panfrost_regs.h (15936B)
1/* SPDX-License-Identifier: GPL-2.0 */ 2/* Copyright 2018 Marty E. Plummer <hanetzer@startmail.com> */ 3/* Copyright 2019 Linaro, Ltd, Rob Herring <robh@kernel.org> */ 4/* 5 * Register definitions based on mali_midg_regmap.h 6 * (C) COPYRIGHT 2010-2018 ARM Limited. All rights reserved. 7 */ 8#ifndef __PANFROST_REGS_H__ 9#define __PANFROST_REGS_H__ 10 11#define GPU_ID 0x00 12#define GPU_L2_FEATURES 0x004 /* (RO) Level 2 cache features */ 13#define GPU_CORE_FEATURES 0x008 /* (RO) Shader Core Features */ 14#define GPU_TILER_FEATURES 0x00C /* (RO) Tiler Features */ 15#define GPU_MEM_FEATURES 0x010 /* (RO) Memory system features */ 16#define GROUPS_L2_COHERENT BIT(0) /* Cores groups are l2 coherent */ 17 18#define GPU_MMU_FEATURES 0x014 /* (RO) MMU features */ 19#define GPU_AS_PRESENT 0x018 /* (RO) Address space slots present */ 20#define GPU_JS_PRESENT 0x01C /* (RO) Job slots present */ 21 22#define GPU_INT_RAWSTAT 0x20 23#define GPU_INT_CLEAR 0x24 24#define GPU_INT_MASK 0x28 25#define GPU_INT_STAT 0x2c 26#define GPU_IRQ_FAULT BIT(0) 27#define GPU_IRQ_MULTIPLE_FAULT BIT(7) 28#define GPU_IRQ_RESET_COMPLETED BIT(8) 29#define GPU_IRQ_POWER_CHANGED BIT(9) 30#define GPU_IRQ_POWER_CHANGED_ALL BIT(10) 31#define GPU_IRQ_PERFCNT_SAMPLE_COMPLETED BIT(16) 32#define GPU_IRQ_CLEAN_CACHES_COMPLETED BIT(17) 33#define GPU_IRQ_MASK_ALL \ 34 (GPU_IRQ_FAULT |\ 35 GPU_IRQ_MULTIPLE_FAULT |\ 36 GPU_IRQ_RESET_COMPLETED |\ 37 GPU_IRQ_POWER_CHANGED |\ 38 GPU_IRQ_POWER_CHANGED_ALL |\ 39 GPU_IRQ_PERFCNT_SAMPLE_COMPLETED |\ 40 GPU_IRQ_CLEAN_CACHES_COMPLETED) 41#define GPU_IRQ_MASK_ERROR \ 42 ( \ 43 GPU_IRQ_FAULT |\ 44 GPU_IRQ_MULTIPLE_FAULT) 45#define GPU_CMD 0x30 46#define GPU_CMD_SOFT_RESET 0x01 47#define GPU_CMD_PERFCNT_CLEAR 0x03 48#define GPU_CMD_PERFCNT_SAMPLE 0x04 49#define GPU_CMD_CLEAN_CACHES 0x07 50#define GPU_CMD_CLEAN_INV_CACHES 0x08 51#define GPU_STATUS 0x34 52#define GPU_STATUS_PRFCNT_ACTIVE BIT(2) 53#define GPU_LATEST_FLUSH_ID 0x38 54#define GPU_PWR_KEY 0x50 /* (WO) Power manager key register */ 55#define GPU_PWR_KEY_UNLOCK 0x2968A819 56#define GPU_PWR_OVERRIDE0 0x54 /* (RW) Power manager override settings */ 57#define GPU_PWR_OVERRIDE1 0x58 /* (RW) Power manager override settings */ 58#define GPU_FAULT_STATUS 0x3C 59#define GPU_FAULT_ADDRESS_LO 0x40 60#define GPU_FAULT_ADDRESS_HI 0x44 61 62#define GPU_PERFCNT_BASE_LO 0x60 63#define GPU_PERFCNT_BASE_HI 0x64 64#define GPU_PERFCNT_CFG 0x68 65#define GPU_PERFCNT_CFG_MODE(x) (x) 66#define GPU_PERFCNT_CFG_MODE_OFF 0 67#define GPU_PERFCNT_CFG_MODE_MANUAL 1 68#define GPU_PERFCNT_CFG_MODE_TILE 2 69#define GPU_PERFCNT_CFG_AS(x) ((x) << 4) 70#define GPU_PERFCNT_CFG_SETSEL(x) ((x) << 8) 71#define GPU_PRFCNT_JM_EN 0x6c 72#define GPU_PRFCNT_SHADER_EN 