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
|
/*
* drivers/video/tegra/host/gk20a/ltc_common.c
*
* GK20A Graphics
*
* Copyright (c) 2011-2017, NVIDIA CORPORATION. All rights reserved.
*
* This program is free software; you can redistribute it and/or modify it
* under the terms and conditions of the GNU General Public License,
* version 2, as published by the Free Software Foundation.
*
* This program is distributed in the hope it will be useful, but WITHOUT
* ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
* FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for
* more details.
*
* You should have received a copy of the GNU General Public License
* along with this program. If not, see <http://www.gnu.org/licenses/>.
*/
#include <nvgpu/dma.h>
#include <nvgpu/enabled.h>
#include "gk20a.h"
#include "gr_gk20a.h"
/*
* Sets the ZBC color for the passed index.
*/
static void gk20a_ltc_set_zbc_color_entry(struct gk20a *g,
struct zbc_entry *color_val,
u32 index)
{
u32 i;
u32 real_index = index + GK20A_STARTOF_ZBC_TABLE;
gk20a_writel(g, ltc_ltcs_ltss_dstg_zbc_index_r(),
ltc_ltcs_ltss_dstg_zbc_index_address_f(real_index));
for (i = 0;
i < ltc_ltcs_ltss_dstg_zbc_color_clear_value__size_1_v(); i++) {
gk20a_writel(g, ltc_ltcs_ltss_dstg_zbc_color_clear_value_r(i),
color_val->color_l2[i]);
}
gk20a_readl(g, ltc_ltcs_ltss_dstg_zbc_index_r());
}
/*
* Sets the ZBC depth for the passed index.
*/
static void gk20a_ltc_set_zbc_depth_entry(struct gk20a *g,
struct zbc_entry *depth_val,
u32 index)
{
u32 real_index = index + GK20A_STARTOF_ZBC_TABLE;
gk20a_writel(g, ltc_ltcs_ltss_dstg_zbc_index_r(),
ltc_ltcs_ltss_dstg_zbc_index_address_f(real_index));
gk20a_writel(g, ltc_ltcs_ltss_dstg_zbc_depth_clear_value_r(),
depth_val->depth);
gk20a_readl(g, ltc_ltcs_ltss_dstg_zbc_index_r());
}
static int gk20a_ltc_alloc_phys_cbc(struct gk20a *g,
size_t compbit_backing_size)
{
struct gr_gk20a *gr = &g->gr;
return nvgpu_dma_alloc_flags_sys(g, NVGPU_DMA_FORCE_CONTIGUOUS,
compbit_backing_size,
&gr->compbit_store.mem);
}
static int gk20a_ltc_alloc_virt_cbc(struct gk20a *g,
size_t compbit_backing_size)
{
struct gr_gk20a *gr = &g->gr;
return nvgpu_dma_alloc_flags_sys(g, NVGPU_DMA_NO_KERNEL_MAPPING,
compbit_backing_size,
&gr->compbit_store.mem);
}
static void gk20a_ltc_init_cbc(struct gk20a *g, struct gr_gk20a *gr)
{
u32 max_size = gr->max_comptag_mem;
u32 max_comptag_lines = max_size << 3;
u32 compbit_base_post_divide;
u64 compbit_base_post_multiply64;
u64 compbit_store_iova;
u64 compbit_base_post_divide64;
if (nvgpu_is_enabled(g, NVGPU_IS_FMODEL))
compbit_store_iova = gk20a_mem_phys(&gr->compbit_store.mem);
else
compbit_store_iova = g->ops.mm.get_iova_addr(g,
gr->compbit_store.mem.priv.sgt->sgl, 0);
compbit_base_post_divide64 = compbit_store_iova >>
ltc_ltcs_ltss_cbc_base_alignment_shift_v();
do_div(compbit_base_post_divide64, g->ltc_count);
compbit_base_post_divide = u64_lo32(compbit_base_post_divide64);
compbit_base_post_multiply64 = ((u64)compbit_base_post_divide *
g->ltc_count) << ltc_ltcs_ltss_cbc_base_alignment_shift_v();
if (compbit_base_post_multiply64 < compbit_store_iova)
compbit_base_post_divide++;
/* Bug 1477079 indicates sw adjustment on the posted divided base. */
if (g->ops.ltc.cbc_fix_config)
compbit_base_post_divide =
g->ops.ltc.cbc_fix_config(g, compbit_base_post_divide);
gk20a_writel(g, ltc_ltcs_ltss_cbc_base_r(),
compbit_base_post_divide);
gk20a_dbg(gpu_dbg_info | gpu_dbg_map_v | gpu_dbg_pte,
"compbit base.pa: 0x%x,%08x cbc_base:0x%08x\n",
(u32)(compbit_store_iova >> 32),
(u32)(compbit_store_iova & 0xffffffff),
compbit_base_post_divide);
gr->compbit_store.base_hw = compbit_base_post_divide;
g->ops.ltc.cbc_ctrl(g, gk20a_cbc_op_invalidate,
0, max_comptag_lines - 1);
}
#ifdef CONFIG_DEBUG_FS
static void gk20a_ltc_sync_debugfs(struct gk20a *g)
{
u32 reg_f = ltc_ltcs_ltss_tstg_set_mgmt_2_l2_bypass_mode_enabled_f();
nvgpu_spinlock_acquire(&g->debugfs_lock);
if (g->mm.ltc_enabled != g->mm.ltc_enabled_debug) {
u32 reg = gk20a_readl(g, ltc_ltcs_ltss_tstg_set_mgmt_2_r());
if (g->mm.ltc_enabled_debug)
/* bypass disabled (normal caching ops)*/
reg &= ~reg_f;
else
/* bypass enabled (no caching) */
reg |= reg_f;
gk20a_writel(g, ltc_ltcs_ltss_tstg_set_mgmt_2_r(), reg);
g->mm.ltc_enabled = g->mm.ltc_enabled_debug;
}
nvgpu_spinlock_release(&g->debugfs_lock);
}
#endif
|
