diff options
Diffstat (limited to 'drivers/gpu/drm/i915')
-rw-r--r-- | drivers/gpu/drm/i915/i915_cmd_parser.c | 616 | ||||
-rw-r--r-- | drivers/gpu/drm/i915/i915_drv.c | 5 | ||||
-rw-r--r-- | drivers/gpu/drm/i915/i915_drv.h | 147 | ||||
-rw-r--r-- | drivers/gpu/drm/i915/i915_gem.c | 25 | ||||
-rw-r--r-- | drivers/gpu/drm/i915/i915_gem_context.c | 5 | ||||
-rw-r--r-- | drivers/gpu/drm/i915/i915_gem_execbuffer.c | 110 | ||||
-rw-r--r-- | drivers/gpu/drm/i915/i915_gem_gtt.c | 63 | ||||
-rw-r--r-- | drivers/gpu/drm/i915/i915_gem_gtt.h | 3 | ||||
-rw-r--r-- | drivers/gpu/drm/i915/i915_gem_request.c | 4 | ||||
-rw-r--r-- | drivers/gpu/drm/i915/i915_params.c | 6 | ||||
-rw-r--r-- | drivers/gpu/drm/i915/i915_params.h | 2 | ||||
-rw-r--r-- | drivers/gpu/drm/i915/i915_reg.h | 11 | ||||
-rw-r--r-- | drivers/gpu/drm/i915/intel_csr.c | 18 | ||||
-rw-r--r-- | drivers/gpu/drm/i915/intel_drv.h | 3 | ||||
-rw-r--r-- | drivers/gpu/drm/i915/intel_lrc.c | 27 | ||||
-rw-r--r-- | drivers/gpu/drm/i915/intel_pm.c | 160 | ||||
-rw-r--r-- | drivers/gpu/drm/i915/intel_ringbuffer.c | 11 | ||||
-rw-r--r-- | drivers/gpu/drm/i915/intel_ringbuffer.h | 18 |
18 files changed, 840 insertions, 394 deletions
diff --git a/drivers/gpu/drm/i915/i915_cmd_parser.c b/drivers/gpu/drm/i915/i915_cmd_parser.c index 70980f82a15b..1e104518192d 100644 --- a/drivers/gpu/drm/i915/i915_cmd_parser.c +++ b/drivers/gpu/drm/i915/i915_cmd_parser.c @@ -26,6 +26,7 @@ */ #include "i915_drv.h" +#include "intel_ringbuffer.h" /** * DOC: batch buffer command parser @@ -50,13 +51,11 @@ * granting userspace undue privileges. There are three categories of privilege. * * First, commands which are explicitly defined as privileged or which should - * only be used by the kernel driver. The parser generally rejects such - * commands, though it may allow some from the drm master process. + * only be used by the kernel driver. The parser rejects such commands * * Second, commands which access registers. To support correct/enhanced * userspace functionality, particularly certain OpenGL extensions, the parser - * provides a whitelist of registers which userspace may safely access (for both - * normal and drm master processes). + * provides a whitelist of registers which userspace may safely access * * Third, commands which access privileged memory (i.e. GGTT, HWS page, etc). * The parser always rejects such commands. @@ -81,11 +80,104 @@ * in the per-engine command tables. * * Other command table entries map fairly directly to high level categories - * mentioned above: rejected, master-only, register whitelist. The parser - * implements a number of checks, including the privileged memory checks, via a - * general bitmasking mechanism. + * mentioned above: rejected, register whitelist. The parser implements a number + * of checks, including the privileged memory checks, via a general bitmasking + * mechanism. */ +/* + * A command that requires special handling by the command parser. + */ +struct drm_i915_cmd_descriptor { + /* + * Flags describing how the command parser processes the command. + * + * CMD_DESC_FIXED: The command has a fixed length if this is set, + * a length mask if not set + * CMD_DESC_SKIP: The command is allowed but does not follow the + * standard length encoding for the opcode range in + * which it falls + * CMD_DESC_REJECT: The command is never allowed + * CMD_DESC_REGISTER: The command should be checked against the + * register whitelist for the appropriate ring + */ + u32 flags; +#define CMD_DESC_FIXED (1<<0) +#define CMD_DESC_SKIP (1<<1) +#define CMD_DESC_REJECT (1<<2) +#define CMD_DESC_REGISTER (1<<3) +#define CMD_DESC_BITMASK (1<<4) + + /* + * The command's unique identification bits and the bitmask to get them. + * This isn't strictly the opcode field as defined in the spec and may + * also include type, subtype, and/or subop fields. + */ + struct { + u32 value; + u32 mask; + } cmd; + + /* + * The command's length. The command is either fixed length (i.e. does + * not include a length field) or has a length field mask. The flag + * CMD_DESC_FIXED indicates a fixed length. Otherwise, the command has + * a length mask. All command entries in a command table must include + * length information. + */ + union { + u32 fixed; + u32 mask; + } length; + + /* + * Describes where to find a register address in the command to check + * against the ring's register whitelist. Only valid if flags has the + * CMD_DESC_REGISTER bit set. + * + * A non-zero step value implies that the command may access multiple + * registers in sequence (e.g. LRI), in that case step gives the + * distance in dwords between individual offset fields. + */ + struct { + u32 offset; + u32 mask; + u32 step; + } reg; + +#define MAX_CMD_DESC_BITMASKS 3 + /* + * Describes command checks where a particular dword is masked and + * compared against an expected value. If the command does not match + * the expected value, the parser rejects it. Only valid if flags has + * the CMD_DESC_BITMASK bit set. Only entries where mask is non-zero + * are valid. + * + * If the check specifies a non-zero condition_mask then the parser + * only performs the check when the bits specified by condition_mask + * are non-zero. + */ + struct { + u32 offset; + u32 mask; + u32 expected; + u32 condition_offset; + u32 condition_mask; + } bits[MAX_CMD_DESC_BITMASKS]; +}; + +/* + * A table of commands requiring special handling by the command parser. + * + * Each engine has an array of tables. Each table consists of an array of + * command descriptors, which must be sorted with command opcodes in + * ascending order. + */ +struct drm_i915_cmd_table { + const struct drm_i915_cmd_descriptor *table; + int count; +}; + #define STD_MI_OPCODE_SHIFT (32 - 9) #define STD_3D_OPCODE_SHIFT (32 - 16) #define STD_2D_OPCODE_SHIFT (32 - 10) @@ -95,7 +187,7 @@ #define CMD(op, opm, f, lm, fl, ...) \ { \ .flags = (fl) | ((f) ? CMD_DESC_FIXED : 0), \ - .cmd = { (op), ~0u << (opm) }, \ + .cmd = { (op & ~0u << (opm)), ~0u << (opm) }, \ .length = { (lm) }, \ __VA_ARGS__ \ } @@ -110,14 +202,13 @@ #define R CMD_DESC_REJECT #define W CMD_DESC_REGISTER #define B CMD_DESC_BITMASK -#define M CMD_DESC_MASTER /* Command Mask Fixed Len Action ---------------------------------------------------------- */ -static const struct drm_i915_cmd_descriptor common_cmds[] = { +static const struct drm_i915_cmd_descriptor gen7_common_cmds[] = { CMD( MI_NOOP, SMI, F, 1, S ), CMD( MI_USER_INTERRUPT, SMI, F, 1, R ), - CMD( MI_WAIT_FOR_EVENT, SMI, F, 1, M ), + CMD( MI_WAIT_FOR_EVENT, SMI, F, 1, R ), CMD( MI_ARB_CHECK, SMI, F, 1, S ), CMD( MI_REPORT_HEAD, SMI, F, 1, S ), CMD( MI_SUSPEND_FLUSH, SMI, F, 1, S ), @@ -147,7 +238,7 @@ static const struct drm_i915_cmd_descriptor common_cmds[] = { CMD( MI_BATCH_BUFFER_START, SMI, !F, 0xFF, S ), }; -static const struct drm_i915_cmd_descriptor render_cmds[] = { +static const struct drm_i915_cmd_descriptor gen7_render_cmds[] = { CMD( MI_FLUSH, SMI, F, 1, S ), CMD( MI_ARB_ON_OFF, SMI, F, 1, R ), CMD( MI_PREDICATE, SMI, F, 1, S ), @@ -214,7 +305,7 @@ static const struct drm_i915_cmd_descriptor hsw_render_cmds[] = { CMD( MI_URB_ATOMIC_ALLOC, SMI, F, 1, S ), CMD( MI_SET_APPID, SMI, F, 1, S ), CMD( MI_RS_CONTEXT, SMI, F, 1, S ), - CMD( MI_LOAD_SCAN_LINES_INCL, SMI, !F, 0x3F, M ), + CMD( MI_LOAD_SCAN_LINES_INCL, SMI, !F, 0x3F, R ), CMD( MI_LOAD_SCAN_LINES_EXCL, SMI, !F, 0x3F, R ), CMD( MI_LOAD_REGISTER_REG, SMI, !F, 0xFF, W, .reg = { .offset = 1, .mask = 0x007FFFFC, .step = 1 } ), @@ -231,7 +322,7 @@ static const struct drm_i915_cmd_descriptor hsw_render_cmds[] = { CMD( GFX_OP_3DSTATE_BINDING_TABLE_EDIT_PS, S3D, !F, 0x1FF, S ), }; -static const struct drm_i915_cmd_descriptor video_cmds[] = { +static const struct drm_i915_cmd_descriptor gen7_video_cmds[] = { CMD( MI_ARB_ON_OFF, SMI, F, 1, R ), CMD( MI_SET_APPID, SMI, F, 1, S ), CMD( MI_STORE_DWORD_IMM, SMI, !F, 0xFF, B, @@ -275,7 +366,7 @@ static const struct drm_i915_cmd_descriptor video_cmds[] = { CMD( MFX_WAIT, SMFX, F, 1, S ), }; -static const struct drm_i915_cmd_descriptor vecs_cmds[] = { +static const struct drm_i915_cmd_descriptor gen7_vecs_cmds[] = { CMD( MI_ARB_ON_OFF, SMI, F, 1, R ), CMD( MI_SET_APPID, SMI, F, 1, S ), CMD( MI_STORE_DWORD_IMM, SMI, !F, 0xFF, B, @@ -313,7 +404,7 @@ static const struct drm_i915_cmd_descriptor vecs_cmds[] = { }}, ), }; -static const struct drm_i915_cmd_descriptor blt_cmds[] = { +static const struct drm_i915_cmd_descriptor gen7_blt_cmds[] = { CMD( MI_DISPLAY_FLIP, SMI, !F, 0xFF, R ), CMD( MI_STORE_DWORD_IMM, SMI, !F, 0x3FF, B, .bits = {{ @@ -347,10 +438,64 @@ static const struct drm_i915_cmd_descriptor blt_cmds[] = { }; static const struct drm_i915_cmd_descriptor hsw_blt_cmds[] = { - CMD( MI_LOAD_SCAN_LINES_INCL, SMI, !F, 0x3F, M ), + CMD( MI_LOAD_SCAN_LINES_INCL, SMI, !F, 0x3F, R ), CMD( MI_LOAD_SCAN_LINES_EXCL, SMI, !F, 0x3F, R ), }; +/* + * For Gen9 we can still rely on the h/w to enforce cmd security, and only + * need to re-enforce the register access checks. We therefore only need to + * teach the cmdparser how to find the end of each command, and identify + * register accesses. The table doesn't need to reject any commands, and so + * the