| Age | Commit message (Collapse) | Author |
|---|
|
|
|
device)
|
|
the intended meaning.
|
|
on amd64.
Evaluating the _PDC seems to lead to instability (PR5878 and a report by
jmc@) on lower end machines its too close to release and I don't have
enough nickels.
ok deraadt@
|
|
controller.
Found, diagnosed and patch from Stefan Sperling <stsp@stsp.name>
Minor KNF changes while here.
Tested by many on a variaty of IBM laptops and others with and without
acpiec.
ok deraadt
|
|
The current method results in a MAC address of all zeros.
Tested by bernd@/jcs@ with 8102E and fkr@ with 8168C.
From FreeBSD
ok jsg@
|
|
to cause issues as per netbsd pr/39023. deraadt@ yuo@ ok.
|
|
"fragment" gets replaced by "sample".
"sample" gets replaced by "frame".
A sample is a peace of a frame (final image). No functional changes.
OK deraadt@
|
|
|
|
|
|
|
|
for some devices and we will fail back to GET_CUR then.
|
|
the DRM_{INFO,ERROR,DEBUG} printf macros still need some __VA_ARGS__ love.
reyk@ and marco@ liked this, prompted by deraadt.
|
|
work.
ok deraadt@
|
|
claims to (as seen on the M$ LifeCam NX6000), abort further processing.
Safes us from potential NULL pointer derefs.
|
|
|
|
custom ones.
- Define functions which use USB return values with usbd_status
instead int.
- Define functions which always return 0 with void instead int.
- (void) function calls which don't check the return code.
|
|
DPRINTF and printf messages while here.
|
|
From jcs@ and in accordance with what the Intel Linux driver does.
|
|
field is unreliable and can start with any number. Use an own internal
array index instead.
- If the read buffer is too small, return a propper error to the calling
functions. Just check the buffer size if we use the read(2) method since
it doesn't affect mmap(2).
Fixes kernel crashes seen with the M$ LifeCam NX-6000 and internal
(laptop) Sonix chipsets.
Tested by jcs@ (Sonix) and myself (NX-6000).
OK deraadt@
|
|
ok deraadt@
|
|
prints it. These should be the last offenders.
|
|
handle in the detach function. Fixes compilation on sparc64.
ok reyk@
|
|
|
|
|
|
|
|
|
|
in a Lenovo T60, reported by Juan Delgado. ok you@, jsg@ deraadt@
|
|
|
|
with different versions for various ar5212 variants and add an extra
table for PCI-E devices. This fixes support for various newer devices
(like the 1st generation MacBook, T61 variants) but it still does not
work on a number of other devices.
Tested by many
ok deraadt@
|
|
From Andrew Thompson at FreeBSD.
|
|
|
|
Some stability fixes for radeon. The most part of this diff is related
to fixing up the VBLANK (vertical blank interrupt) handling. Now, if the
X driver supports the DRM_IOCTL_MODESET_CTL ioctl, (to be used when
changing the video modes), then allow the vblank to be disabled once
that ioctl has been called. Otherwise, keep the interrupt enabled at all
time, since disabling it otherwise will lead to problems.
Tested by a few. "no problem" on API/ABI deraadt@.
|
|
pointers we don't own, but the autoconf machinery does.
|
|
work and this way we fall back on PIO which does work. I'm pretty sure
this is caused by bugs in our code, but I cannot find them.
ok deraadt@, marco@
|
|
and drm_calloc.
With the recent change to check overflow in drm_calloc, this means that
all allocations that require multiplication are now checked. Also use
drm_calloc() when zeroing is needed and drop the bzero/memset
afterwards. Finally, make drm_free() check for NULL, so we don't need
to do so every time.
ok miod@, deraadt@
|
|
which may come from userland via ioctls. ok oga@
|
|
|
|
|
|
and just return;
OK mglocker@
|
|
properly in audio_rint(), the periodic boundary conditions aren't met. This
causes, later read(2) to return EFAULT while trying to access unmapped
regions of the kernel address space. Fix this by using the correct pointer
arithmetic.
ok jakemsr@
|
|
on an attachment dependent flag; the Tadople SPARCLE needs it.
ok kettenis@ deraadt@
|
|
Fix a typo while here.
The rebuild field was appended to the metadata so that we can automatically
update from v2 to v3. The only visible difference will be that in the dmesg
the SCSI version number will go from 002 to 003.
ok deraad.
|
|
RX/TX configuration registers to 128 instead of 512 bytes.
- add a few more MAC/RF id strings for the dmesg.
- check for PCI-E instead of single chip variants in a few places.
ok deraadt@
|
|
and then the whole string based on the returned size, not just one byte.
The Lexar card reader kindly loaned by
Denis Doroshenko (denis dot doroshenko at gmail.com) it's happy now, and
after further inspection it's what other OSs do.
Discussed with yuo@. krw@ yuo@ ok.
|
|
|
|
it while atheros' driver says something wrong (even if it is
technically correct). also rename AR9280 PCIE to AR9281 (guessed).
triggered by kettenis@
|
|
|
|
numerical order.
|
|
|
