You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.

719 lines
21 KiB

#include "rendercopy.h"
#include "intel_chipset.h"
#include "gen4_render.h"
#include "surfaceformat.h"
#include <assert.h>
#define VERTEX_SIZE (3*4)
#define URB_VS_ENTRY_SIZE 1
#define URB_GS_ENTRY_SIZE 0
#define URB_CL_ENTRY_SIZE 0
#define URB_SF_ENTRY_SIZE 2
#define URB_CS_ENTRY_SIZE 1
#define GEN4_GRF_BLOCKS(nreg) (((nreg) + 15) / 16 - 1)
#define SF_KERNEL_NUM_GRF 16
#define PS_KERNEL_NUM_GRF 32
static const uint32_t gen4_sf_kernel_nomask[][4] = {
{ 0x00400031, 0x20c01fbd, 0x0069002c, 0x01110001 },
{ 0x00600001, 0x206003be, 0x00690060, 0x00000000 },
{ 0x00600040, 0x20e077bd, 0x00690080, 0x006940a0 },
{ 0x00600041, 0x202077be, 0x008d00e0, 0x000000c0 },
{ 0x00600040, 0x20e077bd, 0x006900a0, 0x00694060 },
{ 0x00600041, 0x204077be, 0x008d00e0, 0x000000c8 },
{ 0x00600031, 0x20001fbc, 0x008d0000, 0x8640c800 },
};
static const uint32_t gen5_sf_kernel_nomask[][4] = {
{ 0x00400031, 0x20c01fbd, 0x1069002c, 0x02100001 },
{ 0x00600001, 0x206003be, 0x00690060, 0x00000000 },
{ 0x00600040, 0x20e077bd, 0x00690080, 0x006940a0 },
{ 0x00600041, 0x202077be, 0x008d00e0, 0x000000c0 },
{ 0x00600040, 0x20e077bd, 0x006900a0, 0x00694060 },
{ 0x00600041, 0x204077be, 0x008d00e0, 0x000000c8 },
{ 0x00600031, 0x20001fbc, 0x648d0000, 0x8808c800 },
};
static const uint32_t gen4_ps_kernel_nomask_affine[][4] = {
{ 0x00800040, 0x23c06d29, 0x00480028, 0x10101010 },
{ 0x00800040, 0x23806d29, 0x0048002a, 0x11001100 },
{ 0x00802040, 0x2100753d, 0x008d03c0, 0x00004020 },
{ 0x00802040, 0x2140753d, 0x008d0380, 0x00004024 },
{ 0x00802059, 0x200077bc, 0x00000060, 0x008d0100 },
{ 0x00802048, 0x204077be, 0x00000064, 0x008d0140 },
{ 0x00802059, 0x200077bc, 0x00000070, 0x008d0100 },
{ 0x00802048, 0x208077be, 0x00000074, 0x008d0140 },
{ 0x00600201, 0x20200022, 0x008d0000, 0x00000000 },
{ 0x00000201, 0x20280062, 0x00000000, 0x00000000 },
{ 0x01800031, 0x21801d09, 0x008d0000, 0x02580001 },
{ 0x00600001, 0x204003be, 0x008d0180, 0x00000000 },
{ 0x00601001, 0x20c003be, 0x008d01a0, 0x00000000 },
{ 0x00600001, 0x206003be, 0x008d01c0, 0x00000000 },
{ 0x00601001, 0x20e003be, 0x008d01e0, 0x00000000 },
{ 0x00600001, 0x208003be, 0x008d0200, 0x00000000 },
{ 0x00601001, 0x210003be, 0x008d0220, 0x00000000 },
{ 0x00600001, 0x20a003be, 0x008d0240, 0x00000000 },
{ 0x00601001, 0x212003be, 0x008d0260, 0x00000000 },
{ 0x00600201, 0x202003be, 0x008d0020, 0x00000000 },
{ 0x00800031, 0x20001d28, 0x008d0000, 0x85a04800 },
};
static const uint32_t gen5_ps_kernel_nomask_affine[][4] = {
{ 0x00800040, 0x23c06d29, 0x00480028, 0x10101010 },
