diff options
Diffstat (limited to 'drivers/gpu/drm/i915/intel_uncore.c')
| -rw-r--r-- | drivers/gpu/drm/i915/intel_uncore.c | 1417 |
1 files changed, 1018 insertions, 399 deletions
diff --git a/drivers/gpu/drm/i915/intel_uncore.c b/drivers/gpu/drm/i915/intel_uncore.c index 1bed8f666048..4adeb271fcbf 100644 --- a/drivers/gpu/drm/i915/intel_uncore.c +++ b/drivers/gpu/drm/i915/intel_uncore.c @@ -22,54 +22,100 @@ */ #include <linux/pm_runtime.h> -#include <asm/iosf_mbi.h> + +#include <drm/drm_managed.h> +#include <drm/drm_print.h> + +#include "display/intel_display_core.h" +#include "gt/intel_engine_regs.h" +#include "gt/intel_gt.h" +#include "gt/intel_gt_regs.h" #include "i915_drv.h" -#include "i915_trace.h" +#include "i915_iosf_mbi.h" +#include "i915_reg.h" #include "i915_vgpu.h" -#include "intel_pm.h" +#include "i915_wait_util.h" +#include "i915_mmio_range.h" +#include "intel_uncore_trace.h" #define FORCEWAKE_ACK_TIMEOUT_MS 50 #define GT_FIFO_TIMEOUT_MS 10 +struct intel_uncore *to_intel_uncore(struct drm_device *drm) +{ + return &to_i915(drm)->uncore; +} + #define __raw_posting_read(...) ((void)__raw_uncore_read32(__VA_ARGS__)) +static void +fw_domains_get(struct intel_uncore *uncore, enum forcewake_domains fw_domains) +{ + uncore->fw_get_funcs->force_wake_get(uncore, fw_domains); +} + void -intel_uncore_mmio_debug_init_early(struct intel_uncore_mmio_debug *mmio_debug) +intel_uncore_mmio_debug_init_early(struct drm_i915_private *i915) { - spin_lock_init(&mmio_debug->lock); - mmio_debug->unclaimed_mmio_check = 1; + spin_lock_init(&i915->mmio_debug.lock); + i915->mmio_debug.unclaimed_mmio_check = 1; + + i915->uncore.debug = &i915->mmio_debug; } -static void mmio_debug_suspend(struct intel_uncore_mmio_debug *mmio_debug) +static void mmio_debug_suspend(struct intel_uncore *uncore) { - lockdep_assert_held(&mmio_debug->lock); + if (!uncore->debug) + return; + + spin_lock(&uncore->debug->lock); /* Save and disable mmio debugging for the user bypass */ - if (!mmio_debug->suspend_count++) { - mmio_debug->saved_mmio_check = mmio_debug->unclaimed_mmio_check; - mmio_debug->unclaimed_mmio_check = 0; + if (!uncore->debug->suspend_count++) { + uncore->debug->saved_mmio_check = uncore->debug->unclaimed_mmio_check; + uncore->debug->unclaimed_mmio_check = 0; } + + spin_unlock(&uncore->debug->lock); } -static void mmio_debug_resume(struct intel_uncore_mmio_debug *mmio_debug) +static bool check_for_unclaimed_mmio(struct intel_uncore *uncore); + +static void mmio_debug_resume(struct intel_uncore *uncore) { - lockdep_assert_held(&mmio_debug->lock); + if (!uncore->debug) + return; + + spin_lock(&uncore->debug->lock); + + if (!--uncore->debug->suspend_count) + uncore->debug->unclaimed_mmio_check = uncore->debug->saved_mmio_check; - if (!--mmio_debug->suspend_count) - mmio_debug->unclaimed_mmio_check = mmio_debug->saved_mmio_check; + if (check_for_unclaimed_mmio(uncore)) + drm_info(&uncore->i915->drm, + "Invalid mmio detected during user access\n"); + + spin_unlock(&uncore->debug->lock); } static const char * const forcewake_domain_names[] = { "render", - "blitter", + "gt", "media", "vdbox0", "vdbox1", "vdbox2", "vdbox3", + "vdbox4", + "vdbox5", + "vdbox6", + "vdbox7", "vebox0", "vebox1", + "vebox2", + "vebox3", + "gsc", }; const char * @@ -97,8 +143,11 @@ fw_domain_reset(const struct intel_uncore_forcewake_domain *d) * trying to reset here does exist at this point (engines could be fused * off in ICL+), so no waiting for acks */ - /* WaRsClearFWBitsAtReset:bdw,skl */ - fw_clear(d, 0xffff); + /* WaRsClearFWBitsAtReset */ + if (GRAPHICS_VER(d->uncore->i915) >= 12) + fw_clear(d, 0xefff); + else + fw_clear(d, 0xffff); } static inline void @@ -139,11 +188,21 @@ wait_ack_set(const struct intel_uncore_forcewake_domain *d, static inline void fw_domain_wait_ack_clear(const struct intel_uncore_forcewake_domain *d) { - if (wait_ack_clear(d, FORCEWAKE_KERNEL)) { - DRM_ERROR("%s: timed out waiting for forcewake ack to clear.\n", - intel_uncore_forcewake_domain_to_str(d->id)); - add_taint_for_CI(d->uncore->i915, TAINT_WARN); /* CI now unreliable */ + if (!wait_ack_clear(d, FORCEWAKE_KERNEL)) + return; + + if (fw_ack(d) == ~0) { + drm_err(&d->uncore->i915->drm, + "%s: MMIO unreliable (forcewake register returns 0xFFFFFFFF)!\n", + intel_uncore_forcewake_domain_to_str(d->id)); + intel_gt_set_wedged_async(d->uncore->gt); + } else { + drm_err(&d->uncore->i915->drm, + "%s: timed out waiting for forcewake ack to clear.\n", + intel_uncore_forcewake_domain_to_str(d->id)); } + + add_taint_for_CI(d->uncore->i915, TAINT_WARN); /* CI now unreliable */ } enum ack_type { @@ -188,11 +247,12 @@ fw_domain_wait_ack_with_fallback(const struct intel_uncore_forcewake_domain *d, fw_clear(d, FORCEWAKE_KERNEL_FALLBACK); } while (!ack_detected && pass++ < 10); - DRM_DEBUG_DRIVER("%s had to use fallback to %s ack, 0x%x (passes %u)\n", - intel_uncore_forcewake_domain_to_str(d->id), - type == ACK_SET ? "set" : "clear", - fw_ack(d), - pass); + drm_dbg(&d->uncore->i915->drm, + "%s had to use fallback to %s ack, 0x%x (passes %u)\n", + intel_uncore_forcewake_domain_to_str(d->id), + type == ACK_SET ? "set" : "clear", + fw_ack(d), + pass); return ack_detected ? 0 : -ETIMEDOUT; } @@ -217,8 +277,9 @@ static inline void fw_domain_wait_ack_set(const struct intel_uncore_forcewake_domain *d) { if (wait_ack_set(d, FORCEWAKE_KERNEL)) { - DRM_ERROR("%s: timed out waiting for forcewake ack request.\n", - intel_uncore_forcewake_domain_to_str(d->id)); + drm_err(&d->uncore->i915->drm, + "%s: timed out waiting for forcewake ack request.\n", + intel_uncore_forcewake_domain_to_str(d->id)); add_taint_for_CI(d->uncore->i915, TAINT_WARN); /* CI now unreliable */ } } @@ -240,7 +301,7 @@ fw_domain_put(const struct intel_uncore_forcewake_domain *d) } static void -fw_domains_get(struct intel_uncore *uncore, enum forcewake_domains fw_domains) +fw_domains_get_normal(struct intel_uncore *uncore, enum forcewake_domains fw_domains) { struct intel_uncore_forcewake_domain *d; unsigned int tmp; @@ -332,7 +393,7 @@ static void __gen6_gt_wait_for_thread_c0(struct intel_uncore *uncore) static void fw_domains_get_with_thread_status(struct intel_uncore *uncore, enum forcewake_domains fw_domains) { - fw_domains_get(uncore, fw_domains); + fw_domains_get_normal(uncore, fw_domains); /* WaRsForcewakeWaitTC0:snb,ivb,hsw,bdw,vlv */ __gen6_gt_wait_for_thread_c0(uncore); @@ -388,7 +449,7 @@ intel_uncore_fw_release_timer(struct hrtimer *timer) GEM_BUG_ON(!domain->wake_count); if (--domain->wake_count == 0) - uncore->funcs.force_wake_put(uncore, domain->mask); + fw_domains_put(uncore, domain->mask); spin_unlock_irqrestore(&uncore->lock, irqflags); @@ -446,7 +507,7 @@ intel_uncore_forcewake_reset(struct intel_uncore *uncore) fw = uncore->fw_domains_active; if (fw) - uncore->funcs.force_wake_put(uncore, fw); + fw_domains_put(uncore, fw); fw_domains_reset(uncore, uncore->fw_domains); assert_forcewakes_inactive(uncore); @@ -554,7 +615,7 @@ static void forcewake_early_sanitize(struct intel_uncore *uncore, intel_uncore_forcewake_reset(uncore); if (restore_forcewake) { spin_lock_irq(&uncore->lock); - uncore->funcs.force_wake_get(uncore, restore_forcewake); + fw_domains_get(uncore, restore_forcewake); if (intel_uncore_has_fifo(uncore)) uncore->fifo_count = fifo_free_entries(uncore); @@ -615,7 +676,7 @@ static void __intel_uncore_forcewake_get(struct intel_uncore *uncore, } if (fw_domains) - uncore->funcs.force_wake_get(uncore, fw_domains); + fw_domains_get(uncore, fw_domains); } /** @@ -636,7 +697,7 @@ void intel_uncore_forcewake_get(struct intel_uncore *uncore, { unsigned long irqflags; - if (!uncore->funcs.force_wake_get) + if (!uncore->fw_get_funcs) return; assert_rpm_wakelock_held(uncore->rpm); @@ -659,9 +720,7 @@ void intel_uncore_forcewake_user_get(struct intel_uncore *uncore) spin_lock_irq(&uncore->lock); if (!uncore->user_forcewake_count++) { intel_uncore_forcewake_get__locked(uncore, FORCEWAKE_ALL); - spin_lock(&uncore->debug->lock); - mmio_debug_suspend(uncore->debug); - spin_unlock(&uncore->debug->lock); + mmio_debug_suspend(uncore); } spin_unlock_irq(&uncore->lock); } @@ -677,14 +736,7 @@ void intel_uncore_forcewake_user_put(struct intel_uncore *uncore) { spin_lock_irq(&uncore->lock); if (!--uncore->user_forcewake_count) { - spin_lock(&uncore->debug->lock); - mmio_debug_resume(uncore->debug); - - if (check_for_unclaimed_mmio(uncore)) - drm_info(&uncore->i915->drm, - "Invalid mmio detected during user access\n"); - spin_unlock(&uncore->debug->lock); - + mmio_debug_resume(uncore); intel_uncore_forcewake_put__locked(uncore, FORCEWAKE_ALL); } spin_unlock_irq(&uncore->lock); @@ -703,14 +755,15 @@ void intel_uncore_forcewake_get__locked(struct intel_uncore *uncore, { lockdep_assert_held(&uncore->lock); - if (!uncore->funcs.force_wake_get) + if (!uncore->fw_get_funcs) return; __intel_uncore_forcewake_get(uncore, fw_domains); } static void __intel_uncore_forcewake_put(struct intel_uncore *uncore, - enum forcewake_domains fw_domains) + enum forcewake_domains fw_domains, + bool delayed) { struct intel_uncore_forcewake_domain *domain; unsigned int tmp; @@ -725,7 +778,11 @@ static void __intel_uncore_forcewake_put(struct intel_uncore *uncore, continue; } - uncore->funcs.force_wake_put(uncore, domain->mask); + if (delayed && + !(domain->uncore->fw_domains_timer & domain->mask)) + fw_domain_arm_timer(domain); + else + fw_domains_put(uncore, domain->mask); } } @@ -742,11 +799,24 @@ void intel_uncore_forcewake_put(struct intel_uncore *uncore, { unsigned long irqflags; - if (!uncore->funcs.force_wake_put) + if (!uncore->fw_get_funcs) return; spin_lock_irqsave(&uncore->lock, irqflags); - __intel_uncore_forcewake_put(uncore, fw_domains); + __intel_uncore_forcewake_put(uncore, fw_domains, false); + spin_unlock_irqrestore(&uncore->lock, irqflags); +} + +void intel_uncore_forcewake_put_delayed(struct intel_uncore *uncore, + enum forcewake_domains fw_domains) +{ + unsigned long irqflags; + + if (!uncore->fw_get_funcs) + return; + + spin_lock_irqsave(&uncore->lock, irqflags); + __intel_uncore_forcewake_put(uncore, fw_domains, true); spin_unlock_irqrestore(&uncore->lock, irqflags); } @@ -761,7 +831,7 @@ void intel_uncore_forcewake_flush(struct intel_uncore *uncore, struct intel_uncore_forcewake_domain *domain; unsigned int tmp; - if (!uncore->funcs.force_wake_put) + if (!uncore->fw_get_funcs) return; fw_domains &= uncore->fw_domains; @@ -773,9 +843,9 @@ void intel_uncore_forcewake_flush(struct intel_uncore *uncore, } /** - * intel_uncore_forcewake_put__locked - grab forcewake domain references + * intel_uncore_forcewake_put__locked - release forcewake domain references * @uncore: the intel_uncore structure - * @fw_domains: forcewake domains to get reference on + * @fw_domains: forcewake domains to put references * * See intel_uncore_forcewake_put(). This variant places the onus * on the caller to explicitly handle the dev_priv->uncore.lock spinlock. @@ -785,15 +855,15 @@ void intel_uncore_forcewake_put__locked(struct intel_uncore *uncore, { lockdep_assert_held(&uncore->lock); - if (!uncore->funcs.force_wake_put) + if (!uncore->fw_get_funcs) return; - __intel_uncore_forcewake_put(uncore, fw_domains); + __intel_uncore_forcewake_put(uncore, fw_domains, false); } void assert_forcewakes_inactive(struct intel_uncore *uncore) { - if (!uncore->funcs.force_wake_get) + if (!uncore->fw_get_funcs) return; drm_WARN(&uncore->i915->drm, uncore->fw_domains_active, @@ -810,7 +880,7 @@ void assert_forcewakes_active(struct intel_uncore *uncore, if (!IS_ENABLED(CONFIG_DRM_I915_DEBUG_RUNTIME_PM)) return; - if (!uncore->funcs.force_wake_get) + if (!uncore->fw_get_funcs) return; spin_lock_irq(&uncore->lock); @@ -842,17 +912,13 @@ void assert_forcewakes_active(struct intel_uncore *uncore, spin_unlock_irq(&uncore->lock); } -/* We give fast paths for the really cool registers */ -#define NEEDS_FORCE_WAKE(reg) ((reg) < 0x40000) - -#define __gen6_reg_read_fw_domains(uncore, offset) \ -({ \ - enum forcewake_domains __fwd; \ - if (NEEDS_FORCE_WAKE(offset)) \ - __fwd = FORCEWAKE_RENDER; \ - else \ - __fwd = 0; \ - __fwd; \ +/* + * We give fast paths for the really cool registers. The second range includes + * media domains (and the GSC starting from Xe_LPM+) + */ +#define NEEDS_FORCE_WAKE(reg) ({ \ + u32 __reg = (reg); \ + __reg < 0x40000 || __reg >= 0x116000; \ }) static int fw_range_cmp(u32 offset, const struct intel_forcewake_range *entry) @@ -889,6 +955,9 @@ find_fw_domain(struct intel_uncore *uncore, u32 offset) { const struct intel_forcewake_range *entry; + if (IS_GSI_REG(offset)) + offset += uncore->gsi_offset; + entry = BSEARCH(offset, uncore->fw_domains_table, uncore->fw_domains_table_entries, @@ -912,96 +981,206 @@ find_fw_domain(struct intel_uncore *uncore, u32 offset) return entry->domains; } -#define GEN_FW_RANGE(s, e, d) \ - { .start = (s), .end = (e), .domains = (d) } +/* + * Shadowed register tables describe special register ranges that i915 is + * allowed to write to without acquiring forcewake. If these registers' power + * wells are down, the hardware will save values written by i915 to a shadow + * copy and automatically transfer them into the real register the next time + * the power well is woken up. Shadowing only applies to writes; forcewake + * must still be acquired when reading from registers in these ranges. + * + * The documentation for shadowed registers is somewhat spotty on older + * platforms. However missing registers from these lists is non-fatal; it just + * means we'll wake up the hardware for some register accesses where we didn't + * really need to. + * + * The ranges listed in these tables must be sorted by offset. + * + * When adding new tables here, please also add them to + * intel_shadow_table_check() in selftests/intel_uncore.c so that they will be + * scanned for obvious mistakes or typos by the selftests. + */ -/* *Must* be sorted by offset ranges! See intel_fw_table_check(). */ -static const struct intel_forcewake_range __vlv_fw_ranges[] = { - GEN_FW_RANGE(0x2000, 0x3fff, FORCEWAKE_RENDER), - GEN_FW_RANGE(0x5000, 0x7fff, FORCEWAKE_RENDER), - GEN_FW_RANGE(0xb000, 0x11fff, FORCEWAKE_RENDER), - GEN_FW_RANGE(0x12000, 0x13fff, FORCEWAKE_MEDIA), - GEN_FW_RANGE(0x22000, 0x23fff, FORCEWAKE_MEDIA), - GEN_FW_RANGE(0x2e000, 0x2ffff, FORCEWAKE_RENDER), - GEN_FW_RANGE(0x30000, 0x3ffff, FORCEWAKE_MEDIA), +static const struct i915_mmio_range gen8_shadowed_regs[] = { + { .start = 0x2030, .end = 0x2030 }, + { .start = 0xA008, .end = 0xA00C }, + { .start = 0x12030, .end = 0x12030 }, + { .start = 0x1a030, .end = 0x1a030 }, + { .start = 0x22030, .end = 0x22030 }, }; -#define __fwtable_reg_read_fw_domains(uncore, offset) \ -({ \ - enum forcewake_domains __fwd = 0; \ - if (NEEDS_FORCE_WAKE((offset))) \ - __fwd = find_fw_domain(uncore, offset); \ - __fwd; \ -}) +static const struct i915_mmio_range gen11_shadowed_regs[] = { + { .start = 0x2030, .end = 0x2030 }, + { .start = 0x2550, .end = 0x2550 }, + { .start = 0xA008, .end = 0xA00C }, + { .start = 0x22030, .end = 0x22030 }, + { .start = 0x22230, .end = 0x22230 }, + { .start = 0x22510, .end = 0x22550 }, + { .start = 0x1C0030, .end = 0x1C0030 }, + { .start = 0x1C0230, .end = 0x1C0230 }, + { .start = 0x1C0510, .end = 0x1C0550 }, + { .start = 0x1C4030, .end = 0x1C4030 }, + { .start = 0x1C4230, .end = 0x1C4230 }, + { .start = 0x1C4510, .end = 0x1C4550 }, + { .start = 0x1C8030, .end = 0x1C8030 }, + { .start = 0x1C8230, .end = 0x1C8230 }, + { .start = 0x1C8510, .end = 0x1C8550 }, + { .start = 0x1D0030, .end = 0x1D0030 }, + { .start = 0x1D0230, .end = 0x1D0230 }, + { .start = 0x1D0510, .end = 0x1D0550 }, + { .start = 0x1D4030, .end = 0x1D4030 }, + { .start = 0x1D4230, .end = 0x1D4230 }, + { .start = 0x1D4510, .end = 0x1D4550 }, + { .start = 0x1D8030, .end = 0x1D8030 }, + { .start = 0x1D8230, .end = 0x1D8230 }, + { .start = 0x1D8510, .end = 0x1D8550 }, +}; + +static const struct i915_mmio_range gen12_shadowed_regs[] = { + { .start = 0x2030, .end = 0x2030 }, + { .start = 0x2510, .end = 0x2550 }, + { .start = 0xA008, .end = 0xA00C }, + { .start = 0xA188, .end = 0xA188 }, + { .start = 0xA278, .end = 0xA278 }, + { .start = 0xA540, .end = 0xA56C }, + { .start = 0xC4C8, .end = 0xC4C8 }, + { .start = 0xC4D4, .end = 0xC4D4 }, + { .start = 0xC600, .end = 0xC600 }, + { .start = 0x22030, .end = 0x22030 }, + { .start = 0x22510, .end = 0x22550 }, + { .start = 0x1C0030, .end = 0x1C0030 }, + { .start = 0x1C0510, .end = 0x1C0550 }, + { .start = 0x1C4030, .end = 0x1C4030 }, + { .start = 0x1C4510, .end = 0x1C4550 }, + { .start = 0x1C8030, .end = 0x1C8030 }, + { .start = 0x1C8510, .end = 0x1C8550 }, + { .start = 0x1D0030, .end = 0x1D0030 }, + { .start = 0x1D0510, .end = 0x1D0550 }, + { .start = 0x1D4030, .end = 0x1D4030 }, + { .start = 0x1D4510, .end = 0x1D4550 }, + { .start = 0x1D8030, .end = 0x1D8030 }, + { .start = 0x1D8510, .end = 0x1D8550 }, + + /* + * The rest of these ranges are specific to Xe_HP and beyond, but + * are reserved/unused ranges on earlier gen12 platforms, so they can + * be safely added to the gen12 table. + */ + { .start = 0x1E0030, .end = 0x1E0030 }, + { .start = 0x1E0510, .end = 0x1E0550 }, + { .start = 0x1E4030, .end = 0x1E4030 }, + { .start = 0x1E4510, .end = 0x1E4550 }, + { .start = 0x1E8030, .end = 0x1E8030 }, + { .start = 0x1E8510, .end = 0x1E8550 }, + { .start = 0x1F0030, .end = 0x1F0030 }, + { .start = 0x1F0510, .end = 0x1F0550 }, + { .start = 0x1F4030, .end = 0x1F4030 }, + { .start = 0x1F4510, .end = 0x1F4550 }, + { .start = 0x1F8030, .end = 0x1F8030 }, + { .start = 0x1F8510, .end = 0x1F8550 }, +}; -#define __gen11_fwtable_reg_read_fw_domains(uncore, offset) \ - find_fw_domain(uncore, offset) - -#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) */ - GEN6_RPNSWREQ, /* 0xA008 */ - GEN6_RC_VIDEO_FREQ, /* 0xA00C */ - RING_TAIL(GEN6_BSD_RING_BASE), /* 0x12000 (base) */ - RING_TAIL(VEBOX_RING_BASE), /* 0x1a000 (base) */ - RING_TAIL(BLT_RING_BASE), /* 0x22000 (base) */ - /* TODO: Other registers are not yet used */ +static const struct i915_mmio_range dg2_shadowed_regs[] = { + { .start = 0x2030, .end = 0x2030 }, + { .start = 0x2510, .end = 0x2550 }, + { .start = 0xA008, .end = 0xA00C }, + { .start = 0xA188, .end = 0xA188 }, + { .start = 0xA278, .end = 0xA278 }, + { .start = 0xA540, .end = 0xA56C }, + { .start = 0xC4C8, .end = 0xC4C8 }, + { .start = 0xC4E0, .end = 0xC4E0 }, + { .start = 0xC600, .end = 0xC600 }, + { .start = 0xC658, .end = 0xC658 }, + { .start = 0x22030, .end = 0x22030 }, + { .start = 0x22510, .end = 0x22550 }, + { .start = 0x1C0030, .end = 0x1C0030 }, + { .start = 0x1C0510, .end = 0x1C0550 }, + { .start = 0x1C4030, .end = 0x1C4030 }, + { .start = 0x1C4510, .end = 0x1C4550 }, + { .start = 0x1C8030, .end = 0x1C8030 }, + { .start = 0x1C8510, .end = 0x1C8550 }, + { .start = 0x1D0030, .end = 0x1D0030 }, + { .start = 0x1D0510, .end = 0x1D0550 }, + { .start = 0x1D4030, .end = 0x1D4030 }, + { .start = 0x1D4510, .end = 0x1D4550 }, + { .start = 0x1D8030, .end = 0x1D8030 }, + { .start = 0x1D8510, .end = 0x1D8550 }, + { .start = 0x1E0030, .end = 0x1E0030 }, + { .start = 0x1E0510, .end = 0x1E0550 }, + { .start = 0x1E4030, .end = 0x1E4030 }, + { .start = 0x1E4510, .end = 0x1E4550 }, + { .start = 0x1E8030, .end = 0x1E8030 }, + { .start = 0x1E8510, .end = 0x1E8550 }, + { .start = 0x1F0030, .end = 0x1F0030 }, + { .start = 0x1F0510, .end = 0x1F0550 }, + { .start = 0x1F4030, .end = 0x1F4030 }, + { .start = 0x1F4510, .end = 0x1F4550 }, + { .start = 0x1F8030, .end = 0x1F8030 }, + { .start = 0x1F8510, .end = 0x1F8550 }, }; -static const i915_reg_t gen11_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), /* 0x1C0000 (base) */ - RING_TAIL(GEN11_BSD2_RING_BASE), /* 0x1C4000 (base) */ - RING_TAIL(GEN11_VEBOX_RING_BASE), /* 0x1C8000 (base) */ - RING_TAIL(GEN11_BSD3_RING_BASE), /* 0x1D0000 (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 const struct i915_mmio_range mtl_shadowed_regs[] = { + { .start = 0x2030, .end = 0x2030 }, + { .start = 0x2510, .end = 0x2550 }, + { .start = 0xA008, .end = 0xA00C }, + { .start = 0xA188, .end = 0xA188 }, + { .start = 0xA278, .end = 0xA278 }, + { .start = 0xA540, .end = 0xA56C }, + { .start = 0xC050, .end = 0xC050 }, + { .start = 0xC340, .end = 0xC340 }, + { .start = 0xC4C8, .end = 0xC4C8 }, + { .start = 0xC4E0, .end = 0xC4E0 }, + { .start = 0xC600, .end = 0xC600 }, + { .start = 0xC658, .end = 0xC658 }, + { .start = 0xCFD4, .end = 0xCFDC }, + { .start = 0x22030, .end = 0x22030 }, + { .start = 0x22510, .end = 0x22550 }, }; -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), /* 0x1C0000 (base) */ - RING_TAIL(GEN11_BSD2_RING_BASE), /* 0x1C4000 (base) */ - RING_TAIL(GEN11_VEBOX_RING_BASE), /* 0x1C8000 (base) */ - RING_TAIL(GEN11_BSD3_RING_BASE), /* 0x1D0000 (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 const struct i915_mmio_range xelpmp_shadowed_regs[] = { + { .start = 0x1C0030, .end = 0x1C0030 }, + { .start = 0x1C0510, .end = 0x1C0550 }, + { .start = 0x1C8030, .end = 0x1C8030 }, + { .start = 0x1C8510, .end = 0x1C8550 }, + { .start = 0x1D0030, .end = 0x1D0030 }, + { .start = 0x1D0510, .end = 0x1D0550 }, + { .start = 0x38A008, .end = 0x38A00C }, + { .start = 0x38A188, .end = 0x38A188 }, + { .start = 0x38A278, .end = 0x38A278 }, + { .start = 0x38A540, .end = 0x38A56C }, + { .start = 0x38A618, .end = 0x38A618 }, + { .start = 0x38C050, .end = 0x38C050 }, + { .start = 0x38C340, .end = 0x38C340 }, + { .start = 0x38C4C8, .end = 0x38C4C8 }, + { .start = 0x38C4E0, .end = 0x38C4E4 }, + { .start = 0x38C600, .end = 0x38C600 }, + { .start = 0x38C658, .end = 0x38C658 }, + { .start = 0x38CFD4, .end = 0x38CFDC }, }; -static int mmio_reg_cmp(u32 key, const i915_reg_t *reg) +static int mmio_range_cmp(u32 key, const struct i915_mmio_range *range) { - u32 offset = i915_mmio_reg_offset(*reg); - - if (key < offset) + if (key < range->start) return -1; - else if (key > offset) + else if (key > range->end) return 1; else return 0; } -#define __is_genX_shadowed(x) \ -static bool is_gen##x##_shadowed(u32 offset) \ -{ \ - const i915_reg_t *regs = gen##x##_shadowed_regs; \ - return BSEARCH(offset, regs, ARRAY_SIZE(gen##x##_shadowed_regs), \ - mmio_reg_cmp); \ -} +static bool is_shadowed(struct intel_uncore *uncore, u32 offset) +{ + if (drm_WARN_ON(&uncore->i915->drm, !uncore->shadowed_reg_table)) + return false; + + if (IS_GSI_REG(offset)) + offset += uncore->gsi_offset; -__is_genX_shadowed(8) -__is_genX_shadowed(11) -__is_genX_shadowed(12) + return BSEARCH(offset, + uncore->shadowed_reg_table, + uncore->shadowed_reg_table_entries, + mmio_range_cmp); +} static enum forcewake_domains gen6_reg_write_fw_domains(struct intel_uncore *uncore, i915_reg_t reg) @@ -1009,17 +1188,70 @@ gen6_reg_write_fw_domains(struct intel_uncore *uncore, i915_reg_t reg) return FORCEWAKE_RENDER; } -#define __gen8_reg_write_fw_domains(uncore, offset) \ +#define __fwtable_reg_read_fw_domains(uncore, offset) \ +({ \ + enum forcewake_domains __fwd = 0; \ + if (NEEDS_FORCE_WAKE((offset))) \ + __fwd = find_fw_domain(uncore, offset); \ + __fwd; \ +}) + +#define __fwtable_reg_write_fw_domains(uncore, offset) \ ({ \ - enum forcewake_domains __fwd; \ - if (NEEDS_FORCE_WAKE(offset) && !is_gen8_shadowed(offset)) \ - __fwd = FORCEWAKE_RENDER; \ - else \ - __fwd = 0; \ + enum forcewake_domains __fwd = 0; \ + const u32 __offset = (offset); \ + if (NEEDS_FORCE_WAKE((__offset)) && !is_shadowed(uncore, __offset)) \ + __fwd = find_fw_domain(uncore, __offset); \ __fwd; \ }) -/* *Must* be sorted by offset ranges! See intel_fw_table_check(). */ +#define GEN_FW_RANGE(s, e, d) \ + { .start = (s), .end = (e), .domains = (d) } + +/* + * All platforms' forcewake tables below must be sorted by offset ranges. + * Furthermore, new forcewake tables added should be "watertight" and have + * no gaps between ranges. + * + * When there are multiple consecutive ranges listed in the bspec with + * the same forcewake domain, it is customary to combine them into a single + * row in the tables below to keep the tables small and lookups fast. + * Likewise, reserved/unused ranges may be combined with the preceding and/or + * following ranges since the driver will never be making MMIO accesses in + * those ranges. + * + * For example, if the bspec were to list: + * + * ... + * 0x1000 - 0x1fff: GT + * 0x2000 - 0x2cff: GT + * 0x2d00 - 0x2fff: unused/reserved + * 0x3000 - 0xffff: GT + * ... + * + * these could all be represented by a single line in the code: + * + * GEN_FW_RANGE(0x1000, 0xffff, FORCEWAKE_GT) + * + * When adding new forcewake tables here, please also add them to + * intel_uncore_mock_selftests in selftests/intel_uncore.c so that they will be + * scanned for obvious mistakes or typos by the selftests. + */ + +static const struct intel_forcewake_range __gen6_fw_ranges[] = { + GEN_FW_RANGE(0x0, 0x3ffff, FORCEWAKE_RENDER), +}; + +static const struct intel_forcewake_range __vlv_fw_ranges[] = { + GEN_FW_RANGE(0x2000, 0x3fff, FORCEWAKE_RENDER), + GEN_FW_RANGE(0x5000, 0x7fff, FORCEWAKE_RENDER), + GEN_FW_RANGE(0xb000, 0x11fff, FORCEWAKE_RENDER), + GEN_FW_RANGE(0x12000, 0x13fff, FORCEWAKE_MEDIA), + GEN_FW_RANGE(0x22000, 0x23fff, FORCEWAKE_MEDIA), + GEN_FW_RANGE(0x2e000, 0x2ffff, FORCEWAKE_RENDER), + GEN_FW_RANGE(0x30000, 0x3ffff, FORCEWAKE_MEDIA), +}; + static const struct intel_forcewake_range __chv_fw_ranges[] = { GEN_FW_RANGE(0x2000, 0x3fff, FORCEWAKE_RENDER), GEN_FW_RANGE(0x4000, 0x4fff, FORCEWAKE_RENDER | FORCEWAKE_MEDIA), @@ -1039,33 +1271,6 @@ static const struct intel_forcewake_range __chv_fw_ranges[] = { GEN_FW_RANGE(0x30000, 0x37fff, FORCEWAKE_MEDIA), }; -#define __fwtable_reg_write_fw_domains(uncore, offset) \ -({ \ - enum forcewake_domains __fwd = 0; \ - if (NEEDS_FORCE_WAKE((offset)) && !is_gen8_shadowed(offset)) \ - __fwd = find_fw_domain(uncore, offset); \ - __fwd; \ -}) - -#define __gen11_fwtable_reg_write_fw_domains(uncore, offset) \ -({ \ - enum forcewake_domains __fwd = 0; \ - const u32 __offset = (offset); \ - if (!is_gen11_shadowed(__offset)) \ - __fwd = find_fw_domain(uncore, __offset); \ - __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_GT), GEN_FW_RANGE(0xb00, 0x1fff, 0), /* uncore range */ @@ -1101,7 +1306,6 @@ static const struct intel_forcewake_range __gen9_fw_ranges[] = { GEN_FW_RANGE(0x30000, 0x3ffff, FORCEWAKE_MEDIA), }; -/* *Must* be sorted by offset ranges! See intel_fw_table_check(). */ static const struct intel_forcewake_range __gen11_fw_ranges[] = { GEN_FW_RANGE(0x0, 0x1fff, 0), /* uncore range */ GEN_FW_RANGE(0x2000, 0x26ff, FORCEWAKE_RENDER), @@ -1140,14 +1344,6 @@ static const struct intel_forcewake_range __gen11_fw_ranges[] = { GEN_FW_RANGE(0x1d4000, 0x1dbfff, 0) }; -/* - * *Must* be sorted by offset ranges! See intel_fw_table_check(). - * - * Note that the spec lists several reserved/unused ranges that don't - * actually contain any registers. In the table below we'll combine those - * reserved ranges with either the preceding or following range to keep the - * table small and lookups fast. - */ static const struct intel_forcewake_range __gen12_fw_ranges[] = { GEN_FW_RANGE(0x0, 0x1fff, 0), /* 0x0 - 0xaff: reserved @@ -1249,23 +1445,329 @@ static const struct intel_forcewake_range __gen12_fw_ranges[] = { 0x1d3f00 - 0x1d3fff: VD2 */ }; +static const struct intel_forcewake_range __dg2_fw_ranges[] = { + GEN_FW_RANGE(0x0, 0x1fff, 0), /* + 0x0 - 0xaff: reserved + 0xb00 - 0x1fff: always on */ + GEN_FW_RANGE(0x2000, 0x26ff, FORCEWAKE_RENDER), + GEN_FW_RANGE(0x2700, 0x4aff, FORCEWAKE_GT), + GEN_FW_RANGE(0x4b00, 0x51ff, 0), /* + 0x4b00 - 0x4fff: reserved + 0x5000 - 0x51ff: always on */ + GEN_FW_RANGE(0x5200, 0x7fff, FORCEWAKE_RENDER), + GEN_FW_RANGE(0x8000, 0x813f, FORCEWAKE_GT), + GEN_FW_RANGE(0x8140, 0x815f, FORCEWAKE_RENDER), + GEN_FW_RANGE(0x8160, 0x81ff, 0), /* + 0x8160 - 0x817f: reserved + 0x8180 - 0x81ff: always on */ + GEN_FW_RANGE(0x8200, 0x82ff, FORCEWAKE_GT), + GEN_FW_RANGE(0x8300, 0x84ff, FORCEWAKE_RENDER), + GEN_FW_RANGE(0x8500, 0x8cff, FORCEWAKE_GT), /* + 0x8500 - 0x87ff: gt + 0x8800 - 0x8c7f: reserved + 0x8c80 - 0x8cff: gt (DG2 only) */ + GEN_FW_RANGE(0x8d00, 0x8fff, FORCEWAKE_RENDER), /* + 0x8d00 - 0x8dff: render (DG2 only) + 0x8e00 - 0x8fff: reserved */ + GEN_FW_RANGE(0x9000, 0x94cf, FORCEWAKE_GT), /* + 0x9000 - 0x947f: gt + 0x9480 - 0x94cf: reserved */ + GEN_FW_RANGE(0x94d0, 0x955f, FORCEWAKE_RENDER), + GEN_FW_RANGE(0x9560, 0x967f, 0), /* + 0x9560 - 0x95ff: always on + 0x9600 - 0x967f: reserved */ + GEN_FW_RANGE(0x9680, 0x97ff, FORCEWAKE_RENDER), /* + 0x9680 - 0x96ff: render + 0x9700 - 0x97ff: reserved */ + GEN_FW_RANGE(0x9800, 0xcfff, FORCEWAKE_GT), /* + 0x9800 - 0xb4ff: gt + 0xb500 - 0xbfff: reserved + 0xc000 - 0xcfff: gt */ + GEN_FW_RANGE(0xd000, 0xd7ff, 0), + GEN_FW_RANGE(0xd800, 0xd87f, FORCEWAKE_RENDER), + GEN_FW_RANGE(0xd880, 0xdbff, FORCEWAKE_GT), + GEN_FW_RANGE(0xdc00, 0xdcff, FORCEWAKE_RENDER), + GEN_FW_RANGE(0xdd00, 0xde7f, FORCEWAKE_GT), /* + 0xdd00 - 0xddff: gt + 0xde00 - 0xde7f: reserved */ + GEN_FW_RANGE(0xde80, 0xe8ff, FORCEWAKE_RENDER), /* + 0xde80 - 0xdfff: render + 0xe000 - 0xe0ff: reserved + 0xe100 - 0xe8ff: render */ + GEN_FW_RANGE(0xe900, 0xffff, FORCEWAKE_GT), /* + 0xe900 - 0xe9ff: gt + 0xea00 - 0xefff: reserved + 0xf000 - 0xffff: gt */ + GEN_FW_RANGE(0x10000, 0x12fff, 0), /* + 0x10000 - 0x11fff: reserved + 0x12000 - 0x127ff: always on + 0x12800 - 0x12fff: reserved */ + GEN_FW_RANGE(0x13000, 0x131ff, FORCEWAKE_MEDIA_VDBOX0), + GEN_FW_RANGE(0x13200, 0x147ff, FORCEWAKE_MEDIA_VDBOX2), /* + 0x13200 - 0x133ff: VD2 (DG2 only) + 0x13400 - 0x147ff: reserved */ + GEN_FW_RANGE(0x14800, 0x14fff, FORCEWAKE_RENDER), + GEN_FW_RANGE(0x15000, 0x16dff, FORCEWAKE_GT), /* + 0x15000 - 0x15fff: gt (DG2 only) + 0x16000 - 0x16dff: reserved */ + GEN_FW_RANGE(0x16e00, 0x21fff, FORCEWAKE_RENDER), /* + 0x16e00 - 0x1ffff: render + 0x20000 - 0x21fff: reserved */ + GEN_FW_RANGE(0x22000, 0x23fff, FORCEWAKE_GT), + GEN_FW_RANGE(0x24000, 0x2417f, 0), /* + 0x24000 - 0x2407f: always on + 0x24080 - 0x2417f: reserved */ + GEN_FW_RANGE(0x24180, 0x249ff, FORCEWAKE_GT), /* + 0x24180 - 0x241ff: gt + 0x24200 - 0x249ff: reserved */ + GEN_FW_RANGE(0x24a00, 0x251ff, FORCEWAKE_RENDER), /* + 0x24a00 - 0x24a7f: render + 0x24a80 - 0x251ff: reserved */ + GEN_FW_RANGE(0x25200, 0x25fff, FORCEWAKE_GT), /* + 0x25200 - 0x252ff: gt + 0x25300 - 0x25fff: reserved */ + GEN_FW_RANGE(0x26000, 0x2ffff, FORCEWAKE_RENDER), /* + 0x26000 - 0x27fff: render + 0x28000 - 0x29fff: reserved + 0x2a000 - 0x2ffff: undocumented */ + GEN_FW_RANGE(0x30000, 0x3ffff, FORCEWAKE_GT), + GEN_FW_RANGE(0x40000, 0x1bffff, 0), + GEN_FW_RANGE(0x1c0000, 0x1c3fff, FORCEWAKE_MEDIA_VDBOX0), /* + 0x1c0000 - 0x1c2bff: VD0 + 0x1c2c00 - 0x1c2cff: reserved + 0x1c2d00 - 0x1c2dff: VD0 + 0x1c2e00 - 0x1c3eff: VD0 + 0x1c3f00 - 0x1c3fff: VD0 */ + GEN_FW_RANGE(0x1c4000, 0x1c7fff, FORCEWAKE_MEDIA_VDBOX1), /* + 0x1c4000 - 0x1c6bff: VD1 + 0x1c6c00 - 0x1c6cff: reserved + 0x1c6d00 - 0x1c6dff: VD1 + 0x1c6e00 - 0x1c7fff: reserved */ + GEN_FW_RANGE(0x1c8000, 0x1cbfff, FORCEWAKE_MEDIA_VEBOX0), /* + 0x1c8000 - 0x1ca0ff: VE0 + 0x1ca100 - 0x1cbfff: reserved */ + GEN_FW_RANGE(0x1cc000, 0x1ccfff, FORCEWAKE_MEDIA_VDBOX0), + GEN_FW_RANGE(0x1cd000, 0x1cdfff, FORCEWAKE_MEDIA_VDBOX2), + GEN_FW_RANGE(0x1ce000, 0x1cefff, FORCEWAKE_MEDIA_VDBOX4), + GEN_FW_RANGE(0x1cf000, 0x1cffff, FORCEWAKE_MEDIA_VDBOX6), + GEN_FW_RANGE(0x1d0000, 0x1d3fff, FORCEWAKE_MEDIA_VDBOX2), /* + 0x1d0000 - 0x1d2bff: VD2 + 0x1d2c00 - 0x1d2cff: reserved + 0x1d2d00 - 0x1d2dff: VD2 + 0x1d2e00 - 0x1d3dff: VD2 + 0x1d3e00 - 0x1d3eff: reserved + 0x1d3f00 - 0x1d3fff: VD2 */ + GEN_FW_RANGE(0x1d4000, 0x1d7fff, FORCEWAKE_MEDIA_VDBOX3), /* + 0x1d4000 - 0x1d6bff: VD3 + 0x1d6c00 - 0x1d6cff: reserved + 0x1d6d00 - 0x1d6dff: VD3 + 0x1d6e00 - 0x1d7fff: reserved */ + GEN_FW_RANGE(0x1d8000, 0x1dffff, FORCEWAKE_MEDIA_VEBOX1), /* + 0x1d8000 - 0x1da0ff: VE1 + 0x1da100 - 0x1dffff: reserved */ + GEN_FW_RANGE(0x1e0000, 0x1e3fff, FORCEWAKE_MEDIA_VDBOX4), /* + 0x1e0000 - 0x1e2bff: VD4 + 0x1e2c00 - 0x1e2cff: reserved + 0x1e2d00 - 0x1e2dff: VD4 + 0x1e2e00 - 0x1e3eff: reserved + 0x1e3f00 - 0x1e3fff: VD4 */ + GEN_FW_RANGE(0x1e4000, 0x1e7fff, FORCEWAKE_MEDIA_VDBOX5), /* + 0x1e4000 - 0x1e6bff: VD5 + 0x1e6c00 - 0x1e6cff: reserved + 0x1e6d00 - 0x1e6dff: VD5 + 0x1e6e00 - 0x1e7fff: reserved */ + GEN_FW_RANGE(0x1e8000, 0x1effff, FORCEWAKE_MEDIA_VEBOX2), /* + 0x1e8000 - 0x1ea0ff: VE2 + 0x1ea100 - 0x1effff: reserved */ + GEN_FW_RANGE(0x1f0000, 0x1f3fff, FORCEWAKE_MEDIA_VDBOX6), /* + 0x1f0000 - 0x1f2bff: VD6 + 0x1f2c00 - 0x1f2cff: reserved + 0x1f2d00 - 0x1f2dff: VD6 + 0x1f2e00 - 0x1f3eff: reserved + 0x1f3f00 - 0x1f3fff: VD6 */ + GEN_FW_RANGE(0x1f4000, 0x1f7fff, FORCEWAKE_MEDIA_VDBOX7), /* + 0x1f4000 - 0x1f6bff: VD7 + 0x1f6c00 - 0x1f6cff: reserved + 0x1f6d00 - 0x1f6dff: VD7 + 0x1f6e00 - 0x1f7fff: reserved */ + GEN_FW_RANGE(0x1f8000, 0x1fa0ff, FORCEWAKE_MEDIA_VEBOX3), +}; + +static const struct intel_forcewake_range __mtl_fw_ranges[] = { + GEN_FW_RANGE(0x0, 0xaff, 0), + GEN_FW_RANGE(0xb00, 0xbff, FORCEWAKE_GT), + GEN_FW_RANGE(0xc00, 0xfff, 0), + GEN_FW_RANGE(0x1000, 0x1fff, FORCEWAKE_GT), + GEN_FW_RANGE(0x2000, 0x26ff, FORCEWAKE_RENDER), + GEN_FW_RANGE(0x2700, 0x2fff, FORCEWAKE_GT), + GEN_FW_RANGE(0x3000, 0x3fff, FORCEWAKE_RENDER), + GEN_FW_RANGE(0x4000, 0x51ff, FORCEWAKE_GT), /* + 0x4000 - 0x48ff: render + 0x4900 - 0x51ff: reserved */ + GEN_FW_RANGE(0x5200, 0x7fff, FORCEWAKE_RENDER), /* + 0x5200 - 0x53ff: render + 0x5400 - 0x54ff: reserved + 0x5500 - 0x7fff: render */ + GEN_FW_RANGE(0x8000, 0x813f, FORCEWAKE_GT), + GEN_FW_RANGE(0x8140, 0x817f, FORCEWAKE_RENDER), /* + 0x8140 - 0x815f: render + 0x8160 - 0x817f: reserved */ + GEN_FW_RANGE(0x8180, 0x81ff, 0), + GEN_FW_RANGE(0x8200, 0x94cf, FORCEWAKE_GT), /* + 0x8200 - 0x87ff: gt + 0x8800 - 0x8dff: reserved + 0x8e00 - 0x8f7f: gt + 0x8f80 - 0x8fff: reserved + 0x9000 - 0x947f: gt + 0x9480 - 0x94cf: reserved */ + GEN_FW_RANGE(0x94d0, 0x955f, FORCEWAKE_RENDER), + GEN_FW_RANGE(0x9560, 0x967f, 0), /* + 0x9560 - 0x95ff: always on + 0x9600 - 0x967f: reserved */ + GEN_FW_RANGE(0x9680, 0x97ff, FORCEWAKE_RENDER), /* + 0x9680 - 0x96ff: render + 0x9700 - 0x97ff: reserved */ + GEN_FW_RANGE(0x9800, 0xcfff, FORCEWAKE_GT), /* + 0x9800 - 0xb4ff: gt + 0xb500 - 0xbfff: reserved + 0xc000 - 0xcfff: gt */ + GEN_FW_RANGE(0xd000, 0xd7ff, 0), /* + 0xd000 - 0xd3ff: always on + 0xd400 - 0xd7ff: reserved */ + GEN_FW_RANGE(0xd800, 0xd87f, FORCEWAKE_RENDER), + GEN_FW_RANGE(0xd880, 0xdbff, FORCEWAKE_GT), + GEN_FW_RANGE(0xdc00, 0xdcff, FORCEWAKE_RENDER), + GEN_FW_RANGE(0xdd00, 0xde7f, FORCEWAKE_GT), /* + 0xdd00 - 0xddff: gt + 0xde00 - 0xde7f: reserved */ + GEN_FW_RANGE(0xde80, 0xe8ff, FORCEWAKE_RENDER), /* + 0xde80 - 0xdfff: render + 0xe000 - 0xe0ff: reserved + 0xe100 - 0xe8ff: render */ + GEN_FW_RANGE(0xe900, 0xe9ff, FORCEWAKE_GT), + GEN_FW_RANGE(0xea00, 0x147ff, 0), /* + 0xea00 - 0x11fff: reserved + 0x12000 - 0x127ff: always on + 0x12800 - 0x147ff: reserved */ + GEN_FW_RANGE(0x14800, 0x19fff, FORCEWAKE_GT), /* + 0x14800 - 0x153ff: gt + 0x15400 - 0x19fff: reserved */ + GEN_FW_RANGE(0x1a000, 0x21fff, FORCEWAKE_RENDER), /* + 0x1a000 - 0x1bfff: render + 0x1c000 - 0x21fff: reserved */ + GEN_FW_RANGE(0x22000, 0x23fff, FORCEWAKE_GT), + GEN_FW_RANGE(0x24000, 0x2ffff, 0), /* + 0x24000 - 0x2407f: always on + 0x24080 - 0x2ffff: reserved */ + GEN_FW_RANGE(0x30000, 0x3ffff, FORCEWAKE_GT), + GEN_FW_RANGE(0x40000, 0x1901ef, 0), + GEN_FW_RANGE(0x1901f0, 0x1901f3, FORCEWAKE_GT) + /* FIXME: WA to wake GT while triggering H2G */ +}; + +/* + * Note that the register ranges here are the final offsets after + * translation of the GSI block to the 0x380000 offset. + * + * NOTE: There are a couple MCR ranges near the bottom of this table + * that need to power up either VD0 or VD2 depending on which replicated + * instance of the register we're trying to access. Our forcewake logic + * at the moment doesn't have a good way to take steering into consideration, + * and the driver doesn't even access any registers in those ranges today, + * so for now we just mark those ranges as FORCEWAKE_ALL. That will ensure + * proper operation if we do start using the ranges in the future, and we + * can determine at that time whether it's worth adding extra complexity to + * the forcewake handling to take steering into consideration. + */ +static const struct intel_forcewake_range __xelpmp_fw_ranges[] = { + GEN_FW_RANGE(0x0, 0x115fff, 0), /* render GT range */ + GEN_FW_RANGE(0x116000, 0x11ffff, FORCEWAKE_GSC), /* + 0x116000 - 0x117fff: gsc + 0x118000 - 0x119fff: reserved + 0x11a000 - 0x11efff: gsc + 0x11f000 - 0x11ffff: reserved */ + GEN_FW_RANGE(0x120000, 0x1bffff, 0), /* non-GT range */ + GEN_FW_RANGE(0x1c0000, 0x1c7fff, FORCEWAKE_MEDIA_VDBOX0), /* + 0x1c0000 - 0x1c3dff: VD0 + 0x1c3e00 - 0x1c3eff: reserved + 0x1c3f00 - 0x1c3fff: VD0 + 0x1c4000 - 0x1c7fff: reserved */ + GEN_FW_RANGE(0x1c8000, 0x1cbfff, FORCEWAKE_MEDIA_VEBOX0), /* + 0x1c8000 - 0x1ca0ff: VE0 + 0x1ca100 - 0x1cbfff: reserved */ + GEN_FW_RANGE(0x1cc000, 0x1cffff, FORCEWAKE_MEDIA_VDBOX0), /* + 0x1cc000 - 0x1cdfff: VD0 + 0x1ce000 - 0x1cffff: reserved */ + GEN_FW_RANGE(0x1d0000, 0x1d7fff, FORCEWAKE_MEDIA_VDBOX2), /* + 0x1d0000 - 0x1d3dff: VD2 + 0x1d3e00 - 0x1d3eff: reserved + 0x1d4000 - 0x1d7fff: VD2 */ + GEN_FW_RANGE(0x1d8000, 0x1da0ff, FORCEWAKE_MEDIA_VEBOX1), + GEN_FW_RANGE(0x1da100, 0x380aff, 0), /* + 0x1da100 - 0x23ffff: reserved + 0x240000 - 0x37ffff: non-GT range + 0x380000 - 0x380aff: reserved */ + GEN_FW_RANGE(0x380b00, 0x380bff, FORCEWAKE_GT), + GEN_FW_RANGE(0x380c00, 0x380fff, 0), + GEN_FW_RANGE(0x381000, 0x38817f, FORCEWAKE_GT), /* + 0x381000 - 0x381fff: gt + 0x382000 - 0x383fff: reserved + 0x384000 - 0x384aff: gt + 0x384b00 - 0x3851ff: reserved + 0x385200 - 0x3871ff: gt + 0x387200 - 0x387fff: reserved + 0x388000 - 0x38813f: gt + 0x388140 - 0x38817f: reserved */ + GEN_FW_RANGE(0x388180, 0x3882ff, 0), /* + 0x388180 - 0x3881ff: always on + 0x388200 - 0x3882ff: reserved */ + GEN_FW_RANGE(0x388300, 0x38955f, FORCEWAKE_GT), /* + 0x388300 - 0x38887f: gt + 0x388880 - 0x388fff: reserved + 0x389000 - 0x38947f: gt + 0x389480 - 0x38955f: reserved */ + GEN_FW_RANGE(0x389560, 0x389fff, 0), /* + 0x389560 - 0x3895ff: always on + 0x389600 - 0x389fff: reserved */ + GEN_FW_RANGE(0x38a000, 0x38cfff, FORCEWAKE_GT), /* + 0x38a000 - 0x38afff: gt + 0x38b000 - 0x38bfff: reserved + 0x38c000 - 0x38cfff: gt */ + GEN_FW_RANGE(0x38d000, 0x38d11f, 0), + GEN_FW_RANGE(0x38d120, 0x391fff, FORCEWAKE_GT), /* + 0x38d120 - 0x38dfff: gt + 0x38e000 - 0x38efff: reserved + 0x38f000 - 0x38ffff: gt + 0x389000 - 0x391fff: reserved */ + GEN_FW_RANGE(0x392000, 0x392fff, 0), /* + 0x392000 - 0x3927ff: always on + 0x392800 - 0x292fff: reserved */ + GEN_FW_RANGE(0x393000, 0x3931ff, FORCEWAKE_GT), + GEN_FW_RANGE(0x393200, 0x39323f, FORCEWAKE_ALL), /* instance-based, see note above */ + GEN_FW_RANGE(0x393240, 0x3933ff, FORCEWAKE_GT), + GEN_FW_RANGE(0x393400, 0x3934ff, FORCEWAKE_ALL), /* instance-based, see note above */ + GEN_FW_RANGE(0x393500, 0x393c7f, 0), /* + 0x393500 - 0x393bff: reserved + 0x393c00 - 0x393c7f: always on */ + GEN_FW_RANGE(0x393c80, 0x393dff, FORCEWAKE_GT), +}; + static void ilk_dummy_write(struct intel_uncore *uncore) { /* WaIssueDummyWriteToWakeupFromRC6:ilk Issue a dummy write to wake up * the chip from rc6 before touching it for real. MI_MODE is masked, * hence harmless to write 0 into. */ - __raw_uncore_write32(uncore, MI_MODE, 0); + __raw_uncore_write32(uncore, RING_MI_MODE(RENDER_RING_BASE), 0); } static void __unclaimed_reg_debug(struct intel_uncore *uncore, const i915_reg_t reg, - const bool read, - const bool before) + const bool read) { if (drm_WARN(&uncore->i915->drm, - check_for_unclaimed_mmio(uncore) && !before, + check_for_unclaimed_mmio(uncore), "Unclaimed %s register 0x%x\n", read ? "read from" : "write to", i915_mmio_reg_offset(reg))) @@ -1273,25 +1775,43 @@ __unclaimed_reg_debug(struct intel_uncore *uncore, uncore->i915->params.mmio_debug--; } -static inline void -unclaimed_reg_debug(struct intel_uncore *uncore, - const i915_reg_t reg, - const bool read, - const bool before) +static void +__unclaimed_previous_reg_debug(struct intel_uncore *uncore, + const i915_reg_t reg, + const bool read) { - if (likely(!uncore->i915->params.mmio_debug)) - return; + if (check_for_unclaimed_mmio(uncore)) + drm_dbg(&uncore->i915->drm, + "Unclaimed access detected before %s register 0x%x\n", + read ? "read from" : "write to", + i915_mmio_reg_offset(reg)); +} + +static inline bool __must_check +unclaimed_reg_debug_header(struct intel_uncore *uncore, + const i915_reg_t reg, const bool read) +{ + if (likely(!uncore->i915->params.mmio_debug) || !uncore->debug) + return false; /* interrupts are disabled and re-enabled around uncore->lock usage */ lockdep_assert_held(&uncore->lock); - if (before) - spin_lock(&uncore->debug->lock); + spin_lock(&uncore->debug->lock); + __unclaimed_previous_reg_debug(uncore, reg, read); - __unclaimed_reg_debug(uncore, reg, read, before); + return true; +} - if (!before) - spin_unlock(&uncore->debug->lock); +static inline void +unclaimed_reg_debug_footer(struct intel_uncore *uncore, + const i915_reg_t reg, const bool read) +{ + /* interrupts are disabled and re-enabled around uncore->lock usage */ + lockdep_assert_held(&uncore->lock); + + __unclaimed_reg_debug(uncore, reg, read); + spin_unlock(&uncore->debug->lock); } #define __vgpu_read(x) \ @@ -1349,13 +1869,15 @@ __gen2_read(64) #define GEN6_READ_HEADER(x) \ u32 offset = i915_mmio_reg_offset(reg); \ unsigned long irqflags; \ + bool unclaimed_reg_debug; \ u##x val = 0; \ assert_rpm_wakelock_held(uncore->rpm); \ spin_lock_irqsave(&uncore->lock, irqflags); \ - unclaimed_reg_debug(uncore, reg, true, true) + unclaimed_reg_debug = unclaimed_reg_debug_header(uncore, reg, true) #define GEN6_READ_FOOTER \ - unclaimed_reg_debug(uncore, reg, true, false); \ + if (unclaimed_reg_debug) \ + unclaimed_reg_debug_footer(uncore, reg, true); \ spin_unlock_irqrestore(&uncore->lock, irqflags); \ trace_i915_reg_rw(false, reg, val, sizeof(val), trace); \ return val @@ -1371,7 +1893,7 @@ static noinline void ___force_wake_auto(struct intel_uncore *uncore, for_each_fw_domain_masked(domain, fw_domains, uncore, tmp) fw_domain_arm_timer(domain); - uncore->funcs.force_wake_get(uncore, fw_domains); + fw_domains_get(uncore, fw_domains); } static inline void __force_wake_auto(struct intel_uncore *uncore, @@ -1387,35 +1909,30 @@ static inline void __force_wake_auto(struct intel_uncore *uncore, ___force_wake_auto(uncore, fw_domains); } -#define __gen_read(func, x) \ +#define __gen_fwtable_read(x) \ static u##x \ -func##_read##x(struct intel_uncore *uncore, i915_reg_t reg, bool trace) { \ +fwtable_read##x(struct intel_uncore *uncore, i915_reg_t reg, bool trace) \ +{ \ enum forcewake_domains fw_engine; \ GEN6_READ_HEADER(x); \ - fw_engine = __##func##_reg_read_fw_domains(uncore, offset); \ + fw_engine = __fwtable_reg_read_fw_domains(uncore, offset); \ if (fw_engine) \ __force_wake_auto(uncore, fw_engine); \ val = __raw_uncore_read##x(uncore, reg); \ GEN6_READ_FOOTER; \ } -#define __gen_reg_read_funcs(func) \ -static enum forcewake_domains \ -func##_reg_read_fw_domains(struct intel_uncore *uncore, i915_reg_t reg) { \ - return __##func##_reg_read_fw_domains(uncore, i915_mmio_reg_offset(reg)); \ -} \ -\ -__gen_read(func, 8) \ -__gen_read(func, 16) \ -__gen_read(func, 32) \ -__gen_read(func, 64) +static enum forcewake_domains +fwtable_reg_read_fw_domains(struct intel_uncore *uncore, i915_reg_t reg) { + return __fwtable_reg_read_fw_domains(uncore, i915_mmio_reg_offset(reg)); +} -__gen_reg_read_funcs(gen12_fwtable); -__gen_reg_read_funcs(gen11_fwtable); -__gen_reg_read_funcs(fwtable); -__gen_reg_read_funcs(gen6); +__gen_fwtable_read(8) +__gen_fwtable_read(16) +__gen_fwtable_read(32) +__gen_fwtable_read(64) -#undef __gen_reg_read_funcs +#undef __gen_fwtable_read #undef GEN6_READ_FOOTER #undef GEN6_READ_HEADER @@ -1458,13 +1975,15 @@ __gen2_write(32) #define GEN6_WRITE_HEADER \ u32 offset = i915_mmio_reg_offset(reg); \ unsigned long irqflags; \ + bool unclaimed_reg_debug; \ trace_i915_reg_rw(true, reg, val, sizeof(val), trace); \ assert_rpm_wakelock_held(uncore->rpm); \ spin_lock_irqsave(&uncore->lock, irqflags); \ - unclaimed_reg_debug(uncore, reg, false, true) + unclaimed_reg_debug = unclaimed_reg_debug_header(uncore, reg, false) #define GEN6_WRITE_FOOTER \ - unclaimed_reg_debug(uncore, reg, false, false); \ + if (unclaimed_reg_debug) \ + unclaimed_reg_debug_footer(uncore, reg, false); \ spin_unlock_irqrestore(&uncore->lock, irqflags) #define __gen6_write(x) \ @@ -1480,34 +1999,29 @@ __gen6_write(8) __gen6_write(16) __gen6_write(32) -#define __gen_write(func, x) \ +#define __gen_fwtable_write(x) \ static void \ -func##_write##x(struct intel_uncore *uncore, i915_reg_t reg, u##x val, bool trace) { \ +fwtable_write##x(struct intel_uncore *uncore, i915_reg_t reg, u##x val, bool trace) { \ enum forcewake_domains fw_engine; \ GEN6_WRITE_HEADER; \ - fw_engine = __##func##_reg_write_fw_domains(uncore, offset); \ + fw_engine = __fwtable_reg_write_fw_domains(uncore, offset); \ if (fw_engine) \ __force_wake_auto(uncore, fw_engine); \ __raw_uncore_write##x(uncore, reg, val); \ GEN6_WRITE_FOOTER; \ } -#define __gen_reg_write_funcs(func) \ -static enum forcewake_domains \ -func##_reg_write_fw_domains(struct intel_uncore *uncore, i915_reg_t reg) { \ - return __##func##_reg_write_fw_domains(uncore, i915_mmio_reg_offset(reg)); \ -} \ -\ -__gen_write(func, 8) \ -__gen_write(func, 16) \ -__gen_write(func, 32) +static enum forcewake_domains +fwtable_reg_write_fw_domains(struct intel_uncore *uncore, i915_reg_t reg) +{ + return __fwtable_reg_write_fw_domains(uncore, i915_mmio_reg_offset(reg)); +} -__gen_reg_write_funcs(gen12_fwtable); -__gen_reg_write_funcs(gen11_fwtable); -__gen_reg_write_funcs(fwtable); -__gen_reg_write_funcs(gen8); +__gen_fwtable_write(8) +__gen_fwtable_write(16) +__gen_fwtable_write(32) -#undef __gen_reg_write_funcs +#undef __gen_fwtable_write #undef GEN6_WRITE_FOOTER #undef GEN6_WRITE_HEADER @@ -1570,8 +2084,8 @@ static int __fw_domain_init(struct intel_uncore *uncore, d->uncore = uncore; d->wake_count = 0; - d->reg_set = uncore->regs + i915_mmio_reg_offset(reg_set); - d->reg_ack = uncore->regs + i915_mmio_reg_offset(reg_ack); + d->reg_set = uncore->regs + i915_mmio_reg_offset(reg_set) + uncore->gsi_offset; + d->reg_ack = uncore->regs + i915_mmio_reg_offset(reg_ack) + uncore->gsi_offset; d->id = domain_id; @@ -1582,13 +2096,19 @@ static int __fw_domain_init(struct intel_uncore *uncore, BUILD_BUG_ON(FORCEWAKE_MEDIA_VDBOX1 != (1 << FW_DOMAIN_ID_MEDIA_VDBOX1)); BUILD_BUG_ON(FORCEWAKE_MEDIA_VDBOX2 != (1 << FW_DOMAIN_ID_MEDIA_VDBOX2)); BUILD_BUG_ON(FORCEWAKE_MEDIA_VDBOX3 != (1 << FW_DOMAIN_ID_MEDIA_VDBOX3)); + BUILD_BUG_ON(FORCEWAKE_MEDIA_VDBOX4 != (1 << FW_DOMAIN_ID_MEDIA_VDBOX4)); + BUILD_BUG_ON(FORCEWAKE_MEDIA_VDBOX5 != (1 << FW_DOMAIN_ID_MEDIA_VDBOX5)); + BUILD_BUG_ON(FORCEWAKE_MEDIA_VDBOX6 != (1 << FW_DOMAIN_ID_MEDIA_VDBOX6)); + BUILD_BUG_ON(FORCEWAKE_MEDIA_VDBOX7 != (1 << FW_DOMAIN_ID_MEDIA_VDBOX7)); BUILD_BUG_ON(FORCEWAKE_MEDIA_VEBOX0 != (1 << FW_DOMAIN_ID_MEDIA_VEBOX0)); BUILD_BUG_ON(FORCEWAKE_MEDIA_VEBOX1 != (1 << FW_DOMAIN_ID_MEDIA_VEBOX1)); + BUILD_BUG_ON(FORCEWAKE_MEDIA_VEBOX2 != (1 << FW_DOMAIN_ID_MEDIA_VEBOX2)); + BUILD_BUG_ON(FORCEWAKE_MEDIA_VEBOX3 != (1 << FW_DOMAIN_ID_MEDIA_VEBOX3)); + BUILD_BUG_ON(FORCEWAKE_GSC != (1 << FW_DOMAIN_ID_GSC)); d->mask = BIT(domain_id); - hrtimer_init(&d->timer, CLOCK_MONOTONIC, HRTIMER_MODE_REL); - d->timer.function = intel_uncore_fw_release_timer; + hrtimer_setup(&d->timer, intel_uncore_fw_release_timer, CLOCK_MONOTONIC, HRTIMER_MODE_REL); uncore->fw_domains |= BIT(domain_id); @@ -1625,6 +2145,18 @@ static void intel_uncore_fw_domains_fini(struct intel_uncore *uncore) fw_domain_fini(uncore, d->id); } +static const struct intel_uncore_fw_get uncore_get_fallback = { + .force_wake_get = fw_domains_get_with_fallback +}; + +static const struct intel_uncore_fw_get uncore_get_normal = { + .force_wake_get = fw_domains_get_normal, +}; + +static const struct intel_uncore_fw_get uncore_get_thread_status = { + .force_wake_get = fw_domains_get_with_thread_status +}; + static int intel_uncore_fw_domains_init(struct intel_uncore *uncore) { struct drm_i915_private *i915 = uncore->i915; @@ -1636,18 +2168,26 @@ static int intel_uncore_fw_domains_init(struct intel_uncore *uncore) (ret ?: (ret = __fw_domain_init((uncore__), (id__), (set__), (ack__)))) if (GRAPHICS_VER(i915) >= 11) { - /* we'll prune the domains of missing engines later */ - intel_engine_mask_t emask = INTEL_INFO(i915)->platform_engine_mask; + intel_engine_mask_t emask; int i; - uncore->funcs.force_wake_get = fw_domains_get_with_fallback; - uncore->funcs.force_wake_put = fw_domains_put; - fw_domain_init(uncore, FW_DOMAIN_ID_RENDER, - FORCEWAKE_RENDER_GEN9, - FORCEWAKE_ACK_RENDER_GEN9); - fw_domain_init(uncore, FW_DOMAIN_ID_GT, - FORCEWAKE_GT_GEN9, - FORCEWAKE_ACK_GT_GEN9); + /* we'll prune the domains of missing engines later */ + emask = uncore->gt->info.engine_mask; + + uncore->fw_get_funcs = &uncore_get_fallback; + if (GRAPHICS_VER_FULL(i915) >= IP_VER(12, 70)) + fw_domain_init(uncore, FW_DOMAIN_ID_GT, + FORCEWAKE_GT_GEN9, + FORCEWAKE_ACK_GT_MTL); + else + fw_domain_init(uncore, FW_DOMAIN_ID_GT, + FORCEWAKE_GT_GEN9, + FORCEWAKE_ACK_GT_GEN9); + + if (RCS_MASK(uncore->gt) || CCS_MASK(uncore->gt)) + fw_domain_init(uncore, FW_DOMAIN_ID_RENDER, + FORCEWAKE_RENDER_GEN9, + FORCEWAKE_ACK_RENDER_GEN9); for (i = 0; i < I915_MAX_VCS; i++) { if (!