| Commit message (Expand) | Author | Age | Files | Lines |
|---|
| * | return ENOTTY rather than EINVAL to indicate an ioctl hasn't been handled |   jmatthew | 2018-08-14 | 1 | -2/+2 |
| * | Make error handling (esp. DATA_UNDERRUN) clearer and avoid losing resid |   krw | 2016-08-17 | 1 | -7/+8 |
| * | change some types in bio from u_quad_t to uint64_t, and fix casts in |   dlg | 2016-08-14 | 1 | -3/+3 |
| * | replace pointer arithmetic and casts with offsetof |   jsg | 2015-10-23 | 1 | -5/+5 |
| * | sizes for free(); ok sthen |   deraadt | 2015-09-09 | 1 | -11/+12 |
| * | Fix some misuse of the | operator. In particular | has higher | jsg | 2015-05-04 | 1 | -2/+2 |
| * | Remove some includes include-what-you-use claims don't | jsg | 2015-03-14 | 1 | -2/+1 |
| * | remove the second void * argument on tasks. | dlg | 2015-01-27 | 1 | -4/+4 |
| * | be less confusing to the compiler when setting up the sge. | dlg | 2014-09-17 | 1 | -2/+3 |
| * | mark the interrupt handler mpsafe, and drop the kernel lock in the scs_cmd | dlg | 2014-09-15 | 1 | -15/+26 |
| * | remove uneeded proc.h includes | jsg | 2014-09-14 | 1 | -2/+1 |
| * | tasks dont need a mutex and a semaphore to protect against multiple uses | dlg | 2014-09-03 | 1 | -27/+3 |
| * | move mpi(4) from workq to taskq |   blambert | 2014-09-01 | 1 | -3/+5 |
| * | Some reallocarray() use; review Jean-Philippe Ouellet, patrick keshishian | deraadt | 2014-07-13 | 1 | -6/+6 |
| * | add a size argument to free. will be used soon, but for now default to 0. |   tedu | 2014-07-12 | 1 | -14/+14 |
| * | print the board name and firmware revision like we do on mpii and | dlg | 2014-04-16 | 1 | -1/+48 |
| * | use lemtohXX and htolemXX as much as possible | dlg | 2014-03-25 | 1 | -41/+39 |
| * | when reading from the chip you use letoh32, not htole32. | dlg | 2014-03-25 | 1 | -2/+2 |
| * | use htolemXX and lemtohXX in the scsi_cmd paths. | dlg | 2014-03-24 | 1 | -6/+6 |
| * | mpi scatter gather lists are 24 bytes long and next to each other in | dlg | 2014-03-24 | 1 | -27/+20 |
| * | factor out the setting of the request context field so mpi_start does it | dlg | 2014-03-24 | 1 | -16/+7 |
| * | bzero to memset | dlg | 2014-01-20 | 1 | -13/+13 |
| * | bcopy to memcpy | dlg | 2014-01-20 | 1 | -6/+8 |
| * | data transfers suit STREAMING dma maps. | dlg | 2013-01-18 | 1 | -2/+2 |
| * | mikeb pointed at that the interrupt status register read in mpi_intr was | dlg | 2013-01-17 | 1 | -2/+6 |
| * | we dont need to read the interrupt status register to know if we have | dlg | 2013-01-17 | 1 | -7/+7 |
| * | treat DVAs as 64bits all the time so we can avoid ugly casts and shifts in | dlg | 2013-01-17 | 1 | -2/+2 |
| * | several tweaks to make mpi(4) work for vmware emulated sas adapters. | dlg | 2012-09-18 | 1 | -12/+14 |
| * | Use sg_addr instead of sg_lo_addr, leftovers from last commit. |   haesbaert | 2012-09-12 | 1 | -13/+9 |
| * | Make sure we don't sleep on autoconf. | haesbaert | 2012-09-12 | 1 | -2/+2 |
| * | htole64 works as good as htole32 twice for dma virtual addresses. | dlg | 2012-08-26 | 1 | -20/+7 |
| * | mpi_get_ccb and mpi_put_ccb are only called via iopools now, so change | dlg | 2012-01-16 | 1 | -9/+11 |
| * | Backout a bunch of my SCSI commits from c2k11. At least one of these |   matthew | 2011-07-17 | 1 | -2/+2 |
| * | First batch of converting SCSI HBAs from setting saa_targets and | matthew | 2011-07-08 | 1 | -2/+2 |
| * | M_WAITOK cleanup of two cases: |   mk | 2011-06-17 | 1 | -2/+2 |
| * | if getting the RAID header fails, dont stop the midlayer from trying to | dlg | 2011-04-27 | 1 | -2/+2 |
| * | configure fc controllers to fail io as fast as possible when cables are | dlg | 2011-04-27 | 1 | -16/+37 |
| * | return XS_RESET to the midlayer if the command was killed for some reason | dlg | 2011-04-27 | 1 | -1/+6 |
| * | rework the scanning of fibre channel ports to match how linux does it. | dlg | 2011-04-27 | 1 | -9/+20 |
| * | Peek at the interrupt status register before poking with the reply post |   mikeb | 2011-03-04 | 1 | -1/+4 |
| * | back out r1.162, the one that bumps openings up on sas and fc | dlg | 2011-03-01 | 1 | -5/+2 |
| * | vol_list in mpi_get_raid is never used. | dlg | 2010-09-24 | 1 | -3/+1 |
| * | tweak the sas io unit to use 32 openings when talking to sata disks if the | dlg | 2010-09-21 | 1 | -1/+42 |
| * | Use SSD_ERRCODE_CURRENT instead of magic 0x70. | krw | 2010-09-20 | 1 | -2/+2 |
| * | allow devices on fc and sas adapters to use all the openings the chip can | dlg | 2010-09-14 | 1 | -2/+5 |
| * | if a busy sas device is unplugged, the pending io on that device will | dlg | 2010-09-13 | 1 | -13/+100 |
| * | dont reuse the event notifications ccbs id for the acknowledgement. | dlg | 2010-09-13 | 1 | -2/+2 |
| * | im not convinced we only have one outstanding event to ack at a time. this | dlg | 2010-09-13 | 1 | -12/+38 |
| * | implement handling of rescan events on fc controllers. allows "hotplug" of | dlg | 2010-09-10 | 1 | -4/+76 |
| * | get some format strings and variables right in debug output | dlg | 2010-08-27 | 1 | -3/+3 |