{ 0x00800040, 0x23806d29, 0x0048002a, 0x11001100 },
{ 0x00802040, 0x2100753d, 0x008d03c0, 0x00004020 },
{ 0x00802040, 0x2140753d, 0x008d0380, 0x00004024 },
{ 0x00802059, 0x200077bc, 0x00000060, 0x008d0100 },
{ 0x00802048, 0x204077be, 0x00000064, 0x008d0140 },
{ 0x00802059, 0x200077bc, 0x00000070, 0x008d0100 },
{ 0x00802048, 0x208077be, 0x00000074, 0x008d0140 },
{ 0x01800031, 0x21801fa9, 0x208d0000, 0x0a8a0001 },
{ 0x00802001, 0x304003be, 0x008d0180, 0x00000000 },
{ 0x00802001, 0x306003be, 0x008d01c0, 0x00000000 },
{ 0x00802001, 0x308003be, 0x008d0200, 0x00000000 },
{ 0x00802001, 0x30a003be, 0x008d0240, 0x00000000 },
{ 0x00600201, 0x202003be, 0x008d0020, 0x00000000 },
{ 0x00800031, 0x20001d28, 0x548d0000, 0x94084800 },
};
static uint32_t
batch_used(struct intel_batchbuffer *batch)
{
return batch->ptr - batch->buffer;
}
static uint32_t
batch_round_upto(struct intel_batchbuffer *batch, uint32_t divisor)
{
uint32_t offset = batch_used(batch);
offset = (offset + divisor - 1) / divisor * divisor;
batch->ptr = batch->buffer + offset;
return offset;
}
static int gen4_max_vs_nr_urb_entries(uint32_t devid)
{
return IS_GEN5(devid) ? 256 : 32;
}
static int gen4_max_sf_nr_urb_entries(uint32_t devid)
{
return IS_GEN5(devid) ? 128 : 64;
}
static int gen4_urb_size(uint32_t devid)
{
return IS_GEN5(devid) ? 1024 : IS_G4X(devid) ? 384 : 256;
}
static int gen4_max_sf_threads(uint32_t devid)
{
return IS_GEN5(devid) ? 48 : 24;
}
static int gen4_max_wm_threads(uint32_t devid)
{
return IS_GEN5(devid) ? 72 : IS_G4X(devid) ? 50 : 32;
}
static void
gen4_render_flush(struct intel_batchbuffer *batch,
drm_intel_context *context, uint32_t batch_end)
{
int ret;
ret = drm_intel_bo_subdata(batch->bo, 0, 4096, batch->buffer);
if (ret == 0)
ret = drm_intel_gem_bo_context_exec(batch->bo, context,
batch_end, 0);
assert(ret == 0);
}
static uint32_t
gen4_bind_buf(struct intel_batchbuffer *batch,
const struct igt_buf *buf,
int is_dst)
{
struct gen4_surface_state *ss;
uint32_t write_domain, read_domain;
int ret;
igt_assert_lte(buf->stride, 128*1024);
igt_assert_lte(igt_buf_width(buf), 8192);
igt_assert_lte(igt_buf_height(buf), 8192);
if (is_dst) {
write_domain = read_domain = I915_GEM_DOMAIN_RENDER;
} else {
write_domain = 0;
read_domain = I915_GEM_DOMAIN_SAMPLER;
}
ss = intel_batchbuffer_subdata_alloc(batch, sizeof(*ss), 32);
ss->ss0.surface_type = SURFACE_2D;
switch (buf->bpp) {
case 8: ss->ss0.surface_format = SURFACEFORMAT_R8_UNORM; break;
case 16: ss->ss0.surface_format = SURFACEFORMAT_R8G8_UNORM; break;
case 32: ss->ss0.surface_format = SURFACEFORMAT_B8G8R8A8_UNORM; break;
case 64: ss->ss0.surface_format = SURFACEFORMAT_R16G16B16A16_FLOAT; break;
default: igt_assert(0);
}
ss->ss0.data_return_format = SURFACERETURNFORMAT_FLOAT32;
ss->ss0.color_blend = 1;
ss->ss1.base_addr = buf->bo->offset;
ret = drm_intel_bo_emit_reloc(batch->bo,
intel_batchbuffer_subdata_offset(batch, ss) + 4,
buf->bo, 0,
read_domain, write_domain);
assert(ret == 0);
ss->ss2.height = igt_buf_height(buf) - 1;
ss->ss2.width = igt_buf_width(buf) - 1;
ss->ss3.pitch = buf->stride - 1;
ss->ss3.tiled_surface = buf->tiling != I915_TILING_NONE;
ss->ss3.tile_walk = buf->tiling == I915_TILING_Y;
return intel_batchbuffer_subdata_offset(batch, ss);
}
static uint32_t
gen4_bind_surfaces(struct intel_batchbuffer *batch,
const struct igt_buf *src,
const struct igt_buf *dst)
{
uint32_t *binding_table;
binding_table = intel_batchbuffer_subdata_alloc(batch, 32, 32);
binding_table[0] = gen4_bind_buf(batch, dst, 1);
binding_table[1] = gen4_bind_buf(batch, src, 0);
return intel_batchbuffer_subdata_offset(batch, binding_table);
}
static void
gen4_emit_sip(struct intel_batchbuffer *batch)
{
OUT_BATCH(GEN4_STATE_SIP | (2 - 2));
OUT_BATCH(0);
}
static void
gen4_emit_state_base_address(struct intel_batchbuffer *batch)
{
if (IS_GEN5(batch->devid)) {
OUT_BATCH(GEN4_STATE_BASE_ADDRESS | (8 - 2));
OUT_RELOC(batch->bo, /* general */
I915_GEM_DOMAIN_INSTRUCTION, 0,
BASE_ADDRESS_MODIFY);
OUT_RELOC(batch->bo, /* surface */
I915_GEM_DOMAIN_INSTRUCTION, 0,
BASE_ADDRESS_MODIFY);
OUT_BATCH(0); /* media */
OUT_RELOC(batch->bo, /* instruction */
I915_GEM_DOMAIN_INSTRUCTION, 0,
BASE_ADDRESS_MODIFY);
/* upper bounds, disable */
OUT_BATCH(BASE_ADDRESS_MODIFY); /* general */
OUT_BATCH(0); /* media */
OUT_BATCH(BASE_ADDRESS_MODIFY); /* instruction */
} else {
OUT_BATCH(GEN4_STATE_BASE_ADDRESS | (6 - 2));
OUT_RELOC(batch->bo, /* general */
I915_GEM_DOMAIN_INSTRUCTION, 0,
BASE_ADDRESS_MODIFY);
OUT_RELOC(batch->bo, /* surface */
I915_GEM_DOMAIN_INSTRUCTION, 0,
BASE_ADDRESS_MODIFY);
OUT_BATCH(0); /* media */
/* upper bounds, disable */
OUT_BATCH(BASE_ADDRESS_MODIFY); /* general */
OUT_BATCH(0); /* media */
}
}
static void
gen4_emit_pipelined_pointers(struct intel_batchbuffer *batch,
uint32_t vs, uint32_t sf,
uint32_t wm, uint32_t cc)
{
OUT_BATCH(GEN4_3DSTATE_PIPELINED_POINTERS | (7 - 2));
OUT_BATCH(vs);
OUT_BATCH(GEN4_GS_DISABLE);
OUT_BATCH(GEN4_CLIP_DISABLE);
OUT_BATCH(sf);
OUT_BATCH(wm);
OUT_BATCH(cc);
}
static void
gen4_emit_urb(struct intel_batchbuffer *batch)
{
int vs_entries = gen4_max_vs_nr_urb_entries(batch->devid);
int gs_entries = 0;
int cl_entries = 0;
int sf_entries = gen4_max_sf_nr_urb_entries(batch->devid);
int cs_entries = 0;
int urb_vs_end = vs_entries * URB_VS_ENTRY_SIZE;
int urb_gs_end = urb_vs_end + gs_entries * URB_GS_ENTRY_SIZE;
int urb_cl_end = urb_gs_end + cl_entries * URB_CL_ENTRY_SIZE;
int urb_sf_end = urb_cl_end + sf_entries * URB_SF_ENTRY_SIZE;
int urb_cs_end = urb_sf_end + cs_entries * URB_CS_ENTRY_SIZE;
assert(urb_cs_end <= gen4_urb_size(batch->devid));
intel_batchbuffer_align(batch, 16);
OUT_BATCH(GEN4_URB_FENCE |
UF0_CS_REALLOC |
UF0_SF_REALLOC |
UF0_CLIP_REALLOC |
UF0_GS_REALLOC |
UF0_VS_REALLOC |
(3 - 2));
OUT_BATCH(urb_cl_end << UF1_CLIP_FENCE_SHIFT |
urb_gs_end << UF1_GS_FENCE_SHIFT |
urb_vs_end << UF1_VS_FENCE_SHIFT);
OUT_BATCH(urb_cs_end << UF2_CS_FENCE_SHIFT |
urb_sf_end << UF2_SF_FENCE_SHIFT);
OUT_BATCH(GEN4_CS_URB_STATE | (2 - 2));
OUT_BATCH((URB_CS_ENTRY_SIZE - 1) << 4 | cs_entries << 0);
}
static void
gen4_emit_null_depth_buffer(struct intel_batchbuffer *batch)
{
if (IS_G4X(batch->devid) || IS_GEN5(batch->devid)) {
OUT_BATCH(GEN4_3DSTATE_DEPTH_BUFFER | (6 - 2));
OUT_BATCH(SURFACE_NULL << GEN4_3DSTATE_DEPTH_BUFFER_TYPE_SHIFT |
GEN4_DEPTHFORMAT_D32_FLOAT << GEN4_3DSTATE_DEPTH_BUFFER_FORMAT_SHIFT);
OUT_BATCH(0);
OUT_BATCH(0);
OUT_BATCH(0);
OUT_BATCH(0);
} else {
OUT_BATCH(GEN4_3DSTATE_DEPTH_BUFFER | (5 - 2));
OUT_BATCH(SURFACE_NULL << GEN4_3DSTATE_DEPTH_BUFFER_TYPE_SHIFT |
GEN4_DEPTHFORMAT_D32_FLOAT << GEN4_3DSTATE_DEPTH_BUFFER_FORMAT_SHIFT);
OUT_BATCH(0);
OUT_BATCH(0);
OUT_BATCH(0);
}
if (IS_GEN5(batch->devid)) {
OUT_BATCH(GEN4_3DSTATE_CLEAR_PARAMS | (2 - 2));
OUT_BATCH(0);
}
}
static void
gen4_emit_invariant(struct intel_batchbuffer *batch)
{
OUT_BATCH(MI_FLUSH | MI_INHIBIT_RENDER_CACHE_FLUSH);
if (IS_GEN5(batch->devid) || IS_G4X(batch->devid))
OUT_BATCH(G4X_PIPELINE_SELECT | PIPELINE_SELECT_3D);
else
OUT_BATCH(GEN4_PIPELINE_SELECT | PIPELINE_SELECT_3D);
}
static uint32_t
gen4_create_vs_state(struct intel_batchbuffer *batch)
{
struct gen4_vs_state *vs;
int nr_urb_entries;
vs = intel_batchbuffer_subdata_alloc(batch, sizeof(*vs), 32);
/* Set up the vertex shader to be disabled (passthrough) */
nr_urb_entries = gen4_max_vs_nr_urb_entries(batch->devid);
if (IS_GEN5(batch->devid))
nr_urb_entries >>= 2;
vs->vs4.nr_urb_entries = nr_urb_entries;
vs->vs4.urb_entry_allocation_size = URB_VS_ENTRY_SIZE - 1;
vs->vs6.vs_enable = 0;
vs->vs6.vert_cache_disable = 1;
return intel_batchbuffer_subdata_offset(batch, vs);
}
static uint32_t
gen4_create_sf_state(struct intel_batchbuffer *batch,
uint32_t kernel)
{
struct gen4_sf_state *sf;
sf = intel_batchbuffer_subdata_alloc(batch, sizeof(*sf), 32);
sf->sf0.grf_reg_count = GEN4_GRF_BLOCKS(SF_KERNEL_NUM_GRF);
sf->sf0.kernel_start_pointer = kernel >> 6;
sf->sf3.urb_entry_read_length = 1; /* 1 URB per vertex */
/* don't smash vertex header, read start from dw8 */
sf->sf3.urb_entry_read_offset = 1;
sf->sf3.dispatch_grf_start_reg = 3;
sf->sf4.max_threads = gen4_max_sf_threads(batch->devid) - 1;
sf->sf4.urb_entry_allocation_size = URB_SF_ENTRY_SIZE - 1;
sf->sf4.nr_urb_entries = gen4_max_sf_nr_urb_entries(batch->devid);
sf->sf6.cull_mode = GEN4_CULLMODE_NONE;
sf->sf6.dest_org_vbias = 0x8;
sf->sf6.dest_org_hbias = 0x8;
return intel_batchbuffer_subdata_offset(batch, sf);
}
static uint32_t
gen4_create_wm_state(struct intel_batchbuffer *batch,
uint32_t kernel,
uint32_t sampler)
{
struct gen4_wm_state *wm;
wm = intel_batchbuffer_subdata_alloc(batch, sizeof(*wm), 32);
assert((kernel & 63) == 0);
wm->wm0.kernel_start_pointer = kernel >> 6;
wm->wm0.grf_reg_count = GEN4_GRF_BLOCKS(PS_KERNEL_NUM_GRF);
wm->wm3.urb_entry_read_offset = 0;
wm->wm3.dispatch_grf_start_reg = 3;
assert((sampler & 31) == 0);
wm->wm4.sampler_state_pointer = sampler >> 5;
wm->wm4.sampler_count = 1;
wm->wm5.max_threads = gen4_max_wm_threads(batch->devid);
wm->wm5.thread_dispatch_enable = 1;
wm->wm5.enable_16_pix = 1;
wm->wm5.early_depth_test = 1;
if (IS_GEN5(batch->devid))
wm->wm1.binding_table_entry_count = 0;
else
wm->wm1.binding_table_entry_count = 2;
wm->wm3.urb_entry_read_length = 2;
return intel_batchbuffer_subdata_offset(batch, wm);
}
static void
gen4_emit_binding_table(struct intel_batchbuffer *batch,
uint32_t wm_table)
{
OUT_BATCH(GEN4_3DSTATE_BINDING_TABLE_POINTERS | (6 - 2));
OUT_BATCH(0); /* vs */
OUT_BATCH(0); /* gs */
OUT_BATCH(0); /* clip */
OUT_BATCH(0); /* sf */
OUT_BATCH(wm_table); /* ps */
}
static void
gen4_emit_drawing_rectangle(struct intel_batchbuffer *batch,
const struct igt_buf *dst)
{
OUT_BATCH(GEN4_3DSTATE_DRAWING_RECTANGLE | (4 - 2));
OUT_BATCH(0);
OUT_BATCH((igt_buf_height(dst) - 1) << 16 |
(igt_buf_width(dst) - 1));
OUT_BATCH(0);
}
static void
gen4_emit_vertex_elements(struct intel_batchbuffer *batch)
{
if (IS_GEN5(batch->devid)) {
/* The VUE layout
* dword 0-3: pad (0.0, 0.0, 0.0, 0.0),
* dword 4-7: position (x, y, 1.0, 1.0),
* dword 8-11: texture coordinate 0 (u0, v0, 0, 0)
*
* dword 4-11 are fetched from vertex buffer
*/
OUT_BATCH(GEN4_3DSTATE_VERTEX_ELEMENTS | (3 * 2 + 1 - 2));
/* pad */
OUT_BATCH(0 << GEN4_VE0_VERTEX_BUFFER_INDEX_SHIFT | GEN4_VE0_VALID |
SURFACEFORMAT_R32G32B32A32_FLOAT << VE0_FORMAT_SHIFT |
0 << VE0_OFFSET_SHIFT);
OUT_BATCH(GEN4_VFCOMPONENT_STORE_0 << VE1_VFCOMPONENT_0_SHIFT |
