| Age | Commit message (Collapse) | Author | |
|---|---|---|---|
| 2008-11-06 | Oops. Forgot to commit this chunk with the last commit. | Owain Ainsworth | |
| Remove softc members and prototypes related to the tasklet system. | |||
| 2008-11-06 | Make sure we do not unlock twice. | David Hill | |
| nice catch, oga@ | |||
| 2008-11-06 | Remove the drm_locked_tasklet interface. The only consumer that used it | Owain Ainsworth | |
| was the i915 vblank swap ioctl, which just went kaput. | |||
| 2008-11-06 | The i915 vblank swap ioctl is fundamentally racy. | Owain Ainsworth | |
| using it allowed rendering to continue while waiting for a vblank swap, and often this lead to flickering and rendering a new scene before the swap. this broke a lot of things. With the removal of this swap, userland falls back to the old way of waiting for the vblank then doing the swap itself, this is smooth enough. I decided independantly to kill this, but the intel guys recently concurred. Comment change comes from Eric Anholt at intel. | |||
| 2008-11-06 | add a description for the recently added c-mode. | Igor Sobrado | |
| ok jmc@ | |||
| 2008-11-06 | if the freeprot flag (F) is set, do not do delayed frees for chunks | Otto Moerbeek | |
| (might catch errors closer to the trouble spot) and junk fill pages just before reuse instead of immediate (we can't access the page anyway) since we set PROT_NONE in the F case. ok djm@ | |||
| 2008-11-06 | regen | Brad Smith | |
| 2008-11-06 | Add the PCI id for the Broadcom BCM5903F Ethernet chipset. | Brad Smith | |
| 2008-11-06 | use spacing consistently to make example shorter and more readable; | Igor Sobrado | |
| improve indentation. | |||
| 2008-11-06 | input and volume twiddles | Todd T. Fries | |
| o set init data to match snapper_set_foo* o set volume to 190 (audible compared to the default) o set input to mic (most common usage) ok jakemsr@ | |||
| 2008-11-06 | More complete example, for e.g., usb devices that need wep/wpa keys. | Ian Darwin | |
| OK (with typo corrected) sthen@ | |||
| 2008-11-06 | -Wall -Wshadow clean, no binary change | Damien Miller | |
| 2008-11-06 | Set the IFCAP_VLAN_MTU capabilities flag so these interfaces are allowed | Brad Smith | |
| to transmit full sized VLAN tagged frames. | |||
| 2008-11-06 | add support for iso8859-13 | Martynas Venckus | |
| 2008-11-05 | return an error when trying to get stereo values from a mono | Jacob Meuser | |
| control. makes mixerctl(1) output more sensible. ok drahn, todd | |||
| 2008-11-05 | add missing header needed by strcmp/strerror/memset functions. | Charles Longeau | |
| ok canacar@ | |||
| 2008-11-05 | Fix some function prototypes, remove unused variables and commented out code | Can Erkin Acar | |
| 2008-11-05 | add gilles@ as copyright holder for this file. | Igor Sobrado | |
| 2008-11-05 | add a few missing id tags; there are a bunch of files, and developers | Igor Sobrado | |
| will probably miss this change when working on more important matters, so it is probably better to sort them now. there is a risk of losing the tags if a change needs to be reverted too. written with excellent advice from jmc@ ok gilles@ | |||
| 2008-11-05 | passord -> password; | Jason McIntyre | |
| fixes user/5975 from Rene Maroufi | |||
| 2008-11-05 | wrap use of KNOTE macro arguments in () to prevent potential strange | David Gwynne | |
| expansion. requested by otto@ | |||
| 2008-11-05 | wrap an if statement in a macro up with do { } while (0) so it is safe to | David Gwynne | |
| use in other if/else blocks. "yeah" deraadt@ | |||
| 2008-11-05 | since mmclose() is only called once for the final close, | Matthieu Herrb | |
| set ap_open_count = 0 in mmclose() instread of decrementing it. ok miod@, oga@. | |||
| 2008-11-05 | sync | Theo de Raadt | |
| 2008-11-05 | Remove some PCI macros no longer being used. | Brad Smith | |
| 2008-11-05 | add dynamic forward escape command line; ok djm@ | Kevin Steves | |
| 2008-11-05 | remove unused macros, partly from Alexey Suslikov | Jacob Meuser | |
| 2008-11-05 | * use uint8_t instead of int to hold the value of an 8-bit register. | Jacob Meuser | |
| * for outputs.{bass,treble}, 0 dB corresponds to mixer value 128, not 0. * line-in is configured as the default recording source; set sc->sc_record_source accordingly. * add missing register to initialization code. from Marco (marco2z at arsystel com), thanks testing/ok drahn@ | |||
| 2008-11-05 | * remove two custom unsolicited event handlers in favor of generic | Jacob Meuser | |
| jack sense handling. makes azalia_generic_mixer_pinctrl useless, so zap it. * azalia_generic_mixer_pin_sense works well enough to not need guessing of pin direction in azalia_generic_mixer_default. from Alexey Suslikov, thanks | |||
| 2008-11-04 | man page states that alpha is fully ieee compliant, but it does not | Martynas Venckus | |
| maintain inexact flag. some platforms implement -mieee-with-inexact; however it's not practical since it slows down the code, with very small benefit. note the fact in the man page that inexact flag is not maintained. w/ a tweak from jmc@. ok jmc@, millert@, merdely@ | |||
| 2008-11-04 | Use defines for constants. Use __attribute__ bounded. | Hans-Joerg Hoexer | |
| ok markus@ (quite some time ago) | |||
| 2008-11-04 | conversions are smarter now, remove corresponding ``BUGS'' item | Alexandre Ratchov | |
| 2008-11-04 | split code that converts any->any in two parts: one to encode | Alexandre Ratchov | |
| native->any and one to decode any->native. It is simpler and faster this way. | |||
| 2008-11-04 | word better than number here. | Igor Sobrado | |
| 2008-11-04 | uvmspace_unshare() is never used; ok miod | Theo de Raadt | |
| 2008-11-04 | implement azalia_create_encodings, which scans through the supported | Jacob Meuser | |
| formats and creates an array of unique encodings. use the array in azalia_query_encoding. in other words, no more duplicate encodings in `audioctl encodings`. | |||
| 2008-11-04 | Reintroduce the extendsidi2 optimized sequence, but as a define_split | Miod Vallat | |
| construct, so that it is still written in rtl statements, and part of it can be put in a delay slot. And the way it's written now, it does not create bogus uninitialized warnings. | |||
| 2008-11-04 | need to take mode into account when verifying/setting parameters | Jacob Meuser | |
| 2008-11-04 | sync | Theo de Raadt | |
| 2008-11-04 | fix potential null dereference. | Charles Longeau | |
| Found by LLVM/Clang Static Analyzer. initial patch from me, rewritten by jsing@ ok jsg@ | |||
| 2008-11-04 | because parse_forward() is now used to parse all forward types (DLR), | Kevin Steves | |
| and it malloc's space for host variables, we don't need to malloc here. fixes small memory leaks. previously dynamic forwards were not parsed in parse_forward() and space was not malloc'd in that case. ok djm@ | |||
| 2008-11-04 | spelling: ambigious -> ambiguous | Marc Espie | |
| 2008-11-04 | cleanup: remove channel mapping bits from encoding conversion code. | Alexandre Ratchov | |
| As a side effect, reduce CPU usage by ~10% on envy(4) devices. | |||
| 2008-11-04 | add "new" conversion code to map any channel range into any other | Alexandre Ratchov | |
| channel range (based on the encoding conversion code). Will allow to unentangle channel mapping from encoding conversion. As a side effect, greatly reduces CPU usage for channel mapping. No functionnal change. | |||
| 2008-11-04 | sync | Theo de Raadt | |
| 2008-11-04 | new devs | Theo de Raadt | |
| 2008-11-04 | If we need to call the tasklet function on unlock, we don't need to hold | Owain Ainsworth | |
| tsk_lock (which blocks irqs) for the whole call of the function, just when we manipulate the function pointer. | |||
| 2008-11-04 | cleanup: resample stream while in native format and remove resampling | Alexandre Ratchov | |
| bits from the encoding conversion code. As a side effect this reduces CPU usage by 30% on envy(4). | |||
| 2008-11-04 | missing setipl in the 'busy processing soft interrupts state' | Dale Rahn | |
| causes high priority interrupts to be delayed until that state is cleared. | |||
| 2008-11-04 | optimization: add "new" resampling code (actually based on the existing | Alexandre Ratchov | |
| conversion bits) and use it when resampling only is required (ie for clients using s16 encoding), this is the most common case. Reduces CPU usage by ~50%. No functional change. | |||
 |
index : src | |
| OpenBSD base system |
| summaryrefslogtreecommitdiff |