only commands listed here are: + * 1) Those that touch registers + * 2) Those that do not have the default 8-bit length + * + * Note that the default MI length mask chosen for this table is 0xFF, not + * the 0x3F used on older devices. This is because the vast majority of MI + * cmds on Gen9 use a standard 8-bit Length field. + * All the Gen9 blitter instructions are standard 0xFF length mask, and + * none allow access to non-general registers, so in fact no BLT cmds are + * included in the table at all. + * + */ +static const struct drm_i915_cmd_descriptor gen9_blt_cmds[] = { + CMD( MI_NOOP, SMI, F, 1, S ), + CMD( MI_USER_INTERRUPT, SMI, F, 1, S ), + CMD( MI_WAIT_FOR_EVENT, SMI, F, 1, S ), + CMD( MI_FLUSH, SMI, F, 1, S ), + CMD( MI_ARB_CHECK, SMI, F, 1, S ), + CMD( MI_REPORT_HEAD, SMI, F, 1, S ), + CMD( MI_ARB_ON_OFF, SMI, F, 1, S ), + CMD( MI_SUSPEND_FLUSH, SMI, F, 1, S ), + CMD( MI_LOAD_SCAN_LINES_INCL, SMI, !F, 0x3F, S ), + CMD( MI_LOAD_SCAN_LINES_EXCL, SMI, !F, 0x3F, S ), + CMD( MI_STORE_DWORD_IMM, SMI, !F, 0x3FF, S ), + CMD( MI_LOAD_REGISTER_IMM(1), SMI, !F, 0xFF, W, + .reg = { .offset = 1, .mask = 0x007FFFFC, .step = 2 } ), + CMD( MI_UPDATE_GTT, SMI, !F, 0x3FF, S ), + CMD( MI_STORE_REGISTER_MEM_GEN8, SMI, F, 4, W, + .reg = { .offset = 1, .mask = 0x007FFFFC } ), + CMD( MI_FLUSH_DW, SMI, !F, 0x3F, S ), + CMD( MI_LOAD_REGISTER_MEM_GEN8, SMI, F, 4, W, + .reg = { .offset = 1, .mask = 0x007FFFFC } ), + CMD( MI_LOAD_REGISTER_REG, SMI, !F, 0xFF, W, + .reg = { .offset = 1, .mask = 0x007FFFFC, .step = 1 } ), + + /* + * We allow BB_START but apply further checks. We just sanitize the + * basic fields here. + */ +#define MI_BB_START_OPERAND_MASK GENMASK(SMI-1, 0) +#define MI_BB_START_OPERAND_EXPECT (MI_BATCH_PPGTT_HSW | 1) + CMD( MI_BATCH_BUFFER_START_GEN8, SMI, !F, 0xFF, B, + .bits = {{ + .offset = 0, + .mask = MI_BB_START_OPERAND_MASK, + .expected = MI_BB_START_OPERAND_EXPECT, + }}, ), +}; + static const struct drm_i915_cmd_descriptor noop_desc = CMD(MI_NOOP, SMI, F, 1, S); @@ -364,40 +509,44 @@ static const struct drm_i915_cmd_descriptor noop_desc = #undef R #undef W #undef B -#undef M -static const struct drm_i915_cmd_table gen7_render_cmds[] = { - { common_cmds, ARRAY_SIZE(common_cmds) }, - { render_cmds, ARRAY_SIZE(render_cmds) }, +static const struct drm_i915_cmd_table gen7_render_cmd_table[] = { + { gen7_common_cmds, ARRAY_SIZE(gen7_common_cmds) }, + { gen7_render_cmds, ARRAY_SIZE(gen7_render_cmds) }, }; -static const struct drm_i915_cmd_table hsw_render_ring_cmds[] = { - { common_cmds, ARRAY_SIZE(common_cmds) }, - { render_cmds, ARRAY_SIZE(render_cmds) }, +static const struct drm_i915_cmd_table hsw_render_ring_cmd_table[] = { + { gen7_common_cmds, ARRAY_SIZE(gen7_common_cmds) }, + { gen7_render_cmds, ARRAY_SIZE(gen7_render_cmds) }, { hsw_render_cmds, ARRAY_SIZE(hsw_render_cmds) }, }; -static const struct drm_i915_cmd_table gen7_video_cmds[] = { - { common_cmds, ARRAY_SIZE(common_cmds) }, - { video_cmds, ARRAY_SIZE(video_cmds) }, +static const struct drm_i915_cmd_table gen7_video_cmd_table[] = { + { gen7_common_cmds, ARRAY_SIZE(gen7_common_cmds) }, + { gen7_video_cmds, ARRAY_SIZE(gen7_video_cmds) }, }; -static const struct drm_i915_cmd_table hsw_vebox_cmds[] = { - { common_cmds, ARRAY_SIZE(common_cmds) }, - { vecs_cmds, ARRAY_SIZE(vecs_cmds) }, +static const struct drm_i915_cmd_table hsw_vebox_cmd_table[] = { + { gen7_common_cmds, ARRAY_SIZE(gen7_common_cmds) }, + { gen7_vecs_cmds, ARRAY_SIZE(gen7_vecs_cmds) }, }; -static const struct drm_i915_cmd_table gen7_blt_cmds[] = { - { common_cmds, ARRAY_SIZE(common_cmds) }, - { blt_cmds, ARRAY_SIZE(blt_cmds) }, +static const struct drm_i915_cmd_table gen7_blt_cmd_table[] = { + { gen7_common_cmds, ARRAY_SIZE(gen7_common_cmds) }, + { gen7_blt_cmds, ARRAY_SIZE(gen7_blt_cmds) }, }; -static const struct drm_i915_cmd_table hsw_blt_ring_cmds[] = { - { common_cmds, ARRAY_SIZE(common_cmds) }, - { blt_cmds, ARRAY_SIZE(blt_cmds) }, +static const struct drm_i915_cmd_table hsw_blt_ring_cmd_table[] = { + { gen7_common_cmds, ARRAY_SIZE(gen7_common_cmds) }, + { gen7_blt_cmds, ARRAY_SIZE(gen7_blt_cmds) }, { hsw_blt_cmds, ARRAY_SIZE(hsw_blt_cmds) }, }; +static const struct drm_i915_cmd_table gen9_blt_cmd_table[] = { + { gen9_blt_cmds, ARRAY_SIZE(gen9_blt_cmds) }, +}; + + /* * Register whitelists, sorted by increasing register offset. */ @@ -450,7 +599,6 @@ static const struct drm_i915_reg_descriptor gen7_render_regs[] = { REG64(PS_INVOCATION_COUNT), REG64(PS_DEPTH_COUNT), REG64_IDX(RING_TIMESTAMP, RENDER_RING_BASE), - REG32(OACONTROL), /* Only allowed for LRI and SRM. See below. */ REG64(MI_PREDICATE_SRC0), REG64(MI_PREDICATE_SRC1), REG32(GEN7_3DPRIM_END_OFFSET), @@ -514,17 +662,27 @@ static const struct drm_i915_reg_descriptor gen7_blt_regs[] = { REG64_IDX(RING_TIMESTAMP, BLT_RING_BASE), }; -static const struct drm_i915_reg_descriptor ivb_master_regs[] = { - REG32(FORCEWAKE_MT), - REG32(DERRMR), - REG32(GEN7_PIPE_DE_LOAD_SL(PIPE_A)), - REG32(GEN7_PIPE_DE_LOAD_SL(PIPE_B)), - REG32(GEN7_PIPE_DE_LOAD_SL(PIPE_C)), -}; - -static const struct drm_i915_reg_descriptor hsw_master_regs[] = { - REG32(FORCEWAKE_MT), - REG32(DERRMR), +static const struct drm_i915_reg_descriptor gen9_blt_regs[] = { + REG64_IDX(RING_TIMESTAMP, RENDER_RING_BASE), + REG64_IDX(RING_TIMESTAMP, BSD_RING_BASE), + REG32(BCS_SWCTRL), + REG64_IDX(RING_TIMESTAMP, BLT_RING_BASE), + REG64_IDX(BCS_GPR, 0), + REG64_IDX(BCS_GPR, 1), + REG64_IDX(BCS_GPR, 2), + REG64_IDX(BCS_GPR, 3), + REG64_IDX(BCS_GPR, 4), + REG64_IDX(BCS_GPR, 5), + REG64_IDX(BCS_GPR, 6), + REG64_IDX(BCS_GPR, 7), + REG64_IDX(BCS_GPR, 8), + REG64_IDX(BCS_GPR, 9), + REG64_IDX(BCS_GPR, 10), + REG64_IDX(BCS_GPR, 11), + REG64_IDX(BCS_GPR, 12), + REG64_IDX(BCS_GPR, 13), + REG64_IDX(BCS_GPR, 14), + REG64_IDX(BCS_GPR, 15), }; #undef REG64 @@ -533,33 +691,32 @@ static const struct drm_i915_reg_descriptor hsw_master_regs[] = { struct drm_i915_reg_table { const struct drm_i915_reg_descriptor *regs; int num_regs; - bool master; }; static const struct drm_i915_reg_table ivb_render_reg_tables[] = { - { gen7_render_regs, ARRAY_SIZE(gen7_render_regs), false }, - { ivb_master_regs, ARRAY_SIZE(ivb_master_regs), true }, + { gen7_render_regs, ARRAY_SIZE(gen7_render_regs) }, }; static const struct drm_i915_reg_table ivb_blt_reg_tables[] = { - { gen7_blt_regs, ARRAY_SIZE(gen7_blt_regs), false }, - { ivb_master_regs, ARRAY_SIZE(ivb_master_regs), true }, + { gen7_blt_regs, ARRAY_SIZE(gen7_blt_regs) }, }; static const struct drm_i915_reg_table hsw_render_reg_tables[] = { - { gen7_render_regs, ARRAY_SIZE(gen7_render_regs), false }, - { hsw_render_regs, ARRAY_SIZE(hsw_render_regs), false }, - { hsw_master_regs, ARRAY_SIZE(hsw_master_regs), true }, + { gen7_render_regs, ARRAY_SIZE(gen7_render_regs) }, + { hsw_render_regs, ARRAY_SIZE(hsw_render_regs) }, }; static const struct drm_i915_reg_table hsw_blt_reg_tables[] = { - { gen7_blt_regs, ARRAY_SIZE(gen7_blt_regs), false }, - { hsw_master_regs, ARRAY_SIZE(hsw_master_regs), true }, + { gen7_blt_regs, ARRAY_SIZE(gen7_blt_regs) }, +}; + +static const struct drm_i915_reg_table gen9_blt_reg_tables[] = { + { gen9_blt_regs, ARRAY_SIZE(gen9_blt_regs) }, }; static u32 gen7_render_get_cmd_length_mask(u32 cmd_header) { - u32 client = (cmd_header & INSTR_CLIENT_MASK) >> INSTR_CLIENT_SHIFT; + u32 client = cmd_header >> INSTR_CLIENT_SHIFT; u32 subclient = (cmd_header & INSTR_SUBCLIENT_MASK) >> INSTR_SUBCLIENT_SHIFT; @@ -578,7 +735,7 @@ static u32 gen7_render_get_cmd_length_mask(u32 cmd_header) static u32 gen7_bsd_get_cmd_length_mask(u32 cmd_header) { - u32 client = (cmd_header & INSTR_CLIENT_MASK) >> INSTR_CLIENT_SHIFT; + u32 client = cmd_header >> INSTR_CLIENT_SHIFT; u32 subclient = (cmd_header & INSTR_SUBCLIENT_MASK) >> INSTR_SUBCLIENT_SHIFT; u32 op = (cmd_header & INSTR_26_TO_24_MASK) >> INSTR_26_TO_24_SHIFT; @@ -601,7 +758,7 @@ static u32 gen7_bsd_get_cmd_length_mask(u32 cmd_header) static u32 gen7_blt_get_cmd_length_mask(u32 cmd_header) { - u32 client = (cmd_header & INSTR_CLIENT_MASK) >> INSTR_CLIENT_SHIFT; + u32 client = cmd_header >> INSTR_CLIENT_SHIFT; if (client == INSTR_MI_CLIENT) return 0x3F; @@ -612,6 +769,17 @@ static u32 gen7_blt_get_cmd_length_mask(u32 cmd_header) return 0; } +static u32 gen9_blt_get_cmd_length_mask(u32 cmd_header) +{ + u32 client = cmd_header >> INSTR_CLIENT_SHIFT; + + if (client == INSTR_MI_CLIENT || client == INSTR_BC_CLIENT) + return 0xFF; + + DRM_DEBUG_DRIVER("CMD: Abnormal blt cmd length! 0x%08X\n", cmd_header); + return 0; +} + static bool validate_cmds_sorted(const struct intel_engine_cs *engine, const struct drm_i915_cmd_table *cmd_tables, int cmd_table_count) @@ -703,22 +871,15 @@ struct cmd_node { */ static inline u32 cmd_header_key(u32 x) { - u32 shift; - switch (x >> INSTR_CLIENT_SHIFT) { default: case INSTR_MI_CLIENT: - shift = STD_MI_OPCODE_SHIFT; - break; + return x >> STD_MI_OPCODE_SHIFT; case INSTR_RC_CLIENT: - shift = STD_3D_OPCODE_SHIFT; - break; + return x >> STD_3D_OPCODE_SHIFT; case INSTR_BC_CLIENT: - shift = STD_2D_OPCODE_SHIFT; - break; + return x >> STD_2D_OPCODE_SHIFT; } - - return x >> shift; } static int init_hash_table(struct intel_engine_cs *engine, @@ -776,18 +937,19 @@ void intel_engine_init_cmd_parser(struct intel_engine_cs *engine) int cmd_table_count; int ret; - if (!IS_GEN7(engine->i915)) + if (!IS_GEN7(engine->i915) && !