Compile time per kernel increased alot after recent image commits, re-shuffle some code to fix this.
Patch by "LazyDodo".
Differential Revision: https://developer.blender.org/D2012
Once and for all - I hope!
So, for the records: Xinput1 events' valuators have a first_axis and axis_count defining the 'range'
of valid axes for that event - but valid data in the valuator's array **always** start at zero
(which means that, if event reports first axis as being axis 2, axis 2's value is in axis_data[0]).
This adds support for CUDA Texture objects (also known as Bindless textures) for Kepler GPUs (Geforce 6xx and above).
This is used for all 2D/3D textures, data still uses arrays as before.
User benefits:
* No more limits of image textures on Kepler.
We had 5 float4 and 145 byte4 slots there before, now we have 1024 float4 and 1024 byte4.
This can be extended further if we need to (just change the define).
* Single channel textures slots (byte and float) are now supported on Kepler as well (1024 slots for each type).
ToDo / Issues:
* 3D textures don't work yet, at least don't show up during render. I have no idea whats wrong yet.
* Dynamically allocate bindless_mapping array?
I hope Fermi still works fine, but that should be tested on a Fermi card before pushing to master.
Part of my GSoC 2016.
Reviewers: sergey, #cycles, brecht
Subscribers: swerner, jtheninja, brecht, sergey
Differential Revision: https://developer.blender.org/D1999
The idea of pole merge is to fade interocular distance after a certain
altitude to zero when altitude goes closer to a pole. This should prevent
annoyances looking up in the sky or down to the bottom.
Works for both panorama and perspective cameras when Spherical Stereo
is enabled.
Reviewers: dfelinto, brecht
Reviewed By: brecht
Subscribers: sebastian_k
Differential Revision: https://developer.blender.org/D1998
This commit makes it so malloc() is only happening once per volume and
once per transparent shadow query (per thread), improving scalability of
the code to multiple CPU cores.
Hard to measure this with a low-bottom i7 here currently, but from quick
tests seems volume sampling gave about 3-5% speedup.
The idea is to store allocated memory in kernel globals, which are per
thread on CPU already.
Reviewers: dingto, juicyfruit, lukasstockner97, maiself, brecht
Reviewed By: brecht
Subscribers: Blendify, nutel
Differential Revision: https://developer.blender.org/D1996
This way, we also save 3/4th of memory for single channel byte textures (e.g. Bump Maps).
Note: In order for this to work, the texture *must* have 1 channel only.
In Gimp you can e.g. do that via the menu: Image -> Mode -> Grayscale
Until now, single channel textures were packed into a float4, wasting 3 floats per pixel. Memory usage of such textures is now reduced by 3/4.
Voxel Attributes such as density, flame and heat benefit from this, but also Bumpmaps with one channel.
This commit also includes some cleanup and code deduplication for image loading.
Example Smoke render from Cosmos Laundromat: http://www.pasteall.org/pic/show.php?id=102972
Memory here went down from ~600MB to ~300MB.
Reviewers: #cycles, brecht
Differential Revision: https://developer.blender.org/D1981
Many keyboard layouts (italian, spanish, german...) have direct access to '+' key on main
keyboard area (not the numpad one), ans x11 has own define for this key, so use it instead
of generating an unkown key event.
Note that we most likely have much more missing 'specific' keycodes for non-US keyboard layout,
but think since we already had a 'minus' keyevent, supporting 'plus' one is totally consistent.
And we had a spare space in our defined values just for it even!
This keyevent is only supported/generated by x11 and cocoa Ghost backends for now,
neither SDL nor win32 seem to have matching key events...
Title says it all, this adds OpenCL float4 texture support.
There is a bug in the code still, I get a "Out of ressources error" on nvidia hardware here, not sure whats wrong yet.
Will investigate further, but maybe someone else has an idea. :)
Reviewers: #cycles, brecht
Subscribers: brecht, candreacchio
Differential Revision: https://developer.blender.org/D1983
This commit:
* Removes most of all dirty internal details from public atomi_ops.h file, and move them into /intern private subdir.
* Removes unused 'architectures' (__apple__ and jemalloc).
* Split each implementation into its own file.
* Makes use of C99's limits.h system header to determine pointer and int size, instead of using fix hardcoded list of architectures.
* Introduces new 'faked' atomics ops for floats.
Note that we may add a lot more real and 'faked' atomic operations over integers and floats
(multiplication, division, bitshift, bitwise booleans, etc.), as needs arise.
Reviewers: sergey, campbellbarton
Differential Revision: https://developer.blender.org/D1982
This commit makes remove_table skip the freeing if the offset is
already set to invalid - or, if it wasn't, set it to invalid after freeing.
That's what the current code was already doing in the Manager classes,
this change allows them to just call remove without the additional code.
Also, two potential memory leaks where new tables were always allocated
without freeing the old ones are fixed.
Reviewers: sergey, dingto, brecht
Differential Revision: https://developer.blender.org/D1974
Swap the for-loops in the RenderBuffer reset code to follow the convention
of looping over y in the outer loop.
The improved cache performance won't really be noticable here, but it's nicer
if it follows the usual style.
