping, medium devsel, latency 64, IRQ 18
Memory at fd000000 (32-bit, non-prefetchable) [size=16M]
I/O ports at c800 [size=256]
Memory at fe7ff000 (32-bit, non-prefetchable) [size=4K]
Expansion ROM at fe7c0000 [disabled] [size=128K]
Capabilities: <available only to root>
00:04.0 Class 0200: 8086:1229 (rev 0d)
Subsystem: 8086:1050
Flags: bus master, medium devsel, latency 64, IRQ 17
Memory at fe7fc000 (32-bit, non-prefetchable) [size=4K]
I/O ports at ce80 [size=64]
Memory at fe780000 (32-bit, non-prefetchable) [size=128K]
Expansion ROM at fe770000 [disabled] [size=64K]
Capabilities: <available only to root>
00:05.0 Class 0200: 8086:1229 (rev 0d)
Subsystem: 8086:1050
Flags: bus master, medium devsel, latency 64, IRQ 19
Memory at fe7fd000 (32-bit, non-prefetchable) [size=4K]
I/O ports at cf00 [size=64]
Memory at fe7a0000 (32-bit, non-prefetchable) [size=128K]
Expansion ROM at fe7e0000 [disabled] [size=64K]
Capabilities: <available only to root>
00:0f.0 Class 0601: 1166:0201 (rev 93)
Subsystem: d915:5539
Flags: bus master, medium devsel, latency 64
00:0f.1 Class 0101: 1166:0212 (rev 93) (prog-if 8a [Master SecP PriP])
Subsystem: 1166:0212
Flags: bus master, medium devsel, latency 64
I/O ports at 01f0 [size=8]
I/O ports at 03f4
I/O ports at 0170 [size=8]
I/O ports at 0374
I/O ports at ffa0 [size=16]
00:0f.2 Class 0c03: 1166:0220 (rev 05) (prog-if 10)
Subsystem: 1166:0220
Flags: bus master, medium devsel, latency 64, IRQ 10
Memory at fe7fe000 (32-bit, non-prefetchable) [size=4K]
00:0f.3 Class 0600: 1166:0225
Subsystem: d915:5539
Flags: bus master, medium devsel, latency 0
00:11.0 Class 0600: 1166:0101 (rev 03)
Flags: 66Mhz, medium devsel
Capabilities: <available only to root>
00:11.2 Class 0600: 1166:0101 (rev 03)
Flags: 66Mhz, medium devsel
Capabilities: <available only to root>
02:02.0 Class 0100: 9005:00cf (rev 01)
Subsystem: d915:5539
Flags: bus master, 66Mhz, medium devsel, latency 64, IRQ 30
BIST result: 00
I/O ports at e400 [disabled] [size=256]
Memory at febfe000 (64-bit, non-prefetchable) [size=4K]
Expansion ROM at feba0000 [disabled] [size=128K]
Capabilities: <available only to root>
02:02.1 Class 0100: 9005:00cf (rev 01)
Subsystem: d915:5539
Flags: bus master, 66Mhz, medium devsel, latency 64, IRQ 31
BIST result: 00
I/O ports at e800 [disabled] [size=256]
Memory at febff000 (64-bit, non-prefetchable) [size=4K]
Expansion ROM at febc0000 [disabled] [size=128K]
Capabilities: <available only to root>
marge1$ lspci -n
00:00.0 Class 0600: 1166:0012 (rev 13)
00:00.1 Class 0600: 1166:0012
00:00.2 Class 0600: 1166:0000
00:02.0 Class 0300: 1002:4752 (rev 27)
00:04.0 Class 0200: 8086:1229 (rev 0d)
00:05.0 Class 0200: 8086:1229 (rev 0d)
00:0f.0 Class 0601: 1166:0201 (rev 93)
00:0f.1 Class 0101: 1166:0212 (rev 93)
00:0f.2 Class 0c03: 1166:0220 (rev 05)
00:0f.3 Class 0600: 1166:0225
00:11.0 Class 0600: 1166:0101 (rev 03)
00:11.2 Class 0600: 1166:0101 (rev 03)
02:02.0 Class 0100: 9005:00cf (rev 01)
02:02.1 Class 0100: 9005:00cf (rev 01)
^ permalink raw reply [flat|nested] 9+ messages in thread
* RE: Memory performance on Serverworks GC-LE based system poor?