(IS_GEN9(engine->i915) && + engine->id == BCS)) return; switch (engine->id) { case RCS: if (IS_HASWELL(engine->i915)) { - cmd_tables = hsw_render_ring_cmds; + cmd_tables = hsw_render_ring_cmd_table; cmd_table_count = - ARRAY_SIZE(hsw_render_ring_cmds); + ARRAY_SIZE(hsw_render_ring_cmd_table); } else { - cmd_tables = gen7_render_cmds; - cmd_table_count = ARRAY_SIZE(gen7_render_cmds); + cmd_tables = gen7_render_cmd_table; + cmd_table_count = ARRAY_SIZE(gen7_render_cmd_table); } if (IS_HASWELL(engine->i915)) { @@ -797,36 +959,46 @@ void intel_engine_init_cmd_parser(struct intel_engine_cs *engine) engine->reg_tables = ivb_render_reg_tables; engine->reg_table_count = ARRAY_SIZE(ivb_render_reg_tables); } - engine->get_cmd_length_mask = gen7_render_get_cmd_length_mask; break; case VCS: - cmd_tables = gen7_video_cmds; - cmd_table_count = ARRAY_SIZE(gen7_video_cmds); + cmd_tables = gen7_video_cmd_table; + cmd_table_count = ARRAY_SIZE(gen7_video_cmd_table); engine->get_cmd_length_mask = gen7_bsd_get_cmd_length_mask; break; case BCS: - if (IS_HASWELL(engine->i915)) { - cmd_tables = hsw_blt_ring_cmds; - cmd_table_count = ARRAY_SIZE(hsw_blt_ring_cmds); + engine->get_cmd_length_mask = gen7_blt_get_cmd_length_mask; + if (IS_GEN9(engine->i915)) { + cmd_tables = gen9_blt_cmd_table; + cmd_table_count = ARRAY_SIZE(gen9_blt_cmd_table); + engine->get_cmd_length_mask = + gen9_blt_get_cmd_length_mask; + + /* BCS Engine unsafe without parser */ + engine->flags |= I915_ENGINE_REQUIRES_CMD_PARSER; + } else if (IS_HASWELL(engine->i915)) { + cmd_tables = hsw_blt_ring_cmd_table; + cmd_table_count = ARRAY_SIZE(hsw_blt_ring_cmd_table); } else { - cmd_tables = gen7_blt_cmds; - cmd_table_count = ARRAY_SIZE(gen7_blt_cmds); + cmd_tables = gen7_blt_cmd_table; + cmd_table_count = ARRAY_SIZE(gen7_blt_cmd_table); } - if (IS_HASWELL(engine->i915)) { + if (IS_GEN9(engine->i915)) { + engine->reg_tables = gen9_blt_reg_tables; + engine->reg_table_count = + ARRAY_SIZE(gen9_blt_reg_tables); + } else if (IS_HASWELL(engine->i915)) { engine->reg_tables = hsw_blt_reg_tables; engine->reg_table_count = ARRAY_SIZE(hsw_blt_reg_tables); } else { engine->reg_tables = ivb_blt_reg_tables; engine->reg_table_count = ARRAY_SIZE(ivb_blt_reg_tables); } - - engine->get_cmd_length_mask = gen7_blt_get_cmd_length_mask; break; case VECS: - cmd_tables = hsw_vebox_cmds; - cmd_table_count = ARRAY_SIZE(hsw_vebox_cmds); + cmd_tables = hsw_vebox_cmd_table; + cmd_table_count = ARRAY_SIZE(hsw_vebox_cmd_table); /* VECS can use the same length_mask function as VCS */ engine->get_cmd_length_mask = gen7_bsd_get_cmd_length_mask; break; @@ -852,7 +1024,7 @@ void intel_engine_init_cmd_parser(struct intel_engine_cs *engine) return; } - engine->needs_cmd_parser = true; + engine->flags |= I915_ENGINE_USING_CMD_PARSER; } /** @@ -864,7 +1036,7 @@ void intel_engine_init_cmd_parser(struct intel_engine_cs *engine) */ void intel_engine_cleanup_cmd_parser(struct intel_engine_cs *engine) { - if (!engine->needs_cmd_parser) + if (!intel_engine_using_cmd_parser(engine)) return; fini_hash_table(engine); @@ -938,22 +1110,16 @@ __find_reg(const struct drm_i915_reg_descriptor *table, int count, u32 addr) } static const struct drm_i915_reg_descriptor * -find_reg(const struct intel_engine_cs *engine, bool is_master, u32 addr) +find_reg(const struct intel_engine_cs *engine, u32 addr) { const struct drm_i915_reg_table *table = engine->reg_tables; + const struct drm_i915_reg_descriptor *reg = NULL; int count = engine->reg_table_count; - do { - if (!table->master || is_master) { - const struct drm_i915_reg_descriptor *reg; - - reg = __find_reg(table->regs, table->num_regs, addr); - if (reg != NULL) - return reg; - } - } while (table++, --count); + for (; !reg && (count > 0); ++table, --count) + reg = __find_reg(table->regs, table->num_regs, addr); - return NULL; + return reg; } /* Returns a vmap'd pointer to dst_obj, which the caller must unmap */ @@ -1036,32 +1202,9 @@ unpin_src: return dst; } -/** - * intel_engine_needs_cmd_parser() - should a given engine use software - * command parsing? - * @engine: the engine in question - * - * Only certain platforms require software batch buffer command parsing, and - * only when enabled via module parameter. - * - * Return: true if the engine requires software command parsing - */ -bool intel_engine_needs_cmd_parser(struct intel_engine_cs *engine) -{ - if (!engine->needs_cmd_parser) - return false; - - if (!USES_PPGTT(engine->i915)) - return false; - - return (i915.enable_cmd_parser == 1); -} - static bool check_cmd(const struct intel_engine_cs *engine, const struct drm_i915_cmd_descriptor *desc, - const u32 *cmd, u32 length, - const bool is_master, - bool *oacontrol_set) + const u32 *cmd, u32 length) { if (desc->flags & CMD_DESC_SKIP) return true; @@ -1071,12 +1214,6 @@ static bool check_cmd(const struct intel_engine_cs *engine, return false; } - if ((desc->flags & CMD_DESC_MASTER) && !is_master) { - DRM_DEBUG_DRIVER("CMD: Rejected master-only command: 0x%08X\n", - *cmd); - return false; - } - if (desc->flags & CMD_DESC_REGISTER) { /* * Get the distance between individual register offset @@ -1090,7 +1227,7 @@ static bool check_cmd(const struct intel_engine_cs *engine, offset += step) { const u32 reg_addr = cmd[offset] & desc->reg.mask; const struct drm_i915_reg_descriptor *reg = - find_reg(engine, is_master, reg_addr); + find_reg(engine, reg_addr); if (!reg) { DRM_DEBUG_DRIVER("CMD: Rejected register 0x%08X in command: 0x%08X (exec_id=%d)\n", @@ -1099,31 +1236,6 @@ static bool check_cmd(const struct intel_engine_cs *engine, } /* - * OACONTROL requires some special handling for - * writes. We want to make sure that any batch which - * enables OA also disables it before the end of the - * batch. The goal is to prevent one process from - * snooping on the perf data from another process. To do - * that, we need to check the value that will be written - * to the register. Hence, limit OACONTROL writes to - * only MI_LOAD_REGISTER_IMM commands. - */ - if (reg_addr == i915_mmio_reg_offset(OACONTROL)) { - if (desc->cmd.value == MI_LOAD_REGISTER_MEM) { - DRM_DEBUG_DRIVER("CMD: Rejected LRM to OACONTROL\n"); - return false; - } - - if (desc->cmd.value == MI_LOAD_REGISTER_REG) { - DRM_DEBUG_DRIVER("CMD: Rejected LRR to OACONTROL\n"); - return false; - } - - if (desc->cmd.value == MI_LOAD_REGISTER_IMM(1)) - *oacontrol_set = (cmd[offset + 1] != 0); - } - - /* * Check the value written to the register against the * allowed mask/value pair given in the whitelist entry. */ @@ -1170,6 +1282,12 @@ static bool check_cmd(const struct intel_engine_cs *engine, continue; } + if (desc->bits[i].offset >= length) { + DRM_DEBUG_DRIVER("CMD: Rejected command 0x%08X, too short to check bitmask (%s)\n", + *cmd, engine->name); + return false; + } + dword = cmd[desc->bits[i].offset] & desc->bits[i].mask; @@ -1187,16 +1305,112 @@ static bool check_cmd(const struct intel_engine_cs *engine, return true; } +static int check_bbstart(const struct i915_gem_context *ctx, + u32 *cmd, u32 offset, u32 length, + u32 batch_len, + u64 batch_start, + u64 shadow_batch_start) +{ + u64 jump_offset, jump_target; + u32 target_cmd_offset, target_cmd_index; + + /* For igt compatibility on older platforms */ + if (CMDPARSER_USES_GGTT(ctx->i915)) { + DRM_DEBUG("CMD: Rejecting BB_START for ggtt based submission\n"); + return -EACCES; + } + + if (length != 3) { + DRM_DEBUG("CMD: Recursive BB_START with bad length(%u)\n", + length); + return -EINVAL; + } + + jump_target = *(u64*)(cmd+1); + jump_offset = jump_target - batch_start; + + /* + * Any underflow of jump_target is guaranteed to be outside the range + * of a u32, so >= test catches both too large and too small + */ + if (jump_offset >= batch_len) { + DRM_DEBUG("CMD: BB_START to 0x%llx jumps out of BB\n", + jump_target); + return -EINVAL; + } + + /* + * This cannot overflow a u32 because we already checked jump_offset + * is within the BB, and the batch_len is a u32 + */ + target_cmd_offset = lower_32_bits(jump_offset); + target_cmd_index = target_cmd_offset / sizeof(u32); + + *(u64*)(cmd + 1) = shadow_batch_start + target_cmd_offset; + + if (target_cmd_index == offset) + return 0; + + if (ctx->jump_whitelist_cmds <= target_cmd_index) { + DRM_DEBUG("CMD: Rejecting BB_START - truncated whitelist array\n"); + return -EINVAL; + } else if (!test_bit(target_cmd_index, ctx->jump_whitelist)) { + DRM_DEBUG("CMD: BB_START to 0x%llx not a previously executed cmd\n", + jump_target); + return -EINVAL; + } + + return 0; +} + +static void init_whitelist(struct i915_gem_context *ctx, u32 batch_len) +{ + const u32 batch_cmds = DIV_ROUND_UP(batch_len, sizeof(u32)); + const u32 exact_size = BITS_TO_LONGS(batch_cmds); + u32 next_size = BITS_TO_LONGS(roundup_pow_of_two(batch_cmds)); + unsigned long *next_whitelist; + + if (CMDPARSER_USES_GGTT(ctx->i915)) + return; + + if (batch_cmds <= ctx->jump_whitelist_cmds) { + bitmap_zero(ctx->jump_whitelist, batch_cmds); + return; + } + +again: + next_whitelist = kcalloc(next_size, sizeof(long), GFP_KERNEL); + if (next_whitelist) { + kfree(ctx->jump_whitelist); + ctx->jump_whitelist = next_whitelist; + ctx->jump_whitelist_cmds = + next_size * BITS_PER_BYTE * sizeof(long); + return; + } + + if (next_size > exact_size) { + next_size = exact_size; + goto again; + } + + DRM_DEBUG("CMD: Failed to extend whitelist. BB_START may be disallowed\n"); + bitmap_zero(ctx->jump_whitelist, ctx->jump_whitelist_cmds); + + return; +} + #define LENGTH_BIAS 2 /** * i915_parse_cmds() - parse a submitted batch buffer for privilege violations + * @ctx: the context in which the batch is to execute * @engine: the engine on which the batch is to execute * @batch_obj: the batch buffer in question - * @shadow_batch_obj: copy of the batch buffer in question + * @batch_start: Canonical base address of batch * @batch_start_offset: byte offset in the batch at which execution starts * @batch_len: length of the commands in batch_obj - * @is_master: is the submitting process the drm master? + * @shadow_batch_obj: copy of the batch buffer in question + * @shadow_batch_start: Canonical base address of shadow_batch_obj * * Parses the specified batch buffer looking for privilege violations as * described in the overview. @@ -1204,17 +1418,19 @@ static bool check_cmd(const struct intel_engine_cs *engine, * Return: non-zero if the parser finds violations or otherwise fails; -EACCES * if the batch appears legal but should use hardware parsing */ -int intel_engine_cmd_parser(struct intel_engine_cs *engine, + +int intel_engine_cmd_parser(struct i915_gem_context *ctx, + struct intel_engine_cs *engine, struct drm_i915_gem_object *batch_obj, - struct drm_i915_gem_object *shadow_batch_obj, + u64 batch_start, u32 batch_start_offset, u32 batch_len, - bool is_master) + struct drm_i915_gem_object *shadow_batch_obj, + u64 shadow_batch_start) { - u32 *cmd, *batch_end; + u32 *cmd, *batch_end, offset = 0; struct drm_i915_cmd_descriptor default_desc = noop_desc; const struct drm_i915_cmd_descriptor *desc = &default_desc; - bool oacontrol_set = false; /* OACONTROL tracking. See check_cmd() */ bool needs_clflush_after = false; int ret = 0; @@ -1226,13 +1442,15 @@ int intel_engine_cmd_parser(struct intel_engine_cs *engine, return PTR_ERR(cmd); } + init_whitelist(ctx, batch_len); + /* * We use the batch length as size because the shadow object is as * large or larger and copy_batch() will write MI_NOPs to the extra * space. Parsing should be faster in some cases this way. */ batch_end = cmd + (batch_len / sizeof(*batch_end)); - while (cmd < batch_end) { + do { u32 length; if (*cmd == MI_BATCH_BUFFER_END) @@ -1243,17 +1461,7 @@ int intel_engine_cmd_parser(struct intel_engine_cs *engine, DRM_DEBUG_DRIVER("CMD: Unrecognized command: 0x%08X\n", *cmd); ret = -EINVAL; - break; - } - - /* - * If the batch buffer contains a chained batch, return an - * error that tells the caller to abort and dispatch the - * workload as a non-secure batch. - */ - if (desc->cmd.value == MI_BATCH_BUFFER_START) { - ret = -EACCES; - break; + goto err; } if (desc->flags & CMD_DESC_FIXED) @@ -1267,32 +1475,44 @@ int intel_engine_cmd_parser(struct intel_engine_cs *engine, length, batch_end - cmd); ret = -EINVAL; - break; + goto err; } - if (!check_cmd(engine, desc, cmd, length, is_master, - &oacontrol_set)) { - ret = -EINVAL; + if (!check_cmd(engine, desc, cmd, length)) { + ret = -EACCES; + goto err; + } + + if (desc->cmd.value == MI_BATCH_BUFFER_START) { + ret = check_bbstart(ctx, cmd, offset, length, + batch_len, batch_start, + shadow_batch_start); + + if (ret) + goto err; break; } - cmd += length; - } + if (ctx->jump_whitelist_cmds > offset) + set_bit(offset, ctx->jump_whitelist); - if (oacontrol_set) { - DRM_DEBUG_DRIVER("CMD: batch set OACONTROL but did not clear it\n"); - ret = -EINVAL; - } + cmd += length; + offset += length; + if (cmd >= batch_end) { + DRM_DEBUG_DRIVER("CMD: Got to the end of the buffer w/o a BBE cmd!\n"); + ret = -EINVAL; + goto err; + } + } while (1); - if (cmd >= batch_end) { - DRM_DEBUG_DRIVER("CMD: Got to the end of the buffer w/o a BBE cmd!\n"); - ret = -EINVAL; + if (needs_clflush_after) { + void *ptr = ptr_mask_bits(shadow_batch_obj->mapping); + drm_clflush_virt_range(ptr, + (void *)(cmd + 1) - ptr); } - if (ret == 0 && needs_clflush_after) - drm_clflush_virt_range(shadow_batch_obj->mapping, batch_len); +err: i915_gem_object_unpin_map(shadow_batch_obj); - return ret; } @@ -1312,7 +1532,7 @@ int i915_cmd_parser_get_version(struct drm_i915_private *dev_priv) /* If the command parser is not enabled, report 0 - unsupported */ for_each_engine(engine, dev_priv) { - if (intel_engine_needs_cmd_parser(engine)) { + if (intel_engine_using_cmd_parser(engine)) { active = true; break; } @@ -1332,6 +1552,12 @@ int i915_cmd_parser_get_version(struct drm_i915_private *dev_priv) * 5. GPGPU dispatch compute indirect registers. * 6. TIMESTAMP register and Haswell CS GPR registers * 7. Allow MI_LOAD_REGISTER_REG between whitelisted registers. + * 8. Don't report cmd_check() failures as EINVAL errors to userspace; + * rely on the HW to NOOP disallowed commands as it would without + * the parser enabled. + * 9. Don't whitelist or handle oacontrol specially, as ownership + * for oacontrol state is moving to i915-perf. + * 10. Support for Gen9 BCS Parsing */ - return 7; + return 10; } diff --git a/drivers/gpu/drm/i915/i915_drv.c b/drivers/gpu/drm/i915/i915_drv.c index bae62cf934cf..ff61229d963b 100644 --- a/drivers/gpu/drm/i915/i915_drv.c +++ b/drivers/gpu/drm/i915/i915_drv.c @@ -280,7 +280,7 @@ static int i915_getparam(struct drm_device *dev, void *data, value = i915.semaphores; break; case I915_PARAM_HAS_SECURE_BATCHES: - value = capable(CAP_SYS_ADMIN); + value = HAS_SECURE_BATCHES(dev_priv) && capable(CAP_SYS_ADMIN); break; case I915_PARAM_CMD_PARSER_VERSION: value = i915_cmd_parser_get_version(dev_priv); @@ -1470,6 +1470,7 @@ static int i915_drm_suspend_late(struct drm_device *dev, bool hibernation) disable_rpm_wakeref_asserts(dev_priv); intel_display_set_init_power(dev_priv, false); + i915_rc6_ctx_wa_suspend(dev_priv); fw_csr = !IS_BROXTON(dev_priv) && suspend_to_idle(dev_priv) && dev_priv->csr.dmc_payload; @@ -1706,6 +1707,8 @@ static int i915_drm_resume_early(struct drm_device *dev) else intel_display_set_init_power(dev_priv, true); + i915_rc6_ctx_wa_resume(dev_priv); + enable_rpm_wakeref_asserts(dev_priv); out: diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h index e23748cca0c0..c4f155663ca9 100644 --- a/drivers/gpu/drm/i915/i915_drv.h +++ b/drivers/gpu/drm/i915/i915_drv.h @@ -943,6 +943,13 @@ struct i915_gem_context { struct list_head link; u8 remap_slice; + + /** jump_whitelist: Bit array for tracking cmds during cmdparsing */ + unsigned long *jump_whitelist; + + /** jump_whitelist_cmds: No of cmd slots available */ + u32 jump_whitelist_cmds; + bool closed:1; }; @@ -1221,6 +1228,7 @@ struct intel_gen6_power_mgmt { bool client_boost; bool enabled; + bool ctx_corrupted; struct delayed_work autoenable_work; unsigned boosts; @@ -2339,6 +2347,18 @@ i915_gem_object_put_unlocked(struct drm_i915_gem_object *obj) __deprecated extern void drm_gem_object_unreference_unlocked(struct drm_gem_object *); +static inline void +i915_gem_object_set_readonly(struct drm_i915_gem_object *obj) +{ + obj->base.vma_node.readonly = true; +} + +static inline bool +i915_gem_object_is_readonly(const struct drm_i915_gem_object *obj) +{ + return obj->base.vma_node.readonly; +} + static inline bool i915_gem_object_has_struct_page(const struct drm_i915_gem_object *obj) { @@ -2476,102 +2496,6 @@ static inline struct scatterlist *__sg_next(struct scatterlist *sg) (((__iter).curr += PAGE_SIZE) < (__iter).max) || \ ((__iter) = __sgt_iter(__sg_next((__iter).sgp), false), 0)) -/* - * A command that requires special handling by the command parser. - */ -struct drm_i915_cmd_descriptor { - /* - * Flags describing how the command parser processes the command. - * - * CMD_DESC_FIXED: The command has a fixed length if this is set, - * a length mask if not set - * CMD_DESC_SKIP: The command is allowed but does not follow the - * standard length encoding for the opcode range in - * which it falls - * CMD_DESC_REJECT: The command is never allowed - * CMD_DESC_REGISTER: The command should be checked against the - * register whitelist for the appropriate ring - * CMD_DESC_MASTER: The command is allowed if the submitting process - * is the DRM master - */ - u32 flags; -#define CMD_DESC_FIXED (1<<0) -#define CMD_DESC_SKIP (1<<1) -#define CMD_DESC_REJECT (1<<2) -#define CMD_DESC_REGISTER (1<<3) -#define CMD_DESC_BITMASK (1<<4) -#define CMD_DESC_MASTER (1<<5) - - /* - * The command's unique identification bits and the bitmask to get them. - * This isn't strictly the opcode field as defined in the spec and may - * also include type, subtype, and/or subop fields. - */ - struct { - u32 value; - u32 mask; - } cmd; - - /* - * The command's length. The command is either fixed length (i.e. does - * not include a length field) or has a length field mask. The flag - * CMD_DESC_FIXED indicates a fixed length. Otherwise, the command has - * a length mask. All command entries in a command table must include - * length information. - */ - union { - u32 fixed; - u32 mask; - } length; - - /* - * Describes where to find a register address in the command to check - * against the ring's register whitelist. Only valid if flags has the - * CMD_DESC_REGISTER bit set. - * - * A non-zero step value implies that the command may access multiple - * registers in sequence (e.g. LRI), in that case step gives the - * distance in dwords between individual offset fields. - */ - struct { - u32 offset; - u32 mask; - u32 step; - } reg; - -#define MAX_CMD_DESC_BITMASKS 3 - /* - * Describes command checks where a particular dword is masked and - * compared against an expected value. If the command does not match - * the expected value, the parser rejects it. Only valid if flags has - * the CMD_DESC_BITMASK bit set. Only entries where mask is non-zero - * are valid. - * - * If the check specifies a non-zero condition_mask then the parser - * only performs the check when the bits specified by condition_mask - * are non-zero. - */ - struct { - u32 offset; - u32 mask; - u32 expected; - u32 condition_offset; - u32 condition_mask; - } bits[MAX_CMD_DESC_BITMASKS]; -}; - -/* - * A table of commands requiring special handling by the command parser. - * - * Each engine has an array of tables. Each table consists of an array of - * command descriptors, which must be sorted with command opcodes in - * ascending order. - */ -struct drm_i915_cmd_table { - const struct drm_i915_cmd_descriptor *table; - int count; -}; - /* Note that the (struct drm_i915_private *) cast is just to shut up gcc. */ #define __I915__(p) ({ \ struct drm_i915_private *__p; \ @@ -2729,6 +2653,12 @@ struct drm_i915_cmd_table { #define IS_GEN8(dev) (!!(INTEL_INFO(dev)->gen_mask & BIT(7))) #define IS_GEN9(dev) (!!(INTEL_INFO(dev)->gen_mask & BIT(8))) +/* + * The Gen7 cmdparser copies the scanned buffer to the ggtt for execution + * All later gens can run the final buffer from the ppgtt + */ +#define CMDPARSER_USES_GGTT(dev_priv) IS_GEN7(dev_priv) + #define ENGINE_MASK(id) BIT(id) #define RENDER_RING ENGINE_MASK(RCS) #define BSD_RING ENGINE_MASK(VCS) @@ -2745,6 +2675,8 @@ struct drm_i915_cmd_table { #define HAS_BLT(dev_priv) HAS_ENGINE(dev_priv, BCS) #define HAS_VEBOX(dev_priv) HAS_ENGINE(dev_priv, VECS) +#define HAS_SECURE_BATCHES(dev_priv) (INTEL_GEN(dev_priv) < 6) + #define HAS_LLC(dev) (INTEL_INFO(dev)->has_llc) #define HAS_SNOOP(dev) (INTEL_INFO(dev)->has_snoop) #define HAS_EDRAM(dev) (!!(__I915__(dev)->edram_cap & EDRAM_ENABLED)) @@ -2764,11 +2696,13 @@ struct drm_i915_cmd_table { /* Early gen2 have a totally busted CS tlb and require pinned batches. */ #define HAS_BROKEN_CS_TLB(dev) (IS_I830(dev) || IS_845G(dev)) +#define NEEDS_RC6_CTX_CORRUPTION_WA(dev_priv) \ + (IS_BROADWELL(dev_priv) || INTEL_GEN(dev_priv) == 9) + /* WaRsDisableCoarsePowerGating:skl,bxt */ #define NEEDS_WaRsDisableCoarsePowerGating(dev_priv) \ (IS_BXT_REVID(dev_priv, 0, BXT_REVID_A1) || \ - IS_SKL_GT3(dev_priv) || \ - IS_SKL_GT4(dev_priv)) + (INTEL_GEN(dev_priv) == 9)) /* * dp aux and gmbus irq on gen4 seems to be able to generate legacy interrupts @@ -3098,6 +3032,14 @@ i915_gem_object_ggtt_pin(struct drm_i915_gem_object *obj, u64 alignment, u64 flags); +struct i915_vma * __must_check +i915_gem_object_pin(struct drm_i915_gem_object *obj, + struct i915_address_space *vm, + const struct i915_ggtt_view *view, + u64 size, + u64 alignment, + u64 flags); + int i915_vma_bind(struct i915_vma *vma, enum i915_cache_level cache_level, u32 flags); void __i915_vma_set_map_and_fenceable(struct i915_vma *vma); @@ -3551,13 +3493,14 @@ const char *i915_cache_level_str(struct drm_i915_private *i915, int type); int i915_cmd_parser_get_version(struct drm_i915_private *dev_priv); void intel_engine_init_cmd_parser(struct intel_engine_cs *engine); void intel_engine_cleanup_cmd_parser(struct intel_engine_cs *engine); -bool intel_engine_needs_cmd_parser(struct intel_engine_cs *engine); -int intel_engine_cmd_parser(struct intel_engine_cs *engine, +int intel_engine_cmd_parser(struct i915_gem_context *cxt, + struct intel_engine_cs *engine, struct drm_i915_gem_object *batch_obj, - struct drm_i915_gem_object *shadow_batch_obj, + u64 user_batch_start, u32 batch_start_offset, u32 batch_len, - bool is_master); + struct drm_i915_gem_object *shadow_batch_obj, + u64 shadow_batch_start); /* i915_suspend.c */ extern int i915_save_state(struct drm_device *dev); diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c index 26c4befcd234..3fb4f9acacba 100644 --- a/drivers/gpu/drm/i915/i915_gem.c +++ b/drivers/gpu/drm/i915/i915_gem.c @@ -1773,6 +1773,10 @@ int i915_gem_fault(struct vm_area_struct *area, struct vm_fault *vmf) unsigned int flags; int ret; + /* Sanity check that we allow writing into this object */ + if (i915_gem_object_is_readonly(obj) && write) + return VM_FAULT_SIGBUS; + /* We don't use vmf->pgoff since that has the fake offset */ page_offset = ((unsigned long)vmf->virtual_address - area->vm_start) >> PAGE_SHIFT; @@ -2759,6 +2763,12 @@ i915_gem_idle_work_handler(struct work_struct *work) if (INTEL_GEN(dev_priv) >= 6) gen6_rps_idle(dev_priv); + + if (NEEDS_RC6_CTX_CORRUPTION_WA(dev_priv)) { + i915_rc6_ctx_wa_check(dev_priv); + intel_uncore_forcewake_put(dev_priv, FORCEWAKE_ALL); + } + intel_runtime_pm_put(dev_priv); out_unlock: mutex_unlock(&dev->struct_mutex); @@ -3822,6 +3832,19 @@ i915_gem_object_ggtt_pin(struct drm_i915_gem_object *obj, u64 flags) { struct i915_address_space *vm = &to_i915(obj->base.dev)->ggtt.base; + + return i915_gem_object_pin(obj, vm, view, size, alignment, + flags | PIN_GLOBAL); +} + +struct i915_vma * +i915_gem_object_pin(struct drm_i915_gem_object *obj, + struct i915_address_space *vm, + const struct i915_ggtt_view *view, + u64 size, + u64 alignment, + u64 flags) +{ struct i915_vma *vma; int ret; @@ -3846,7 +3869,7 @@ i915_gem_object_ggtt_pin(struct drm_i915_gem_object *obj, return ERR_PTR(ret); } - ret = i915_vma_pin(vma, size, alignment, flags | PIN_GLOBAL); + ret = i915_vma_pin(vma, size, alignment, flags); if (ret) return ERR_PTR(ret); diff --git a/drivers/gpu/drm/i915/i915_gem_context.c b/drivers/gpu/drm/i915/i915_gem_context.c index df10f4e95736..5d55cd159e89 100644 --- a/drivers/gpu/drm/i915/i915_gem_context.c +++ b/drivers/gpu/drm/i915/i915_gem_context.c @@ -158,6 +158,8 @@ void i915_gem_context_free(struct kref *ctx_ref) i915_vma_put(ce->state); } + kfree(ctx->jump_whitelist); + put_pid(ctx->pid); list_del(&ctx->link); @@ -327,6 +329,9 @@ __create_hw_context(struct drm_device *dev, GEN8_CTX_ADDRESSING_MODE_SHIFT; ATOMIC_INIT_NOTIFIER_HEAD(&ctx->status_notifier); + ctx->jump_whitelist = NULL; + ctx->jump_whitelist_cmds = 0; + return ctx; err_out: diff --git a/drivers/gpu/drm/i915/i915_gem_execbuffer.c b/drivers/gpu/drm/i915/i915_gem_execbuffer.c index 2117f172d7a2..4548d89abcdc 100644 --- a/drivers/gpu/drm/i915/i915_gem_execbuffer.c +++ b/drivers/gpu/drm/i915/i915_gem_execbuffer.c @@ -55,6 +55,7 @@ struct i915_execbuffer_params { struct i915_vma *batch; u32 dispatch_flags; u32 args_batch_start_offset; + u64 args_batch_len; struct intel_engine_cs *engine; struct i915_gem_context *ctx; struct drm_i915_gem_request *request; @@ -1401,41 +1402,85 @@ i915_reset_gen7_sol_offsets(struct drm_i915_gem_request *req) return 0; } +static struct i915_vma* +shadow_batch_pin(struct drm_i915_gem_object *obj, struct i915_address_space *vm) +{ + struct drm_i915_private *dev_priv = to_i915(obj->base.dev); + u64 flags; + + /* + * PPGTT backed shadow buffers must be mapped RO, to prevent + * post-scan tampering + */ + if (CMDPARSER_USES_GGTT(dev_priv)) { + flags = PIN_GLOBAL; + vm = &dev_priv->ggtt.base; + } else if (vm->has_read_only) { + flags = PIN_USER; + i915_gem_object_set_readonly(obj); + } else { + DRM_DEBUG("Cannot prevent post-scan tampering without RO capable vm\n"); + return ERR_PTR(-EINVAL); + } + + return i915_gem_object_pin(obj, vm, NULL, 0, 0, flags); +} + static struct i915_vma * i915_gem_execbuffer_parse(struct intel_engine_cs *engine, struct drm_i915_gem_exec_object2 *shadow_exec_entry, - struct drm_i915_gem_object *batch_obj, + struct i915_execbuffer_params *params, struct eb_vmas *eb, - u32 batch_start_offset, - u32 batch_len, - bool is_master) + struct i915_address_space *vm) { + struct drm_i915_gem_object *batch_obj = params->batch->obj; struct drm_i915_gem_object *shadow_batch_obj; struct i915_vma *vma; + u64 batch_start; + u32 batch_start_offset = params->args_batch_start_offset; + u32 batch_len = params->args_batch_len; + u64 shadow_batch_start; int ret; + shadow_batch_obj = i915_gem_batch_pool_get(&engine->batch_pool, PAGE_ALIGN(batch_len)); if (IS_ERR(shadow_batch_obj)) return ERR_CAST(shadow_batch_obj); - ret = intel_engine_cmd_parser(engine, + vma = shadow_batch_pin(shadow_batch_obj, vm); + if (IS_ERR(vma)) + goto out; + + batch_start = gen8_canonical_addr(params->batch->node.start) + + batch_start_offset; + shadow_batch_start = gen8_canonical_addr(vma->node.start); + + ret = intel_engine_cmd_parser(params->ctx, + engine, batch_obj, - shadow_batch_obj, + batch_start, batch_start_offset, batch_len, - is_master); + shadow_batch_obj, + shadow_batch_start); if (ret) { - if (ret == -EACCES) /* unhandled chained batch */ + i915_vma_unpin(vma); + + /* + * Unsafe GGTT-backed buffers can still be submitted safely + * as non-secure. + * For PPGTT backing however, we have no choice but to forcibly + * reject unsafe buffers + */ + if (CMDPARSER_USES_GGTT(eb->i915) && (ret == -EACCES)) + /* Execute original buffer non-secure */ vma = NULL; else vma = ERR_PTR(ret); - goto out; - } - vma = i915_gem_object_ggtt_pin(shadow_batch_obj, NULL, 0, 0, 0); - if (IS_ERR(vma)) goto out; + } memset(shadow_exec_entry, 0, sizeof(*shadow_exec_entry)); @@ -1476,13 +1521,10 @@ execbuf_submit(struct i915_execbuffer_params *params, return ret; } - exec_len = args->batch_len; + exec_len = params->args_batch_len; exec_start = params->batch->node.start + params->args_batch_start_offset; - if (exec_len == 0) - exec_len = params->batch->size - params->args_batch_start_offset; - ret = params->engine->emit_bb_start(params->request, exec_start, exec_len, params->dispatch_flags); @@ -1601,8 +1643,15 @@ i915_gem_do_execbuffer(struct drm_device *dev, void *data, dispatch_flags = 0; if (args->flags & I915_EXEC_SECURE) { + if (INTEL_GEN(dev_priv) >= 11) + return -ENODEV; + + /* Return -EPERM to trigger fallback code on old binaries. */ + if (!HAS_SECURE_BATCHES(dev_priv)) + return -EPERM; + if (!drm_is_current_master(file) || !capable(CAP_SYS_ADMIN)) - return -EPERM; + return -EPERM; dispatch_flags |= I915_DISPATCH_SECURE; } @@ -1710,32 +1759,26 @@ i915_gem_do_execbuffer(struct drm_device *dev, void *data, goto err; } + params->ctx = ctx; params->args_batch_start_offset = args->batch_start_offset; - if (intel_engine_needs_cmd_parser(engine) && args->batch_len) { + params->args_batch_len = args->batch_len; + if (args->batch_len == 0) + params->args_batch_len = params->batch->size - params->args_batch_start_offset; + + if (intel_engine_requires_cmd_parser(engine) || + (intel_engine_using_cmd_parser(engine) && args->batch_len)) { struct i915_vma *vma; vma = i915_gem_execbuffer_parse(engine, &shadow_exec_entry, - params->batch->obj, - eb, - args->batch_start_offset, - args->batch_len, - drm_is_current_master(file)); + params, eb, vm); if (IS_ERR(vma)) { ret = PTR_ERR(vma); goto err; } if (vma) { - /* - * Batch parsed and accepted: - * - * Set the DISPATCH_SECURE bit to remove the NON_SECURE - * bit from MI_BATCH_BUFFER_START commands issued in - * the dispatch_execbuffer implementations. We - * specifically don't want that set on batches the - * command parser has accepted. - */ - dispatch_flags |= I915_DISPATCH_SECURE; + if (CMDPARSER_USES_GGTT(dev_priv)) + dispatch_flags |= I915_DISPATCH_SECURE; params->args_batch_start_offset = 0; params->batch = vma; } @@ -1798,7 +1841,6 @@ i915_gem_do_execbuffer(struct drm_device *dev, void *data, params->file = file; params->engine = engine; params->dispatch_flags = dispatch_flags; - params->ctx = ctx; ret = execbuf_submit(params, args, &eb->vmas); err_request: diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.c b/drivers/gpu/drm/i915/i915_gem_gtt.c index 0bb4232f66bc..16f56f14f4d0 100644 --- a/drivers/gpu/drm/i915/i915_gem_gtt.c +++ b/drivers/gpu/drm/i915/i915_gem_gtt.c @@ -140,7 +140,8 @@ int intel_sanitize_enable_ppgtt(struct drm_i915_private *dev_priv, if (enable_ppgtt == 0 && INTEL_GEN(dev_priv) < 9) return 0; - if (enable_ppgtt == 1) + /* Full PPGTT is required by the Gen9 cmdparser */ + if (enable_ppgtt == 1 && INTEL_GEN(dev_priv) != 9) return 1; if (enable_ppgtt == 2 && has_full_ppgtt) @@ -177,8 +178,8 @@ static int ppgtt_bind_vma(struct i915_vma *vma, vma->pages = vma->obj->pages; - /* Currently applicable only to VLV */ - if (vma->obj->gt_ro) + /* Applicable to VLV, and gen8+ */ + if (i915_gem_object_is_readonly(vma->obj)) pte_flags |= PTE_READ_ONLY; vma->vm->insert_entries(vma->vm, vma->pages, vma->node.start, @@ -197,11 +198,14 @@ static void ppgtt_unbind_vma(struct i915_vma *vma) static gen8_pte_t gen8_pte_encode(dma_addr_t addr, enum i915_cache_level level, - bool valid) + bool valid, u32 flags) { gen8_pte_t pte = valid ? _PAGE_PRESENT | _PAGE_RW : 0; pte |= addr; + if (unlikely(flags & PTE_READ_ONLY)) + pte &= ~_PAGE_RW; + switch (level) { case I915_CACHE_NONE: pte |= PPAT_UNCACHED_INDEX; @@ -472,7 +476,7 @@ static void gen8_initialize_pt(struct i915_address_space *vm, gen8_pte_t scratch_pte; scratch_pte = gen8_pte_encode(vm->scratch_page.daddr, - I915_CACHE_LLC, true); + I915_CACHE_LLC, true, 0); fill_px(vm->dev, pt, scratch_pte); } @@ -769,7 +773,7 @@ static void gen8_ppgtt_clear_range(struct i915_address_space *vm, { struct i915_hw_ppgtt *ppgtt = i915_vm_to_ppgtt(vm); gen8_pte_t scratch_pte = gen8_pte_encode(vm->scratch_page.daddr, - I915_CACHE_LLC, use_scratch); + I915_CACHE_LLC, use_scratch, 0); if (!USES_FULL_48BIT_PPGTT(vm->dev)) { gen8_ppgtt_clear_pte_range(vm, &ppgtt->pdp, start, length, @@ -790,7 +794,8 @@ gen8_ppgtt_insert_pte_entries(struct i915_address_space *vm, struct i915_page_directory_pointer *pdp, struct sg_page_iter *sg_iter, uint64_t start, - enum i915_cache_level cache_level) + enum i915_cache_level cache_level, + u32 flags) { struct i915_hw_ppgtt *ppgtt = i915_vm_to_ppgtt(vm); gen8_pte_t *pt_vaddr; @@ -809,7 +814,7 @@ gen8_ppgtt_insert_pte_entries(struct i915_address_space *vm, pt_vaddr[pte] = gen8_pte_encode(sg_page_iter_dma_address(sg_iter), - cache_level, true); + cache_level, true, flags); if (++pte == GEN8_PTES) { kunmap_px(ppgtt, pt_vaddr); pt_vaddr = NULL; @@ -830,7 +835,7 @@ static void gen8_ppgtt_insert_entries(struct i915_address_space *vm, struct sg_table *pages, uint64_t start, enum i915_cache_level cache_level, - u32 unused) + u32 flags) { struct i915_hw_ppgtt *ppgtt = i915_vm_to_ppgtt(vm); struct sg_page_iter sg_iter; @@ -839,7 +844,7 @@ static void gen8_ppgtt_insert_entries(struct i915_address_space *vm, if (!USES_FULL_48BIT_PPGTT(vm->dev)) { gen8_ppgtt_insert_pte_entries(vm, &ppgtt->pdp, &sg_iter, start, - cache_level); + cache_level, flags); } else { struct i915_page_directory_pointer *pdp; uint64_t pml4e; @@ -847,7 +852,7 @@ static void gen8_ppgtt_insert_entries(struct i915_address_space *vm, gen8_for_each_pml4e(pdp, &ppgtt->pml4, start, length, pml4e) { gen8_ppgtt_insert_pte_entries(vm, pdp, &sg_iter, - start, cache_level); + start, cache_level, flags); } } } @@ -1452,7 +1457,7 @@ static void gen8_dump_ppgtt(struct i915_hw_ppgtt *ppgtt, struct seq_file *m) uint64_t start = ppgtt->base.start; uint64_t length = ppgtt->base.total; gen8_pte_t scratch_pte = gen8_pte_encode(vm->scratch_page.daddr, - I915_CACHE_LLC, true); + I915_CACHE_LLC, true, 0); if (!USES_FULL_48BIT_PPGTT(vm->dev)) { gen8_dump_pdp(&ppgtt->pdp, start, length, scratch_pte, m); @@ -1520,6 +1525,14 @@ static int gen8_ppgtt_init(struct i915_hw_ppgtt *ppgtt) ppgtt->base.clear_range = gen8_ppgtt_clear_range; ppgtt->base.unbind_vma = ppgtt_unbind_vma; ppgtt->base.bind_vma = ppgtt_bind_vma; + + /* + * From bdw, there is support for read-only pages in the PPGTT. + * + * XXX GVT is not honouring the lack of RW in the PTE bits. + */ + ppgtt->base.has_read_only = !intel_vgpu_active(to_i915(ppgtt->base.dev)); + ppgtt->debug_dump = gen8_dump_ppgtt; if (USES_FULL_48BIT_PPGTT(ppgtt->base.dev)) { @@ -2321,7 +2334,7 @@ static void gen8_ggtt_insert_page(struct i915_address_space *vm, rpm_atomic_seq = assert_rpm_atomic_begin(dev_priv); - gen8_set_pte(pte, gen8_pte_encode(addr, level, true)); + gen8_set_pte(pte, gen8_pte_encode(addr, level, true, 0)); I915_WRITE(GFX_FLSH_CNTL_GEN6, GFX_FLSH_CNTL_EN); POSTING_READ(GFX_FLSH_CNTL_GEN6); @@ -2332,7 +2345,7 @@ static void gen8_ggtt_insert_page(struct i915_address_space *vm, static void gen8_ggtt_insert_entries(struct i915_address_space *vm, struct sg_table *st, uint64_t start, - enum i915_cache_level level, u32 unused) + enum i915_cache_level level, u32 flags) { struct drm_i915_private *dev_priv = to_i915(vm->dev); struct i915_ggtt *ggtt = i915_vm_to_ggtt(vm); @@ -2343,12 +2356,20 @@ static void gen8_ggtt_insert_entries(struct i915_address_space *vm, int rpm_atomic_seq; int i = 0; + /* The GTT does not support read-only mappings */ + GEM_BUG_ON(flags & PTE_READ_ONLY); + rpm_atomic_seq = assert_rpm_atomic_begin(dev_priv); + /* + * Note that we ignore PTE_READ_ONLY here. The caller must be careful + * not to allow the user to override access to a read only page. + */ + gtt_entries = (gen8_pte_t __iomem *)ggtt->gsm + (start >> PAGE_SHIFT); for_each_sgt_dma(addr, sgt_iter, st) { - gtt_entry = gen8_pte_encode(addr, level, true); + gtt_entry = gen8_pte_encode(addr, level, true, 0); gen8_set_pte(>t_entries[i++], gtt_entry); } @@ -2499,7 +2520,7 @@ static void gen8_ggtt_clear_range(struct i915_address_space *vm, scratch_pte = gen8_pte_encode(vm->scratch_page.daddr, I915_CACHE_LLC, - use_scratch); + use_scratch, 0); for (i = 0; i < num_entries; i++) gen8_set_pte(>t_base[i], scratch_pte); readl(gtt_base); @@ -2604,8 +2625,8 @@ static int ggtt_bind_vma(struct i915_vma *vma, if (ret) return ret; - /* Currently applicable only to VLV */ - if (obj->gt_ro) + /* Applicable to VLV (gen8+ do not support RO in the GGTT) */ + if (i915_gem_object_is_readonly(obj)) pte_flags |= PTE_READ_ONLY; vma->vm->insert_entries(vma->vm, vma->pages, vma->node.start, @@ -2634,7 +2655,7 @@ static int aliasing_gtt_bind_vma(struct i915_vma *vma, /* Currently applicable only to VLV */ pte_flags = 0; - if (vma->obj->gt_ro) + if (i915_gem_object_is_readonly(vma->obj)) pte_flags |= PTE_READ_ONLY; @@ -3193,6 +3214,10 @@ int i915_ggtt_init_hw(struct drm_i915_private *dev_priv) ggtt->base.total -= PAGE_SIZE; i915_address_space_init(&ggtt->base, dev_priv); ggtt->base.total += PAGE_SIZE; + + /* Only VLV supports read-only GGTT mappings */ + ggtt->base.has_read_only = IS_VALLEYVIEW(dev_priv); + if (!HAS_LLC(dev_priv)) ggtt->base.mm.color_adjust = i915_gtt_color_adjust; diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.h b/drivers/gpu/drm/i915/i915_gem_gtt.h index ec78be2f8c77..43a0192242eb 100644 --- a/drivers/gpu/drm/i915/i915_gem_gtt.h +++ b/drivers/gpu/drm/i915/i915_gem_gtt.h @@ -392,6 +392,9 @@ struct i915_address_space { */ struct list_head unbound_list; + /* Some systems support read-only mappings for GGTT and/or PPGTT */ + bool has_read_only:1; + /* FIXME: Need a more generic return type */ gen6_pte_t (*pte_encode)(dma_addr_t addr, enum i915_cache_level level, diff --git a/drivers/gpu/drm/i915/i915_gem_request.c b/drivers/gpu/drm/i915/i915_gem_request.c index 8832f8ec1583..f597261c264f 100644 --- a/drivers/gpu/drm/i915/i915_gem_request.c +++ b/drivers/gpu/drm/i915/i915_gem_request.c @@ -558,6 +558,10 @@ static void i915_gem_mark_busy(const struct intel_engine_cs *engine) return; intel_runtime_pm_get_noresume(dev_priv); + + if (NEEDS_RC6_CTX_CORRUPTION_WA(dev_priv)) + intel_uncore_forcewake_get(dev_priv, FORCEWAKE_ALL); + dev_priv->gt.awake = true; intel_enable_gt_powersave(dev_priv); diff --git a/drivers/gpu/drm/i915/i915_params.c b/drivers/gpu/drm/i915/i915_params.c index 768ad89d9cd4..9d9dfe194b9b 100644 --- a/drivers/gpu/drm/i915/i915_params.c +++ b/drivers/gpu/drm/i915/i915_params.c @@ -49,7 +49,7 @@ struct i915_params i915 __read_mostly = { .reset = true, .invert_brightness = 0, .disable_display = 0, - .enable_cmd_parser = 1, + .enable_cmd_parser = true, .use_mmio_flip = 0, .mmio_debug = 0, .verbose_state_checks = 1, @@ -178,9 +178,9 @@ MODULE_PARM_DESC(invert_brightness, module_param_named(disable_display, i915.disable_display, bool, 0400); MODULE_PARM_DESC(disable_display, "Disable display (default: false)"); -module_param_named_unsafe(enable_cmd_parser, i915.enable_cmd_parser, int, 0600); +module_param_named_unsafe(enable_cmd_parser, i915.enable_cmd_parser, bool, 0400); MODULE_PARM_DESC(enable_cmd_parser, - "Enable command parsing (1=enabled [default], 0=disabled)"); + "Enable command parsing (true=enabled [default], false=disabled)"); module_param_named_unsafe(use_mmio_flip, i915.use_mmio_flip, int, 0600); MODULE_PARM_DESC(use_mmio_flip, diff --git a/drivers/gpu/drm/i915/i915_params.h b/drivers/gpu/drm/i915/i915_params.h index 3a0dd78ddb38..82ac6e886eed 100644 --- a/drivers/gpu/drm/i915/i915_params.h +++ b/drivers/gpu/drm/i915/i915_params.h @@ -44,7 +44,6 @@ struct i915_params { int disable_power_well; int enable_ips; int invert_brightness; - int enable_cmd_parser; int enable_guc_loading; int enable_guc_submission; int guc_log_level; @@ -53,6 +52,7 @@ struct i915_params { int edp_vswing; unsigned int inject_load_failure; /* leave bools at the end to not create holes */ + bool enable_cmd_parser; bool enable_hangcheck; bool fastboot; bool prefault_disable; diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h index 70d96162def6..5468e69bf520 100644 --- a/drivers/gpu/drm/i915/i915_reg.h +++ b/drivers/gpu/drm/i915/i915_reg.h @@ -223,6 +223,8 @@ static inline bool i915_mmio_reg_valid(i915_reg_t reg) #define GEN8_CONFIG0 _MMIO(0xD00) #define GEN9_DEFAULT_FIXES (1 << 3 | 1 << 2 | 1 << 1) +#define GEN8_RC6_CTX_INFO _MMIO(0x8504) + #define GAC_ECO_BITS _MMIO(0x14090) #define ECOBITS_SNB_BIT (1<<13) #define ECOBITS_PPGTT_CACHE64B (3<<8) @@ -295,7 +297,6 @@ static inline bool i915_mmio_reg_valid(i915_reg_t reg) * Instruction field definitions used by the command parser */ #define INSTR_CLIENT_SHIFT 29 -#define INSTR_CLIENT_MASK 0xE0000000 #define INSTR_MI_CLIENT 0x0 #define INSTR_BC_CLIENT 0x2 #define INSTR_RC_CLIENT 0x3 @@ -569,6 +570,10 @@ static inline bool i915_mmio_reg_valid(i915_reg_t reg) */ #define BCS_SWCTRL _MMIO(0x22200) +/* There are 16 GPR registers */ +#define BCS_GPR(n) _MMIO(0x22600 + (n) * 8) +#define BCS_GPR_UDW(n) _MMIO(0x22600 + (n) * 8 + 4) + #define GPGPU_THREADS_DISPATCHED _MMIO(0x2290) #define GPGPU_THREADS_DISPATCHED_UDW _MMIO(0x2290 + 4) #define HS_INVOCATION_COUNT _MMIO(0x2300) @@ -5936,6 +5941,10 @@ enum { #define SKL_CSR_DC5_DC6_COUNT _MMIO(0x8002C) #define BXT_CSR_DC3_DC5_COUNT _MMIO(0x80038) +/* Display Internal Timeout Register */ +#define RM_TIMEOUT _MMIO(0x42060) +#define MMIO_TIMEOUT_US(us) ((us) << 0) + /* interrupts */ #define DE_MASTER_IRQ_CONTROL (1 << 31) #define DE_SPRITEB_FLIP_DONE (1 << 29) diff --git a/drivers/gpu/drm/i915/intel_csr.c b/drivers/gpu/drm/i915/intel_csr.c index 1ea0e1f43397..54d878cb458f 100644 --- a/drivers/gpu/drm/i915/intel_csr.c +++ b/drivers/gpu/drm/i915/intel_csr.c @@ -280,10 +280,17 @@ static uint32_t *parse_csr_fw(struct drm_i915_private *dev_priv, uint32_t i; uint32_t *dmc_payload; uint32_t required_version; + size_t fsize; if (!fw) return NULL; + fsize = sizeof(struct intel_css_header) + + sizeof(struct intel_package_header) + + sizeof(struct intel_dmc_header); + if (fsize > fw->size) + goto error_truncated; + /* Extract CSS Header information*/ css_header = (struct intel_css_header *)fw->data; if (sizeof(struct intel_css_header) != @@ -349,6 +356,9 @@ static uint32_t *parse_csr_fw(struct drm_i915_private *dev_priv, return NULL; } readcount += dmc_offset; + fsize += dmc_offset; + if (fsize > fw->size) + goto error_truncated; /* Extract dmc_header information. */ dmc_header = (struct intel_dmc_header *)&fw->data[readcount]; @@ -379,6 +389,10 @@ static uint32_t *parse_csr_fw(struct drm_i915_private *dev_priv, /* fw_size is in dwords, so multiplied by 4 to convert into bytes. */ nbytes = dmc_header->fw_size * 4; + fsize += nbytes; + if (fsize > fw->size) + goto error_truncated; + if (nbytes > CSR_MAX_FW_SIZE) { DRM_ERROR("CSR firmware too big (%u) bytes\n", nbytes); return NULL; @@ -392,6 +406,10 @@ static uint32_t *parse_csr_fw(struct drm_i915_private *dev_priv, } return memcpy(dmc_payload, &fw->data[readcount], nbytes); + +error_truncated: + DRM_ERROR("Truncated DMC firmware, rejecting.\n"); + return NULL; } static void csr_load_work_fn(struct work_struct *work) diff --git a/drivers/gpu/drm/i915/intel_drv.h b/drivers/gpu/drm/i915/intel_drv.h index 8aafb9601540..b3af565b7027 100644 --- a/drivers/gpu/drm/i915/intel_drv.h +++ b/drivers/gpu/drm/i915/intel_drv.h @@ -1730,6 +1730,9 @@ void intel_enable_gt_powersave(struct drm_i915_private *dev_priv); void intel_autoenable_gt_powersave(struct drm_i915_private *dev_priv); void intel_disable_gt_powersave(struct drm_i915_private *dev_priv); void intel_suspend_gt_powersave(struct drm_i915_private *dev_priv); +bool i915_rc6_ctx_wa_check(struct drm_i915_private *i915); +void i915_rc6_ctx_wa_suspend(struct drm_i915_private *i915); +void i915_rc6_ctx_wa_resume(struct drm_i915_private *i915); void gen6_rps_busy(struct drm_i915_private *dev_priv); void gen6_rps_reset_ei(struct drm_i915_private *dev_priv); void gen6_rps_idle(struct drm_i915_private *dev_priv); diff --git a/drivers/gpu/drm/i915/intel_lrc.c b/drivers/gpu/drm/i915/intel_lrc.c index 62bcc770a181..b935d62be918 100644 --- a/drivers/gpu/drm/i915/intel_lrc.c +++ b/drivers/gpu/drm/i915/intel_lrc.c @@ -1017,6 +1017,8 @@ static int gen9_init_indirectctx_bb(struct intel_engine_cs *engine, int ret; struct drm_i915_private *dev_priv = engine->i915; uint32_t index = wa_ctx_start(wa_ctx, *offset, CACHELINE_DWORDS); + u32 scratch_addr = + i915_ggtt_offset(engine->scratch) + 2 * CACHELINE_BYTES; /* WaDisableCtxRestoreArbitration:skl,bxt */ if (IS_SKL_REVID(dev_priv, 0, SKL_REVID_D0) || @@ -1036,22 +1038,17 @@ static int gen9_init_indirectctx_bb(struct intel_engine_cs *engine, GEN9_DISABLE_GATHER_AT_SET_SHADER_COMMON_SLICE)); wa_ctx_emit(batch, index, MI_NOOP); - /* WaClearSlmSpaceAtContextSwitch:kbl */ + /* WaClearSlmSpaceAtContextSwitch:skl,bxt,kbl,glk,cfl */ /* Actual scratch location is at 128 bytes offset */ - if (IS_KBL_REVID(dev_priv, 0, KBL_REVID_A0)) { - u32 scratch_addr = - i915_ggtt_offset(engine->scratch) + 2 * CACHELINE_BYTES; - - wa_ctx_emit(batch, index, GFX_OP_PIPE_CONTROL(6)); - wa_ctx_emit(batch, index, (PIPE_CONTROL_FLUSH_L3 | - PIPE_CONTROL_GLOBAL_GTT_IVB | - PIPE_CONTROL_CS_STALL | - PIPE_CONTROL_QW_WRITE)); - wa_ctx_emit(batch, index, scratch_addr); - wa_ctx_emit(batch, index, 0); - wa_ctx_emit(batch, index, 0); - wa_ctx_emit(batch, index, 0); - } + wa_ctx_emit(batch, index, GFX_OP_PIPE_CONTROL(6)); + wa_ctx_emit(batch, index, (PIPE_CONTROL_FLUSH_L3 | + PIPE_CONTROL_GLOBAL_GTT_IVB | + PIPE_CONTROL_CS_STALL | + PIPE_CONTROL_QW_WRITE)); + wa_ctx_emit(batch, index, scratch_addr); + wa_ctx_emit(batch, index, 0); + wa_ctx_emit(batch, index, 0); + wa_ctx_emit(batch, index, 0); /* WaMediaPoolStateCmdInWABB:bxt */ if (HAS_POOLED_EU(engine->i915)) { diff --git a/drivers/gpu/drm/i915/intel_pm.c