> Mike Christie wrote: > David C Somayajulu wrote: > > This patch OBSOLETES previous patches 0/5 thru 5/5 titled qla4xxx: Add IPv6 support and misc. It > incorporates the feedback received from Mike Christie and others, and encapsulates everything into a > single patch. > > > > The patch contains the following: > > 1. clean up and make firmware related structures in ql4_fw.h to confirm to linux coding style. > > 2. add support for IPv6 for QLA4032 > > 3. add support for capturing all AEN types in a ring buffer, to aid troubleshooting. > > 4. abstract the functions for sending commands to the HBA (i.e., sending via Request Queue or > Mailbox registers) so that the underlying functions can be different in future hardware. Also mailbox > commands which are not performance critical may be sent via Request Queues (using mailbox iocbs). > > 5. In ql4_init.c fix qla4x00_pci_config() > > 6. In ql4_init.c fix in qla4xxx_add_device_dynamically(), so that a new target coming online > and existing targets which went offline and then came back online are differentiated. > > 7. In ql4_init.c in qla4xxx_build_ddb_list() the firmware ddb entry's state needs to be updated > by calling qla4xxx_get_fwddb_entry() after qla4xxx_set_ddb_entry(). > > 8. remove dead code in ql4_dbg.c and ql4_mbx.c > > 9. Since qla4xxx is a multifunction device having one iSCSI and one Ethernet function per > physical port, when one of the functions resets the HBA all others need to acknowledge it and wait > till the reset is complete and each of them need to re-initialize. A bug in qla4xxx_free_adapter() > was resulting in where the resources pertaining to a function were getting released due to module > unload, when the function was recovering from reset issued by one of the other functions. > > 10. In qla4xxx_free_adapter() memory resources needed to be freed after free_irq() is invoked. > > Thanks > > David Somayajulu > > QLogic Corporation > > > > Looks ok, but it is best to send updates with one change per patch and > each patch having some change info. It makes it nice for when searching > git logs for problems later. Maybe James wants you to resubmit a proper > patchset though. Thanks Mike. I decided to put everything in one patch, since there were too many interdependencies. Anyway I can resubmit if James does require it that way. - 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
