summaryrefslogtreecommitdiff
path: root/sys
AgeCommit message (Collapse)Author
2010-08-30disable SVR4 compat by default; ok kettenisTheo de Raadt
2010-08-30Convert the pcmcia power hook to an activate function. Then change theTheo de Raadt
model so that DVACT_SUSPEND methods call sub-drivers with DVACT_SUSPEND not DVACT_DEACTIVATE, and such. This creates a whole bunch of fallout, which needs fixing. wdc(4), ne(4), and wi(4) have been fully tested and actually work better than they ever have before (no more wdc timeouts on resume!) malo(4) has some other nasty bug which has not yet been spotted which causes crazy panics. The other drivers are more rare and will be fixed if people show up with hardware; for now at least they have been adapted to the new model. ok kettenis
2010-08-30fix tailq use; ok thibTheo de Raadt
2010-08-30delete PROCFS we will never supportTheo de Raadt
2010-08-30During boot probe walk disklist rather than alldevs. This avoids a raceJoel Sing
where disks can be on alldevs, however they are not yet initialised and have not yet called disk_attach() (in particular this means that dk_label is a null pointer). Also, if we sleep restart the scan from the top of the disklist in case things have changed whilst we slept. ok marco@
2010-08-30Add a device number to struct disk and populate it on disk attach. WhilstJoel Sing
here also expose disklist for future use. ok deraadt@ miod@
2010-08-30ddbcpu for sgi. ok miod@Takuya ASADA
2010-08-30Avoid possible aliasing-related compiler issues by always writing toMatthew Dempsky
cmd->opcode through the type casted command pointer rather than through xs->cmd->opcode. Requested by deraadt@; ok krw@.
2010-08-29the scsi completion code thinks that if you dont get MFI_STAT_OKDavid Gwynne
back from the chip on a command then the command completely failed. MFI_STAT_SCSI_DONE_WITH_ERROR really means the command completed fine, but there's some sense data too. this tweaks the handling to be more appropriate, as per the linux and solaris drivers. timed out waiting for beck@ putting this in cos its obviously more correct than the current code.
2010-08-29s/wdcactivate/config_activate_children/Theo de Raadt
2010-08-29wdcactivate() can go since it is just a wrapper aroundTheo de Raadt
config_activate_children (same change made to wdc.c and wdcvar.h) ok kettenis
2010-08-29- add a check for FWRITE here too.Jasper Lievisse Adriaanse
ok miod@
2010-08-29pcmcia/wdc_pcmcia.cTheo de Raadt
2010-08-29expose dp8390_enable/disable to the higher level driversTheo de Raadt
2010-08-29Massage the powerhook functions into activate functions, and then callTheo de Raadt
them from the powerhook. Fix a few quibbles about the things done for the IFF_RUNNING and IFF_UP cases ok kettenis
2010-08-29regenMiod Vallat
2010-08-29Translation code for Fn-key sequences on the Gdium Liberty internal keyboard,Miod Vallat
which reports Fn as a regular key.
2010-08-29- add a missing ioctl FWRITE checkJasper Lievisse Adriaanse
ok miod@
2010-08-29Fix kernel compiling with disabled IPSEC and enabled GIF/MPLSJoerg Goltermann
OK: claudio@, bob@, sthen@, thib@
2010-08-29put the bufq_restart() in the right place, so that a 2nd resume works.Theo de Raadt
how did noone notice for so long? tsk, tsk, tsk. of course, I broke it. tested by jakemsr
2010-08-28typoTheo de Raadt
2010-08-28Correctly compute memory size if less than 256MB.Miod Vallat
2010-08-28Fix an uninitialized value leading to bogus KASSERT in uvm_pmr_use_inc().Miod Vallat
2010-08-28Garbage collect struct dkdriver.Matthew Dempsky
ok miod@; "please go ahead" jsing@
2010-08-28rename the *_power functions to *_powerhook to indicate that we've beenTheo de Raadt
here already for the activate/powerhook cleanup
2010-08-28powerhooks in acpi drivers? oh boy. change this into a (not yet called)Theo de Raadt
activate function ok kettenis
2010-08-28regenMiod Vallat
2010-08-28Provide a way to get < and > symbols on 101-key slovenian layouts; fromMiod Vallat
Zavisa Nikac (firstname dot lastname at inet dot hr)
2010-08-28In pckbd_scancode_translate(), correctly translate the break sequenceMiod Vallat
between scan code sets 1 and 2.
