Re: [Intel-gfx] [PATCH 1/4] drm/i915/tgl: Introduce gen12 forcewake ranges
On 9/12/19 6:38 AM, Mika Kuoppala wrote: From: Michel Thierry The media ranges extend beyond what gen11 gives so we can't piggypack on gen11 ranges, even on read side. Introduce a table for gen12 and accessors for it. v2: correctly implement gen12_fwtable_write/read (Daniele) v3: update with ranges from bspec. v4: avoid GEN11_NEEDS_FORCEWAKE on read size (Mika) If we think the overhead of the unneeded BSEARCH is not too bad, should we remove this from the other accessors as well? Having only the gen12 paths behave differently feels inconsistent. We should recoup the cost later with the display uncore split (if I manage to find time to get back to it). BSpec: 18331. This should be 52078 for TGL. Cc: Tvrtko Ursulin Cc: Daniele Ceraolo Spurio Signed-off-by: Michel Thierry Signed-off-by: Mika Kuoppala --- drivers/gpu/drm/i915/intel_uncore.c | 75 ++- drivers/gpu/drm/i915/selftests/intel_uncore.c | 2 + 2 files changed, 76 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/i915/intel_uncore.c b/drivers/gpu/drm/i915/intel_uncore.c index 9e583f13a9e4..18e8314641a8 100644 --- a/drivers/gpu/drm/i915/intel_uncore.c +++ b/drivers/gpu/drm/i915/intel_uncore.c @@ -910,6 +910,9 @@ static const struct intel_forcewake_range __vlv_fw_ranges[] = { __fwd; \ }) +#define __gen12_fwtable_reg_read_fw_domains(uncore, offset) \ + find_fw_domain(uncore, offset) + /* *Must* be sorted by offset! See intel_shadow_table_check(). */ static const i915_reg_t gen8_shadowed_regs[] = { RING_TAIL(RENDER_RING_BASE),/* 0x2000 (base) */ @@ -935,6 +938,20 @@ static const i915_reg_t gen11_shadowed_regs[] = { /* TODO: Other registers are not yet used */ }; +static const i915_reg_t gen12_shadowed_regs[] = { + RING_TAIL(RENDER_RING_BASE),/* 0x2000 (base) */ + GEN6_RPNSWREQ, /* 0xA008 */ + GEN6_RC_VIDEO_FREQ, /* 0xA00C */ + RING_TAIL(BLT_RING_BASE), /* 0x22000 (base) */ + RING_TAIL(GEN11_BSD_RING_BASE), /* 0x1C (base) */ + RING_TAIL(GEN11_BSD2_RING_BASE),/* 0x1C4000 (base) */ + RING_TAIL(GEN11_VEBOX_RING_BASE), /* 0x1C8000 (base) */ + RING_TAIL(GEN11_BSD3_RING_BASE),/* 0x1D (base) */ + RING_TAIL(GEN11_BSD4_RING_BASE),/* 0x1D4000 (base) */ + RING_TAIL(GEN11_VEBOX2_RING_BASE), /* 0x1D8000 (base) */ + /* TODO: Other registers are not yet used */ +}; + static int mmio_reg_cmp(u32 key, const i915_reg_t *reg) { u32 offset = i915_mmio_reg_offset(*reg); @@ -957,6 +974,7 @@ static bool is_gen##x##_shadowed(u32 offset) \ __is_genX_shadowed(8) __is_genX_shadowed(11) +__is_genX_shadowed(12) static enum forcewake_domains gen6_reg_write_fw_domains(struct intel_uncore *uncore, i915_reg_t reg) @@ -1010,6 +1028,15 @@ static const struct intel_forcewake_range __chv_fw_ranges[] = { __fwd; \ }) +#define __gen12_fwtable_reg_write_fw_domains(uncore, offset) \ +({ \ + enum forcewake_domains __fwd = 0; \ + const u32 __offset = (offset); \ + if (!is_gen12_shadowed(__offset)) \ + __fwd = find_fw_domain(uncore, __offset); \ + __fwd; \ +}) + /* *Must* be sorted by