FreeBSD source tree https://cgit-dev.freebsd.org/src/atom?h=releng%2F4.7 2002-07-29T04:25:59Z 2002-07-29T04:25:59Z Matt Jacob mjacob@FreeBSD.org 2002-07-29T04:25:59Z urn:sha1:43fc07cf065969bd452819f551da24e3d1135ed4 2002-07-29T04:25:40Z Matt Jacob mjacob@FreeBSD.org 2002-07-29T04:25:40Z urn:sha1:d39851d81644781ab2da5ff31abf247cab78bf5a and a couple new ISPASYNC defines. 2002-07-29T04:25:02Z Matt Jacob mjacob@FreeBSD.org 2002-07-29T04:25:02Z urn:sha1:210b8f077d35d45483e3eac0d03a6f9e9f768177 2002-07-29T04:24:28Z Matt Jacob mjacob@FreeBSD.org 2002-07-29T04:24:28Z urn:sha1:c4bffcd1a06159258761b9cf84fd59ee26b772e6 Set all 23XX cards as 'touched' (we have trouble, unpredictably, about running ABOUT FIRMWARE with some that were started by BIOS downloads). Redo CTIO2 dma mapping- use continuation segments instead of multiple CTIO2s. Thanks to Veritas for sponsoring this work (in a different context). RELENG_4 only: preserve unit numbering for 'disabled' HBAs by returning success in attachment (but not connecting the dots to CAM). 2002-07-29T04:21:23Z Matt Jacob mjacob@FreeBSD.org 2002-07-29T04:21:23Z urn:sha1:135d445ab029b8fb4882a260c91c93fbb448ef1f 2002-07-29T04:20:46Z Matt Jacob mjacob@FreeBSD.org 2002-07-29T04:20:46Z urn:sha1:432d73e704cb8914f82ef75538ad0c7ec50b482e Go back to *not* fully evaluating loop/fabric state if our role is ISP_ROLE_NONE. 2002-07-29T04:20:16Z Matt Jacob mjacob@FreeBSD.org 2002-07-29T04:20:16Z urn:sha1:a088b43887964f1ea2a9eef666b64714c1e0aefc Clean up some internal target mode stuff plus internal flags. Up MAXISPREQUEST to be 1024 for FC && ULTRA2 or better cards (up fromn 256) Use real bus_dmamap_sync stuff instead of platform foobs. Complain if somebody reuses FC scratch area unexpectedly. 2002-07-29T04:18:09Z Matt Jacob mjacob@FreeBSD.org 2002-07-29T04:18:09Z urn:sha1:67198a155e144d0b9b2cbe317d4c757f5a3f9268 Add support for ISP_FC_GETHINFO, which returns current connection topology, speed, loopid, WWPN/WWNN, etc., plus other ioctls. Coalesece all the freeze simq for loopdown into a isp_freeze_loopdown inline. Beef up target mode. Add isp_handle_platform_notify_scsi and isp_handle_platform_notify_fc platform handlers to handle immediate notifies (isp_handle_platform_notify_scsi is still stubbed out). In implementation of isp_handle_platform_notify_fc, for IN_ABORT_TASK, peel off a pending XPT_IMMED_NOTIFY and call xpt_done on it and hope that somebody upstream is listening. Make sure on final CTIO2s that we set residual correctly. These are absolutely crucial. Make sure we set relative offset for each CTIO2 based upon bytes we've already xferred. This is what the private adjunct datat to the original ATIO is. Note state of command so we can figure out where to find it if we get an ABORT from the firmware. Make sure we *always* set CAM_TAG_ACTION_VALID for ATIO2s. Make sure we keep track of the original lun. 2002-07-29T04:11:47Z Matt Jacob mjacob@FreeBSD.org 2002-07-29T04:11:47Z urn:sha1:f24de6a5ed9ed1c2286c3f2d549e8b8c33274fea Remove the 'bogus registrant' hack for fabric searches. If we've overrrides marked, don't set some default values from reading NVRAM. Switch to using USEC_DELAY rather than USEC_SLEEP at isp_reset time. If the HBA is already 'touched', still set maxluns. Fix some bugs in the non-HANDLE_LOOPSTATE_IN_OUTER_LAYERS vase. Clean up some target mode stuff. 2002-05-02T04:03:35Z Matt Jacob mjacob@FreeBSD.org 2002-05-02T04:03:35Z urn:sha1:3a7c66e90f36255532f85933f798f822b34ff65b If we get a DATA UNDERRUN error from QLogic FC cards, but the RQCS_RU bit is not set in the scsi completion status, or if the residual is clearly nonsense, then this was a command that suffered the loss of one or more FC frames in the middle of the exchange. Set HBA_BOTCH and hope it will get retried. It's the only thing we can do. Approved by: re@freebsd.org