Linux SCSI subsystem development
 help / color / mirror / Atom feed
From: Joe Jin <joe.jin@oracle.com>
To: huangj@Brocade.COM
Cc: linux-scsi@vger.kernel.org, soeren.binner@oracle.com,
	haobo.zhou@oracle.com
Subject: 答复: RE: bfa driver on xen
Date: Fri, 30 Apr 2010 17:40:45 -0700 (PDT)	[thread overview]
Message-ID: <be817d0d-1139-4f0f-86e6-0dd709e38f40@default> (raw)

Jing, thanks for you suggestion.

----- 原始邮件 -----
发件人: huangj@Brocade.COM
收件人: joe.jin@oracle.com, linux-scsi@vger.kernel.org
抄送: soeren.binner@oracle.com, haobo.zhou@oracle.com
发送时间: 2010年 5 月 01日 星期六  上午 5:13:36 GMT +08:00 北京/重庆/香港/乌鲁木齐
主题: RE: bfa driver on xen

Hi Joe,

Thanks for reporting the issue. We have noticed this during the netapp/oracal ovm test to include bfa driver as part of ovm distro. As a workaround, we suggested to set num_sgpg=512 and num_ios=512. 

We are working on a fix and it will be available in our 2.3 driver.

Regards,

Jing

>-----Original Message-----
>From: Joe Jin [mailto:joe.jin@oracle.com]
>Sent: Friday, April 30, 2010 2:34 AM
>To: Jing Huang; linux-scsi@vger.kernel.org
>Cc: soeren.binner@oracle.com; haobo.zhou@oracle.com; joe.jin@oracle.com
>Subject: bfa driver on xen
>
>Hi Jing Huang,
>
>I'm Joe Jin from Oracle Linux team, recently we tried brocade device
>on HP DL385 with OracleVM and  found the device could not worked for
>driver could not applied enough memory with below error message:
>
>bfad[0]: memory allocation failed with num_sgpgs: 2048
>bfad[0]: trying to allocate memory with num_sgpgs: 1024
>bfad[0]: memory allocation failed with num_sgpgs: 1024
>bfad[0]: trying to allocate memory with num_sgpgs: 512
>bfad0 bfad_hal_mem_alloc failure
>Not enough memory to attach all Brocade HBA ports, System may need more
>
>I have did some investigated, here are some information:
>OracleVM xen it 64bit and Dom0 is 32bit, the root cause is when 32bit Dom0
>running on 64bit hypervisor(xen), it on compat mode and xen just allowed
>to allocate 1M(order = 8) contiguous memory, if not of compat mode,
>Dom0 would successed applied 2M(order=9) contiguous memory.
>
>To let it the device works, we have to decrease driver config params
>like num_sgpgs, reqq_size, rspq_size and so on, I'm not sure if
>there are some releations among them, also did not known what values
>is good enough for performance, woule you like give me some advices
>for those? we have to reduce bfa driver DMA memory no more than 1M.
>
>Thanks,
>Joe
>

--
To unsubscribe from this list: send the line "unsubscribe linux-scsi" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

                 reply	other threads:[~2010-05-01  0:41 UTC|newest]

Thread overview: [no followups] expand[flat|nested]  mbox.gz  Atom feed

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=be817d0d-1139-4f0f-86e6-0dd709e38f40@default \
    --to=joe.jin@oracle.com \
    --cc=haobo.zhou@oracle.com \
    --cc=huangj@Brocade.COM \
    --cc=linux-scsi@vger.kernel.org \
    --cc=soeren.binner@oracle.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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox