summaryrefslogtreecommitdiff
path: root/drivers/gpu/drm/i915/i915_gem_render_state.c
blob: 7032c542a9b1d007c4bc69e328db040b88efc15d (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
/*
 * Copyright © 2014 Intel Corporation
 *
 * Permission is hereby granted, free of charge, to any person obtaining a
 * copy of this software and associated documentation files (the "Software"),
 * to deal in the Software without restriction, including without limitation
 * the rights to use, copy, modify, merge, publish, distribute, sublicense,
 * and/or sell copies of the Software, and to permit persons to whom the
 * Software is furnished to do so, subject to the following conditions:
 *
 * The above copyright notice and this permission notice (including the next
 * paragraph) shall be included in all copies or substantial portions of the
 * Software.
 *
 * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
 * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
 * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
 * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
 * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
 * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
 * IN THE SOFTWARE.
 *
 * Authors:
 *    Mika Kuoppala <mika.kuoppala@intel.com>
 *
 */

#include "i915_drv.h"
#include "intel_renderstate.h"

struct intel_render_state {
	const struct intel_renderstate_rodata *rodata;
	struct i915_vma *vma;
	u32 batch_offset;
	u32 batch_size;
	u32 aux_offset;
	u32 aux_size;
};

static const struct intel_renderstate_rodata *
render_state_get_rodata(const struct intel_engine_cs *engine)
{
	switch (INTEL_GEN(engine->i915)) {
	case 6:
		return &gen6_null_state;
	case 7:
		return &gen7_null_state;
	case 8:
		return &gen8_null_state;
	case 9:
		return &gen9_null_state;
	}

	return NULL;
}

/*
 * Macro to add commands to auxiliary batch.
 * This macro only checks for page overflow before inserting the commands,
 * this is sufficient as the null state generator makes the final batch
 * with two passes to build command and state separately. At this point
 * the size of both are known and it compacts them by relocating the state
 * right after the commands taking care of alignment so we should sufficient
 * space below them for adding new commands.
 */
#define OUT_BATCH(batch, i, val)				\
	do {							\
		if ((i) >= PAGE_SIZE / sizeof(u32))		\
			goto err;				\
		(batch)[(i)++] = (val);				\
	} while(0)

static int render_state_setup(struct intel_render_state *so,
			      struct drm_i915_private *i915)
{
	const struct intel_renderstate_rodata *rodata = so->rodata;
	struct drm_i915_gem_object *obj = so->vma->obj;
	unsigned int i = 0, reloc_index = 0;
	unsigned int needs_clflush;
	u32 *d;
	int ret;

	ret = i915_gem_obj_prepare_shmem_write(obj, &needs_clflush);
	if (ret)
		return ret;

	d = kmap_atomic(i915_gem_object_get_dirty_page(obj, 0));

	while (i < rodata->batch_items) {
		u32 s = rodata->batch[i];

		if (i * 4  == rodata->reloc[reloc_index]) {
			u64 r = s + so->vma->node.start;
			s = lower_32_bits(r);
			if (HAS_64BIT_RELOC(i915)) {
				if (i + 1 >= rodata->batch_items ||
				    rodata->batch[i + 1] != 0)
					goto err;

				d[i++] = s;
				s = upper_32_bits(r);
			}

			reloc_index++;
		}

		d[i++] = s;
	}

	if (rodata->reloc[reloc_index] != -1) {
		DRM_ERROR("only %d relocs resolved\n", reloc_index);
		goto err;
	}

	so->batch_offset = so->vma->node.start;
	so->batch_size = rodata->batch_items * sizeof(u32);

	while (i % CACHELINE_DWORDS)
		OUT_BATCH(d, i, MI_NOOP);

	so->aux_offset = i * sizeof(u32);

	if (HAS_POOLED_EU(i915)) {
		/*
		 * We always program 3x6 pool config but depending upon which
		 * subslice is disabled HW drops down to appropriate config
		 * shown below.
		 *
		 * In the below table 2x6 config always refers to
		 * fused-down version, native 2x6 is not available and can
		 * be ignored
		 *
		 * SNo  subslices config                eu pool configuration
		 * -----------------------------------------------------------
		 * 1    3 subslices enabled (3x6)  -    0x00777000  (9+9)
		 * 2    ss0 disabled (2x6)         -    0x00777000  (3+9)
		 * 3    ss1 disabled (2x6)         -    0x00770000  (6+6)
		 * 4    ss2 disabled (2x6)         -    0x00007000  (9+3)
		 */
		u32 eu_pool_config = 0x00777000;

