From f01db988ef6f6c70a6cc36ee71e4a98a68901229 Mon Sep 17 00:00:00 2001 From: Sean Paul Date: Fri, 16 Mar 2012 12:43:22 -0400 Subject: drm/i915: Add wait_for in init_ring_common I have seen a number of "blt ring initialization failed" messages where the ctl or start registers are not the correct value. Upon further inspection, if the code just waited a little bit, it would read the correct value. Adding the wait_for to these reads should eliminate the issue. Signed-off-by: Sean Paul Reviewed-by: Ben Widawsky Signed-off-by: Daniel Vetter --- drivers/gpu/drm/i915/intel_ringbuffer.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) (limited to 'drivers/gpu/drm/i915/intel_ringbuffer.c') diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.c b/drivers/gpu/drm/i915/intel_ringbuffer.c index ca3972f2c6f..b7b50a5b047 100644 --- a/drivers/gpu/drm/i915/intel_ringbuffer.c +++ b/drivers/gpu/drm/i915/intel_ringbuffer.c @@ -290,9 +290,9 @@ static int init_ring_common(struct intel_ring_buffer *ring) | RING_REPORT_64K | RING_VALID); /* If the head is still not zero, the ring is dead */ - if ((I915_READ_CTL(ring) & RING_VALID) == 0 || - I915_READ_START(ring) != obj->gtt_offset || - (I915_READ_HEAD(ring) & HEAD_ADDR) != 0) { + if (wait_for((I915_READ_CTL(ring) & RING_VALID) != 0 && + I915_READ_START(ring) == obj->gtt_offset && + (I915_READ_HEAD(ring) & HEAD_ADDR) == 0, 50)) { DRM_ERROR("%s initialization failed " "ctl %08x head %08x tail %08x start %08x\n", ring->name, -- cgit v1.2.3-70-g09d2 From 7e231dbe0c2a4359472f929732d4200479d8f939 Mon Sep 17 00:00:00 2001 From: Jesse Barnes Date: Wed, 28 Mar 2012 13:39:38 -0700 Subject: drm/i915: ValleyView IRQ support ValleyView has a new interrupt architecture; best to put it in a new set of functions. Also make sure the ring mask functions handle ValleyView. FIXME: fix flipping; need to enable interrupts and call prepare/finish Signed-off-by: Jesse Barnes Signed-off-by: Daniel Vetter --- drivers/gpu/drm/i915/i915_debugfs.c | 40 +++- drivers/gpu/drm/i915/i915_irq.c | 338 +++++++++++++++++++++++++++++++- drivers/gpu/drm/i915/i915_reg.h | 7 +- drivers/gpu/drm/i915/intel_ringbuffer.c | 4 +- 4 files changed, 383 insertions(+), 6 deletions(-) (limited to 'drivers/gpu/drm/i915/intel_ringbuffer.c') diff --git a/drivers/gpu/drm/i915/i915_debugfs.c b/drivers/gpu/drm/i915/i915_debugfs.c index e74674b3097..d226f2f2f7b 100644 --- a/drivers/gpu/drm/i915/i915_debugfs.c +++ b/drivers/gpu/drm/i915/i915_debugfs.c @@ -468,7 +468,45 @@ static int i915_interrupt_info(struct seq_file *m, void *data) if (ret) return ret; - if (!HAS_PCH_SPLIT(dev)) { + if (IS_VALLEYVIEW(dev)) { + seq_printf(m, "Display IER:\t%08x\n", + I915_READ(VLV_IER)); + seq_printf(m, "Display IIR:\t%08x\n", + I915_READ(VLV_IIR)); + seq_printf(m, "Display IIR_RW:\t%08x\n", + I915_READ(VLV_IIR_RW)); + seq_printf(m, "Display