Skip site navigation (1)Skip section navigation (2)
Date:      Fri, 8 Mar 2002 16:09:10 -0500 (EST)
From:      kf <ken@cleveland.lug.net>
To:        Eric Dantan Rzewnicki <rzewnickie@rfa.org>
Cc:        <gibbs@FreeBSD.ORG>, <aic7xxx@FreeBSD.ORG>, <network@rfa.org>
Subject:   Re: performance issues: linux aic7xxx, 29160, Radion IFT-7200
Message-ID:  <Pine.LNX.4.33.0203081600250.10521-100000@cleveland.lug.net>
In-Reply-To: <3C891A1D.D757AC00@rfa.org>

next in thread | previous in thread | raw e-mail | index | archive | help

I'm not fluent enough in SCSI to say for sure, but there's been reports of
difficulties on machines using AMD CPUs, specifically the classic Athlons
(pre-XP models).  One problem is that it seems there was a change in the
memory paging routines on the Athlons which weren't incorporated into the
kernel.  Given that you're talking about a running machine, you're
probably not having these difficulties specifically, or else you've
already overcome them.  Can you try this configuration with a different
CPU installed?


kf

On Fri, 8 Mar 2002, Eric Dantan Rzewnicki wrote:

= Hello Justin,
=
= We are experiencing some performance issues with our external Radion
= raid device and the "new" aic7xxx drivers in linux kernel 2.4.x when
= compared with either 2.2.x kernels or 2.4.x kernels using the
= aic7xxx_old driver. Our hardware:
=
= AMD Athlon 1.3 GHz
= ASUS A7A266 motherboard (ALiMAGiK1 chipset: M1647 north bridge and
= M1535D+ south bridge)
= 1GB PC133 RAM
= Apaptec 29160 host adaptor
= Seagate ST318405LW 16GB internal system disk
= Radion IFT-7200 external hardware RAID (configured as RAID 5 w/5 150GB
= maxtor IDE drives)
=
=
= The performance of the Seagate drive on the internal connector of the
= 29160 is fairly consistent between the two driver versions. The Radion
= on the 29160's external connector is more than 3 times as fast with the
= old driver when compared to the new driver.
=
= Here are the relevent sections of /var/log/dmesg and hdparm output for
= driver version 6.2.5 and linux 2.4.18. (We got consistently similar
= results with aic7xxx version 6.2.4 with both linux 2.4.18 and 2.4.16):
=
= newbunyan:~# uname -r
= 2.4.18-aic7xxx-6.2.5
=
= SCSI subsystem driver Revision: 1.00
= PCI: Found IRQ 11 for device 00:0c.0
= PCI: Sharing IRQ 11 with 01:00.0
= scsi0 : Adaptec AIC7XXX EISA/VLB/PCI SCSI HBA DRIVER, Rev 6.2.5
=         <Adaptec 29160 Ultra160 SCSI adapter>
=         aic7892: Ultra160 Wide Channel A, SCSI Id=7, 32/253 SCBs
=
=   Vendor: SEAGATE   Model: ST318405LW        Rev: 5063
=   Type:   Direct-Access                      ANSI SCSI revision: 03
= (scsi0:A:0): 160.000MB/s transfers (80.000MHz DT, offset 63, 16bit)
=   Vendor: IFT       Model: IFT-7200          Rev: 0144
=   Type:   Direct-Access                      ANSI SCSI revision: 04
= (scsi0:A:1): 160.000MB/s transfers (80.000MHz DT, offset 62, 16bit)
=   Vendor: PLEXTOR   Model: CD-ROM PX-40TW    Rev: 1.04
=   Type:   CD-ROM                             ANSI SCSI revision: 02
= (scsi0:A:8): 40.000MB/s transfers (20.000MHz, offset 15, 16bit)
= scsi0:A:0:0: Tagged Queuing enabled.  Depth 8
