summaryrefslogtreecommitdiff
path: root/sys
AgeCommit message (Collapse)Author
2011-09-13Fix up alc_encap() / alc_start():Kevin Lo
- Remove unnecessary nsegs variable from alc_encap() and use map->dm_nsegs. Also remove unnecessary FreeBSD check for 0 DMA segments check. - Remove printfs in alc_encap() failure paths that shouldn't be there. - Add missing IF_PREPEND() from failure path coming off of alc_encap() within alc_start(). - Fix error handling within alc_encap(). Previously alc_encap() was attempting to unload a DMA map upon failure from bus_dmamap_load_mbuf() even though one wasn't loaded at that point and then always forcing mbufs through the EFBIG path. Tested by Gabriel Linder. From Brad
2011-09-13Do not invoke hardclock() until cpu_initclocks() has been invoked; avoidsMiod Vallat
a NULL pointer dereference because the softclock interrupt handler cookie is not initialized yet. Found the hard way and fix tested by aoyama@
2011-09-12Clamp a maximum number of outstanding requests at 40 as it wasMike Belopuhov
figured out the hard way by senior Pedro. Performance decreased severely with newer firmwares when ioc was stuffed with requests. Unfortunately it still remains a magic number as LSI doesn't clearly document what it has to be (controllers report some meaningless values like 3167 and 1952). ok dlg
2011-09-11Define symbolic constants for the addresses of the vsbus devices onMiod Vallat
512KB PROM systems (e.g. KA49), or 256KB PROM systems built after KA49 (e.g. KA5[234]), and use them instead of pouring magic numbers all over the (md) tree. No functional change.
2011-09-11Fix inverted test when selecting resolution on the 1024x864 VLC frame buffer.Miod Vallat
2011-09-08Provide namespace-safe alignment macros in <machine/_types.h>, withPhilip Guenthe
compat names kept in <machine/param.h>. In <sys/socket.h>, pull in <sys/_types.h> instead of the namespace polluting <machine/param.h> and completely eliminate __CMSG_ALIGN, replaced by _ALIGN ok deraadt@
2011-09-08Make the INT_FAST*_{MIN,MAX} macros match the types they're defined to.Philip Guenthe
Since the underlying types of the int_fast types are set by machine/_types.h, put internal macros in that same file and define the exposed INT_FAST*_{MIN,MAX} macros from those. ok millert@, kettenis@
2011-09-06Properly initialize struct carp_if (especially vhif_nvrs) with M_ZERO.Marco Pfatschbacher
This lets carp delete IFF_PROMISC on its carpdev upon destroy. Fix from Stefan Rinkes. OK sthen, bluhm, deraadt.
2011-09-05No need to check for the size to be a multiple of 4 in the alpha-specificMiod Vallat
code to process four samples at a time - we can do as much sets of 4 as possible before falling back to the remainder processed one byte at a time.
2011-09-05Enable MSI support and fix establishing a link to a GigE link partnerKevin Lo
from a FastE ale(4) controller. Previously only link parters which support down-shifting were able to establish a link. Tested by roberth, Johan Torin and sthen@ From Brad
2011-09-04Driver for the TURBOchannel alpha baseboard audio. Ported from NetBSD, withMiod Vallat
changes to allow using non-native encodings, and alpha specific code to try and avoid doing 8 bit memory access. Tested on 3000/300LX, deaf testing on 3000/600.
2011-09-04Add the possibility for a MD attachment to stack a stream filter (sw_code)Miod Vallat
if it needs to rewrite the audio stream; adapted from NetBSD
2011-09-04oops, this DPRINTF flavour takes no parentheses.Miod Vallat
2011-09-04Silence awacs_trigger_input by turning a noisy printf into the DPRINTFMiod Vallat
it was intended to be; reported by Donovan Watteau
2011-09-04another udav device from Loganaden VelvindronJonathan Gray
2011-09-04regenJonathan Gray
2011-09-04Another udav from Loganaden Velvindron andJonathan Gray
Intel 6 Series/C200 Rate Matching Hub from Martin Pelikan
2011-09-04regenMark Kettenis
2011-09-04There is no such thing as the "Intel Centrino Advanced-N 6005".Mark Kettenis
2011-09-04regenMark Kettenis
2011-09-04Add entry for Intel Z68 LPC and correct the H61 LPC entry (pointed out byMark Kettenis
Martin Pelikan).
2011-09-04syncStuart Henderson
2011-09-04Add PCI ids for D-Link DGE-530T C1 GigE NIC and the ServerWorks HT-1100 SATAStuart Henderson
controllers. From Brad.
2011-09-04Add emulation support for slinear:8, ulinear:8 and alaw. Written by jason@Miod Vallat
more than 8 years ago; allows aucat (and sndio users) to happily use amd7930-based audio devices. Verified by playing an mp3 file on vax.
2011-09-03Driver for the audio chip found on VAXstation 4000. Tested on VLC, /60 andMiod Vallat
/90.
2011-09-03Switch the sparc audioamd(4) code to the MI driver; tested on SPARCclassicMiod Vallat
with and without option AUDIO_C_HANDLER.