GEN4_VFCOMPONENT_STORE_0 << VE1_VFCOMPONENT_1_SHIFT |
GEN4_VFCOMPONENT_STORE_0 << VE1_VFCOMPONENT_2_SHIFT |
GEN4_VFCOMPONENT_STORE_0 << VE1_VFCOMPONENT_3_SHIFT);
/* x,y */
OUT_BATCH(0 << GEN4_VE0_VERTEX_BUFFER_INDEX_SHIFT | GEN4_VE0_VALID |
SURFACEFORMAT_R16G16_SSCALED << VE0_FORMAT_SHIFT |
0 << VE0_OFFSET_SHIFT);
OUT_BATCH(GEN4_VFCOMPONENT_STORE_SRC << VE1_VFCOMPONENT_0_SHIFT |
GEN4_VFCOMPONENT_STORE_SRC << VE1_VFCOMPONENT_1_SHIFT |
GEN4_VFCOMPONENT_STORE_1_FLT << VE1_VFCOMPONENT_2_SHIFT |
GEN4_VFCOMPONENT_STORE_1_FLT << VE1_VFCOMPONENT_3_SHIFT);
/* u0, v0 */
OUT_BATCH(0 << GEN4_VE0_VERTEX_BUFFER_INDEX_SHIFT | GEN4_VE0_VALID |
SURFACEFORMAT_R32G32_FLOAT << VE0_FORMAT_SHIFT |
4 << VE0_OFFSET_SHIFT);
OUT_BATCH(GEN4_VFCOMPONENT_STORE_SRC << VE1_VFCOMPONENT_0_SHIFT |
GEN4_VFCOMPONENT_STORE_SRC << VE1_VFCOMPONENT_1_SHIFT |
GEN4_VFCOMPONENT_STORE_0 << VE1_VFCOMPONENT_2_SHIFT |
GEN4_VFCOMPONENT_STORE_0 << VE1_VFCOMPONENT_3_SHIFT);
} else {
/* The VUE layout
* dword 0-3: position (x, y, 1.0, 1.0),
* dword 4-7: texture coordinate 0 (u0, v0, 0, 0)
*
* dword 0-7 are fetched from vertex buffer
*/
OUT_BATCH(GEN4_3DSTATE_VERTEX_ELEMENTS | (2 * 2 + 1 - 2));
/* x,y */
OUT_BATCH(0 << GEN4_VE0_VERTEX_BUFFER_INDEX_SHIFT | GEN4_VE0_VALID |
SURFACEFORMAT_R16G16_SSCALED << VE0_FORMAT_SHIFT |
0 << VE0_OFFSET_SHIFT);
OUT_BATCH(GEN4_VFCOMPONENT_STORE_SRC << VE1_VFCOMPONENT_0_SHIFT |
GEN4_VFCOMPONENT_STORE_SRC << VE1_VFCOMPONENT_1_SHIFT |
GEN4_VFCOMPONENT_STORE_1_FLT << VE1_VFCOMPONENT_2_SHIFT |
GEN4_VFCOMPONENT_STORE_1_FLT << VE1_VFCOMPONENT_3_SHIFT |
4 << VE1_DESTINATION_ELEMENT_OFFSET_SHIFT);
/* u0, v0 */
OUT_BATCH(0 << GEN4_VE0_VERTEX_BUFFER_INDEX_SHIFT | GEN4_VE0_VALID |
SURFACEFORMAT_R32G32_FLOAT << VE0_FORMAT_SHIFT |
4 << VE0_OFFSET_SHIFT);
OUT_BATCH(GEN4_VFCOMPONENT_STORE_SRC << VE1_VFCOMPONENT_0_SHIFT |
GEN4_VFCOMPONENT_STORE_SRC << VE1_VFCOMPONENT_1_SHIFT |
GEN4_VFCOMPONENT_STORE_0 << VE1_VFCOMPONENT_2_SHIFT |
GEN4_VFCOMPONENT_STORE_0 << VE1_VFCOMPONENT_3_SHIFT |
8 << VE1_DESTINATION_ELEMENT_OFFSET_SHIFT);
}
}
static uint32_t
gen4_create_cc_viewport(struct intel_batchbuffer *batch)
{
struct gen4_cc_viewport *vp;
vp = intel_batchbuffer_subdata_alloc(batch, sizeof(*vp), 32);
vp->min_depth = -1.e35;
vp->max_depth = 1.e35;
return intel_batchbuffer_subdata_offset(batch, vp);
}
static uint32_t
gen4_create_cc_state(struct intel_batchbuffer *batch,
uint32_t cc_vp)
{
struct gen4_color_calc_state *cc;
cc = intel_batchbuffer_subdata_alloc(batch, sizeof(*cc), 64);
cc->cc4.cc_viewport_state_offset = cc_vp;
return intel_batchbuffer_subdata_offset(batch, cc);
}
static uint32_t
