| Commit message (Expand) | Author | Age | Files | Lines |
* | use ifiq_input and use it's return value to apply backpressure to rxrs. | dlg | 2020-06-22 | 1 | -2/+4 |
* | Ack the interrupt first, so that it can receive interrupts for new | yasuoka | 2019-11-09 | 1 | -3/+3 |
* | Reshuffle vic_start and get rid of the dequeue begin/rollback/commit dance | mikeb | 2017-07-12 | 1 | -14/+6 |
* | move counting if_opackets next to counting if_obytes in if_enqueue. | dlg | 2017-01-22 | 1 | -2/+1 |
* | G/C IFQ_SET_READY(). | mpi | 2016-04-13 | 1 | -2/+1 |
* | replace IFF_OACTIVE manipulation with mpsafe operations. | dlg | 2015-11-25 | 1 | -6/+7 |
* | The only network driver needing <net/if_types.h> is upl(4) for IFT_OTHER. | mpi | 2015-11-24 | 1 | -2/+1 |
* | shuffle struct ifqueue so in flight mbufs are protected by a mutex. | dlg | 2015-11-20 | 1 | -3/+5 |
* | arp_ifinit() is no longer needed. | mpi | 2015-10-25 | 1 | -4/+1 |
* | Increment if_ipackets in if_input(). | mpi | 2015-06-24 | 1 | -3/+1 |
* | Revert unrelated changes in previous. | uebayasi | 2015-05-29 | 1 | -11/+1 |
* | Initial addition of ``Patrol Read'' support in bio(4), biocto(8), and | uebayasi | 2015-05-29 | 1 | -1/+11 |
* | No need to set `rcvif', if_input() does it for you! | mpi | 2015-04-30 | 1 | -2/+1 |
* | Correct comments. | uebayasi | 2015-04-01 | 1 | -3/+3 |
* | Remove some includes include-what-you-use claims don't | jsg | 2015-03-14 | 1 | -2/+1 |
* | Replace hand rolled code with m_defrag(). | brad | 2015-02-10 | 1 | -26/+8 |
* | convert VMXNET drivers to ml_enqueue + if_input | pelikan | 2015-02-10 | 1 | -8/+5 |
* | unifdef INET | tedu | 2014-12-22 | 1 | -5/+1 |
* | rxr ioctl handling. | brad | 2014-12-01 | 1 | -1/+22 |
* | Some reallocarray() use; review Jean-Philippe Ouellet, patrick keshishian | deraadt | 2014-07-13 | 1 | -4/+4 |
* | add a size argument to free. will be used soon, but for now default to 0. | tedu | 2014-07-12 | 1 | -3/+3 |
* | record the size of the rx rings so we can wrap around them correctly. | dlg | 2014-07-10 | 1 | -1/+2 |
* | cut things that relied on mclgeti for rx ring accounting/restriction over | dlg | 2014-07-08 | 1 | -15/+12 |
* | Do not blindly return 1 from the interrupt handler since we do not know | jsing | 2011-11-29 | 1 | -2/+2 |
* | BUS_DMA_ZERO instead of alloc, map, bzero. | oga | 2010-05-19 | 1 | -4/+2 |
* | Tidy up promisc/multicast handling. Tested by myself (and earlier | sthen | 2009-11-17 | 1 | -30/+29 |
* | - consistify cfdriver for the ethernet drivers (0 -> NULL) | jasper | 2009-08-13 | 1 | -2/+2 |
* | delete xxshutdown handlers that are never even hooked up | deraadt | 2009-08-10 | 1 | -10/+1 |
* | MCLGETI() will now allocate a mbuf header if it is not provided, thus | deraadt | 2009-08-09 | 1 | -9/+3 |
* | Fix some minor format string problems found in a maze of false positives | deraadt | 2009-06-02 | 1 | -2/+2 |
* | use m_clsetwms to tell the allocator how big the rings are. | dlg | 2009-02-01 | 1 | -1/+4 |
* | Add missing newline to error message printf. | reyk | 2008-12-30 | 1 | -2/+2 |
* | matthieu and reyk say that using 9k frames for rx breaks vic on esx. i can | dlg | 2008-12-12 | 1 | -2/+2 |
* | enable 9k frames on the "jumbo" ring | dlg | 2008-12-05 | 1 | -2/+2 |
* | allocate the right number of entries in the "jumbo" rx ring now that it is | dlg | 2008-12-03 | 1 | -7/+7 |
* | use the right variable when looping over rxqs. | dlg | 2008-12-03 | 1 | -2/+2 |
* | Eliminate the redundant bits of code for MTU and multicast handling | brad | 2008-11-28 | 1 | -27/+4 |
* | backout large cluster allocators. | dlg | 2008-11-25 | 1 | -2/+2 |
* | put 9k frames on the jumbo ring | dlg | 2008-11-25 | 1 | -2/+2 |
* | no 9k cluster allocator yet, switch back to 4k. | dlg | 2008-11-24 | 1 | -2/+2 |
* | switch from 4k to 9k frames on the jumbo ring now the cluster allocator is | dlg | 2008-11-24 | 1 | -2/+2 |
* | enable the use of the second rx ring. populate it with 4k frames until the | dlg | 2008-11-24 | 1 | -131/+133 |
* | switch from using MCLGET to MCLGETI for clusters going onto the rx ring. | dlg | 2008-11-24 | 1 | -2/+2 |
* | drop the requirement that the rx ring has to be filled with mbufs. we only | dlg | 2008-11-24 | 1 | -50/+87 |
* | dlg says "well, that is embarassing" | deraadt | 2008-10-29 | 1 | -3/+3 |
* | First step towards cleaning up the Ethernet driver ioctl handling. | brad | 2008-10-02 | 1 | -9/+2 |
* | Convert timeout_add() calls using multiples of hz to timeout_add_sec() | blambert | 2008-09-10 | 1 | -3/+3 |
* | tweak the alignment of the rx buffers so the headers in the frame are | dlg | 2008-07-07 | 1 | -2/+3 |
* | use correct data type for the pci address (bus_addr_t) | reyk | 2007-11-28 | 1 | -2/+2 |
* | let vic attach to the virtual pcnet hardware in vmware. | dlg | 2007-10-28 | 1 | -8/+85 |