0x70 73#define GPU_PRFCNT_TILER_EN 0x74 74#define GPU_PRFCNT_MMU_L2_EN 0x7c 75 76#define GPU_THREAD_MAX_THREADS 0x0A0 /* (RO) Maximum number of threads per core */ 77#define GPU_THREAD_MAX_WORKGROUP_SIZE 0x0A4 /* (RO) Maximum workgroup size */ 78#define GPU_THREAD_MAX_BARRIER_SIZE 0x0A8 /* (RO) Maximum threads waiting at a barrier */ 79#define GPU_THREAD_FEATURES 0x0AC /* (RO) Thread features */ 80#define GPU_THREAD_TLS_ALLOC 0x310 /* (RO) Number of threads per core that 81 * TLS must be allocated for */ 82 83#define GPU_TEXTURE_FEATURES(n) (0x0B0 + ((n) * 4)) 84#define GPU_JS_FEATURES(n) (0x0C0 + ((n) * 4)) 85#define GPU_AFBC_FEATURES (0x4C) /* (RO) AFBC support on Bifrost */ 86 87#define GPU_SHADER_PRESENT_LO 0x100 /* (RO) Shader core present bitmap, low word */ 88#define GPU_SHADER_PRESENT_HI 0x104 /* (RO) Shader core present bitmap, high word */ 89#define GPU_TILER_PRESENT_LO 0x110 /* (RO) Tiler core present bitmap, low word */ 90#define GPU_TILER_PRESENT_HI 0x114 /* (RO) Tiler core present bitmap, high word */ 91 92#define GPU_L2_PRESENT_LO 0x120 /* (RO) Level 2 cache present bitmap, low word */ 93#define GPU_L2_PRESENT_HI 0x124 /* (RO) Level 2 cache present bitmap, high word */ 94 95#define GPU_COHERENCY_FEATURES 0x300 /* (RO) Coherency features present */ 96#define COHERENCY_ACE_LITE BIT(0) 97#define COHERENCY_ACE BIT(1) 98 99#define GPU_STACK_PRESENT_LO 0xE00 /* (RO) Core stack present bitmap, low word */ 100#define GPU_STACK_PRESENT_HI 0xE04 /* (RO) Core stack present bitmap, high word */ 101 102#define SHADER_READY_LO 0x140 /* (RO) Shader core ready bitmap, low word */ 103#define SHADER_READY_HI 0x144 /* (RO) Shader core ready bitmap, high word */ 104 105#define TILER_READY_LO 0x150 /* (RO) Tiler core ready bitmap, low word */ 106#define TILER_READY_HI 0x154 /* (RO) Tiler core ready bitmap, high word */ 107 108#define L2_READY_LO 0x160 /* (RO) Level 2 cache ready bitmap, low word */ 109#define L2_READY_HI 0x164 /* (RO) Level 2 cache ready bitmap, high word */ 110 111#define STACK_READY_LO 0xE10 /* (RO) Core stack ready bitmap, low word */ 112#define STACK_READY_HI 0xE14 /* (RO) Core stack ready bitmap, high word */ 113 114 115#define SHADER_PWRON_LO 0x180 /* (WO) Shader core power on bitmap, low word */ 116#define SHADER_PWRON_HI 0x184 /* (WO) Shader core power on bitmap, high word */ 117 118#define TILER_PWRON_LO 0x190 /* (WO) Tiler core power on bitmap, low word */ 119#define TILER_PWRON_HI 0x194 /* (WO) Tiler core power on bitmap, high word */ 120 121#define L2_PWRON_LO 0x1A0 /* (WO) Level 2 cache power on bitmap, low word */ 122#define L2_PWRON_HI 0x1A4 /* (WO) Level 2 cache power on bitmap, high word */ 123 124#define STACK_PWRON_LO 0xE20 /* (RO) Core stack power