src/sys/dev/oce, branch stable/9

MFC r271946 and r272595:

2014-11-03T12:38:41Z

Improve transmit sending offload, TSO, algorithm in general. This change allows all HCAs from Mellanox Technologies to function properly when TSO is enabled. See r271946 and r272595 for more details about this commit. Sponsored by: Mellanox Technologies

MFC r228478, r263710, r273377, r273378, r273423, r273455 and r273899:

2014-10-31T18:18:04Z

- Reimplement CTASSERT() using _Static_assert(). - De-vnet hash sizes and hash masks. - Fix multiple issues related to arguments passed to SYSCTL macros. Sponsored by: Mellanox Technologies

MFC r258941,267839:

2014-06-27T00:37:03Z

Apply vendor improvements to oce(4) driver: - Add support to 20Gbps, 25Gbps, 40Gbps devices; - Add support to control adaptive interrupt coalescing (AIC) via sysctl; - Improve support of BE3 devices; - Big endian support fixes; Many thanks to Emulex for their continued support of FreeBSD. Submitted by: Venkata Duvvuru

MFC r260110:

2014-01-13T19:42:37Z

Eliminate unused drbr_stats_update implementation in oce(4) driver. Noticed by: dim

MFC r257007,258140:

2013-11-25T20:05:23Z

Update driver to version 10.0.664.0. Many thanks to Emulex for their continued support of FreeBSD. Submitted by: Venkata Duvvuru

MFC r246482 (rrs) + r246581:

2013-11-25T20:01:34Z

This fixes a out-of-order problem with several of the newer drivers. The basic problem was that the driver was pulling the mbuf off the drbr ring and then when sending with xmit(), encounting a full transmit ring. Thus the lower layer xmit() function would return an error, and the drivers would then append the data back on to the ring. For TCP this is a horrible scenario sure to bring on a fast-retransmit. The fix is to use drbr_peek() to pull the data pointer but not remove it from the ring. If it fails then we either call the new drbr_putback or drbr_advance method. Advance moves it forward (we do this sometimes when the xmit() function frees the mbuf). When we succeed we always call advance. The putback will always copy the mbuf back to the top of the ring. Note that the putback *cannot* be used with a drbr_dequeue() only with drbr_peek(). We most of the time, in putback, would not need to copy it back since most likey the mbuf is still the same, but sometimes xmit() functions will change the mbuf via a pullup or other call. So the optimial case for the single consumer is to always copy it back. If we ever do a multiple_consumer (for lagg?) we will need a test and atomic in the put back possibly a seperate putback_mc() in the ring buf. Reviewed by: jhb@freebsd.org, jlv@freebsd.org

MFC r252869:

2013-07-06T23:56:58Z

Update driver with recent vendor improvements, most notably support of Skyhawk adapters. Many thanks to Emulex for their continued support of FreeBSD. Submitted by: "Duvvuru,Venkat Kumar"

MFC: r243857 (partial)

2013-03-09T00:39:54Z

Mechanically substitute flags from historic mbuf allocator with malloc(9) flags in sys/dev.

MFC r231879,246799,247880,248059:

2013-03-08T18:46:21Z

Update driver to version 4.6.95.0. Submitted by: "Duvvuru,Venkat Kumar"

MFC 246799:

2013-03-01T20:33:35Z

Resolve issue that caused WITNESS to report LORs.