IMR:\t%08x\n", + I915_READ(VLV_IMR)); + for_each_pipe(pipe) + seq_printf(m, "Pipe %c stat:\t%08x\n", + pipe_name(pipe), + I915_READ(PIPESTAT(pipe))); + + seq_printf(m, "Master IER:\t%08x\n", + I915_READ(VLV_MASTER_IER)); + + seq_printf(m, "Render IER:\t%08x\n", + I915_READ(GTIER)); + seq_printf(m, "Render IIR:\t%08x\n", + I915_READ(GTIIR)); + seq_printf(m, "Render IMR:\t%08x\n", + I915_READ(GTIMR)); + + seq_printf(m, "PM IER:\t\t%08x\n", + I915_READ(GEN6_PMIER)); + seq_printf(m, "PM IIR:\t\t%08x\n", + I915_READ(GEN6_PMIIR)); + seq_printf(m, "PM IMR:\t\t%08x\n", + I915_READ(GEN6_PMIMR)); + + seq_printf(m, "Port hotplug:\t%08x\n", + I915_READ(PORT_HOTPLUG_EN)); + seq_printf(m, "DPFLIPSTAT:\t%08x\n", + I915_READ(VLV_DPFLIPSTAT)); + seq_printf(m, "DPINVGTT:\t%08x\n", + I915_READ(DPINVGTT)); + + } else if (!HAS_PCH_SPLIT(dev)) { seq_printf(m, "Interrupt enable: %08x\n", I915_READ(IER)); seq_printf(m, "Interrupt identity: %08x\n", diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c index 998116e871f..01fb6509797 100644 --- a/drivers/gpu/drm/i915/i915_irq.c +++ b/drivers/gpu/drm/i915/i915_irq.c @@ -120,6 +120,10 @@ void intel_enable_asle(struct drm_device *dev) drm_i915_private_t *dev_priv = dev->dev_private; unsigned long irqflags; + /* FIXME: opregion/asle for VLV */ + if (IS_VALLEYVIEW(dev)) + return; + spin_lock_irqsave(&dev_priv->irq_lock, irqflags); if (HAS_PCH_SPLIT(dev)) @@ -426,6 +430,119 @@ static void gen6_pm_rps_work(struct work_struct *work) mutex_unlock(&dev_priv->dev->struct_mutex); } +static irqreturn_t valleyview_irq_handler(DRM_IRQ_ARGS) +{ + struct drm_device *dev = (struct drm_device *) arg; + drm_i915_private_t *dev_priv = (drm_i915_private_t *) dev->dev_private; + u32 iir, gt_iir, pm_iir; + irqreturn_t ret = IRQ_NONE; + unsigned long irqflags; + int pipe; + u32 pipe_stats[I915_MAX_PIPES]; + u32 vblank_status; + int vblank = 0; + bool blc_event; + + atomic_inc(&dev_priv->irq_received); + + vblank_status = PIPE_START_VBLANK_INTERRUPT_STATUS | + PIPE_VBLANK_INTERRUPT_STATUS; + + while (true) { + iir = I915_READ(VLV_IIR); + gt_iir = I915_READ(GTIIR); + pm_iir = I915_READ(GEN6_PMIIR); + + if (gt_iir == 0 && pm_iir == 0 && iir == 0) + goto out; + + ret = IRQ_HANDLED; + + if (gt_iir & (GT_USER_INTERRUPT | GT_PIPE_NOTIFY)) + notify_ring(dev, &dev_priv->ring[RCS]); + if (gt_iir & GT_GEN6_BSD_USER_INTERRUPT) + notify_ring(dev, &dev_priv->ring[VCS]); + if (gt_iir & GT_BLT_USER_INTERRUPT) + notify_ring(dev, &dev_priv->ring[BCS]); + + if (gt_iir & (GT_GEN6_BLT_CS_ERROR_INTERRUPT | + GT_GEN6_BSD_CS_ERROR_INTERRUPT | + GT_RENDER_CS_ERROR_INTERRUPT)) { + DRM_ERROR("GT error interrupt 0x%08x\n", gt_iir); + i915_handle_error(dev, false); + } + + spin_lock_irqsave(&dev_priv->irq_lock, irqflags); + for_each_pipe(pipe) { + int reg = PIPESTAT(pipe); + pipe_stats[pipe] = I915_READ(reg); + + /* + * Clear the PIPE*STAT regs before the IIR + */ + if (pipe_stats[pipe] & 0x8000ffff) { + if (pipe_stats[pipe] & PIPE_FIFO_UNDERRUN_STATUS) + DRM_DEBUG_DRIVER("pipe %c underrun\n", + pipe_name(pipe)); + I915_WRITE(reg, pipe_stats[pipe]); + } + } + spin_unlock_irqrestore(&dev_priv->irq_lock, irqflags); + + /* Consume port. Then clear IIR or we'll miss events */ + if (iir & I915_DISPLAY_PORT_INTERRUPT) { + u32 hotplug_status = I915_READ(PORT_HOTPLUG_STAT); + + DRM_DEBUG_DRIVER("hotplug event received, stat 0x%08x\n", + hotplug_status); + if (hotplug_status & dev_priv->hotplug_supported_mask) + queue_work(dev_priv->wq, + &dev_priv->hotplug_work); + + I915_WRITE(PORT_HOTPLUG_STAT, hotplug_status); + I915_READ(PORT_HOTPLUG_STAT); + } + + + if (iir & I915_DISPLAY_PIPE_A_VBLANK_INTERRUPT) { + drm_handle_vblank(dev, 0); + vblank++; + if (!dev_priv->flip_pending_is_done) { + intel_finish_page_flip(dev, 0); + } + } + + if (iir & I915_DISPLAY_PIPE_B_VBLANK_INTERRUPT) { + drm_handle_vblank(dev, 1); + vblank++; + if (!dev_priv->flip_pending_is_done) { + intel_finish_page_flip(dev, 0); + } + } + + if (pipe_stats[pipe] & PIPE_LEGACY_BLC_EVENT_STATUS) + blc_event = true; + + if (pm_iir & GEN6_PM_DEFERRED_EVENTS) { + unsigned long flags; + spin_lock_irqsave(&dev_priv->rps_lock, flags); + WARN(dev_priv->pm_iir & pm_iir, "Missed a PM interrupt\n"); + dev_priv->pm_iir |= pm_iir; + I915_WRITE(GEN6_PMIMR, dev_priv->pm_iir); + POSTING_READ(GEN6_PMIMR); + spin_unlock_irqrestore(&dev_priv->rps_lock, flags); + queue_work(dev_priv->wq, &dev_priv->rps_work); + } + + I915_WRITE(GTIIR, gt_iir); + I915_WRITE(GEN6_PMIIR, pm_iir); + I915_WRITE(VLV_IIR, iir); + } + +out: + return ret; +} + static void pch_irq_handler(struct drm_device *dev) { drm_i915_private_t *dev_priv = (drm_i915_private_t *) dev->dev_private; @@ -1567,6 +1684,32 @@ static int ivybridge_enable_vblank(struct drm_device *dev, int pipe) return 0; } +static int valleyview_enable_vblank(struct drm_device *dev, int pipe) +{ + drm_i915_private_t *dev_priv = (drm_i915_private_t *) dev->dev_private; + unsigned long irqflags; + u32 dpfl, imr; + + if (!i915_pipe_enabled(dev, pipe)) + return -EINVAL; + + spin_lock_irqsave(&dev_priv->irq_lock, irqflags); + dpfl = I915_READ(VLV_DPFLIPSTAT); + imr = I915_READ(VLV_IMR); + if (pipe == 0) { + dpfl |= PIPEA_VBLANK_INT_EN; + imr &= ~I915_DISPLAY_PIPE_A_VBLANK_INTERRUPT; + } else { + dpfl |= PIPEA_VBLANK_INT_EN; + imr &= ~I915_DISPLAY_PIPE_B_VBLANK_INTERRUPT; + } + I915_WRITE(VLV_DPFLIPSTAT, dpfl); + I915_WRITE(VLV_IMR, imr); + spin_unlock_irqrestore(&dev_priv->irq_lock, irqflags); + + return 0; +} + /* Called from drm generic code, passed 'crtc' which * we use as a pipe index */ @@ -1608,6 +1751,28 @@ static void ivybridge_disable_vblank(struct drm_device *dev, int pipe) spin_unlock_irqrestore(&dev_priv->irq_lock, irqflags); } +static void valleyview_disable_vblank(struct drm_device *dev, int pipe) +{ + drm_i915_private_t *dev_priv = (drm_i915_private_t *) dev->dev_private; + unsigned long irqflags; + u32 dpfl, imr; + + spin_lock_irqsave(&dev_priv->irq_lock, irqflags); + dpfl = I915_READ(VLV_DPFLIPSTAT); + imr = I915_READ(VLV_IMR); + if (pipe == 0) { + dpfl &= ~PIPEA_VBLANK_INT_EN; + imr |= I915_DISPLAY_PIPE_A_VBLANK_INTERRUPT; + } else { + dpfl &= ~PIPEB_VBLANK_INT_EN; + imr |= I915_DISPLAY_PIPE_B_VBLANK_INTERRUPT; + } + I915_WRITE(VLV_IMR, imr); + I915_WRITE(VLV_DPFLIPSTAT, dpfl); + spin_unlock_irqrestore(&dev_priv->irq_lock, irqflags); +} + + /* Set the vblank monitor pipe */ int i915_vblank_pipe_set(struct drm_device *dev, void *data, @@ -1817,6 +1982,53 @@ static void ironlake_irq_preinstall(struct drm_device *dev) POSTING_READ(SDEIER); } +static void valleyview_irq_preinstall(struct drm_device *dev) +{ + drm_i915_private_t *dev_priv = (drm_i915_private_t *) dev->dev_private; + int pipe; + + atomic_set(&dev_priv->irq_received, 0); + + INIT_WORK(&dev_priv->hotplug_work, i915_hotplug_work_func); + INIT_WORK(&dev_priv->error_work, i915_error_work_func); + + /* VLV magic */ + I915_WRITE(VLV_IMR, 0); + I915_WRITE(RING_IMR(RENDER_RING_BASE), 0); + I915_WRITE(RING_IMR(GEN6_BSD_RING_BASE), 0); + I915_WRITE(RING_IMR(BLT_RING_BASE), 0); + + if (IS_GEN6(dev) || IS_GEN7(dev)) { + /* Workaround stalls observed on Sandy Bridge GPUs by + * making the blitter command streamer generate a + * write to the Hardware Status Page for + * MI_USER_INTERRUPT. This appears to serialize the + * previous seqno write out before the interrupt + * happens. + */ + I915_WRITE(GEN6_BLITTER_HWSTAM, ~GEN6_BLITTER_USER_INTERRUPT); + I915_WRITE(GEN6_BSD_HWSTAM, ~GEN6_BSD_USER_INTERRUPT); + } + + /* and GT */ + I915_WRITE(GTIIR, I915_READ(GTIIR)); + I915_WRITE(GTIIR, I915_READ(GTIIR)); + I915_WRITE(GTIMR, 0xffffffff); + I915_WRITE(GTIER, 0x0); + POSTING_READ(GTIER); + + I915_WRITE(DPINVGTT, 0xff); + + I915_WRITE(PORT_HOTPLUG_EN, 0); + I915_WRITE(PORT_HOTPLUG_STAT, I915_READ(PORT_HOTPLUG_STAT)); + for_each_pipe(pipe) + I915_WRITE(PIPESTAT(pipe), 0xffff); + I915_WRITE(VLV_IIR, 0xffffffff); + I915_WRITE(VLV_IMR, 0xffffffff); + I915_WRITE(VLV_IER, 0x0); + POSTING_READ(VLV_IER); +} + /* * Enable digital hotplug on the PCH, and configure the DP short pulse * duration to 2ms (which is the minimum in the Display Port spec) @@ -1963,6 +2175,96 @@ static int ivybridge_irq_postinstall(struct drm_device *dev) return 0; } +static int valleyview_irq_postinstall(struct drm_device *dev) +{ + drm_i915_private_t *dev_priv = (drm_i915_private_t *) dev->dev_private; + u32 render_irqs; + u32 enable_mask; + u32 hotplug_en = I915_READ(PORT_HOTPLUG_EN); + u16 msid; + + enable_mask = I915_DISPLAY_PORT_INTERRUPT; + enable_mask |= I915_DISPLAY_PIPE_A_VBLANK_INTERRUPT | + I915_DISPLAY_PIPE_B_VBLANK_INTERRUPT; + + dev_priv->irq_mask = ~enable_mask; + + + DRM_INIT_WAITQUEUE(&dev_priv->ring[RCS].irq_queue); + DRM_INIT_WAITQUEUE(&dev_priv->ring[VCS].irq_queue); + DRM_INIT_WAITQUEUE(&dev_priv->ring[BCS].irq_queue); + + dev_priv->pipestat[0] = 