gen4_create_sf_kernel(struct intel_batchbuffer *batch)
{
if (IS_GEN5(batch->devid))
return intel_batchbuffer_copy_data(batch, gen5_sf_kernel_nomask,
sizeof(gen5_sf_kernel_nomask),
64);
else
return intel_batchbuffer_copy_data(batch, gen4_sf_kernel_nomask,
sizeof(gen4_sf_kernel_nomask),
64);
}
static uint32_t
gen4_create_ps_kernel(struct intel_batchbuffer *batch)
{
if (IS_GEN5(batch->devid))
return intel_batchbuffer_copy_data(batch, gen5_ps_kernel_nomask_affine,
sizeof(gen5_ps_kernel_nomask_affine),
64);
else
return intel_batchbuffer_copy_data(batch, gen4_ps_kernel_nomask_affine,
sizeof(gen4_ps_kernel_nomask_affine),
64);
}
static uint32_t
gen4_create_sampler(struct intel_batchbuffer *batch,
sampler_filter_t filter,
sampler_extend_t extend)
{
struct gen4_sampler_state *ss;
ss = intel_batchbuffer_subdata_alloc(batch, sizeof(*ss), 32);
ss->ss0.lod_preclamp = GEN4_LOD_PRECLAMP_OGL;
/* We use the legacy mode to get the semantics specified by
* the Render extension.
*/
ss->ss0.border_color_mode = GEN4_BORDER_COLOR_MODE_LEGACY;
switch (filter) {
default:
case SAMPLER_FILTER_NEAREST:
ss->ss0.min_filter = GEN4_MAPFILTER_NEAREST;
ss->ss0.mag_filter = GEN4_MAPFILTER_NEAREST;
break;
case SAMPLER_FILTER_BILINEAR:
ss->ss0.min_filter = GEN4_MAPFILTER_LINEAR;
ss->ss0.mag_filter = GEN4_MAPFILTER_LINEAR;
break;
}
switch (extend) {
default:
case SAMPLER_EXTEND_NONE:
ss->ss1.r_wrap_mode = GEN4_TEXCOORDMODE_CLAMP_BORDER;
ss->ss1.s_wrap_mode = GEN4_TEXCOORDMODE_CLAMP_BORDER;
ss->ss1.t_wrap_mode = GEN4_TEXCOORDMODE_CLAMP_BORDER;
break;
case SAMPLER_EXTEND_REPEAT:
ss->ss1.r_wrap_mode = GEN4_TEXCOORDMODE_WRAP;
ss->ss1.s_wrap_mode = GEN4_TEXCOORDMODE_WRAP;
ss->ss1.t_wrap_mode = GEN4_TEXCOORDMODE_WRAP;
break;
case SAMPLER_EXTEND_PAD:
ss->ss1.r_wrap_mode = GEN4_TEXCOORDMODE_CLAMP;
ss->ss1.s_wrap_mode = GEN4_TEXCOORDMODE_CLAMP;
ss->ss1.t_wrap_mode = GEN4_TEXCOORDMODE_CLAMP;
break;
case SAMPLER_EXTEND_REFLECT:
ss->ss1.r_wrap_mode = GEN4_TEXCOORDMODE_MIRROR;
ss->ss1.s_wrap_mode = GEN4_TEXCOORDMODE_MIRROR;
ss->ss1.t_wrap_mode = GEN4_TEXCOORDMODE_MIRROR;
break;
}
return intel_batchbuffer_subdata_offset(batch, ss);
}
static void gen4_emit_vertex_buffer(struct intel_batchbuffer *batch)
{
OUT_BATCH(GEN4_3DSTATE_VERTEX_BUFFERS | (5 - 2));
OUT_BATCH(GEN4_VB0_VERTEXDATA |
0 << GEN4_VB0_BUFFER_INDEX_SHIFT |
VERTEX_SIZE << VB0_BUFFER_PITCH_SHIFT);
OUT_RELOC(batch->bo, I915_GEM_DOMAIN_VERTEX, 0, 0);
if (IS_GEN5(batch->devid))
OUT_RELOC(batch->bo, I915_GEM_DOMAIN_VERTEX, 0,
batch->bo->size - 1);
else
OUT_BATCH(batch->bo->size / VERTEX_SIZE - 1);
OUT_BATCH(0);
}