on bitmap, low word */ 125#define STACK_PWRON_HI 0xE24 /* (RO) Core stack power on bitmap, high word */ 126 127 128#define SHADER_PWROFF_LO 0x1C0 /* (WO) Shader core power off bitmap, low word */ 129#define SHADER_PWROFF_HI 0x1C4 /* (WO) Shader core power off bitmap, high word */ 130 131#define TILER_PWROFF_LO 0x1D0 /* (WO) Tiler core power off bitmap, low word */ 132#define TILER_PWROFF_HI 0x1D4 /* (WO) Tiler core power off bitmap, high word */ 133 134#define L2_PWROFF_LO 0x1E0 /* (WO) Level 2 cache power off bitmap, low word */ 135#define L2_PWROFF_HI 0x1E4 /* (WO) Level 2 cache power off bitmap, high word */ 136 137#define STACK_PWROFF_LO 0xE30 /* (RO) Core stack power off bitmap, low word */ 138#define STACK_PWROFF_HI 0xE34 /* (RO) Core stack power off bitmap, high word */ 139 140 141#define SHADER_PWRTRANS_LO 0x200 /* (RO) Shader core power transition bitmap, low word */ 142#define SHADER_PWRTRANS_HI 0x204 /* (RO) Shader core power transition bitmap, high word */ 143 144#define TILER_PWRTRANS_LO 0x210 /* (RO) Tiler core power transition bitmap, low word */ 145#define TILER_PWRTRANS_HI 0x214 /* (RO) Tiler core power transition bitmap, high word */ 146 147#define L2_PWRTRANS_LO 0x220 /* (RO) Level 2 cache power transition bitmap, low word */ 148#define L2_PWRTRANS_HI 0x224 /* (RO) Level 2 cache power transition bitmap, high word */ 149 150#define STACK_PWRTRANS_LO 0xE40 /* (RO) Core stack power transition bitmap, low word */ 151#define STACK_PWRTRANS_HI 0xE44 /* (RO) Core stack power transition bitmap, high word */ 152 153 154#define SHADER_PWRACTIVE_LO 0x240 /* (RO) Shader core active bitmap, low word */ 155#define SHADER_PWRACTIVE_HI 0x244 /* (RO) Shader core active bitmap, high word */ 156 157#define TILER_PWRACTIVE_LO 0x250 /* (RO) Tiler core active bitmap, low word */ 158#define TILER_PWRACTIVE_HI 0x254 /* (RO) Tiler core active bitmap, high word */ 159 160#define L2_PWRACTIVE_LO 0x260 /* (RO) Level 2 cache active bitmap, low word */ 161#define L2_PWRACTIVE_HI 0x264 /* (RO) Level 2 cache active bitmap, high word */ 162 163#define GPU_JM_CONFIG 0xF00 /* (RW) Job Manager configuration register (Implementation specific register) */ 164#define GPU_SHADER_CONFIG 0xF04 /* (RW) Shader core configuration settings (Implementation specific register) */ 165#define GPU_TILER_CONFIG 0xF08 /* (RW) Tiler core configuration settings (Implementation specific register) */ 166#define GPU_L2_MMU_CONFIG 0xF0C /* (RW) Configuration of the L2 cache and MMU (Implementation specific register) */ 167 168/* L2_MMU_CONFIG register */ 169#define L2_MMU_CONFIG_ALLOW_SNOOP_DISPARITY_SHIFT 23 170#define L2_MMU_CONFIG_ALLOW_SNOOP_DISPARITY (0x1 << L2_MMU_CONFIG_ALLOW_SNOOP_DISPARITY_SHIFT) 171#define