Messages in this thread | | | Subject | Re: [PATCH v2 1/2] tcm: Add support for BIDI-COMMANDS and XDWRITE_READ_10 emulation | From | "Nicholas A. Bellinger" <> | Date | Mon, 20 Sep 2010 02:02:15 -0700 |
| |
On Sun, 2010-09-19 at 15:29 +0200, Boaz Harrosh wrote: > On 09/16/2010 12:41 AM, Nicholas A. Bellinger wrote: > > From: Nicholas Bellinger <nab@linux-iscsi.org> > > > > Hi dear Nicholas > > I still have a few reservations regarding the use of the: > + int t_tasks_bidi:1; > > at struct se_transport_task at minimum I'd use the t_tasks_se_bidi_num > as a look ahead. But I hate that as well. I suspect none of this is > needed. But >
Yes, unfortuately AFAICT this is required by the fabric module during the initial I/O setup phase in order signal TCM Core before the T_TASK(cmd)->t_mem_bidi_list is setup w/o depending upon DMA_BIDIRECTIONAL or something else like it..
> At http://git.kernel.org/?p=linux/kernel/git/nab/lio-4.0.git I still get > an old head without these or the CDB32 stuff. > > Where can I find a git web. With latest bits? I'd like to have a > closer look. >
My 'upstream' v4.0 branch that all of the changes have been going into is here..
http://git.kernel.org/?p=linux/kernel/git/nab/lio-core-2.6.git;a=shortlog;h=refs/heads/lio-4.0
> (You know that I have a vested interest in all this I need it to be solid) >
Many thanks Boaz!
--nab
| |