static uint32_t gen4_emit_primitive(struct intel_batchbuffer *batch)
{
uint32_t offset;
OUT_BATCH(GEN4_3DPRIMITIVE |
GEN4_3DPRIMITIVE_VERTEX_SEQUENTIAL |
_3DPRIM_RECTLIST << GEN4_3DPRIMITIVE_TOPOLOGY_SHIFT |
0 << 9 |
(6 - 2));
OUT_BATCH(3); /* vertex count */
offset = batch_used(batch);
OUT_BATCH(0); /* vertex_index */
OUT_BATCH(1); /* single instance */
OUT_BATCH(0); /* start instance location */
OUT_BATCH(0); /* index buffer offset, ignored */
return offset;
}
void gen4_render_copyfunc(struct intel_batchbuffer *batch,
drm_intel_context *context,
const struct igt_buf *src,
unsigned src_x, unsigned src_y,
unsigned width, unsigned height,
const struct igt_buf *dst,
unsigned dst_x, unsigned dst_y)
{
uint32_t cc, cc_vp;
uint32_t wm, wm_sampler, wm_kernel, wm_table;
uint32_t sf, sf_kernel;
uint32_t vs;
uint32_t offset, batch_end;
igt_assert(src->bpp == dst->bpp);
intel_batchbuffer_flush_with_context(batch, context);
batch->ptr = batch->buffer + 1024;
intel_batchbuffer_subdata_alloc(batch, 64, 64);
vs = gen4_create_vs_state(batch);
sf_kernel = gen4_create_sf_kernel(batch);
sf = gen4_create_sf_state(batch, sf_kernel);
wm_table = gen4_bind_surfaces(batch, src, dst);
wm_kernel = gen4_create_ps_kernel(batch);
wm_sampler = gen4_create_sampler(batch,
SAMPLER_FILTER_NEAREST,
SAMPLER_EXTEND_NONE);
wm = gen4_create_wm_state(batch, wm_kernel, wm_sampler);
cc_vp = gen4_create_cc_viewport(batch);
cc = gen4_create_cc_state(batch, cc_vp);
batch->ptr = batch->buffer;
gen4_emit_invariant(batch);
gen4_emit_state_base_address(batch);
gen4_emit_sip(batch);
gen4_emit_null_depth_buffer(batch);
gen4_emit_drawing_rectangle(batch, dst);
gen4_emit_binding_table(batch, wm_table);
gen4_emit_vertex_elements(batch);
gen4_emit_pipelined_pointers(batch, vs, sf, wm, cc);
gen4_emit_urb(batch);
gen4_emit_vertex_buffer(batch);
offset = gen4_emit_primitive(batch);
OUT_BATCH(MI_BATCH_BUFFER_END);
batch_end = intel_batchbuffer_align(batch, 8);
*(uint32_t *)(batch->buffer + offset) =
batch_round_upto(batch, VERTEX_SIZE)/VERTEX_SIZE;
emit_vertex_2s(batch, dst_x + width, dst_y + height);
emit_vertex_normalized(batch, src_x + width, igt_buf_width(src));
emit_vertex_normalized(batch, src_y + height, igt_buf_height(src));
emit_vertex_2s(batch, dst_x, dst_y + height);
emit_vertex_normalized(batch, src_x, igt_buf_width(src));
emit_vertex_normalized(batch, src_y + height, igt_buf_height(src));
emit_vertex_2s(batch, dst_x, dst_y);
emit_vertex_normalized(batch, src_x, igt_buf_width(src));
emit_vertex_normalized(batch, src_y, igt_buf_height(src));
gen4_render_flush(batch, context, batch_end);
intel_batchbuffer_reset(batch);
}