@ 2002-11-11 12:04 Martin Knoblauch
0 siblings, 0 replies; 9+ messages in thread
From: Martin Knoblauch @ 2002-11-11 12:04 UTC (permalink / raw)
To: manish; +Cc: linux-kernel
>
> I, too get 1600 using the Force 4203 motherboard with GC-LE chipset. Same
> CPU and RAM ...
>
> Thanks
> Manish
hmm. 1600 sounds a lot more like, although not really impressive IMO.
Of course I cwould be happy to see those numbers. :-(
Martin
>
> -----Original Message-----
> From: Craig Whitmore [mailto:linuxkernel@orcon.net.nz]
> Sent: Sunday, November 10, 2002 5:23 PM
> To: Alan Cox; Dave Jones
> Cc: Martin Knoblauch; Linux Kernel Mailing List
> Subject: Re: Memory performance on Serverworks GC-LE based system poor?
>
> I get 1600 on a similar machine but with a Intel SHG2 motherboard
> (ServerWorks GC LE Chipset as well) (Same RAM + Processors)
> Maybe not a very good Motherboard you have?
>
> Thanks
> Craig
>
^ permalink raw reply [flat|nested] 9+ messages in thread
* RE: Memory performance on Serverworks GC-LE based system poor?
@ 2002-11-11 9:47 Martin Knoblauch
0 siblings, 0 replies; 9+ messages in thread
From: Martin Knoblauch @ 2002-11-11 9:47 UTC (permalink / raw)
To: manish; +Cc: linux-kernel
>
> You must have done this, but just wanted to confirm. Did you check the Front
> Side Bus speed in the BIOS?
>
> Thanks
> Manish
>
Actually, we did check a lot of things, but did not see settings for
the FSB. What would the correct speed be?
Thanks
Martin
> -----Original Message-----
> From: Martin Knoblauch [mailto:knobi@knobisoft.de]
> Sent: Sunday, November 10, 2002 4:30 PM
> To: linux-kernel@vger.kernel.org
> Subject: Memory performance on Serverworks GC-LE based system poor?
>
> Hi,
>
> I have experienced extreme low STREAMS numbers (about 600 MB/sec for Triad)
>
> on two dual CPU systems based on the ServerWorks GC-LE chipset (SuperMicro
> P4DLR+ mainboard). Both systems had 2x2.4 GHz XEONs, 4GB of DDR memory and
> were running kernel 2.4.18. I would usually expect STREAMS numbers of about
> 2000 MB/sec for this kind of systems.
>
> Does this ring any bells?
>
> Martin
>
^ permalink raw reply [flat|nested] 9+ messages in thread
* RE: Memory performance on Serverworks GC-LE based system poor?
@ 2002-11-11 1:39 Manish Lachwani
0 siblings, 0 replies; 9+ messages in thread
From: Manish Lachwani @ 2002-11-11 1:39 UTC (permalink / raw)
To: 'knobi@knobisoft.de', linux-kernel
You must have done this, but just wanted to confirm. Did you check the Front
Side Bus speed in the BIOS?
Thanks
Manish
-----Original Message-----
From: Martin Knoblauch [mailto:knobi@knobisoft.de]
Sent: Sunday, November 10, 2002 4:30 PM
To: linux-kernel@vger.kernel.org
Subject: Memory performance on Serverworks GC-LE based system poor?
Hi,
I have experienced extreme low STREAMS numbers (about 600 MB/sec for Triad)
on two dual CPU systems based on the ServerWorks GC-LE chipset (SuperMicro
P4DLR+ mainboard). Both systems had 2x2.4 GHz XEONs, 4GB of DDR memory and
were running kernel 2.4.18. I would usually expect STREAMS numbers of about
2000 MB/sec for this kind of systems.
Does this ring any bells?
