summaryrefslogtreecommitdiff
path: root/sys
AgeCommit message (Collapse)Author
2009-10-04Merge over a bunch of detach logic from msk(4) so that this driver canTheo de Raadt
detach as well ok kettenis
2009-10-04Correctly store the tag results from bus_dmamem_alloc() so that theTheo de Raadt
detach routine can free the right resources ok dlg kettenis
2009-10-04pull in wbsd conditionally; ok kettenisTheo de Raadt
2009-10-04Add (again) support for divert sockets. They allow you to:Michele Marchetto
- queue packets from pf(4) to a userspace application - reinject packets from the application into the kernel stack. The divert socket can be bound to a special "divert port" and will receive every packet diverted to that port by pf(4). The pf syntax is pretty simple, e.g.: pass on em0 inet proto tcp from any to any port 80 divert-packet port 1 A lot of discussion have happened since my last commit that resulted in many changes and improvements. I would *really* like to thank everyone who took part in the discussion especially canacar@ who spotted out which are the limitations of this approach. OpenBSD divert(4) is meant to be compatible with software running on top of FreeBSD's divert sockets even though they are pretty different and will become even more with time. discusses with many, but mainly reyk@ canacar@ deraadt@ dlg@ claudio@ beck@ tested by reyk@ and myself ok reyk@ claudio@ beck@ manpage help and ok by jmc@
2009-10-03Properly handle XMM exceptions instead of panicing the kernel. Fixes theMark Kettenis
issue reported by Slava Pestov. ok deraadt@
2009-10-03Glue to attach the SD/MCC reader of the W83L519D found on some TadpoleMark Kettenis
laptops.
2009-10-03Driver for the SD/MMC part of the Winbond W83L518D/W83L519D Integrated MediaMark Kettenis
Reader.
2009-10-03Add missing <sys/device.h>.Mark Kettenis
ok deraadt@, miod@
2009-10-02remove trash left in debug changes; lindroos@nls.fiTheo de Raadt
2009-10-02syncStuart Henderson
2009-10-02Remove BCM57720 as it's not going to be released. From a BroadcomStuart Henderson
commit to the Linux driver via Brad. Also checked dmesglog.
2009-10-02Remove BCM57720 as it's not going to be released. From a BroadcomStuart Henderson
commit to the Linux driver via Brad.
2009-10-02Support for DS15/DS25/ES45 (Titan) alpha models (rpb family 38). These systemsMiod Vallat
are very close to the Tsunami systems (rpb family 34) and share the existing Tsunami code; the tsc(4) Tsunami controller code is extended to handle the Titan differences, except for the Titan AGP GART which is left unsupported for now. Tested to not cause regressions on DS20 (deraadt@) and ES40 (miod@). Titan support tested by Sergey Prysiazhnyi on DS25, many thanks for your time!
2009-10-02Link the kernel at a higher physical address, in order to accomodate SRMMiod Vallat
with larger footprint than the usual 2MB, as encountered on some of the most recent alpha systems. Tested on many models (DEC 3000, PC164, AS500, ES40, and more)
2009-10-02Some 21264CB pass information, from information provided by Sergey Prysiazhnyi.Miod Vallat
2009-10-02regenMiod Vallat
2009-10-02One more DEC device found on alpha clustersMiod Vallat
2009-10-01add fdt_parent_node() and OF_parent() functions,Dariusz Swiderski
as requested by kettenis@ ok by kettenis@
2009-10-01Generic soft interrupts for macppc. Tested by mk@, deraadt@Mark Kettenis
ok miod@
2009-10-01Needs memcpy in bootxx now, forgotten in previous getenv change.Miod Vallat
2009-09-30Fix various off-by-ones in irq number validation.Miod Vallat
2009-09-30Remove unused last argument of alpha_shared_intr_disestablish().Miod Vallat
2009-09-30Some old SRM require 8-byte alignment of the buffer passed to the promMiod Vallat
getenv routine, so fulfill this requirement. From NetBSD. (Note this only concerns the boot blocks, as the kernel itself uses extremely well aligned addresses for getenv calls).
