src/sys/dev/iscsi, branch releng/10.2 FreeBSD source tree https://cgit-dev.freebsd.org/src/atom?h=releng%2F10.2 2015-05-15T10:50:58Z MFC r278237: 2015-05-15T10:50:58Z Edward Tomasz Napierala trasz@FreeBSD.org 2015-05-15T10:50:58Z urn:sha1:9c812d6665ac2028634c3476eb8a7e364ccf3ba6 Don't call callout_drain() with iscsi mutex held; this fixes a warning that was introduced recently. While here, don't try to access is_terminating without lock. Sponsored by: The FreeBSD Foundation MFC r278233: 2015-05-15T10:47:25Z Edward Tomasz Napierala trasz@FreeBSD.org 2015-05-15T10:47:25Z urn:sha1:f1000399b2087814323f59f5122d3d1b8a777362 Fix error handling. Sponsored by: The FreeBSD Foundation Merge r263233 from HEAD to stable/10: 2015-03-19T13:37:36Z Robert Watson rwatson@FreeBSD.org 2015-03-19T13:37:36Z urn:sha1:8c0b8b26921d2b7336f487bf71f7bf9e555e374e Update kernel inclusions of capability.h to use capsicum.h instead; some further refinement is required as some device drivers intended to be portable over FreeBSD versions rely on __FreeBSD_version to decide whether to include capability.h. Sponsored by: Google, Inc. MFC r274853: 2015-02-19T14:28:47Z Alexander Motin mav@FreeBSD.org 2015-02-19T14:28:47Z urn:sha1:7919f1ea05d548f3ebfa25e1ef20e80fd4e7cbd2 For both iSCSI initiator and target increase socket buffer sizes before establishing connection. This is a workaround for Chelsio TOE driver, that does not update socket buffer size in hardware after connection established, and unless that is done beforehand, kernel code will stuck, attempting to send/receive full PDU at once. MFC r278098, r278099: Add some fields and constants from RFC7144. 2015-02-09T09:08:55Z Alexander Motin mav@FreeBSD.org 2015-02-09T09:08:55Z urn:sha1:17f5abcf4f98bbefedcc627ba1a8bcf83c4e19d0 MFC r272765: Remove one second wait for threads exit from icl_conn_close(). 2015-01-03T18:35:29Z Alexander Motin mav@FreeBSD.org 2015-01-03T18:35:29Z urn:sha1:10ca244b7b97f32985eb0eb7dc951dd59ba642c6 Switch it from polling with pause() to using cv_wait()/cv_signal(). MFC r274033 (by trasz): 2015-01-03T18:27:03Z Alexander Motin mav@FreeBSD.org 2015-01-03T18:27:03Z urn:sha1:9f21ec0ac219b6c405255964a2aec79693e65e7a s/icl_pdu_new/icl_pdu_new_empty/g; no functional changes. MFC r274036 (by trasz): 2015-01-03T13:36:56Z Alexander Motin mav@FreeBSD.org 2015-01-03T13:36:56Z urn:sha1:c3d2f888a65e6614f29b6be8115d2368a2401cfe s/icl_pdu_new_bhs/icl_pdu_new/; no functional changes, just a little nicer code. MFC r275864: Make sequence numbers checks more strict. 2015-01-03T13:08:08Z Alexander Motin mav@FreeBSD.org 2015-01-03T13:08:08Z urn:sha1:cdedd57cd2e41f6b3d43425a46ed48894900230b While we don't support MCS, hole in received sequence numbers may mean only PDU loss. While we don't support lost PDU recovery, terminate the connection to avoid stuck commands. While there, improve handling of sequence numbers wrap after 2^32 PDUs. MFC r275925: Slightly polish iSCSI parameters negotiation. 2014-12-26T09:12:19Z Alexander Motin mav@FreeBSD.org 2014-12-26T09:12:19Z urn:sha1:261678db8f1d29ea102ef70e4577eef64a7beda7