Martin
--
----------------------------------
Martin Knoblauch
knobi@knobisoft.de
http://www.knobisoft.de
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
^ permalink raw reply [flat|nested] 9+ messages in thread
* RE: Memory performance on Serverworks GC-LE based system poor?
@ 2002-11-11 1:37 Manish Lachwani
0 siblings, 0 replies; 9+ messages in thread
From: Manish Lachwani @ 2002-11-11 1:37 UTC (permalink / raw)
To: 'Craig Whitmore', Alan Cox, Dave Jones
Cc: Martin Knoblauch, Linux Kernel Mailing List
I, too get 1600 using the Force 4203 motherboard with GC-LE chipset. Same
CPU and RAM ...
Thanks
Manish
-----Original Message-----
From: Craig Whitmore [mailto:linuxkernel@orcon.net.nz]
Sent: Sunday, November 10, 2002 5:23 PM
To: Alan Cox; Dave Jones
Cc: Martin Knoblauch; Linux Kernel Mailing List
Subject: Re: Memory performance on Serverworks GC-LE based system poor?
I get 1600 on a similar machine but with a Intel SHG2 motherboard
(ServerWorks GC LE Chipset as well) (Same RAM + Processors)
Maybe not a very good Motherboard you have?
Thanks
Craig
> >
> > > I have experienced extreme low STREAMS numbers (about 600 MB/sec for
Triad)
> > > on two dual CPU systems based on the ServerWorks GC-LE chipset
(SuperMicro
> > > P4DLR+ mainboard). Both systems had 2x2.4 GHz XEONs, 4GB of DDR
memory and
> > > were running kernel 2.4.18. I would usually expect STREAMS numbers of
about
> > > 2000 MB/sec for this kind of systems.
> > >
> > > Does this ring any bells?
> >
> > ISTR serverworks LE errata with MTRRs and write-combining.
> > Whether this is biting you or not I can't say.
>
> Write combining would really only bite graphics cards. The only other
> performance errata I know about affects the CIOB20 earlier revisions
> (vendor serverworks id 0x0006)
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
^ permalink raw reply [flat|nested] 9+ messages in thread
* Re: Memory performance on Serverworks GC-LE based system poor?
2002-11-11 0:51 ` Dave Jones
@ 2002-11-11 1:33 ` Alan Cox
2002-11-11 1:22 ` Craig Whitmore
0 siblings, 1 reply; 9+ messages in thread
From: Alan Cox @ 2002-11-11 1:33 UTC (permalink / raw)
To: Dave Jones; +Cc: Martin Knoblauch, Linux Kernel Mailing List
On Mon, 2002-11-11 at 00:51, Dave Jones wrote:
> On Mon, Nov 11, 2002 at 01:30:13AM +0100, Martin Knoblauch wrote:
>
> > I have experienced extreme low STREAMS numbers (about 600 MB/sec for Triad)
> > on two dual CPU systems based on the ServerWorks GC-LE chipset (SuperMicro
> > P4DLR+ mainboard). Both systems had 2x2.4 GHz XEONs, 4GB of DDR memory and
> > were running kernel 2.4.18. I would usually expect STREAMS numbers of about
> > 2000 MB/sec for this kind of systems.
> >
> > Does this ring any bells?
>
> ISTR serverworks LE errata with MTRRs and write-combining.
> Whether this is biting you or not I can't say.
Write combining would really only bite graphics cards. The only other
performance errata I know about affects the CIOB20 earlier revisions
(vendor serverworks id 0x0006)
^ permalink raw reply [flat|nested] 9+ messages in thread
* Re: Memory performance on Serverworks GC-LE based system poor?
2002-11-11 1:33 ` Alan Cox
@ 2002-11-11 1:22 ` Craig Whitmore
0 siblings, 0 replies; 9+ messages in thread
From: Craig Whitmore @ 2002-11-11 1:22 UTC (permalink / raw)
To: Alan Cox, Dave Jones; +Cc: Martin Knoblauch, Linux Kernel Mailing List
I get 1600 on a similar machine but with a Intel SHG2 motherboard
(ServerWorks GC LE Chipset as well) (Same RAM + Processors)
Maybe not a very good Motherboard you have?