b/drivers/gpu/drm/i915/intel_pm.c index 05427d292457..07d2a8e7f78c 100644 --- a/drivers/gpu/drm/i915/intel_pm.c +++ b/drivers/gpu/drm/i915/intel_pm.c @@ -105,6 +105,13 @@ static void bxt_init_clock_gating(struct drm_device *dev) if (IS_BXT_REVID(dev_priv, BXT_REVID_B0, REVID_FOREVER)) I915_WRITE(GEN9_CLKGATE_DIS_0, I915_READ(GEN9_CLKGATE_DIS_0) | PWM1_GATING_DIS | PWM2_GATING_DIS); + /* + * Lower the display internal timeout. + * This is needed to avoid any hard hangs when DSI port PLL + * is off and a MMIO access is attempted by any privilege + * application, using batch buffers or any other means. + */ + I915_WRITE(RM_TIMEOUT, MMIO_TIMEOUT_US(950)); } static void i915_pineview_get_mem_freq(struct drm_device *dev) @@ -5149,19 +5156,23 @@ static void gen9_disable_rps(struct drm_i915_private *dev_priv) I915_WRITE(GEN6_RP_CONTROL, 0); } -static void gen6_disable_rps(struct drm_i915_private *dev_priv) +static void gen6_disable_rc6(struct drm_i915_private *dev_priv) { I915_WRITE(GEN6_RC_CONTROL, 0); +} + +static void gen6_disable_rps(struct drm_i915_private *dev_priv) +{ I915_WRITE(GEN6_RPNSWREQ, 1 << 31); I915_WRITE(GEN6_RP_CONTROL, 0); } -static void cherryview_disable_rps(struct drm_i915_private *dev_priv) +static void cherryview_disable_rc6(struct drm_i915_private *dev_priv) { I915_WRITE(GEN6_RC_CONTROL, 0); } -static void valleyview_disable_rps(struct drm_i915_private *dev_priv) +static void valleyview_disable_rc6(struct drm_i915_private *dev_priv) { /* we're doing forcewake before Disabling RC6, * This what the BIOS expects when going into suspend */ @@ -5426,7 +5437,8 @@ static void gen9_enable_rc6(struct drm_i915_private *dev_priv) I915_WRITE(GEN9_RENDER_PG_IDLE_HYSTERESIS, 25); /* 3a: Enable RC6 */ - if (intel_enable_rc6() & INTEL_RC6_ENABLE) + if (!dev_priv->rps.ctx_corrupted && + intel_enable_rc6() & INTEL_RC6_ENABLE) rc6_mask = GEN6_RC_CTL_RC6_ENABLE; DRM_INFO("RC6 %s\n", onoff(rc6_mask & GEN6_RC_CTL_RC6_ENABLE)); /* WaRsUseTimeoutMode */ @@ -5484,7 +5496,8 @@ static void gen8_enable_rps(struct drm_i915_private *dev_priv) I915_WRITE(GEN6_RC6_THRESHOLD, 50000); /* 50/125ms per EI */ /* 3: Enable RC6 */ - if (intel_enable_rc6() & INTEL_RC6_ENABLE) + if (!dev_priv->rps.ctx_corrupted && + intel_enable_rc6() & INTEL_RC6_ENABLE) rc6_mask = GEN6_RC_CTL_RC6_ENABLE; intel_print_rc6_info(dev_priv, rc6_mask); if (IS_BROADWELL(dev_priv)) @@ -6655,6 +6668,95 @@ static void intel_init_emon(struct drm_i915_private *dev_priv) dev_priv->ips.corr = (lcfuse & LCFUSE_HIV_MASK); } +static bool i915_rc6_ctx_corrupted(struct drm_i915_private *dev_priv) +{ + return !I915_READ(GEN8_RC6_CTX_INFO); +} + +static void i915_rc6_ctx_wa_init(struct drm_i915_private *i915) +{ + if (!NEEDS_RC6_CTX_CORRUPTION_WA(i915)) + return; + + if (i915_rc6_ctx_corrupted(i915)) { + DRM_INFO("RC6 context corrupted, disabling runtime power management\n"); + i915->rps.ctx_corrupted = true; + intel_runtime_pm_get(i915); + } +} + +static void i915_rc6_ctx_wa_cleanup(struct drm_i915_private *i915) +{ + if (i915->rps.ctx_corrupted) { + intel_runtime_pm_put(i915); + i915->rps.ctx_corrupted = false; + } +} + +/** + * i915_rc6_ctx_wa_suspend - system suspend sequence for the RC6 CTX WA + * @i915: i915 device + * + * Perform any steps needed to clean up the RC6 CTX WA before system suspend. + */ +void i915_rc6_ctx_wa_suspend(struct drm_i915_private *i915) +{ + if (i915->rps.ctx_corrupted) + intel_runtime_pm_put(i915); +} + +/** + * i915_rc6_ctx_wa_resume - system resume sequence for the RC6 CTX WA + * @i915: i915 device + * + * Perform any steps needed to re-init the RC6 CTX WA after system resume. + */ +void i915_rc6_ctx_wa_resume(struct drm_i915_private *i915) +{ + if (!i915->rps.ctx_corrupted) + return; + + if (i915_rc6_ctx_corrupted(i915)) { + intel_runtime_pm_get(i915); + return; + } + + DRM_INFO("RC6 context restored, re-enabling runtime power management\n"); + i915->rps.ctx_corrupted = false; +} + +static void intel_disable_rc6(struct drm_i915_private *dev_priv); + +/** + * i915_rc6_ctx_wa_check - check for a new RC6 CTX corruption + * @i915: i915 device + * + * Check if an RC6 CTX corruption has happened since the last check and if so + * disable RC6 and runtime power management. + * + * Return false if no context corruption has happened since the last call of + * this function, true otherwise. +*/ +bool i915_rc6_ctx_wa_check(struct drm_i915_private *i915) +{ + if (!NEEDS_RC6_CTX_CORRUPTION_WA(i915)) + return false; + + if (i915->rps.ctx_corrupted) + return false; + + if (!i915_rc6_ctx_corrupted(i915)) + return false; + + DRM_NOTE("RC6 context corruption, disabling runtime power management\n"); + + intel_disable_rc6(i915); + i915->rps.ctx_corrupted = true; + intel_runtime_pm_get_noresume(i915); + + return true; +} + void intel_init_gt_powersave(struct drm_i915_private *dev_priv) { /* @@ -6669,6 +6771,8 @@ void intel_init_gt_powersave(struct drm_i915_private *dev_priv) mutex_lock(&dev_priv->drm.struct_mutex); mutex_lock(&dev_priv->rps.hw_lock); + i915_rc6_ctx_wa_init(dev_priv); + /* Initialize RPS limits (for userspace) */ if (IS_CHERRYVIEW(dev_priv)) cherryview_init_gt_powersave(dev_priv); @@ -6718,6 +6822,8 @@ void intel_cleanup_gt_powersave(struct drm_i915_private *dev_priv) if (IS_VALLEYVIEW(dev_priv)) valleyview_cleanup_gt_powersave(dev_priv); + i915_rc6_ctx_wa_cleanup(dev_priv); + if (!i915.enable_rc6) intel_runtime_pm_put(dev_priv); } @@ -6749,27 +6855,47 @@ void intel_sanitize_gt_powersave(struct drm_i915_private *dev_priv) gen6_reset_rps_interrupts(dev_priv); } -void intel_disable_gt_powersave(struct drm_i915_private *dev_priv) +static void __intel_disable_rc6(struct drm_i915_private *dev_priv) { - if (!READ_ONCE(dev_priv->rps.enabled)) - return; + if (INTEL_GEN(dev_priv) >= 9) + gen9_disable_rc6(dev_priv); + else if (IS_CHERRYVIEW(dev_priv)) + cherryview_disable_rc6(dev_priv); + else if (IS_VALLEYVIEW(dev_priv)) + valleyview_disable_rc6(dev_priv); + else if (INTEL_GEN(dev_priv) >= 6) + gen6_disable_rc6(dev_priv); +} +static void intel_disable_rc6(struct drm_i915_private *dev_priv) +{ mutex_lock(&dev_priv->rps.hw_lock); + __intel_disable_rc6(dev_priv); + mutex_unlock(&dev_priv->rps.hw_lock); +} - if (INTEL_GEN(dev_priv) >= 9) { - gen9_disable_rc6(dev_priv); +static void intel_disable_rps(struct drm_i915_private *dev_priv) +{ + if (INTEL_GEN(dev_priv) >= 9) gen9_disable_rps(dev_priv); - } else if (IS_CHERRYVIEW(dev_priv)) { - cherryview_disable_rps(dev_priv); - } else if (IS_VALLEYVIEW(dev_priv)) { - valleyview_disable_rps(dev_priv); - } else if (INTEL_GEN(dev_priv) >= 6) { + else if (INTEL_GEN(dev_priv) >= 6) gen6_disable_rps(dev_priv); - } else if (IS_IRONLAKE_M(dev_priv)) { + else if (IS_IRONLAKE_M(dev_priv)) ironlake_disable_drps(dev_priv); - } +} + +void intel_disable_gt_powersave(struct drm_i915_private *dev_priv) +{ + if (!READ_ONCE(dev_priv->rps.enabled)) + return; + + mutex_lock(&dev_priv->rps.hw_lock); + + __intel_disable_rc6(dev_priv); + intel_disable_rps(dev_priv); dev_priv->rps.enabled = false; + mutex_unlock(&dev_priv->rps.hw_lock); } diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.c b/drivers/gpu/drm/i915/intel_ringbuffer.c index 8babfe0ce4e3..29c3123840ae 100644 --- a/drivers/gpu/drm/i915/intel_ringbuffer.c +++ b/drivers/gpu/drm/i915/intel_ringbuffer.c @@ -1951,6 +1951,7 @@ void intel_ring_unpin(struct intel_ring *ring) static struct i915_vma * intel_ring_create_vma(struct drm_i915_private *dev_priv, int size) { + struct i915_address_space *vm = &dev_priv->ggtt.base; struct drm_i915_gem_object *obj; struct i915_vma *vma; @@ -1960,10 +1961,14 @@ intel_ring_create_vma(struct drm_i915_private *dev_priv, int size) if (IS_ERR(obj)) return ERR_CAST(obj); - /* mark ring buffers as read-only from GPU side by default */ - obj->gt_ro = 1; + /* + * Mark ring buffers as read-only from GPU side (so no stray overwrites) + * if supported by the platform's GGTT. + */ + if (vm->has_read_only) + i915_gem_object_set_readonly(obj); - vma = i915_vma_create(obj, &dev_priv->ggtt.base, NULL); + vma = i915_vma_create(obj, vm, NULL); if (IS_ERR(vma)) goto err; diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.h b/drivers/gpu/drm/i915/intel_ringbuffer.h index ec0b4a0c605d..ce14cd8495e8 100644 --- a/drivers/gpu/drm/i915/intel_ringbuffer.h +++ b/drivers/gpu/drm/i915/intel_ringbuffer.h @@ -341,7 +341,9 @@ struct intel_engine_cs { struct intel_engine_hangcheck hangcheck; - bool needs_cmd_parser; +#define I915_ENGINE_USING_CMD_PARSER BIT(0) +#define I915_ENGINE_REQUIRES_CMD_PARSER BIT(3) + unsigned int flags; /* * Table of commands the command parser needs to know about @@ -374,7 +376,19 @@ intel_engine_initialized(const struct intel_engine_cs *engine) return engine->i915 != NULL; } -static inline unsigned +static inline bool +intel_engine_using_cmd_parser(const struct intel_engine_cs *engine) +{ + return engine->flags & I915_ENGINE_USING_CMD_PARSER; +} + +static inline bool +intel_engine_requires_cmd_parser(const struct intel_engine_cs *engine) +{ + return engine->flags & I915_ENGINE_REQUIRES_CMD_PARSER; +} + +static inline unsigned int intel_engine_flag(const struct intel_engine_cs *engine) { return 1 << engine->id; |