mesa/src/gallium
Philipp Zabel dddec9a66d teflon: Support fused ReLU6 activation via output saturation
If the output tensor quantization range does not exceed 6.0, ReLU6 can
be replaced with ReLU: output values larger than 6.0 are clipped by
output saturation.

Reviewed-by: Tomeu Vizoso <tomeu@tomeuvizoso.net>
Signed-off-by: Philipp Zabel <p.zabel@pengutronix.de>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/32388>
2024-11-29 13:32:42 +00:00
..
auxiliary Fix copy-and-paste bug in nir_lower_aapoint_impl 2024-11-26 18:12:07 +00:00
drivers etnaviv/ml: Create combined input tensors for addition first 2024-11-29 10:46:33 +00:00
frontends teflon: Support fused ReLU6 activation via output saturation 2024-11-29 13:32:42 +00:00
include teflon: Add is_signed parameter to ml_subgraph_invoke and ml_subgraph_read_output 2024-11-29 10:52:48 +01:00
targets teflon/tests: Enable int8 tests 2024-11-29 10:53:01 +01:00
tests gallium/meson: Deconflate swrast/softpipe/llvmpipe 2024-07-18 17:48:20 +00:00
tools gallium: remove take_ownership from set_vertex_buffers, assume it's true 2024-02-07 09:19:42 +00:00
winsys winsys/amdgpu: fix FD mismatch 2024-11-26 21:21:29 +00:00
meson.build targets/vdpau: Build vdpau driver into libgallium when building with dri 2024-09-11 08:07:21 +00:00
README.portability util: include "util/compiler.h" instead of "pipe/p_compiler.h" 2023-06-27 18:18:30 +08:00

	      CROSS-PLATFORM PORTABILITY GUIDELINES FOR GALLIUM3D 


= General Considerations =

The frontend and winsys driver support a rather limited number of
platforms. However, the pipe drivers are meant to run in a wide number of
platforms. Hence the pipe drivers, the auxiliary modules, and all public
headers in general, should strictly follow these guidelines to ensure


= Compiler Support =

* Include the util/compiler.h.

* Cast explicitly when converting to integer types of smaller sizes.

* Cast explicitly when converting between float, double and integral types.

* Don't use named struct initializers.

* Don't use variable number of macro arguments. Use static inline functions
instead.

* Don't use C99 features.

= Standard Library =

* Avoid including standard library headers. Most standard library functions are
not available in Windows Kernel Mode. Use the appropriate p_*.h include.

== Memory Allocation ==

* Use MALLOC, CALLOC, FREE instead of the malloc, calloc, free functions.

* Use align_pointer() function defined in u_memory.h for aligning pointers
 in a portable way.

== Debugging ==

* Use the functions/macros in p_debug.h.

* Don't include assert.h, call abort, printf, etc.


= Code Style =

== Inherantice in C ==

The main thing we do is mimic inheritance by structure containment.

Here's a silly made-up example:

/* base class */
struct buffer
{
  int size;
  void (*validate)(struct buffer *buf);
};

/* sub-class of bufffer */
struct texture_buffer
{
  struct buffer base;  /* the base class, MUST COME FIRST! */
  int format;
  int width, height;
};


Then, we'll typically have cast-wrapper functions to convert base-class 
pointers to sub-class pointers where needed:

static inline struct vertex_buffer *vertex_buffer(struct buffer *buf)
{
  return (struct vertex_buffer *) buf;
}


To create/init a sub-classed object:

struct buffer *create_texture_buffer(int w, int h, int format)
{
  struct texture_buffer *t = malloc(sizeof(*t));
  t->format = format;
  t->width = w;
  t->height = h;
  t->base.size = w * h;
  t->base.validate = tex_validate;
  return &t->base;
}

Example sub-class method:

void tex_validate(struct buffer *buf)
{
  struct texture_buffer *tb = texture_buffer(buf);
  assert(tb->format);
  assert(tb->width);
  assert(tb->height);
}


Note that we typically do not use typedefs to make "class names"; we use
'struct whatever' everywhere.

Gallium's pipe_context and the subclassed psb_context, etc are prime examples 
of this.  There's also many examples in Mesa and the Mesa state tracker.