From: James Smart <James.Smart@Emulex.Com>
To: "Mukker, Atul" <Atul.Mukker@lsi.com>
Cc: Brian King <brking@linux.vnet.ibm.com>,
"linux-scsi@vger.kernel.org" <linux-scsi@vger.kernel.org>
Subject: Re: Recommended HBA management interfaces
Date: Tue, 21 Jul 2009 08:29:22 -0400 [thread overview]
Message-ID: <4A65B4A2.70402@emulex.com> (raw)
In-Reply-To: <6C678488C5CEE74F813A4D1948FD2DC7A99EFE88@cosmail02.lsi.com>
Mukker, Atul wrote:
> The management protocol involves significant amount of binary data transfer involving multiple applications, so sysfs and friends are not useful for this particular application. But I gather from your (and Brian's) email that mid-layer SG extension should be used for this particular purpose.
>
> As for asynchronous notifications, netlink seems to be the de-facto choice (or it's mid-layer extensions). But didn't you mention earlier, vmware would not support this?
I answered from a purely linux perspective. VMware is not Linux. VMware
attempts to emulate the linux driver/midlayer api's, but emulation is done in
their own way, with their own semantics, for their own purposes.. Anyone that
believes they just drop a linux driver into vmware and it works w/o change has
a screw loose. It may appear to work, and some subsystems may be better than
others, but there are very subtle and critical differences. As for all the
ancillary interfaces such as sysfs, sgio, and transports: a) they have a hard
time keeping up with the pace of the linux kernel; b) many of the interfaces
are anti their hypervisor management model. Dependence on user-space utils,
sysfs, etc doesn't work with a cos-less environment. Netlink and sockets
opens up security holes, and hypervisor-level socket support brings all kinds
of headaches and memory issues. Netlink is not supported. Sysfs isn't
supported. Even portions of the transports/midlayer are only partially
implemented. Unfortunately, vmware interfaces need to be taken up with vmware.
-- james s
next prev parent reply other threads:[~2009-07-21 12:29 UTC|newest]
Thread overview: 13+ messages / expand[flat|nested] mbox.gz Atom feed top
2009-07-17 13:16 Recommended HBA management interfaces Mukker, Atul
2009-07-17 15:35 ` Brian King
2009-07-20 16:28 ` Mukker, Atul
2009-07-20 16:57 ` James Smart
2009-07-20 18:03 ` Mukker, Atul
2009-07-20 19:08 ` James Smart
2009-07-20 20:33 ` Mukker, Atul
2009-07-21 12:29 ` James Smart [this message]
2009-07-21 13:38 ` Mukker, Atul
2009-07-21 13:48 ` Drew
2009-07-21 13:58 ` Mukker, Atul
2009-07-21 14:59 ` James Smart
2009-07-21 16:27 ` Drew
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=4A65B4A2.70402@emulex.com \
--to=james.smart@emulex.com \
--cc=Atul.Mukker@lsi.com \
--cc=brking@linux.vnet.ibm.com \
--cc=linux-scsi@vger.kernel.org \
/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