offset ranges! See intel_fw_table_check(). */ static const struct intel_forcewake_range __gen9_fw_ranges[] = { GEN_FW_RANGE(0x0, 0xaff, FORCEWAKE_BLITTER), @@ -1080,6 +1107,46 @@ static const struct intel_forcewake_range __gen11_fw_ranges[] = { GEN_FW_RANGE(0x1d8000, 0x1dbfff, FORCEWAKE_MEDIA_VEBOX1) }; +/* *Must* be sorted by offset ranges! See intel_fw_table_check(). */ +static const struct intel_forcewake_range __gen12_fw_ranges[] = { + GEN_FW_RANGE(0x0, 0xaff, FORCEWAKE_BLITTER), + GEN_FW_RANGE(0xb00, 0x1fff, 0), /* uncore range */ + GEN_FW_RANGE(0x2000, 0x26ff, FORCEWAKE_RENDER), + GEN_FW_RANGE(0x2700, 0x2fff, FORCEWAKE_BLITTER), + GEN_FW_RANGE(0x3000, 0x3fff, FORCEWAKE_RENDER), + GEN_FW_RANGE(0x4000, 0x51ff, FORCEWAKE_BLITTER), + GEN_FW_RANGE(0x5200, 0x7fff, FORCEWAKE_RENDER), + GEN_FW_RANGE(0x8000, 0x813f, FORCEWAKE_BLITTER), + GEN_FW_RANGE(0x8140, 0x815f, FORCEWAKE_RENDER), + GEN_FW_RANGE(0x8160, 0x82ff, FORCEWAKE_BLITTER), + GEN_FW_RANGE(0x8300, 0x84ff, FORCEWAKE_RENDER), + GEN_FW_RANGE(0x8500, 0x8bff, FORCEWAKE_BLITTER), + GEN_FW_RANGE(0x8c00, 0x8cff, FORCEWAKE_RENDER), + GEN_FW_RANGE(0x8d00, 0x93ff, FORCEWAKE_BLITTER), + GEN_FW_RANGE(0x9400, 0x97ff, FORCEWAKE_ALL), + GEN_FW_RANGE(0x9800, 0xafff, FORCEWAKE_BLITTER), + GEN_FW_RANGE(0xb000, 0xb47f, FORCEWAKE_RENDER), + GEN_FW_RANGE(0xb480, 0xdfff, FORCEWAKE_BLITTER), + GEN_FW_RANGE(0xe000, 0xe8ff, FORCEWAKE_RENDER), + GEN_FW_RANGE(0xe900, 0x147ff, FORCEWAKE_BLITTER), + GEN_FW_RANGE(0x14800, 0x148ff, FORCEWAKE_RENDER), + GEN_FW_
[Intel-gfx] [PATCH 1/4] drm/i915/tgl: Introduce gen12 forcewake ranges
From: Michel Thierry The media ranges extend beyond what gen11 gives so we can't piggypack on gen11 ranges, even on read side. Introduce a table for gen12 and accessors for it. v2: correctly implement gen12_fwtable_write/read (Daniele) v3: update with ranges from bspec. v4: avoid GEN11_NEEDS_FORCEWAKE on read size (Mika) BSpec: 18331. Cc: Tvrtko Ursulin Cc: Daniele Ceraolo Spurio Signed-off-by: Michel Thierry Signed-off-by: Mika Kuoppala --- drivers/gpu/drm/i915/intel_uncore.c | 75 ++- drivers/gpu/drm/i915/selftests/intel_uncore.c | 2 + 2 files changed, 76 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/i915/intel_uncore.c b/drivers/gpu/drm/i915/intel_uncore.c index 9e583f13a9e4..18e8314641a8 100644 --- a/drivers/gpu/drm/i915/intel_uncore.c +++ b/drivers/gpu/drm/i915/intel_uncore.c @@ -910,6 +910,9 @@ static const struct intel_forcewake_range __vlv_fw_ranges[] = { __fwd; \ }) +#define __gen12_fwtable_reg_read_fw_domains(uncore, offset) \ + find_fw_domain(uncore, offset) + /* *Must* be sorted by offset! See intel_shadow_table_check(). */ static const i915_reg_t gen8_shadowed_regs[] = { RING_TAIL(RENDER_RING_BASE),/* 0x2000 (base) */ @@ -935,6 +938,20 @@ static const i915_reg_t gen11_shadowed_regs[] = { /* TODO: Other registers are not yet used */ }; +static const i915_reg_t gen12_shadowed_regs[] = { + RING_TAIL(RENDER_RING_BASE),/* 0x2000 (base) */ + GEN6_RPNSWREQ, /* 0xA008 */ + GEN6_RC_VIDEO_FREQ, /* 0xA00C */ + RING_TAIL(BLT_RING_BASE), /* 0x22000 (base) */ + RING_TAIL(GEN11_BSD_RING_BASE), /* 0x1C (base) */ + RING_TAIL(GEN11_BSD2_RING_BASE),/* 0x1C4000 (base) */ + RING_TAIL(GEN11_VEBOX_RING_BASE), /* 0x1C8000 (base) */ + RING_TAIL(GEN11_BSD3_RING_BASE),/* 0x1D (base) */ + RING_TAIL(GEN11_BSD4_RING_BASE),/* 0x1D4000 (base) */ + RING_TAIL(GEN11_VEBOX2_RING_BASE), /* 0x1D8000 (base) */ + /* TODO: Other registers are not yet used */ +}; + static int mmio_reg_cmp(u32 key, const i915_reg_t *reg) { u32 offset = i915_mmio_reg_offset(*reg); @@ -957,6 +974,7 @@ static bool is_gen##x##_shadowed(u32 offset) \ __is_genX_shadowed(8) __is_genX_shadowed(11) +__is_genX_shadowed(12) static enum forcewake_domains gen6_reg_write_fw_domains(struct intel_uncore *uncore, i915_reg_t reg) @@ -1010,6 +1028,15 @@ static const struct intel_forcewake_range __chv_fw_ranges[] = { __fwd; \ }) +#define __gen12_fwtable_reg_write_fw_domains(uncore, offset) \ +({ \ + enum forcewake_domains __fwd = 0; \ + const u32 __offset = (offset); \ + if (!is_gen12_shadowed(__offset)) \ + __fwd = find_fw_domain(uncore, __offset); \ + __fwd; \ +}) + /* *Must* be sorted by offset ranges! See intel_fw_table_check(). */ static const struct intel_forcewake_range __gen9_fw_ranges[] = { GEN_FW_RANGE(0x0, 0xaff, FORCEWAKE_BLITTER), @@ -1080,6 +1107,46 @@ static const struct intel_forcewake_range __gen11_fw_ranges[] = { GEN_FW_RANGE(0x1d8000, 0x1dbfff, FORCEWAKE_MEDIA_VEBOX1) }; +/* *Must* be sorted by offset ranges! See intel_fw_table_check(). */ +static const struct intel_forcewake_range __gen12_fw_ranges[] = { + GEN_FW_RANGE(0x0, 0xaff, FORCEWAKE_BLITTER), + GEN_FW_RANGE(0xb00, 0x1fff, 0), /* uncore range */ + GEN_FW_RANGE(0x2000, 0x26ff, FORCEWAKE_RENDER), + GEN_FW_RANGE(0x2700, 0x2fff, FORCEWAKE_BLITTER), + GEN_FW_RANGE(0x3000, 0x3fff, FORCEWAKE_RENDER), + GEN_FW_RANGE(0x4000, 0x51ff, FORCEWAKE_BLITTER), + GEN_FW_RANGE(0x5200, 0x7fff, FORCEWAKE_RENDER), + GEN_FW_RANGE(0x8000, 0x813f, FORCEWAKE_BLITTER), + GEN_FW_RANGE(0x8140, 0x815f, FORCEWAKE_RENDER), + GEN_FW_RANGE(0x8160, 0x82ff, FORCEWAKE_BLITTER), + GEN_FW_RANGE(0x8300, 0x84ff, FORCEWAKE_RENDER), + GEN_FW_RANGE(0x8500, 0x8bff, FORCEWAKE_BLITTER), + GEN_FW_RANGE(0x8c00, 0x8cff, FORCEWAKE_RENDER), + GEN_FW_RANGE(0x8d00, 0x93ff, FORCEWAKE_BLITTER), + GEN_FW_RANGE(0x9400, 0x97ff, FORCEWAKE_ALL), + GEN_FW_RANGE(0x9800, 0xafff, FORCEWAKE_BLITTER), + GEN_FW_RANGE(0xb000, 0xb47f, FORCEWAKE_RENDER), + GEN_FW_RANGE(0xb480, 0xdfff, FORCEWAKE_BLITTER), + GEN_FW_RANGE(0xe000, 0xe8ff, FORCEWAKE_RENDER), + GEN_FW_RANGE(0xe900, 0x147ff, FORCEWAKE_BLITTER), + GEN_FW_RANGE(0x14800, 0x148ff, FORCEWAKE_RENDER), + GEN_FW_RANGE(0x14900, 0x19fff, FORCEWAKE_BLITTER), + GEN_FW_RANGE(0x1a000, 0x1a7ff, FORCEWAKE_RENDER), + GEN_FW_RANGE(0x1a800, 0x1afff, FORCEWAKE_BLITTER), + GEN_FW_RANGE(0x1b000, 0x1bfff, FORCEWAKE_RENDER), + GEN_FW_RANGE(0x1c000, 0x243ff, FORCEWAKE_BLITTER), + GEN_FW_RANGE(0x24400, 0x247ff, FORCEWAKE_RENDER), + GEN_FW_RANGE(0x24800, 0x3, FORCEWAKE_BLITTER), + GEN_FW_RAN