L2_MMU_CONFIG_LIMIT_EXTERNAL_READS_SHIFT 24 172#define L2_MMU_CONFIG_LIMIT_EXTERNAL_READS (0x3 << L2_MMU_CONFIG_LIMIT_EXTERNAL_READS_SHIFT) 173#define L2_MMU_CONFIG_LIMIT_EXTERNAL_READS_OCTANT (0x1 << L2_MMU_CONFIG_LIMIT_EXTERNAL_READS_SHIFT) 174#define L2_MMU_CONFIG_LIMIT_EXTERNAL_READS_QUARTER (0x2 << L2_MMU_CONFIG_LIMIT_EXTERNAL_READS_SHIFT) 175#define L2_MMU_CONFIG_LIMIT_EXTERNAL_READS_HALF (0x3 << L2_MMU_CONFIG_LIMIT_EXTERNAL_READS_SHIFT) 176 177#define L2_MMU_CONFIG_LIMIT_EXTERNAL_WRITES_SHIFT 26 178#define L2_MMU_CONFIG_LIMIT_EXTERNAL_WRITES (0x3 << L2_MMU_CONFIG_LIMIT_EXTERNAL_WRITES_SHIFT) 179#define L2_MMU_CONFIG_LIMIT_EXTERNAL_WRITES_OCTANT (0x1 << L2_MMU_CONFIG_LIMIT_EXTERNAL_WRITES_SHIFT) 180#define L2_MMU_CONFIG_LIMIT_EXTERNAL_WRITES_QUARTER (0x2 << L2_MMU_CONFIG_LIMIT_EXTERNAL_WRITES_SHIFT) 181#define L2_MMU_CONFIG_LIMIT_EXTERNAL_WRITES_HALF (0x3 << L2_MMU_CONFIG_LIMIT_EXTERNAL_WRITES_SHIFT) 182 183#define L2_MMU_CONFIG_3BIT_LIMIT_EXTERNAL_READS_SHIFT 12 184#define L2_MMU_CONFIG_3BIT_LIMIT_EXTERNAL_READS (0x7 << L2_MMU_CONFIG_LIMIT_EXTERNAL_READS_SHIFT) 185 186#define L2_MMU_CONFIG_3BIT_LIMIT_EXTERNAL_WRITES_SHIFT 15 187#define L2_MMU_CONFIG_3BIT_LIMIT_EXTERNAL_WRITES (0x7 << L2_MMU_CONFIG_LIMIT_EXTERNAL_WRITES_SHIFT) 188 189/* SHADER_CONFIG register */ 190#define SC_ALT_COUNTERS BIT(3) 191#define SC_OVERRIDE_FWD_PIXEL_KILL BIT(4) 192#define SC_SDC_DISABLE_OQ_DISCARD BIT(6) 193#define SC_LS_ALLOW_ATTR_TYPES BIT(16) 194#define SC_LS_PAUSEBUFFER_DISABLE BIT(16) 195#define SC_TLS_HASH_ENABLE BIT(17) 196#define SC_LS_ATTR_CHECK_DISABLE BIT(18) 197#define SC_ENABLE_TEXGRD_FLAGS BIT(25) 198/* End SHADER_CONFIG register */ 199 200/* TILER_CONFIG register */ 201#define TC_CLOCK_GATE_OVERRIDE BIT(0) 202 203/* JM_CONFIG register */ 204#define JM_TIMESTAMP_OVERRIDE BIT(0) 205#define JM_CLOCK_GATE_OVERRIDE BIT(1) 206#define JM_JOB_THROTTLE_ENABLE BIT(2) 207#define JM_JOB_THROTTLE_LIMIT_SHIFT 3 208#define JM_MAX_JOB_THROTTLE_LIMIT 0x3F 209#define JM_FORCE_COHERENCY_FEATURES_SHIFT 2 210#define JM_IDVS_GROUP_SIZE_SHIFT 16 211#define JM_DEFAULT_IDVS_GROUP_SIZE 0xF 212#define JM_MAX_IDVS_GROUP_SIZE 0x3F 213 214 215/* Job Control regs */ 216#define JOB_INT_RAWSTAT 0x1000 217#define JOB_INT_CLEAR 0x1004 218#define JOB_INT_MASK 0x1008 219#define JOB_INT_STAT 0x100c 220#define JOB_INT_JS_STATE 0x1010 221#define JOB_INT_THROTTLE 0x1014 222 223#define MK_JS_MASK(j) (0x10001 << (j)) 224#define JOB_INT_MASK_ERR(j) BIT((j) + 16) 225#define JOB_INT_MASK_DONE(j) BIT(j) 226 227#define JS_BASE 0x1800 228#define JS_HEAD_LO(n) (JS_BASE + ((n) * 0x80) + 0x00) 229#define JS_HEAD_HI(n) (JS_BASE + ((n) * 0x80) + 0x04) 230#define JS_TAIL_LO(n) (JS_BASE + ((n) * 0x80) + 0x08) 231#define JS_TAIL_HI(n) (JS_BASE + ((n) * 0x80) + 0x0c) 232#define JS_AFFINITY_LO(n) (JS_BASE + ((n) * 0x80) + 0x10) 233#define JS_AFFINITY_HI(n) (JS_BASE + ((n) * 0x80) + 0x14) 234#define JS_CONFIG(n) (JS_BASE + ((n) * 0x80) + 0x18) 235#define JS_XAFFINITY(n) (JS_BASE + ((n) * 0x80) + 0x1c) 236#define JS_COMMAND(n) (JS_BASE + ((n) * 0x80) + 0x20) 237#define JS_STATUS(n) (JS_BASE + ((n) * 0x80) + 0x24) 238#define JS_HEAD_NEXT_LO(n) (JS_BASE + ((n) * 0x80) + 0x40) 239#define JS_HEAD_NEXT_HI(n) (JS_BASE + ((n) * 0x80) + 0x44) 240#define JS_AFFINITY_NEXT_LO(n) (JS_BASE + ((n) * 0x80) + 0x50) 241#define JS_AFFINITY_NEXT_HI(n) (JS_BASE + ((n) * 0x80) + 0x54) 242#define JS_CONFIG_NEXT(n) (JS_BASE + ((n) * 0x80) + 0x58) 243#define JS_COMMAND_NEXT(n) (JS_BASE + ((n) * 0x80) + 0x60) 244#define JS_FLUSH_ID_NEXT(n) (JS_BASE + ((n) * 0x80) + 0x70) 245 246/* Possible values of JS_CONFIG and JS_CONFIG_NEXT registers */ 247#define JS_CONFIG_START_FLUSH_CLEAN BIT(8) 248#define JS_CONFIG_START_FLUSH_CLEAN_INVALIDATE (3u << 8) 249#define JS_CONFIG_START_MMU BIT(10) 250#define JS_CONFIG_JOB_CHAIN_FLAG BIT(11) 251#define JS_CONFIG_END_FLUSH_CLEAN BIT(12) 252#define JS_CONFIG_END_FLUSH_CLEAN_INVALIDATE (3u << 12) 253#define JS_CONFIG_ENABLE_FLUSH_REDUCTION BIT(14) 254#define JS_CONFIG_DISABLE_DESCRIPTOR_WR_BK BIT(15) 255#define JS_CONFIG_THREAD_PRI(n) ((n) << 16) 256 257#define JS_COMMAND_NOP 0x00 258#define JS_COMMAND_START 0x01 259#define JS_COMMAND_SOFT_STOP 0x02 /* Gently stop processing a job chain */ 260#define JS_COMMAND_HARD_STOP 0x03 /* Rudely stop processing a job chain */ 261#define JS_COMMAND_SOFT_STOP_0 0x04 /* Execute SOFT_STOP if JOB_CHAIN_FLAG is 0 */ 262#define JS_COMMAND_HARD_STOP_0 0x05 /* Execute HARD_STOP if JOB_CHAIN_FLAG is 0 */ 263#define JS_COMMAND_SOFT_STOP_1 0x06 /* Execute SOFT_STOP if JOB_CHAIN_FLAG is 1 */ 264#define JS_COMMAND_HARD_STOP_1 0x07 /* Execute HARD_STOP if JOB_CHAIN_FLAG is 1 */ 265 266/* MMU regs */ 267#define MMU_INT_RAWSTAT 0x2000 268#define MMU_INT_CLEAR 0x2004 269#define MMU_INT_MASK 0x2008 270#define MMU_INT_STAT 0x200c 271 272/* AS_COMMAND register commands */ 273#define AS_COMMAND_NOP 0x00 /* NOP Operation */ 274#define AS_COMMAND_UPDATE 0x01 /* Broadcasts the values in AS_TRANSTAB and ASn_MEMATTR to all MMUs */ 275#define AS_COMMAND_LOCK 0x02 /* Issue a lock region command to all MMUs */ 276#define AS_COMMAND_UNLOCK 0x03 /* Issue a flush region command to all MMUs */ 277#define AS_COMMAND_FLUSH 0x04 /* Flush all L2 caches then issue a flush region command to all MMUs 278 (deprecated - only for use with T60x) */ 279#define AS_COMMAND_FLUSH_PT 0x04 /* Flush all L2 caches then issue a flush region command to all MMUs */ 280#define AS_COMMAND_FLUSH_MEM 0x05 /* Wait for memory accesses to complete, flush