From: Boaz Harrosh <bharrosh@panasas.com>
To: James Bottomley <James.Bottomley@SteelEye.com>,
FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>,
Andrew Morton <akpm@linux-foundation.org>,
linux-scsi <linux-scsi@vger.kernel.org>
Subject: [RFC 0/7] scsi_sgtable implementation
Date: Thu, 05 Jul 2007 14:51:24 +0300 [thread overview]
Message-ID: <468CDB3C.4060500@panasas.com> (raw)
This is a proposed implementation of the scsi sg tables
solution to scsi-ml in preparations for support of
bidirectional scsi commands.
A complete bidirectional solution on top of these patches
can be found in the usual place:
http://www.bhalevy.com/open-osd/download/sgtable_bidi_varlen
It would be best if these patches can be accepted into Morton's
tree as soon as the open window for 2.6.23. So it can be tested
as much as possible before actually getting accepted into
scsi-misc and eventually into the kernel. The reason I say
that is because there is a small but scary difference between
scsi-ml with accessors over old code and accessors over new
sg-tables code. And that is the absence of the IO members in
the case of a DMA_NONE transfer. An un-careful coding can trip
on it. OK I admit the chances are slim but exist.
The code is kept backward compatible with unconverted drivers
all the way up to the last patch. So it can be run and tested
with some drivers unconverted. Applying the last patch will
remove compatibility.
[patch 1/8] stex driver BROKEN
Once patch 2/8 is applied this driver will no longer compile.
So here I comment out some code. The maintainer of this driver
said that driver will be fixed soon enough in the proper manner.
[patch 2/8] Restrict scsi accessors access to read-only
This patch should be accepted now. Even into 2.6.23.
The rest of the patches are dependent on it.
[patch 3/8] libata-scsi don't set max_phys_segments higher than scsi-ml
Also this one is relevant now. But once scsi_sgtable is used sata
will stop working without this one.
[patch 4/8] RFC scsi-ml: scsi_sgtable implementation
This is the actual patch for review. Please see comments
inside and than fire away ...
(With old code so not to let diff make a mess)
[patch 5/8] Remove old code from scsi_lib.c
For easier viewing old code is removed in second patch
[patch 6/8] RFC scsi_error.c move to sg_table implementation
Logically belongs to 4/8. Kept separate for easier review.
[patch 7/8] sd.c and sr.c move to scsi_sgtable implementaion
And these too.
[patch 8/8] Remove compatibility with unconverted drivers
Once this patch is applied any unconverted files will not
compile.
Free Life
Boaz
next reply other threads:[~2007-07-05 12:06 UTC|newest]
Thread overview: 24+ messages / expand[flat|nested] mbox.gz Atom feed top
2007-07-05 11:51 Boaz Harrosh [this message]
2007-07-05 13:43 ` [RFC 1/8] stex driver BROKEN Boaz Harrosh
2007-07-05 19:12 ` Lin Yu
2007-07-05 13:43 ` [RFC 2/8] Restrict scsi accessors access to read-only Boaz Harrosh
2007-07-05 13:43 ` [RFC 3/8] libata-scsi don't set max_phys_segments higher than scsi-ml Boaz Harrosh
2007-07-05 13:43 ` [RFC 4/8] scsi-ml: scsi_sgtable implementation Boaz Harrosh
2007-07-12 14:43 ` Boaz Harrosh
2007-07-12 19:09 ` Mike Christie
2007-07-13 0:15 ` FUJITA Tomonori
2007-07-18 14:13 ` Boaz Harrosh
2007-07-18 14:19 ` Jens Axboe
2007-07-18 15:00 ` Boaz Harrosh
2007-07-18 18:03 ` Jens Axboe
2007-07-18 19:21 ` Benny Halevy
2007-07-18 20:17 ` Jens Axboe
2007-07-23 14:08 ` [PATCH] sgtable over sglist (Re: [RFC 4/8] scsi-ml: scsi_sgtable implementation) FUJITA Tomonori
2007-07-25 19:53 ` Boaz Harrosh
2007-07-12 22:37 ` [RFC 4/8] scsi-ml: scsi_sgtable implementation FUJITA Tomonori
2007-07-05 13:43 ` [RFC 5/8] Remove old code from scsi_lib.c Boaz Harrosh
2007-07-05 13:43 ` [RFC 6/8] scsi_error.c move to scsi_sgtable implementation Boaz Harrosh
2007-07-05 13:44 ` [RFC 7/8] sd.c and sr.c " Boaz Harrosh
2007-07-26 12:21 ` FUJITA Tomonori
2007-07-29 8:21 ` Benny Halevy
2007-07-05 13:44 ` [RFC 8/8] Remove compatibility with unconverted drivers Boaz Harrosh
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=468CDB3C.4060500@panasas.com \
--to=bharrosh@panasas.com \
--cc=James.Bottomley@SteelEye.com \
--cc=akpm@linux-foundation.org \
--cc=fujita.tomonori@lab.ntt.co.jp \
--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;
as well as URLs for NNTP newsgroup(s).