summaryrefslogtreecommitdiff
path: root/sys/dev
AgeCommit message (Collapse)Author
2011-04-10bring back changes in r 1.192:Jacob Meuser
detach happens after the hardware is gone, so don't try to touch the hardware in the detach path but azalia_pci_detach is called if the device coiuld not be initialized. in that case, shut the hardware down.
2011-04-09add all 64 bit dma plumbing but stick with 32 for now by default.Marco Peereboom
ok deraadt
2011-04-09use unique wait channelsTheo de Raadt
2011-04-09make sure the upper 32 bits are always written to instead of assuming itMarco Peereboom
is 0. ok deraadt.
2011-04-08backout previous (which has some interrupt spin problem, sort of diagnosedTheo de Raadt
by claudio to be related to this commit) until jakemsr has time to fix it
2011-04-08more spacesMarco Peereboom
2011-04-08kill some stray spacesMarco Peereboom
2011-04-08rename wiat channel to something less than 8 chars.Marco Peereboom
prompted by deraadt
2011-04-08use dma_alloc for transient management commands.Marco Peereboom
prodded and ok deraadt
2011-04-08Remove unused code/arrayJordan Hargrave
2011-04-08one dma_free() was mistakenly left as free()Theo de Raadt
2011-04-08Fix raidp/raid6 to work with new iopool code, needed to swap wu'sJordan Hargrave
Update to use dma_malloc for I/O blocks ok marco@
2011-04-07Fix uninitialized name and x variables in setfanJordan Hargrave
2011-04-07Revert prior diff that I ok because I don't know how to read. Real fixMarco Peereboom
coming soon after this revert. ok jordan
2011-04-07Fix device names in acpitz_setfan() printf messages, which got broken inMiod Vallat
1.32. ok marco@
2011-04-07Do not use NULL in integer comparisons. No functional change.Miod Vallat
ok matthew@ tedu@, also eyeballed by at least krw@ oga@ kettenis@ jsg@
2011-04-07Prepare to add support Synaptics touchpadsAlexandr Shadchin
ok miod@
2011-04-07Avoid using an uninitialized variable when downgrading PIO mode too much onMiod Vallat
ITExpress chipsets. (similar to 1.243, with a deja vu)
2011-04-07Make sure the eeprom pointer is correctly initialized inMiod Vallat
ar9380_spur_mitigate_ofdm().
2011-04-07Iopoolification of another once common bus/device.Kenneth R Westerback
2011-04-06Don't enumerate _DOD for attaching acpivout(4), since:Martynas Venckus
- ACPI spec. says _DOD is not required for brightness controls, - The list returned by _DOD might be wrong, - It's an unnecessary work to do. Instead, decision to attach will be based on the actual methods found, similarly like in the other ACPI drivers. Tested by several on tech@. OK kettenis@, marco@, pirofti@. pirofti@ asked me to note here that devices not supporting brightness controls won't attach from now on. This shouldn't be a concern for you, since such devices weren't doing anything at all, anyway.
2011-04-06Correctly protect the arguments of the AIC_ASSERT() macro.Miod Vallat
2011-04-06fix installboot.Marco Peereboom
ok jsing
2011-04-06move ips to iopools. its the usual drill, io between volumes is nowDavid Gwynne
scheduled better, ioctl paths are more reliable, and it removes NO_CCB. ok krw@
2011-04-06add a new "serial" devid type for scsi devices. add code to usb that fakesDavid Gwynne
it up by using the usb devices iSerial thing. ok deraadt@
2011-04-06make crypto work with bigmemMarco Peereboom
2011-04-06fix some spaces while looking for bigmem shizMarco Peereboom
2011-04-06make rebuild also work in bigmemMarco Peereboom
2011-04-06handle bigmem for metadata reads and writes. this should be pre allocatedMarco Peereboom
but for now this allows us to move forward.
2011-04-05pull the dmamam wrapper in from all my other drivers and use it to replaceDavid Gwynne
the siop command buffers and add dmaable buffers for sense data. matthew@ found that siop used to try to dma to the xs->sense bufer, which is not guaranteed to be dma safe. tested by matthew@ kettenis@
2011-04-05binary vs logical or, found by clangJonathan Gray
ok mikeb@ deraadt@
2011-04-05We only use the io space in em_reset_hw() on a few MAC typesJonathan Gray
and it does not exist at all on newer hardware so only map it on those types we are interested in. Fixes "PRO/1000 PT (82575EB)" for Sylvain Desveaux and will also be required for at least 82580. ok claudio@
2011-04-05wd_hibernate_io() is a standalone disk io writer which tries to notTheo de Raadt
damage kernel memory very much while doing the job. It is not very pretty yet, but improving it will need some MI work. ok mlarkin jsg dlg
2011-04-05- use nitems() no binary change on amd64Jasper Lievisse Adriaanse
"reads OK" claudio@
2011-04-05Iopoolification. Testing by marco@.Kenneth R Westerback
ok dlg@ marco@
2011-04-05regenMiod Vallat
2011-04-05Icelandic keyboard mapping. Commited from an Icelandic keyboard.Miod Vallat
2011-04-05mechanic rename M_{TCP|UDP}V4_CSUM_OUT -> M_{TCP|UDP}_CSUM_OUTHenning Brauer
ok claudio krw
2011-04-05provide VSCSI_STAT_RESET so userland can tell us that the connection toDavid Gwynne
the device was reset rather than blindly fail it. requested by claudio@
2011-04-05hid_locate() is supposed to accept the last two arguments being NULL, soMiod Vallat
correctly handle this in the failure case.
2011-04-05Iopoolification to eliminate another few instances of the currentKenneth R Westerback
shibboleth - NO_CCB. This in no way means this driver works. ok gcc@ dlg@
2011-04-05Push COMPAT_FREEBSD in front of a whale. Buggy, out of date, noPhilip Guenthe
one has been weeding it, and it makes life harder. Toasts of Brennivin for its passing from many; diff ok henning@
2011-04-05add a scsi probe routine so isp(4) can reject high luns when the midlayerDavid Gwynne
is probing, rather than every time it issues an io. ok krw@
2011-04-05Deep below wdc_ioc_ata_cmd() it is possible that some controller (present orTheo de Raadt
future) will try to dma the command buffer to the disk in some way. Use dma_alloc() to get ahead of this potential failure path. ok dlg krw
2011-04-05Passing M_WAITOK to mbuf functions is supposed to be a contract betweenBret Lambert
the caller and the function that the function will not fail to allocate memory and return a NULL pointer. However, m_dup_pkthdr() violates this contract, making it possible for functions that pass M_WAITOK to be surprised in ways that hurt. Fix this by passing the wait flag all the way down the functions that actually do the allocation for m_dup_pkthdr() so that we won't be surprised. man page update forthcoming ok claudio@
2011-04-04Kill some #if 0'ed code (the 'raw' scsi path). Move splx()Kenneth R Westerback
to before scsi_done(). This doesn't mean aac will ever enabled! ok dlg@
2011-04-04detach happens after the hardware is gone, so don't try to touchJacob Meuser
the hardware in the detach path
2011-04-04the v4l2 spec says that setting time per frame to zero should reset timeJacob Meuser
per frame to the default, so do that instead of returning EINVAL
2011-04-04Hit this code with a big hammer. It now works on the X201/T510 seriesTheo de Raadt
machines after warm boot or resume. Some of the checks being made were nonsense. Also tested by claudio and guenther. ok jsg
2011-04-04The backplane version of the 82575EB has no link state -- it's alwayswilliam
up. From FreeBSD. ok jsg claudio