On 18/04/17 20:48, Nicolai Hähnle wrote:
On 18.04.2017 02:09, Timothy Arceri wrote:
V2: some bug fixes and suggestions addressed as suggested by Brian and
Nicolai.
This series adds some initial support for the KHR_no_error.
For now it can only be enabled with the MESA_NO_ERROR environment
variable. To start with I've added support to some of the api
calls that were taking a long time (or where caled often) in
the Civ6 benchmark.
Just as a matter of general musing, I wonder if in some cases the
no_error version of an API entry point can be directly configured in the
dispatch table. I.e., have a _mesa_no_error_Foo function which the
_mesa_Foo function chains to. Though often you have lookups in both
cases. Perhaps this is a rare case where templates are actually useful?
Having a "no_error" template argument so that the compiler creates two
function versions for us automatically, if such a thing were possible in
C. Something to consider for later perhaps.
My current thinking was to do this on a case by case basis, adding an
extra attribute to the API XML to flag a function that has a no error
variant. There will be cases like the Uniform functions where it makes
sense to just use the if branch to avoid excessive code duplication.
Other than that I agree it probably makes sense to have the dispatch
table point directly to a no error function where possible.
I think patch #1 needs a security check. I also sent a minor comment on
patch #4, apart from that patches 2-9, 11, 12 are
Reviewed-by: Nicolai Hähnle <nicolai.haeh...@amd.com>
_______________________________________________
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/mesa-dev
_______________________________________________
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/mesa-dev