__HAS_ENGINE(emask, _VCS(i))) @@ -1665,9 +2205,12 @@ static int intel_uncore_fw_domains_init(struct intel_uncore *uncore) FORCEWAKE_MEDIA_VEBOX_GEN11(i), FORCEWAKE_ACK_MEDIA_VEBOX_GEN11(i)); } + + if (uncore->gt->type == GT_MEDIA) + fw_domain_init(uncore, FW_DOMAIN_ID_GSC, + FORCEWAKE_REQ_GSC, FORCEWAKE_ACK_GSC); } else if (IS_GRAPHICS_VER(i915, 9, 10)) { - uncore->funcs.force_wake_get = fw_domains_get_with_fallback; - uncore->funcs.force_wake_put = fw_domains_put; + uncore->fw_get_funcs = &uncore_get_fallback; fw_domain_init(uncore, FW_DOMAIN_ID_RENDER, FORCEWAKE_RENDER_GEN9, FORCEWAKE_ACK_RENDER_GEN9); @@ -1677,16 +2220,13 @@ static int intel_uncore_fw_domains_init(struct intel_uncore *uncore) fw_domain_init(uncore, FW_DOMAIN_ID_MEDIA, FORCEWAKE_MEDIA_GEN9, FORCEWAKE_ACK_MEDIA_GEN9); } else if (IS_VALLEYVIEW(i915) || IS_CHERRYVIEW(i915)) { - uncore->funcs.force_wake_get = fw_domains_get; - uncore->funcs.force_wake_put = fw_domains_put; + uncore->fw_get_funcs = &uncore_get_normal; fw_domain_init(uncore, FW_DOMAIN_ID_RENDER, FORCEWAKE_VLV, FORCEWAKE_ACK_VLV); fw_domain_init(uncore, FW_DOMAIN_ID_MEDIA, FORCEWAKE_MEDIA_VLV, FORCEWAKE_ACK_MEDIA_VLV); } else if (IS_HASWELL(i915) || IS_BROADWELL(i915)) { - uncore->funcs.force_wake_get = - fw_domains_get_with_thread_status; - uncore->funcs.force_wake_put = fw_domains_put; + uncore->fw_get_funcs = &uncore_get_thread_status; fw_domain_init(uncore, FW_DOMAIN_ID_RENDER, FORCEWAKE_MT, FORCEWAKE_ACK_HSW); } else if (IS_IVYBRIDGE(i915)) { @@ -1701,9 +2241,7 @@ static int intel_uncore_fw_domains_init(struct intel_uncore *uncore) * (correctly) interpreted by the test below as MT * forcewake being disabled. */ - uncore->funcs.force_wake_get = - fw_domains_get_with_thread_status; - uncore->funcs.force_wake_put = fw_domains_put; + uncore->fw_get_funcs = &uncore_get_thread_status; /* We need to init first for ECOBUS access and then * determine later if we want to reinit, in case of MT access is @@ -1734,9 +2272,7 @@ static int intel_uncore_fw_domains_init(struct intel_uncore *uncore) FORCEWAKE, FORCEWAKE_ACK); } } else if (GRAPHICS_VER(i915) == 6) { - uncore->funcs.force_wake_get = - fw_domains_get_with_thread_status; - uncore->funcs.force_wake_put = fw_domains_put; + uncore->fw_get_funcs = &uncore_get_thread_status; fw_domain_init(uncore, FW_DOMAIN_ID_RENDER, FORCEWAKE, FORCEWAKE_ACK); } @@ -1760,6 +2296,12 @@ out: (uncore)->fw_domains_table_entries = ARRAY_SIZE((d)); \ } +#define ASSIGN_SHADOW_TABLE(uncore, d) \ +{ \ + (uncore)->shadowed_reg_table = d; \ + (uncore)->shadowed_reg_table_entries = ARRAY_SIZE((d)); \ +} + static int i915_pmic_bus_access_notifier(struct notifier_block *nb, unsigned long action, void *data) { @@ -1793,14 +2335,16 @@ static int i915_pmic_bus_access_notifier(struct notifier_block *nb, return NOTIFY_OK; } -static int uncore_mmio_setup(struct intel_uncore *uncore) +static void uncore_unmap_mmio(struct drm_device *drm, void *regs) +{ + iounmap((void __iomem *)regs); +} + +int intel_uncore_setup_mmio(struct intel_uncore *uncore, phys_addr_t phys_addr) { struct drm_i915_private *i915 = uncore->i915; - struct pci_dev *pdev = to_pci_dev(i915->drm.dev); - int mmio_bar; int mmio_size; - mmio_bar = GRAPHICS_VER(i915) == 2 ? 1 : 0; /* * Before gen4, the registers and the GTT are behind different BARs. * However, from gen4 onwards, the registers and the GTT are shared @@ -1808,38 +2352,33 @@ static int uncore_mmio_setup(struct intel_uncore *uncore) * clobbering the GTT which we want ioremap_wc instead. Fortunately, * the register BAR remains the same size for all the earlier * generations up to Ironlake. - * For dgfx chips register range is expanded to 4MB. + * For dgfx chips register range is expanded to 4MB, and this larger + * range is also used for integrated gpus beginning with Meteor Lake. */ - if (GRAPHICS_VER(i915) < 5) - mmio_size = 512 * 1024; - else if (IS_DGFX(i915)) + if (IS_DGFX(i915) || GRAPHICS_VER_FULL(i915) >= IP_VER(12, 70)) mmio_size = 4 * 1024 * 1024; - else + else if (GRAPHICS_VER(i915) >= 5) mmio_size = 2 * 1024 * 1024; + else + mmio_size = 512 * 1024; - uncore->regs = pci_iomap(pdev, mmio_bar, mmio_size); + uncore->regs = ioremap(phys_addr, mmio_size); if (uncore->regs == NULL) { drm_err(&i915->drm, "failed to map registers\n"); return -EIO; } - return 0; -} - -static void uncore_mmio_cleanup(struct intel_uncore *uncore) -{ - struct pci_dev *pdev = to_pci_dev(uncore->i915->drm.dev); - - pci_iounmap(pdev, uncore->regs); + return drmm_add_action_or_reset(&i915->drm, uncore_unmap_mmio, + (void __force *)uncore->regs); } void intel_uncore_init_early(struct intel_uncore *uncore, - struct drm_i915_private *i915) + struct intel_gt *gt) { spin_lock_init(&uncore->lock); - uncore->i915 = i915; - uncore->rpm = &i915->runtime_pm; - uncore->debug = &i915->mmio_debug; + uncore->i915 = gt->i915; + uncore->gt = gt; + uncore->rpm = >->i915->runtime_pm; } static void uncore_raw_init(struct intel_uncore *uncore) @@ -1858,6 +2397,22 @@ static void uncore_raw_init(struct intel_uncore *uncore) } } +static int uncore_media_forcewake_init(struct intel_uncore *uncore) +{ + struct drm_i915_private *i915 = uncore->i915; + + if (MEDIA_VER(i915) >= 13) { + ASSIGN_FW_DOMAINS_TABLE(uncore, __xelpmp_fw_ranges); + ASSIGN_SHADOW_TABLE(uncore, xelpmp_shadowed_regs); + ASSIGN_WRITE_MMIO_VFUNCS(uncore, fwtable); + } else { + MISSING_CASE(MEDIA_VER(i915)); + return -ENODEV; + } + + return 0; +} + static int uncore_forcewake_init(struct intel_uncore *uncore) { struct drm_i915_private *i915 = uncore->i915; @@ -1870,36 +2425,45 @@ static int uncore_forcewake_init(struct intel_uncore *uncore) return ret; forcewake_early_sanitize(uncore, 0); - if (IS_GRAPHICS_VER(i915, 6, 7)) { - ASSIGN_WRITE_MMIO_VFUNCS(uncore, gen6); + ASSIGN_READ_MMIO_VFUNCS(uncore, fwtable); - if (IS_VALLEYVIEW(i915)) { - ASSIGN_FW_DOMAINS_TABLE(uncore, __vlv_fw_ranges); - ASSIGN_READ_MMIO_VFUNCS(uncore, fwtable); - } else { - ASSIGN_READ_MMIO_VFUNCS(uncore, gen6); - } - } else if (GRAPHICS_VER(i915) == 8) { - if (IS_CHERRYVIEW(i915)) { - ASSIGN_FW_DOMAINS_TABLE(uncore, __chv_fw_ranges); - ASSIGN_WRITE_MMIO_VFUNCS(uncore, fwtable); - ASSIGN_READ_MMIO_VFUNCS(uncore, fwtable); - } else { - ASSIGN_WRITE_MMIO_VFUNCS(uncore, gen8); - ASSIGN_READ_MMIO_VFUNCS(uncore, gen6); - } - } else if (IS_GRAPHICS_VER(i915, 9, 10)) { - ASSIGN_FW_DOMAINS_TABLE(uncore, __gen9_fw_ranges); + if (uncore->gt->type == GT_MEDIA) + return uncore_media_forcewake_init(uncore); + + if (GRAPHICS_VER_FULL(i915) >= IP_VER(12, 70)) { + ASSIGN_FW_DOMAINS_TABLE(uncore, __mtl_fw_ranges); + ASSIGN_SHADOW_TABLE(uncore, mtl_shadowed_regs); + ASSIGN_WRITE_MMIO_VFUNCS(uncore, fwtable); + } else if (GRAPHICS_VER_FULL(i915) >= IP_VER(12, 55)) { + ASSIGN_FW_DOMAINS_TABLE(uncore, __dg2_fw_ranges); + ASSIGN_SHADOW_TABLE(uncore, dg2_shadowed_regs); + ASSIGN_WRITE_MMIO_VFUNCS(uncore, fwtable); + } else if (GRAPHICS_VER(i915) >= 12) { + ASSIGN_FW_DOMAINS_TABLE(uncore, __gen12_fw_ranges); + ASSIGN_SHADOW_TABLE(uncore, gen12_shadowed_regs); ASSIGN_WRITE_MMIO_VFUNCS(uncore, fwtable); - ASSIGN_READ_MMIO_VFUNCS(uncore, fwtable); } else if (GRAPHICS_VER(i915) == 11) { ASSIGN_FW_DOMAINS_TABLE(uncore, __gen11_fw_ranges); - ASSIGN_WRITE_MMIO_VFUNCS(uncore, gen11_fwtable); - ASSIGN_READ_MMIO_VFUNCS(uncore, gen11_fwtable); - } else { - ASSIGN_FW_DOMAINS_TABLE(uncore, __gen12_fw_ranges); - ASSIGN_WRITE_MMIO_VFUNCS(uncore, gen12_fwtable); - ASSIGN_READ_MMIO_VFUNCS(uncore, gen12_fwtable); + ASSIGN_SHADOW_TABLE(uncore, gen11_shadowed_regs); + ASSIGN_WRITE_MMIO_VFUNCS(uncore, fwtable); + } else if (IS_GRAPHICS_VER(i915, 9, 10)) { + ASSIGN_FW_DOMAINS_TABLE(uncore, __gen9_fw_ranges); + ASSIGN_SHADOW_TABLE(uncore, gen8_shadowed_regs); + ASSIGN_WRITE_MMIO_VFUNCS(uncore, fwtable); + } else if (IS_CHERRYVIEW(i915)) { + ASSIGN_FW_DOMAINS_TABLE(uncore, __chv_fw_ranges); + ASSIGN_SHADOW_TABLE(uncore, gen8_shadowed_regs); + ASSIGN_WRITE_MMIO_VFUNCS(uncore, fwtable); + } else if (GRAPHICS_VER(i915) == 8) { + ASSIGN_FW_DOMAINS_TABLE(uncore, __gen6_fw_ranges); + ASSIGN_SHADOW_TABLE(uncore, gen8_shadowed_regs); + ASSIGN_WRITE_MMIO_VFUNCS(uncore, fwtable); + } else if (IS_VALLEYVIEW(i915)) { + ASSIGN_FW_DOMAINS_TABLE(uncore, __vlv_fw_ranges); + ASSIGN_WRITE_MMIO_VFUNCS(uncore, gen6); + } else if (IS_GRAPHICS_VER(i915, 6, 7)) { + ASSIGN_FW_DOMAINS_TABLE(uncore, __gen6_fw_ranges); + ASSIGN_WRITE_MMIO_VFUNCS(uncore, gen6); } uncore->pmic_bus_access_nb.notifier_call = i915_pmic_bus_access_notifier; @@ -1908,12 +2472,43 @@ static int uncore_forcewake_init(struct intel_uncore *uncore) return 0; } +static int sanity_check_mmio_access(struct intel_uncore *uncore) +{ + struct drm_i915_private *i915 = uncore->i915; + + if (GRAPHICS_VER(i915) < 8) + return 0; + + /* + * Sanitycheck that MMIO access to the device is working properly. If + * the CPU is unable to communicate with a PCI device, BAR reads will + * return 0xFFFFFFFF. Let's make sure the device isn't in this state + * before we start trying to access registers. + * + * We use the primary GT's forcewake register as our guinea pig since + * it's been around since HSW and it's a masked register so the upper + * 16 bits can never read back as 1's if device access is operating + * properly. + * + * If MMIO isn't working, we'll wait up to 2 seconds to see if it + * recovers, then give up. + */ +#define COND (__raw_uncore_read32(uncore, FORCEWAKE_MT) != ~0) + if (wait_for(COND, 2000) == -ETIMEDOUT) { + drm_err(&i915->drm, "Device is non-operational; MMIO access returns 0xFFFFFFFF!