Thanks
Craig
> >
> > > I have experienced extreme low STREAMS numbers (about 600 MB/sec for
Triad)
> > > on two dual CPU systems based on the ServerWorks GC-LE chipset
(SuperMicro
> > > P4DLR+ mainboard). Both systems had 2x2.4 GHz XEONs, 4GB of DDR
memory and
> > > were running kernel 2.4.18. I would usually expect STREAMS numbers of
about
> > > 2000 MB/sec for this kind of systems.
> > >
> > > Does this ring any bells?
> >
> > ISTR serverworks LE errata with MTRRs and write-combining.
> > Whether this is biting you or not I can't say.
>
> Write combining would really only bite graphics cards. The only other
> performance errata I know about affects the CIOB20 earlier revisions
> (vendor serverworks id 0x0006)
^ permalink raw reply [flat|nested] 9+ messages in thread
* Re: Memory performance on Serverworks GC-LE based system poor?
2002-11-11 0:30 Martin Knoblauch
@ 2002-11-11 0:51 ` Dave Jones
2002-11-11 1:33 ` Alan Cox
0 siblings, 1 reply; 9+ messages in thread
From: Dave Jones @ 2002-11-11 0:51 UTC (permalink / raw)
To: Martin Knoblauch; +Cc: linux-kernel
On Mon, Nov 11, 2002 at 01:30:13AM +0100, Martin Knoblauch wrote:
> I have experienced extreme low STREAMS numbers (about 600 MB/sec for Triad)
> on two dual CPU systems based on the ServerWorks GC-LE chipset (SuperMicro
> P4DLR+ mainboard). Both systems had 2x2.4 GHz XEONs, 4GB of DDR memory and
> were running kernel 2.4.18. I would usually expect STREAMS numbers of about
> 2000 MB/sec for this kind of systems.
>
> Does this ring any bells?
ISTR serverworks LE errata with MTRRs and write-combining.
Whether this is biting you or not I can't say.
Dave
--
| Dave Jones. http://www.codemonkey.org.uk
| SuSE Labs
^ permalink raw reply [flat|nested] 9+ messages in thread
* Memory performance on Serverworks GC-LE based system poor?
@ 2002-11-11 0:30 Martin Knoblauch
2002-11-11 0:51 ` Dave Jones
0 siblings, 1 reply; 9+ messages in thread
From: Martin Knoblauch @ 2002-11-11 0:30 UTC (permalink / raw)
To: linux-kernel
Hi,
I have experienced extreme low STREAMS numbers (about 600 MB/sec for Triad)
on two dual CPU systems based on the ServerWorks GC-LE chipset (SuperMicro
P4DLR+ mainboard). Both systems had 2x2.4 GHz XEONs, 4GB of DDR memory and
were running kernel 2.4.18. I would usually expect STREAMS numbers of about
2000 MB/sec for this kind of systems.
Does this ring any bells?
Martin
--
----------------------------------
Martin Knoblauch
knobi@knobisoft.de
http://www.knobisoft.de
^ permalink raw reply [flat|nested] 9+ messages in thread
end of thread, other threads:[~2002-11-12 18:36 UTC | newest]
Thread overview: 9+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2002-11-12 18:42 Memory performance on Serverworks GC-LE based system poor? Martin Knoblauch
-- strict thread matches above, loose matches on Subject: below --
2002-11-11 12:04 Martin Knoblauch
2002-11-11 9:47 Martin Knoblauch
2002-11-11 1:39 Manish Lachwani
2002-11-11 1:37 Manish Lachwani
2002-11-11 0:30 Martin Knoblauch
2002-11-11 0:51 ` Dave Jones
2002-11-11 1:33 ` Alan Cox
2002-11-11 1:22 ` Craig Whitmore
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).