All of lore.kernel.org
 help / color / mirror / Atom feed
From: Markus <markus@kola.li>
To: Nagalakshmi Nandigama <Nagalakshmi.Nandigama@lsi.com>,
	Sreekanth Reddy <Sreekanth.Reddy@lsi.com>,
	support@lsi.com, DL-MPTFusionLinux@lsi.com,
	linux-scsi@vger.kernel.org, linux-kernel@vger.kernel.org
Subject: Slow I/O performance on SAS1064
Date: Wed, 5 Mar 2014 18:50:37 +0100	[thread overview]
Message-ID: <20140305175036.GA23885@localhost.localdomain> (raw)

[-- Attachment #1: Type: text/plain, Size: 5346 bytes --]

Hi

I have problem with SATA disks at my Sunfire v245 and its lsi controller they are very slow.

So I test it with dd and while dd in a nother terminal i see dstat results of sda:


Copy from ramdisk to sda.
=== Test ext4 ===
DD

root@outpost:/ramdisk# dd if=debian-7.4.0-sparc-netinst.iso of=/root/test.dd
353344+0 records in
353344+0 records out
180912128 bytes (181 MB) copied, 24.6384 s, 7.3 MB/s

Dstat:

----total-cpu-usage---- --dsk/sda-- ---io/sda--
usr sys idl wai hiq siq| read  writ| read  writ
  1   0 100   0   0   0|   0     0 |   0     0 
  0   0 100   0   0   0|   0     0 |   0     0 
  2   1  97   0   0   0|   0     0 |   0     0 
  1   5  94   0   0   0|   0     0 |   0     0 
  7  43  50   0   0   0|   0     0 |   0     0 
  7  43  50   0   0   0|   0     0 |   0     0 
 10  40  50   0   0   0|   0     0 |   0     0 
 12  45  41   3   0   0|   0     0 |   0     0 
  6  42  36  17   0   1|   0   512k|   0  1.00 
  0   0   0  99   0   0|   0  4096k|   0  8.00 
  0   2   0  98   0   0|   0  5120k|   0  10.0 
  0   1   0  99   0   0|   0  7680k|   0  15.0 
  1   0   0 100   0   0|   0  5632k|   0  11.0 
  0   0   0  99   0   0|   0  7168k|   0  14.0 
  1   1   0  99   0   0|   0  6656k|   0  13.0 
  0   0   0 100   0   0|   0  6144k|   0  12.0 
  1   0   0 100   0   0|   0  7680k|   0  15.0 
  0   1  22  78   0   0|   0  4608k|   0  9.00 
  0   0  50  50   0   0|   0  3072k|   0  6.00 
  1   0  50  50   0   1|   0  7168k|   0  14.0 
  1   1  50  49   0   0|4096B 5120k|1.00  10.0 
  0   0  50  50   0   0|   0  8704k|   0  17.0 
  0   0  50  49   0   0|   0  7680k|   0  15.0 
  0   1  50  50   0   0|   0  7168k|   0  14.0 
  1   0  49  49   0   0|   0  7168k|   0  14.0 
  1   0  50  50   0   0|   0  7680k|   0  15.0 
  0   0  50  50   0   0|   0  7168k|   0  14.0 
  0   0  50  50   0   0|   0  2560k|   0  5.00 
  0   1  30  69   0   0|   0  3624k|   0  8.00 
  1   0   0  99   0   0|   0  4100k|   0  9.00 
  0   0   0 100   0   0|   0  5632k|   0  11.0 
  0   0   0  99   0   0|   0  3584k|   0  7.00 
  1   0   0 100   0   0|   0  4608k|   0  9.00 
  0   0   0  99   0   1|   0  5120k|   0  10.0 
  1   0   4  96   0   0|   0  2560k|   0  6.00 
  0   0  50  50   0   0|   0  3072k|   0  6.00 
  0   0  49  50   0   0|   0  5120k|   0  10.0 
  1   0  50  50   0   0|   0  5120k|   0  10.0 
  0   0  50  50   0   0|   0  4096k|   0  8.00 
  1   0  50  50   0   0|   0  5120k|   0  10.0 
  0   0  50  50   0   0|   0  5120k|   0  10.0 
  0   1  57  42   0   0|   0  1125k|   0  17.0 
  0   0 100   0   0   0|   0     0 |   0     0 
  1   0 100   0   0   0|   0     0 |   0     0 
  1   0 100   0   0   0|   0     0 |   0     0 
  0   0 100   0   0   0|   0     0 |   0     0 
  0   0 100   0   0   0|   0     0 |   0     0

