From: James Bottomley <James.Bottomley@HansenPartnership.com>
To: dgilbert@interlog.com,
SCSI development list <linux-scsi@vger.kernel.org>
Subject: Re: scsi_alloc_target: parent of the target (need not be a scsi host)
Date: Sun, 10 May 2020 11:52:39 -0700 [thread overview]
Message-ID: <1589136759.9701.25.camel@HansenPartnership.com> (raw)
In-Reply-To: <59d462c4-ceab-041a-bbb5-5b509f13a123@interlog.com>
On Sun, 2020-05-10 at 14:32 -0400, Douglas Gilbert wrote:
> This gem is in scsi_scan.c in the documentation of that function's
> first argument. "need not be a scsi host" should read "it damn well
> better be a scsi host" otherwise that function will crash and burn!
It shouldn't: several transport classes, like SAS and FC have
intermediate devices between the host and the target and they all work
just fine using the non-host parent. Since you don't give the error
this is just guesswork, but the host has to be somewhere in the parent
chain otherwise dev_to_shost(parent) will return NULL ... is that your
problem?
> I'm trying to work out why the function: starget_for_each_device() in
> scsi.c does _not_ use that collection right in front of it (i.e.
> scsi_target::devices). Instead, it step up to the host level, and
> iterates over all devices (LUs) on that host and only calls the given
> function for those devices that match the channel and target numbers.
> That is bizarrely wasteful if scsi_target::devices could be iterated
> over instead.
>
> Anyone know why this is?
Best guess would be it wasn't converted over when the target list was
introduced.
James
next prev parent reply other threads:[~2020-05-10 18:52 UTC|newest]
Thread overview: 5+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-05-10 18:32 scsi_alloc_target: parent of the target (need not be a scsi host) Douglas Gilbert
2020-05-10 18:52 ` James Bottomley [this message]
2020-05-10 21:50 ` Douglas Gilbert
2020-05-11 14:41 ` James Bottomley
2020-05-11 18:31 ` Douglas Gilbert
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=1589136759.9701.25.camel@HansenPartnership.com \
--to=james.bottomley@hansenpartnership.com \
--cc=dgilbert@interlog.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