all the L1s cache then 281 flush all L2 caches then issue a flush region command to all MMUs */ 282 283#define MMU_AS(as) (0x2400 + ((as) << 6)) 284 285#define AS_TRANSTAB_LO(as) (MMU_AS(as) + 0x00) /* (RW) Translation Table Base Address for address space n, low word */ 286#define AS_TRANSTAB_HI(as) (MMU_AS(as) + 0x04) /* (RW) Translation Table Base Address for address space n, high word */ 287#define AS_MEMATTR_LO(as) (MMU_AS(as) + 0x08) /* (RW) Memory attributes for address space n, low word. */ 288#define AS_MEMATTR_HI(as) (MMU_AS(as) + 0x0C) /* (RW) Memory attributes for address space n, high word. */ 289#define AS_LOCKADDR_LO(as) (MMU_AS(as) + 0x10) /* (RW) Lock region address for address space n, low word */ 290#define AS_LOCKADDR_HI(as) (MMU_AS(as) + 0x14) /* (RW) Lock region address for address space n, high word */ 291#define AS_COMMAND(as) (MMU_AS(as) + 0x18) /* (WO) MMU command register for address space n */ 292#define AS_FAULTSTATUS(as) (MMU_AS(as) + 0x1C) /* (RO) MMU fault status register for address space n */ 293#define AS_FAULTADDRESS_LO(as) (MMU_AS(as) + 0x20) /* (RO) Fault Address for address space n, low word */ 294#define AS_FAULTADDRESS_HI(as) (MMU_AS(as) + 0x24) /* (RO) Fault Address for address space n, high word */ 295#define AS_STATUS(as) (MMU_AS(as) + 0x28) /* (RO) Status flags for address space n */ 296/* Additional Bifrost AS registers */ 297#define AS_TRANSCFG_LO(as) (MMU_AS(as) + 0x30) /* (RW) Translation table configuration for address space n, low word */ 298#define AS_TRANSCFG_HI(as) (MMU_AS(as) + 0x34) /* (RW) Translation table configuration for address space n, high word */ 299#define AS_FAULTEXTRA_LO(as) (MMU_AS(as) + 0x38) /* (RO) Secondary fault address for address space n, low word */ 300#define AS_FAULTEXTRA_HI(as) (MMU_AS(as) + 0x3C) /* (RO) Secondary fault address for address space n, high word */ 301 302/* 303 * Begin LPAE MMU TRANSTAB register values 304 */ 305#define AS_TRANSTAB_LPAE_ADDR_SPACE_MASK 0xfffffffffffff000 306#define AS_TRANSTAB_LPAE_ADRMODE_IDENTITY 0x2 307#define AS_TRANSTAB_LPAE_ADRMODE_TABLE 0x3 308#define AS_TRANSTAB_LPAE_ADRMODE_MASK 0x3 309#define AS_TRANSTAB_LPAE_READ_INNER BIT(2) 310#define AS_TRANSTAB_LPAE_SHARE_OUTER BIT(4) 311 312#define AS_STATUS_AS_ACTIVE 0x01 313 314#define AS_FAULTSTATUS_ACCESS_TYPE_MASK (0x3 << 8) 315#define AS_FAULTSTATUS_ACCESS_TYPE_ATOMIC (0x0 << 8) 316#define AS_FAULTSTATUS_ACCESS_TYPE_EX (0x1 << 8) 317#define AS_FAULTSTATUS_ACCESS_TYPE_READ (0x2 << 8) 318#define AS_FAULTSTATUS_ACCESS_TYPE_WRITE (0x3 << 8) 319 320#define AS_LOCK_REGION_MIN_SIZE (1ULL << 15) 321 322#define gpu_write(dev, reg, data) writel(data, dev->iomem + reg) 323#define gpu_read(dev, reg) readl(dev->iomem + reg) 324 325#endif