2011-09-03MI driver for AMD79C30 audio chip; from NetBSDMiod Vallat
2011-09-03Help prevent watchdog timeouts.Kevin Lo
Tested by roberth <robert at openbsd dot pap dot st> and Johan Torin. From Brad
2011-09-03Disable PHY hibernation. The PHY will go into sleep state when itKevin Lo
detects no established link and it will re-establish link when the cable is plugged in. Previously it failed to re-establish link when the cable was plugged in such that it required turning the interface down and then up to make it work. This was caused by incorrectly programmed hibernation parameters. Further details regarding PHY setup are necessary to be able to re-enable this feature. Tested by Matteo Filippetto and Gabriel Linder From FreeBSD via Brad
2011-09-03Add a general warning about gdb matching against sigcode instructionsPhilip Guenthe
2011-09-02Differential gain calibration makes the 6005 firmware crap out, so skip it forMark Kettenis
now until we figure out why. This probably means the device won't function optimally, but that's better than not functioning at all. Makes my "Intel Centrinto Advanced-N 6205" work quite well.
2011-09-02generate a devid from vpd page 80 if vpd page 83 doesnt exist or work.David Gwynne
ok krw@
2011-09-01The 6005 and 6050-based parts need DC calibration turned on, otherwise theMark Kettenis
firmware will crap out, at least on the 6005. First step to getting my "Intel Centrino Advanced-N 6205" to work.
2011-08-31Report VXT cpu as `VXT', not `KA48', to match what the PROM says.Miod Vallat
2011-08-31Don't pretend to `enable floating point' and fiddle with PR_ACCS register onMiod Vallat
KA45 and KA48. The FPU is built-in, and PR_ACCS is not implemented (as the PROM will tell you if you try to EX it). This was copied from KA46 and for some reason KA45 and KA48 ignore writes to PR_ACCS (unlike VXT which hangs). Tested on KA45 and KA48.
2011-08-30regenMartin Pieuchot
2011-08-30Add another GeForce 8400 GSMartin Pieuchot
2011-08-30Don't OR the VID, we want the whole TCI, this makes vlanprio (PCP/CF)Christiano F. Haesbaert
available to our network stack. ok mcbride naddy henning
2011-08-30Make ps -N/-M work again by correcting kernel vs user addressPhilip Guenthe
handling in the FILL_KPROC() macro. Problems pointed out by Martin Pelikan (martin.pelikan at gmail.com)
2011-08-30Add support for one shot rules that remove themselves from an activeMike Belopuhov
ruleset after match. In case this is the only rule in the anchor, the anchor will be destroyed automatically after the rule is matched. This is an extremely handy technique for firewall proxies. ok henning, mcbride
2011-08-29Return of the long missing powerpc interrupt rewrite. Was working forDale Rahn
several weeks before release on macppc, socppc bugs just fixed.
2011-08-29protect sc->sc_state access and ncr53c9x_sched() call with splbio inMiod Vallat
ncr53c9x_poll(). Fixes a race causing *some* drives to trigger a panic in ncr53c9x_sched() at attach time on vax. ok dlg@ krw@
2011-08-29Only provide FLT_EVAL_METHOD for C99. Add missing DECIMAL_DIG for C99.Mark Kettenis
ok guenther@
2011-08-29Update firmware to the latest version available from Myricom (1.4.53a).David Gwynne
via freebsd, tested locally in production for a week or so.
2011-08-29dont use a semaphore (which doesnt block interrupts in the criticalDavid Gwynne
section) to protect the completion ring. mpii_poll can be in the middle of the critical section when an interrupt is generated, but because something is already in the critical section the isr cant clear the condition causing the interrupt and you spin forever entering the isr and being forced to exit it without doing any work by the semaphore. this moves to protecting the completion ring with a vanilla mutex. work taken off the ring is stuck on local lists in the isr to be completed outside the mutex in case a completion path issues a polled command which will try to call the isr which tries to mutex_enter while the previous call on the stack would be holding the mutex... this diff is large because of the list changes needed to support the local lists in mpii_intr. issue found by bluhm@ and debugged by Christian Ehrhardt
2011-08-29regenMike Belopuhov
2011-08-29add gpsmap 62sMike Belopuhov
2011-08-28lockmgr() wants to use a different address for the wchan when drainingPhilip Guenthe
the lock, but a change in member ordering meant it was using the same address. Explicitly use different members instead of mixing address of member and address of the lock itself. ok miod@
2011-08-26Be sure to write to the 7990 address register between two consecutive readsMiod Vallat
of the data register, while waiting for the chip initialization to complete. This seems to prevent spurious disappearance of le0 on some systems.
2011-08-26Be sure not to try to match lcg, and to bail out early in cnprobe as well,Miod Vallat
if we are running on a KA45 system. These share the KA48 board type but, unlike KA48, can not sport frame buffers. Prevents a bogus lcg device from attaching on KA45, and also repairs console operation if the S3 switch is down (console on serial #0).
generated by cgit v1.2.3 (git 2.25.1) at 2025-01-09 06:56:27 +0000