Greetings all, I am happy to announce the first release of the tcm_loop.ko virtual SCSI fabric module for TCM Core v3.2 that provides CDB-level SPC-3 and above emulation for SAS Target and Initiator Port Addresses. This allows virtual TCM storage objects like BLOCK, FILEIO and RAMDISK to be presented as locally accessable Linux/SCSI devices (/dev/sdz) using CDB level emulation capable of TCM/LIO v3.2 T10 NAA WWN naming, Persistent Reservations, and ALUA logic. So far, sg_persist is able to span reservations between multiple emulated SAS I_T Nexues, as well as across iSCSI Target Ports for the same TCM backend storage object. Also, the ability to pass the locally accessable TCM_Loop virtual SCSI luns using scsi_generic SG_IO passthrough into VM guests of different operating systems (RHEL5 and NetBSD5 for example) is also up and running on top of the vmw6 hypervisor mpt-fusion HBA emulation on a Linux v2.6.32-rc5 host. For more information and screenshots, please see: http://linux-iscsi.org/index.php/TCM_loop Here is the original running commit for the kernel piece: http://git.kernel.org/?p=linux/kernel/git/nab/lio-core-2.6.git;a=commitdiff;h=10eae38203a01a26fca3b2097f13e48a0ba2d38f And the tcm_loop.py CLI commit: http://git.kernel.org/?p=linux/storage/lio/lio-utils.git;a=commitdiff;h=564dcf1733bb8591f02900c644f0694e1fb2af35 There is still some of work to be done in tcm_loop_configfs.c, and the struct scsi_host_template for the emulated SCSI LLD is pretty much just queuecommand() and proc() currently, so it needs TMR LUN_RESET. Also can_queue needs to be ramped up to 32 and ENABLE_CLUSTERING set, and so far the tcm_loop virtual SCSI LUNs are able to pass simple badblocks tests, and able to push bulk I/O. Thanks! --nab -- To unsubscribe from this list: 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
