test at Target mode

[Justin T. Gibbs]

>Dear ML-members:
>
>I have tested (new) AIC7xxx driver(6.1.13) at target mode.
>
>In this ML, target driver has few disccusion.
>The followings will be unuseful for almost people.
>But I will mail because we need the best target driver for AIC7xxx.

The target mode support for the AIC7XXX driver was designed and
only tested under FreeBSD.  Most if not all of the issues you
list in this mail are due to expectations of how the OSM layer
of the driver will setup SCBs and or handle certain types of
events.

>Changed:
>1)At first,I get compiled errors when I forced targe mode enable.
>    #define AHC_TARGET_MODE 1
>    ->Maybe some definitions for upper driver interface incompletely
>provided.
>    Therefore I left TARGET_MODE disable.

As noted above, the interface is for FreeBSD only.  FreeBSD's SCSI
layer (CAM) has explicit support for target mode operations.  Linux
has yet to gain such features.

>2)Once a BusReset occurred,following BusReset is ignored.
>    ->I forced to set ENSCSIRST bit in SIMODE1(ahc_restart).
>    I set ENRSTI bit in SCSISEQ_TEMPLATE,but that's no effect.

This is by design.  Bus resets are only of interest when:

1) This is the first bus reset since power on.

2) We have outstanding commands as either a target or an initiator.

So, once we see a bus reset, we disable them until the next outgoing
(as initiator) or incoming (as target) command.  This prevents the
driver from being held hostage to a third party holding the reset
line for extrememly long periods of time which would result in an
interrupt storm.  Semantically, there is no difference if we ignore
any of these "extra" bus resets.

>3)Sequencer code)p_data has some routes to go to ITloop at target mode.

I'll have to see if I've introduced any bugs in that area with newer
sequencer drops.  if the TARGET bit is set, we should always return
to the target event loop.

>4)Sequencer code)ENSELO bit has never be set,
>    even if the bit in SCSISEQ_TEMPLATE set.
>    ->when ENSELO bit forced to set with TEMODE,target is selected.
>    Is there another method?

ENSELO will only be set if both initiator and target mode operation
are enabled at the same time.  ENSELO is not used for target mode
operation.

>5)Sequencer code)At no disconnect mode(MsgOut=80),data phase stopped in
>DT mode.
>    ->calling set_trasfer_settings before changing data phase will be
>needed.

Actually, this is handled by the kernel driver rather than consume
code in the sequencer.  The sequencer is already paused and we have
to write to card registers anyway, so setting two more is cheaper
than adding more sequencer instructions.

>6)Sequencer code)When set DISCENB in SCB_CONTROL and reconnected,
>MegOut=8x sent.
>    ->setting MSG_IDENTIFY_DISCFLAG added.

The format of target SCBs is different than for normal SCBs.  See
the FreeBSD driver for details:

http://www.freebsd.org/cgi/cvsweb.cgi/src/sys/dev/aic7xxx/

--
Justin

To Unsubscribe: send mail to majordomo at FreeBSD.org
with "unsubscribe aic7xxx" in the body of the message