= scsi0:A:1:0: Tagged Queuing enabled.  Depth 8
= Attached scsi disk sda at scsi0, channel 0, id 0, lun 0
= Attached scsi disk sdb at scsi0, channel 0, id 1, lun 0
= SCSI device sda: 35843670 512-byte hdwr sectors (18352 MB)
= Partition check:
=  sda: sda1 sda2
= SCSI device sdb: 1280654336 512-byte hdwr sectors (-443816 MB)
=  sdb: sdb1
=
= newbunyan:~# hdparm -t /dev/sda
=
= /dev/sda:
=  Timing buffered disk reads:  64 MB in  1.68 seconds = 38.10 MB/sec
=
= newbunyan:~# hdparm -t /dev/sdb
=
= /dev/sdb:
=  Timing buffered disk reads:  64 MB in  4.16 seconds = 15.38 MB/sec
=
=
= Here are the relevent sections of /var/log/dmesg and hdparm output for
= driver version 5.2.4 and linux 2.4.16:
=
= newbunyan:~# uname -r
= 2.4.16-bunyan-oldscsi-0
=
= SCSI subsystem driver Revision: 1.00
= PCI: Found IRQ 11 for device 00:0c.0
= PCI: Sharing IRQ 11 with 01:00.0
= (scsi0) <Adaptec AIC-7892 Ultra 160/m SCSI host adapter> found at PCI
= 0/12/0
= (scsi0) Wide Channel, SCSI ID=7, 32/255 SCBs
= (scsi0) Downloading sequencer code... 396 instructions downloaded
= scsi0 : Adaptec AHA274x/284x/294x (EISA/VLB/PCI-Fast SCSI) 5.2.4/5.2.0
=        <Adaptec AIC-7892 Ultra 160/m SCSI host adapter>
=   Vendor: SEAGATE   Model: ST318405LW        Rev: 5063
=   Type:   Direct-Access                      ANSI SCSI revision: 03
= (scsi0:0:0:1) Synchronous at 160.0 Mbyte/sec, offset 63.
=   Vendor: IFT       Model: IFT-7200          Rev: 0144
=   Type:   Direct-Access                      ANSI SCSI revision: 04
= (scsi0:0:1:1) Synchronous at 160.0 Mbyte/sec, offset 62.
=   Vendor: PLEXTOR   Model: CD-ROM PX-40TW    Rev: 1.04
=   Type:   CD-ROM                             ANSI SCSI revision: 02
= Attached scsi disk sda at scsi0, channel 0, id 0, lun 0
= Attached scsi disk sdb at scsi0, channel 0, id 1, lun 0
= SCSI device sda: 35843670 512-byte hdwr sectors (18352 MB)
= Partition check:
=  sda: sda1 sda2
= SCSI device sdb: 1280654336 512-byte hdwr sectors (-443816 MB)
=  sdb: sdb1
=
= newbunyan:~# hdparm -t /dev/sda
=
= /dev/sda:
=  Timing buffered disk reads:  64 MB in  1.68 seconds = 38.10 MB/sec
=
= newbunyan:~# hdparm -t /dev/sdb
=
= /dev/sdb:
=  Timing buffered disk reads:  64 MB in  1.35 seconds = 47.41 MB/sec
=
=
=
= Do you have any ideas on what could be causing this performance
= degradation with the new driver? We could conceivably just use the old
= driver, but since it is no longer being developed it seems better to try
= to find a solution for getting better performance from the new driver.
= Is there any other information we can provide to help in sorting this
= out?
=
= Thank you for writing this otherwise excellent driver and thanks in
= advance for any help you can give us.
=
= Sincerely,
= Eric Rzewnicki
= Radio Free Asia
= Network Information Services
=
= To Unsubscribe: send mail to majordomo@FreeBSD.org
= with "unsubscribe aic7xxx" in the body of the message
=

-- 
		 Cleveland Linux Users Group
		  http://cleveland.lug.net/


To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe aic7xxx" in the body of the message




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.LNX.4.33.0203081600250.10521-100000>