\n"); + return -EIO; + } + + return 0; +} + int intel_uncore_init_mmio(struct intel_uncore *uncore) { struct drm_i915_private *i915 = uncore->i915; + struct intel_display *display = i915->display; int ret; - ret = uncore_mmio_setup(uncore); + ret = sanity_check_mmio_access(uncore); if (ret) return ret; @@ -1929,7 +2524,7 @@ int intel_uncore_init_mmio(struct intel_uncore *uncore) return -ENODEV; } - if (INTEL_GEN(i915) > 5 && !intel_vgpu_active(i915)) + if (GRAPHICS_VER(i915) > 5 && !intel_vgpu_active(i915)) uncore->flags |= UNCORE_HAS_FORCEWAKE; if (!intel_uncore_has_forcewake(uncore)) { @@ -1937,16 +2532,15 @@ int intel_uncore_init_mmio(struct intel_uncore *uncore) } else { ret = uncore_forcewake_init(uncore); if (ret) - goto out_mmio_cleanup; + return ret; } /* make sure fw funcs are set if and only if we have fw*/ - GEM_BUG_ON(intel_uncore_has_forcewake(uncore) != !!uncore->funcs.force_wake_get); - GEM_BUG_ON(intel_uncore_has_forcewake(uncore) != !!uncore->funcs.force_wake_put); + GEM_BUG_ON(intel_uncore_has_forcewake(uncore) != !!uncore->fw_get_funcs); GEM_BUG_ON(intel_uncore_has_forcewake(uncore) != !!uncore->funcs.read_fw_domains); GEM_BUG_ON(intel_uncore_has_forcewake(uncore) != !!uncore->funcs.write_fw_domains); - if (HAS_FPGA_DBG_UNCLAIMED(i915)) + if (HAS_FPGA_DBG_UNCLAIMED(display)) uncore->flags |= UNCORE_HAS_FPGA_DBG_UNCLAIMED; if (IS_VALLEYVIEW(i915) || IS_CHERRYVIEW(i915)) @@ -1960,11 +2554,6 @@ int intel_uncore_init_mmio(struct intel_uncore *uncore) drm_dbg(&i915->drm, "unclaimed mmio detected on uncore init, clearing\n"); return 0; - -out_mmio_cleanup: - uncore_mmio_cleanup(uncore); - - return ret; } /* @@ -1988,6 +2577,22 @@ void intel_uncore_prune_engine_fw_domains(struct intel_uncore *uncore, if (HAS_ENGINE(gt, _VCS(i))) continue; + /* + * Starting with XeHP, the power well for an even-numbered + * VDBOX is also used for shared units within the + * media slice such as SFC. So even if the engine + * itself is fused off, we still need to initialize + * the forcewake domain if any of the other engines + * in the same media slice are present. + */ + if (GRAPHICS_VER_FULL(uncore->i915) >= IP_VER(12, 55) && i % 2 == 0) { + if ((i + 1 < I915_MAX_VCS) && HAS_ENGINE(gt, _VCS(i + 1))) + continue; + + if (HAS_ENGINE(gt, _VECS(i / 2))) + continue; + } + if (fw_domains & BIT(domain_id)) fw_domain_fini(uncore, domain_id); } @@ -2001,10 +2606,87 @@ void intel_uncore_prune_engine_fw_domains(struct intel_uncore *uncore, if (fw_domains & BIT(domain_id)) fw_domain_fini(uncore, domain_id); } + + if ((fw_domains & BIT(FW_DOMAIN_ID_GSC)) && !HAS_ENGINE(gt, GSC0)) + fw_domain_fini(uncore, FW_DOMAIN_ID_GSC); +} + +/* + * The driver-initiated FLR is the highest level of reset that we can trigger + * from within the driver. It is different from the PCI FLR in that it doesn't + * fully reset the SGUnit and doesn't modify the PCI config space and therefore + * it doesn't require a re-enumeration of the PCI BARs. However, the + * driver-initiated FLR does still cause a reset of both GT and display and a + * memory wipe of local and stolen memory, so recovery would require a full HW + * re-init and saving/restoring (or re-populating) the wiped memory. Since we + * perform the FLR as the very last action before releasing access to the HW + * during the driver release flow, we don't attempt recovery at all, because + * if/when a new instance of i915 is bound to the device it will do a full + * re-init anyway. + */ +static void driver_initiated_flr(struct intel_uncore *uncore) +{ + struct drm_i915_private *i915 = uncore->i915; + unsigned int flr_timeout_ms; + int ret; + + drm_dbg(&i915->drm, "Triggering Driver-FLR\n"); + + /* + * The specification recommends a 3 seconds FLR reset timeout. To be + * cautious, we will extend this to 9 seconds, three times the specified + * timeout. + */ + flr_timeout_ms = 9000; + + /* + * Make sure any pending FLR requests have cleared by waiting for the + * FLR trigger bit to go to zero. Also clear GU_DEBUG's DRIVERFLR_STATUS + * to make sure it's not still set from a prior attempt (it's a write to + * clear bit). + * Note that we should never be in a situation where a previous attempt + * is still pending (unless the HW is totally dead), but better to be + * safe in case something unexpected happens + */ + ret = intel_wait_for_register_fw(uncore, GU_CNTL, DRIVERFLR, 0, flr_timeout_ms, NULL); + if (ret) { + drm_err(&i915->drm, + "Failed to wait for Driver-FLR bit to clear! %d\n", + ret); + return; + } + intel_uncore_write_fw(uncore, GU_DEBUG, DRIVERFLR_STATUS); + + /* Trigger the actual Driver-FLR */ + intel_uncore_rmw_fw(uncore, GU_CNTL, 0, DRIVERFLR); + + /* Wait for hardware teardown to complete */ + ret = intel_wait_for_register_fw(uncore, GU_CNTL, + DRIVERFLR, 0, + flr_timeout_ms, NULL); + if (ret) { + drm_err(&i915->drm, "Driver-FLR-teardown wait completion failed! %d\n", ret); + return; + } + + /* Wait for hardware/firmware re-init to complete */ + ret = intel_wait_for_register_fw(uncore, GU_DEBUG, + DRIVERFLR_STATUS, DRIVERFLR_STATUS, + flr_timeout_ms, NULL); + if (ret) { + drm_err(&i915->drm, "Driver-FLR-reinit wait completion failed! %d\n", ret); + return; + } + + /* Clear sticky completion status */ + intel_uncore_write_fw(uncore, GU_DEBUG, DRIVERFLR_STATUS); } -void intel_uncore_fini_mmio(struct intel_uncore *uncore) +/* Called via drm-managed action */ +void intel_uncore_fini_mmio(struct drm_device *dev, void *data) { + struct intel_uncore *uncore = data; + if (intel_uncore_has_forcewake(uncore)) { iosf_mbi_punit_acquire(); iosf_mbi_unregister_pmic_bus_access_notifier_unlocked( @@ -2014,77 +2696,8 @@ void intel_uncore_fini_mmio(struct intel_uncore *uncore) iosf_mbi_punit_release(); } - uncore_mmio_cleanup(uncore); -} - -static const struct reg_whitelist { - i915_reg_t offset_ldw; - i915_reg_t offset_udw; - u8 min_graphics_ver; - u8 max_graphics_ver; - u8 size; -} reg_read_whitelist[] = { { - .offset_ldw = RING_TIMESTAMP(RENDER_RING_BASE), - .offset_udw = RING_TIMESTAMP_UDW(RENDER_RING_BASE), - .min_graphics_ver = 4, - .max_graphics_ver = 12, - .size = 8 -} }; - -int i915_reg_read_ioctl(struct drm_device *dev, - void *data, struct drm_file *file) -{ - struct drm_i915_private *i915 = to_i915(dev); - struct intel_uncore *uncore = &i915->uncore; - struct drm_i915_reg_read *reg = data; - struct reg_whitelist const *entry; - intel_wakeref_t wakeref; - unsigned int flags; - int remain; - int ret = 0; - - entry = reg_read_whitelist; - remain = ARRAY_SIZE(reg_read_whitelist); - while (remain) { - u32 entry_offset = i915_mmio_reg_offset(entry->offset_ldw); - - GEM_BUG_ON(!is_power_of_2(entry->size)); - GEM_BUG_ON(entry->size > 8); - GEM_BUG_ON(entry_offset & (entry->size - 1)); - - if (IS_GRAPHICS_VER(i915, entry->min_graphics_ver, entry->max_graphics_ver) && - entry_offset == (reg->offset & -entry->size)) - break; - entry++; - remain--; - } - - if (!remain) - return -EINVAL; - - flags = reg->offset & (entry->size - 1); - - with_intel_runtime_pm(&i915->runtime_pm, wakeref) { - if (entry->size == 8 && flags == I915_REG_READ_8B_WA) - reg->val = intel_uncore_read64_2x32(uncore, - entry->offset_ldw, - entry->offset_udw); - else if (entry->size == 8 && flags == 0) - reg->val = intel_uncore_read64(uncore, - entry->offset_ldw); - else if (entry->size == 4 && flags == 0) - reg->val = intel_uncore_read(uncore, entry->offset_ldw); - else if (entry->size == 2 && flags == 0) - reg->val = intel_uncore_read16(uncore, - entry->offset_ldw); - else if (entry->size == 1 && flags == 0) - reg->val = intel_uncore_read8(uncore, - entry->offset_ldw); - else - ret = -EINVAL; - } - - return ret; + if (intel_uncore_needs_flr_on_fini(uncore)) + driver_initiated_flr(uncore); } /** @@ -2206,6 +2819,9 @@ bool intel_uncore_unclaimed_mmio(struct intel_uncore *uncore) { bool ret; + if (!uncore->debug) + return false; + spin_lock_irq(&uncore->debug->lock); ret = check_for_unclaimed_mmio(uncore); spin_unlock_irq(&uncore->debug->lock); @@ -2218,6 +2834,9 @@ intel_uncore_arm_unclaimed_mmio_detection(struct intel_uncore *uncore) { bool ret = false; + if (drm_WARN_ON(&uncore->i915->drm, !uncore->debug)) + return false; + spin_lock_irq(&uncore->debug->lock); if (unlikely(uncore->debug->unclaimed_mmio_check <= 0)) |