src-test2/sys/netipx, branch release/6.2.0_cvs FreeBSD source tree https://cgit-dev.freebsd.org/src-test2/atom?h=release%2F6.2.0_cvs 2007-01-11T22:54:46Z This commit was manufactured by cvs2svn to create tag 2007-01-11T22:54:46Z cvs2svn cvs2svn@FreeBSD.org 2007-01-11T22:54:46Z urn:sha1:201f78d6b90635d961fa373ecd706e3786333b7e 'RELENG_6_2_0_RELEASE'. This commit was manufactured to restore the state of the 6.2-RELEASE image. Merge ipx_usrreq.c:1.56 from HEAD to RELENG_6: 2006-06-08T23:36:39Z Robert Watson rwatson@FreeBSD.org 2006-06-08T23:36:39Z urn:sha1:6e36e868a33d43fbaec710ff8a0370d99056c925 Make this compile without INVARIANTS. Merge README:1.8 from HEAD to RELENG_6: 2006-06-08T23:36:03Z Robert Watson rwatson@FreeBSD.org 2006-06-08T23:36:03Z urn:sha1:81b33b50dd87d6f32389a783963a03991aea1bb6 Add a simple netipx TODO list to the end of README, since there are a number of problems with netipx that I have not yet resolved, and I don't want them lost track of. Merge ipx_input.c:1.55 from HEAD to RELENG_6: 2006-06-08T23:35:42Z Robert Watson rwatson@FreeBSD.org 2006-06-08T23:35:42Z urn:sha1:1642b197904ede1b2541300e553d417d6b1989eb Canonicalize copyright order in one more file that contains my copyright. Merge spx_usrreq.c:1.82 from HEAD to RELENG_6: 2006-06-08T23:35:16Z Robert Watson rwatson@FreeBSD.org 2006-06-08T23:35:16Z urn:sha1:725748d19b4b14ee49f85d07054ad7a9705529fb In spx_output(), use M_DONTWAIT instead of M_TRYWAIT, as we hold the ipxpcb mutex. Contrary to the comment, even in 4.x this was unsafe, as parallel use of the socket by another process would result in pcb corruption if the mbuf allocation slept. Merge spx_usrreq.c:1.81 from HEAD to RELENG_6: 2006-06-08T23:34:43Z Robert Watson rwatson@FreeBSD.org 2006-06-08T23:34:43Z urn:sha1:33c7ef700f8d800fb81e747507d3a63d5fbbd269 In spx_input(), change a '&&' to a '||', as the spx trace code is able to handle a NULL 'cb' here. Merge spx_usrreq.c:1.80 from HEAD to RELENG_6: 2006-06-08T23:33:58Z Robert Watson rwatson@FreeBSD.org 2006-06-08T23:33:58Z urn:sha1:244f19e2bae0b6313dd3530634a03205d6d39592 In spx_accept, assert ipxp != NULL, not == NULL. Merge spx_usrreq.c:1.79 from HEAD to RELENG_6: 2006-06-08T23:33:24Z Robert Watson rwatson@FreeBSD.org 2006-06-08T23:33:24Z urn:sha1:3c762493f9a208a4353627bae8367ad5dbece3b4 In various SPX protocol entry points from the socket layer, check IPXP_DROPPED before continuing, and return EINVAL or ECONNRESET if it is flagged. It's unclear why each situation should be one or the other, but it is copied from netinet which has the same bugs. Merge ipx_pcb.h:1.26, spx_usrreq.c:1.78 from HEAD to RELENG_6: 2006-06-08T23:32:42Z Robert Watson rwatson@FreeBSD.org 2006-06-08T23:32:42Z urn:sha1:28a46543cdeb9707f21ccfa6ae3400a0f85ba6b1 Add a new ipxpcb flag, IPXP_SPX, which is set on ipxpcb's to mark them as belonging to SPX. This replaces the implicit assumption that the cb pointer for non-SPX pcb's will be NULL. This isn't required in TCP/IP as different pcb lists are maintained for different IP protocols; IPX stores all pcbs on the same global ipxpcb_list. Foot provided by: gnn Merge spx_usrreq.c:1.77 from HEAD to RELENG_6: 2006-06-08T23:31:47Z Robert Watson rwatson@FreeBSD.org 2006-06-08T23:31:47Z urn:sha1:c28def35cdf57fadc573cdf30f33af9946b3a476 Restore original formulation of SPX segment queue draining during SPX PCB detach.