2009-09-30Get rid of prom_open() and implement it directly as a prom_dispatch() wrapper,Miod Vallat
necessary for some old SRM version. From NetBSD.
2009-09-30Clear sticky error status after recoverable machine check.Miod Vallat
From FreeBSD.
2009-09-30Fix MDDT_NONVOLATILE value, it's 10 binary, not 10 hexadecimal.Miod Vallat
While there, add two more RPB system type values.
2009-09-30One missing print -> DPRINTF conversion in dec_kn300_device_register()Miod Vallat
2009-09-30- Sync the BCM5785G PCI id for bge(4) and add BCM5785F / BCM57788.Stuart Henderson
PCI ids from the Linux tg3 driver via Brad.
2009-09-30syncStuart Henderson
2009-09-30- Some new PCI ids for the BCM5717, BCM5718, BCM5785F, and BCM57788 chipsets.Stuart Henderson
- Rename the existing BCM5785 entry to BCM5785G. PCI ids from the Linux tg3 driver via Brad.
2009-09-30Need to scsi_done() before returning COMPLETE for untranslated commands,Miod Vallat
such a Test Unit Ready; found the hard way by jbg@, and inexplicably missed during dlg@ and I audit of hba drivers behaviour.
2009-09-30Fix comments to match reality.Tobias Weingartner
Thank you Vladimir Kirillov <proger () hackndev ! com>
2009-09-30curproc, curprocpaddr moved into cpu_infoTakuya ASADA
OK miod@
2009-09-29Support the Phison PS5000. IO Data and other vendors are now using thisTheo de Raadt
on an expresscard SSD adaptor. ok jsg
2009-09-29Having pty_isfree_locked() be inlined may make pty allocation a tinyTheo de Raadt
bit faster, but come on, inlining is supposed to be reserved only for things which *critically* need it. ok millert
2009-09-29syncTheo de Raadt
2009-09-29phison pciide controller found in expresscard SSD cardsTheo de Raadt
2009-09-28Make video buffer area and BIOS areas unavailable for PCI resource allocation.Mark Kettenis
This should prevent problems on systems where these areas are not reserved in the BIOS memory map. ok miod@, oga@, marco@
2009-09-28when inserting a state, turn the error that pf_state_insert returnsDavid Gwynne
into something usable by ioctl. makes DIOCADDSTATE on /dev/pf cope when inserting an already existing state.
2009-09-27adjust comments; eephy isn't just for Intel GigE. From Brad.Stuart Henderson
2009-09-27In pmap_proc_iflush(), do not bother checking for pmap being curpmap on anyMiod Vallat
processor, since caches are physically addressed and we are working on physical addresses.
2009-09-27Add an implementation of IRIX-compatible cacheflush() routine to mips ports,Miod Vallat
needed for gcc -ftrampoline operation as well as by some third-party software. Although the implementation uses the sysarch() sysctl, the wrapper is added to libc as it was a direct system call (which it is on IRIX).
2009-09-27If no xfer buffer is free for the UDLIO_DAMAGE ioctl, sleep in the kernelMarcus Glocker
until one xfer buffer gets freed again, instead userland needs to handle this. Help and OK miod@, matthieu@
2009-09-27typosMiod Vallat
2009-09-27Print the LOM type (LOMlite/LOMlite2).Mark Kettenis
2009-09-27Add sensor status for fan sensors.Mark Kettenis
2009-09-27Implement queuing of reads and writes. This gets rid of the remaining busyMark Kettenis
waits for LOMlite.
2009-09-26Fix artefacts in compression mode especially seen when running X.Marcus Glocker
2009-09-25- enable axe/aue to make my life a bit easierJasper Lievisse Adriaanse
ok marex@
2009-09-25Fix the value of one voltage sensor in ev6 environmental machine checks.Miod Vallat