2010-08-28ansify function definitions, and constify a few arrays while there.Miod Vallat
no functional change.
2010-08-28Remove dead code, ansify, and clean the match routine; adapted from a diffMiod Vallat
by Alexandr Shadchin (firstname dot lastname at gmail) on tech@
2010-08-27Make PCI Power Management optional, and only enable it when acpi(4) attaches.Mark Kettenis
ok deraadt@
2010-08-27Move the activate function from pci to the the main driver, so that theTheo de Raadt
powerhook can use it ok kettenis
2010-08-27Move the guts of the powerhook function into the activate function and makeTheo de Raadt
it stop calling the powerhook function; then make the powerhook function call activate. This basically inverts the whole goop. ok kettenis
2010-08-27Massage the powerhook function into an activate function, and since we doTheo de Raadt
not neccessarily have a working filesystem for the firmware yet, use a workq to postpone resume. Then make the powerhook function call the activate function. There remains some questions about the sc_power functions... ok kettenis
2010-08-27Make the powerhook use the activate functions, which actually do the fullTheo de Raadt
job.
2010-08-27Move the dc_pci_activate function to dc.c, and mangle it up with some ofTheo de Raadt
the gunk in dc_powerhook. Then make dc_powerhook just call it ok kettenis
2010-08-27The powerhook only did stuff at resume; but suspend matters too!Theo de Raadt
Instead of fixing it, make it call the activate function which does the full job ok kettenis
2010-08-27Get rid of the clumsy pci_activate function and do this job inside ath.cTheo de Raadt
so that the powerhook wrapper can get at it ok kettenis
2010-08-27Run powerhooks with interrupts blocked. This requires a bit of a song andTheo de Raadt
dance since some laptops unsuspend out of apm with interrupts enabled?! That causes problems with shared interrupts... but also it would be best if we treat powerhook functions (now backed by activate functions) the same as acpi treats them, and it means that the powerhook wrappers don't need to do their own spl protection. ok kettenis
2010-08-27caught a small typo while reading the code: senor -> sensorRobert Nagy
2010-08-27Since the changes in neo(4) work, it is very likely that the exact sameTheo de Raadt
mechnical changes work in these drivers too. Testing can occur after this commit.
2010-08-27Mangle neo_power's contents into a neo_activate function, then makeTheo de Raadt
neo_power into a stub which calls it. Tested on an ancient z505r; my first laptop....
2010-08-27Make the powerhook code simply call the activate function (it actually doesTheo de Raadt
all the 7 steps we want done, instead of just 2)
2010-08-27Since the pci-specific activate function does nothing, move it to fxp.c.Theo de Raadt
Then it is possible for fxp_powerhook to use it.
2010-08-27remove the unused if_init callback in struct ifnetJonathan Gray
ok deraadt@ henning@ claudio@
2010-08-27Mutilate the powerhook function into an activate function, then createTheo de Raadt
a stub powerhook function to call it.
2010-08-27Delete #if'd out powerhook code which is just unfinished scaffoldingTheo de Raadt
2010-08-27Stop calling xxx_power() from xxx_resume(); instead do it the other way around.Mark Kettenis
Get rid of the sc_power() callback, which isn't used anymore. ok deraadt@
2010-08-27Have the em powerhook call the activate function, which does a way betterTheo de Raadt
job of taking the chip up and down.