-------------------------------------------------------------------------------
Information:
Kernel 3.13.5 (stable release)
Debian7 (stable)
hdparm v9.39


Controller 
LSI SAS1064
	cat /proc/mpt/ioc0/info 
	ioc0:
	  ProductID = 0x2701 (LSISAS1064 A3)
	  FWVersion = 0x01080400
	  MsgVersion = 0x0105
	  FirstWhoInit = 0x00
	  EventState = 0x00
	  CurrentHostMfaHighAddr = 0x00000000
	  CurrentSenseBufferHighAddr = 0x00000000
	  MaxChainDepth = 0x60 frames
	  MinBlockSize = 0x20 bytes
	  RequestFrames @ 0xfffffc137f602800 (Dma @ 0x00000000c000a800)
	    {CurReqSz=128} x {CurReqDepth=511} = 65408 bytes ^= 0x10000
	    {MaxReqSz=128}   {MaxReqDepth=511}
	  Frames   @ 0xfffffc137f600000 (Dma @ 0x00000000c0008000)
	    {CurRepSz=80} x {CurRepDepth=128} = 10240 bytes ^= 0x2880
	    {MaxRepSz=0}   {MaxRepDepth=511}
	  MaxDevices = 63
	  MaxBuses = 1
	  PortNumber = 1 (of 1)

	cat /proc/mpt/version 
		mptlinux-3.04.20
		  Fusion MPT base driver
		  Fusion MPT SAS host driver
		  Fusion MPT ioctl driver

-------------------------------------------------------------------------------
Hdparm Information:
	hdparm -I /dev/sda
	
	/dev/sda:
	SG_IO: bad/missing sense data, sb[]:  70 00 05 00 00 00 00 0a 00 00 00 00 20 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
	
	ATA device, with non-removable media
	Standards:
	        Likely used: 1
	Configuration:
	        Logical         max     current
	        cylinders       0       0
	        heads           0       0
	        sectors/track   0       0
	        --
	        Logical/Physical Sector size:           512 bytes
	        device size with M = 1024*1024:           0 MBytes
	        device size with M = 1000*1000:           0 MBytes 
	        cache/buffer size  = unknown
	Capabilities:
	        IORDY not likely
	        Cannot perform double-word IO
	        R/W multiple sector transfer: not supported
	        DMA: not supported
			PIO: pio0  

-------------------------------------------------------------------------------
Moduls load:
lsmod | grep mpt
	mptctl                 27359  0 
	mptsas                 45861  5 
	mptscsih               25765  1 mptsas
	mptbase                75530  3 mptctl,mptsas,mptscsih
	scsi_transport_sas     21624  1 mptsas

-------------------------------------------------------------------------------
The hdparm result looks like there is somethink not right . There were no features supported but why ?


Greetings Markus

[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 819 bytes --]

             reply	other threads:[~2014-03-05 17:50 UTC|newest]

Thread overview: 6+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2014-03-05 17:50 Markus [this message]
2014-03-05 18:21 ` Slow I/O performance on SAS1064 Mark Knecht
2014-03-05 21:40   ` markus
2014-03-06 20:51     ` Mark Knecht
2014-03-07  9:46       ` markus
     [not found] <CAN7X1Umb_wyTCk9w590AnSTFgL=3CC5x8zGNiGZoAdNEej9TCQ@mail.gmail.com>
2014-03-08 20:14 ` markus

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20140305175036.GA23885@localhost.localdomain \
    --to=markus@kola.li \
    --cc=DL-MPTFusionLinux@lsi.com \
    --cc=Nagalakshmi.Nandigama@lsi.com \
    --cc=Sreekanth.Reddy@lsi.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-scsi@vger.kernel.org \
    --cc=support@lsi.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.