Andrew, Thanks for the info. I have few more questions based on the review of qla24xx_lun_reset() code, 1) From the code it appears that the LUN-reset uses synchronous context/mailbox command since: qla24xx_lun_reset --> qla2x00_issue_iocb --> qla2x00_mailbox_command so abort-taskset / clear-taskset would also use the same code flow. Just wanted to confirm from u since in ur previous email u mentioned otherwise. 2) Since abort/clear-taskset TMFs operate on the specified LUN's taskset (similar to LUN-reset) I think these TMFs would also need only MK_SYNC_ID_LUN marker IOCB as in qla24xx_lun_reset. Plz confirm. thanks, tkrishna On 2/14/06, Andrew Vasquez <andrew.vasquez@xxxxxxxxxx> wrote: > On Tue, 14 Feb 2006, T.Krishnanand wrote: > > > Does ISP2432 support ABORT_TASK_SET and CLEAR_TASK_SET TMFs ? > > Yes. > > > I understand that Linux SCSI EH strategy doesnt make use of these TMFs > > so QLA2xxx driver doesn't have code related to these. > > Also correct. > > > But If I want to extend the QLA driver to support these TMFs using > > following mailbox commands: > > > > MBC_ABORT_TASK_SET > > MBC_CLEAR_TASK_SET > > > > so that I can invoke them thru some IOCTL based appln... > > > > a) would ISP 2432 firmware/asic support these TMFs ? Are there any > > known gotchas that one should be careful about while implementing > > these? > > It does, but not via a synchronous mailbox command. Instead you need > to use a TASK-management IOCB to send the requests to the firmware via > the request-queue. Take a look at the tsk_mgmt_entry structure in > qla_fw.h. > > > b) While coding the handler for these TMFs which existing EH routine > > would u recommend to be used as template/reference ? > > A reference implementation of the driver issuing a LUN-RESET via the > IOCB can be found in qla24xx_lun_reset(). > > > c) Since current/latest driver doesnt have any code for these TMFs, > > what is the exact mailbox command format for these TMFs ? > > -- > av > - : send the line "unsubscribe linux-scsi" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