		OUT_BATCH(d, i, GEN9_MEDIA_POOL_STATE);
		OUT_BATCH(d, i, GEN9_MEDIA_POOL_ENABLE);
		OUT_BATCH(d, i, eu_pool_config);
		OUT_BATCH(d, i, 0);
		OUT_BATCH(d, i, 0);
		OUT_BATCH(d, i, 0);
	}

	OUT_BATCH(d, i, MI_BATCH_BUFFER_END);
	so->aux_size = i * sizeof(u32) - so->aux_offset;
	so->aux_offset += so->batch_offset;
	/*
	 * Since we are sending length, we need to strictly conform to
	 * all requirements. For Gen2 this must be a multiple of 8.
	 */
	so->aux_size = ALIGN(so->aux_size, 8);

	if (needs_clflush)
		drm_clflush_virt_range(d, i * sizeof(u32));
	kunmap_atomic(d);

	ret = i915_gem_object_set_to_gtt_domain(obj, false);
out:
	i915_gem_obj_finish_shmem_access(obj);
	return ret;

err:
	kunmap_atomic(d);
	ret = -EINVAL;
	goto out;
}

#undef OUT_BATCH

int i915_gem_render_state_init(struct intel_engine_cs *engine)
{
	struct intel_render_state *so;
	const struct intel_renderstate_rodata *rodata;
	struct drm_i915_gem_object *obj;
	int ret;

	if (engine->id != RCS)
		return 0;

	rodata = render_state_get_rodata(engine);
	if (!rodata)
		return 0;

	if (rodata->batch_items * 4 > PAGE_SIZE)
		return -EINVAL;

	so = kmalloc(sizeof(*so), GFP_KERNEL);
	if (!so)
		return -ENOMEM;

	obj = i915_gem_object_create_internal(engine->i915, PAGE_SIZE);
	if (IS_ERR(obj)) {
		ret = PTR_ERR(obj);
		goto err_free;
	}

	so->vma = i915_vma_instance(obj, &engine->i915->ggtt.base, NULL);
	if (IS_ERR(so->vma)) {
		ret = PTR_ERR(so->vma);
		goto err_obj;
	}

	so->rodata = rodata;
	engine->render_state = so;
	return 0;

err_obj:
	i915_gem_object_put(obj);
err_free:
	kfree(so);
	return ret;
}

int i915_gem_render_state_emit(struct drm_i915_gem_request *req)
{
	struct intel_render_state *so;
	int ret;

	lockdep_assert_held(&req->i915->drm.struct_mutex);

	so = req->engine->render_state;
	if (!so)
		return 0;

	/* Recreate the page after shrinking */
	if (!so->vma->obj->mm.pages)
		so->batch_offset = -1;

	ret = i915_vma_pin(so->vma, 0, 0, PIN_GLOBAL | PIN_HIGH);
	if (ret)
		return ret;

	if (so->vma->node.start != so->batch_offset) {
		ret = render_state_setup(so, req->i915);
		if (ret)
			goto err_unpin;
	}

	ret = req->engine->emit_bb_start(req,
					 so->batch_offset, so->batch_size,
					 I915_DISPATCH_SECURE);
	if (ret)
		goto err_unpin;

	if (so->aux_size > 8) {
		ret = req->engine->emit_bb_start(req,
						 so->aux_offset, so->aux_size,
						 I915_DISPATCH_SECURE);
		if (ret)
			goto err_unpin;
	}

	i915_vma_move_to_active(so->vma, req, 0);
err_unpin:
	i915_vma_unpin(so->vma);
	return ret;
}

void i915_gem_render_state_fini(struct intel_engine_cs *engine)
{
	struct intel_render_state *so;
	struct drm_i915_gem_object *obj;

	so = fetch_and_zero(&engine->render_state);
	if (!so)
		return;

	obj = so->vma->obj;

	i915_vma_close(so->vma);
	__i915_gem_object_release_unless_active(obj);

	kfree(so);
}