What about lockedep_is_held() ??

Move code out of locked region.
Add __prefict_true/false.

indentation and style changes,
moved subnet check code from rules_check to parser,

Thank you for working on this!

I don't think the drm kmods are built with INVARIANTS, unless it's somehow passed into the build environment when building out of tree kmods. In general, are there other options added when building the DRM kmods (or other kmods) as part of the kernel build?

Oh, hello Theron.

This comment concerns Mesa drivers, and libGL in general, but not much else:

Incorporated latest feedback :-)

• Simplify COMMENT variable definition.

In D16830#358878, @iwtcex_gmail.com wrote:

made an attempt … according to the initial variant of this patch

In D16698#455371, @wulf wrote:

In D16698#455200, @greg_unrelenting.technology wrote:

We have the wmt driver but right now it's *touchscreen*-only and (of course) USB-only.

I have half-completed (it is working for me!) WIP port of wmt for i2c bus https://github.com/wulf7/iichid.

LGTM, thanks.

Hmmm, arc updated this rather than created a new one. A new one isn't worth creating for something this small...

cem's ideas

In D20983#455261, @rpokala wrote:

Now this includes parts of D20969 😕
The core of this change -- sort and de-dupe getopt() and sort the cases -- is fine. Please try again with just that part, and leave the rest as part of D20969 as intended.

In D16698#455403, @wulf wrote:

OS driver or userland applications just do not have enough information as finger coords or count to support doublefinger scroll or multifinger taps.

I wrote a simple util to dump report descriptor from I2C HID device

fetchdesc.c3 KBDownload

.
It accepts 3 input parameters: iic device path, i2c bus address and i2c HID register and prints report descriptor of given I2C device on stdout. This report descriptor can be further analyzed with any tool (e.g. web-based http://eleccelerator.com/usbdescreqparser/) to examine HID device capabilities.

In D20969#455325, @jilles wrote:

Perhaps it will be more generically useful to add a ${name}_umask variable to rc.subr instead.
Ideally this would go via a login class but the ${name}_login_class variable is already used for "rlimits from this login class".

Reorder to avoid leak.