summaryrefslogtreecommitdiffstats
path: root/sys/dev/bge
Commit message (Expand)AuthorAgeFilesLines
* For the on-board interfaces found in Fujitsu SPARC64 machines obtain themarius2010-04-261-1/+1
* Use pci_get_max_read_req() and pci_set_max_read_req() to set maximimyongari2010-03-251-28/+2
* Revert r205090.yongari2010-03-161-16/+16
* Reorder interrupt handler a bit such that producer/consumeryongari2010-03-121-16/+16
* Fix typo in r204978.yongari2010-03-101-1/+1
* Fix typo in r204975.yongari2010-03-101-1/+1
* Set maximum read byte count to 2048 for PCI-X BCM5703/5704 devices.yongari2010-03-101-1/+29
* Enable hardware fixes for BCM5704 B0 as recommended by data sheet.yongari2010-03-101-0/+12
* Disable TSO on BCM5755M controller until I understand better foryongari2010-02-261-1/+3
* Remove Tx mbuf parsing code for VLAN in TSO path. Controller doesyongari2010-02-221-18/+5
* Add TSO support on VLAN. Controller requires VLAN hardware taggingyongari2010-02-201-8/+10
* Move device specific flag configuration to attach routine.yongari2010-02-091-4/+4
* PCI express device status register has W1C feature. Writing 0 hasyongari2010-02-011-2/+5
* Use new handshake command for BCM5750 or new controllers.yongari2010-01-221-2/+1
* Fix a long standing ASF heartbeat sending bug. The initialyongari2010-01-221-1/+1
* Don't free mbuf chains when bge(4) fails to collapse the mbufyongari2010-01-151-5/+2
* For controllers that has dual mode PHY(copper or fiber) interfacesyongari2010-01-142-12/+23
* o Add PCI ID for BCM 5756.delphij2010-01-132-1/+4
* Fix regression introduced in r198318. BCM5754/BCM5754M uses theyongari2010-01-031-2/+9
* Create sysctl node(dev.bge.%d.focred_collapse) instead ofyongari2009-12-082-21/+25
* Partially revert r200228. For mini RCB case, bge(4) still have toyongari2009-12-081-7/+5
* Don't access jumbo frame related registers if controller lacks theyongari2009-12-071-8/+14
* Remove PHY isolate/power down code in bge_stop(). The isolationyongari2009-12-071-27/+0
* Add workaround to overcome hardware limitation which allows only ayongari2009-12-031-0/+37
* Fix typo which inversed the logic which in turn disabled MSI.yongari2009-11-251-1/+1
* Make sure one shot MSI is enabled.yongari2009-11-251-0/+2
* BGE_FLAG_40BIT_BUG should be set before creating DMA tags.yongari2009-11-241-7/+7
* Reduce status block size DMAed by controller. bge(4) uses singleyongari2009-11-221-8/+26
* Add missing function prototype in r199671.yongari2009-11-221-0/+2
* Implement TSO for BCM5755 or newer controllers. Some controllersyongari2009-11-222-32/+149
* Fix two long standing bugs on bge(4). Most pre BCM5755 controllersyongari2009-11-222-1/+33
* For MSI case, interrupt is not shared and we don't need to forceyongari2009-11-222-10/+106
* Cache Rx producer/Tx consumer index as soon as we know status blockyongari2009-11-221-20/+27
* Rearrange bge_start_locked to see we can send more frames byyongari2009-11-221-24/+26
* Controller does not write Rx descriptors, remove BUS_DMASYNC_PREREAD.yongari2009-11-221-4/+2
* Use capability pointer to access PCIe registers rather thanyongari2009-11-222-31/+45
* Due to newly added PCIe capabilities fallback code for finding theyongari2009-11-221-37/+17
* Fix typo introduced in r199011.yongari2009-11-221-1/+1
* Remove extra white space.yongari2009-11-222-13/+13
* Controller does not update Tx descriptors(send BDs) after sendingyongari2009-11-101-6/+3
* Zero out Tx/Rx descriptors before using them. Also add missingyongari2009-11-091-0/+7
* Add missing bus_dmamap_sync(9) before issuing kick command.yongari2009-11-091-0/+3
* Correct disabling checksum offloading for BCM5700 B0.yongari2009-11-091-1/+1
* Partially revert r199035.yongari2009-11-081-0/+8
* Count number of inbound packets which were chosen to be discardedyongari2009-11-081-0/+2
* Don't count input errors twice, we always read input errors fromyongari2009-11-081-10/+0
* Tell upper layer we support long frames. ether_ifattach()yongari2009-11-071-1/+3
* Fix I mssied in r199011. Rx ring index also should be updated.yongari2009-11-071-0/+6
* Reimplement Rx buffer allocation to handle dma map load failure.yongari2009-11-072-80/+88
* Do bus_dmamap_sync call only if frame size is greater thanyongari2009-11-061-3/+5
OpenPOWER on IntegriCloud