0; + dev_priv->pipestat[1] = 0; + + dev_priv->vblank_pipe = DRM_I915_VBLANK_PIPE_A | DRM_I915_VBLANK_PIPE_B; + + /* Hack for broken MSIs on VLV */ + pci_write_config_dword(dev_priv->dev->pdev, 0x94, 0xfee00000); + pci_read_config_word(dev->pdev, 0x98, &msid); + msid &= 0xff; /* mask out delivery bits */ + msid |= (1<<14); + pci_write_config_word(dev_priv->dev->pdev, 0x98, msid); + + I915_WRITE(VLV_IMR, dev_priv->irq_mask); + I915_WRITE(VLV_IER, enable_mask); + I915_WRITE(VLV_IIR, 0xffffffff); + I915_WRITE(PIPESTAT(0), 0xffff); + I915_WRITE(PIPESTAT(1), 0xffff); + POSTING_READ(VLV_IER); + + I915_WRITE(VLV_IIR, 0xffffffff); + I915_WRITE(VLV_IIR, 0xffffffff); + + render_irqs = GT_GEN6_BLT_FLUSHDW_NOTIFY_INTERRUPT | + GT_GEN6_BLT_CS_ERROR_INTERRUPT | + GT_BLT_USER_INTERRUPT | + GT_GEN6_BSD_USER_INTERRUPT | + GT_GEN6_BSD_CS_ERROR_INTERRUPT | + GT_GEN7_L3_PARITY_ERROR_INTERRUPT | + GT_PIPE_NOTIFY | + GT_RENDER_CS_ERROR_INTERRUPT | + GT_SYNC_STATUS | + GT_USER_INTERRUPT; + + dev_priv->gt_irq_mask = ~render_irqs; + + I915_WRITE(GTIIR, I915_READ(GTIIR)); + I915_WRITE(GTIIR, I915_READ(GTIIR)); + I915_WRITE(GTIMR, 0); + I915_WRITE(GTIER, render_irqs); + POSTING_READ(GTIER); + + /* ack & enable invalid PTE error interrupts */ +#if 0 /* FIXME: add support to irq handler for checking these bits */ + I915_WRITE(DPINVGTT, DPINVGTT_STATUS_MASK); + I915_WRITE(DPINVGTT, DPINVGTT_EN_MASK); +#endif + + I915_WRITE(VLV_MASTER_IER, MASTER_INTERRUPT_ENABLE); +#if 0 /* FIXME: check register definitions; some have moved */ + /* Note HDMI and DP share bits */ + if (dev_priv->hotplug_supported_mask & HDMIB_HOTPLUG_INT_STATUS) + hotplug_en |= HDMIB_HOTPLUG_INT_EN; + if (dev_priv->hotplug_supported_mask & HDMIC_HOTPLUG_INT_STATUS) + hotplug_en |= HDMIC_HOTPLUG_INT_EN; + if (dev_priv->hotplug_supported_mask & HDMID_HOTPLUG_INT_STATUS) + hotplug_en |= HDMID_HOTPLUG_INT_EN; + if (dev_priv->hotplug_supported_mask & SDVOC_HOTPLUG_INT_STATUS) + hotplug_en |= SDVOC_HOTPLUG_INT_EN; + if (dev_priv->hotplug_supported_mask & SDVOB_HOTPLUG_INT_STATUS) + hotplug_en |= SDVOB_HOTPLUG_INT_EN; + if (dev_priv->hotplug_supported_mask & CRT_HOTPLUG_INT_STATUS) { + hotplug_en |= CRT_HOTPLUG_INT_EN; + hotplug_en |= CRT_HOTPLUG_VOLTAGE_COMPARE_50; + } +#endif + + I915_WRITE(PORT_HOTPLUG_EN, hotplug_en); + + return 0; +} + static void i915_driver_irq_preinstall(struct drm_device * dev) { drm_i915_private_t *dev_priv = (drm_i915_private_t *) dev->dev_private; @@ -2066,6 +2368,30 @@ static int i915_driver_irq_postinstall(struct drm_device *dev) return 0; } +static void valleyview_irq_uninstall(struct drm_device *dev) +{ + drm_i915_private_t *dev_priv = (drm_i915_private_t *) dev->dev_private; + int pipe; + + if (!dev_priv) + return; + + dev_priv->vblank_pipe = 0; + + for_each_pipe(pipe) + I915_WRITE(PIPESTAT(pipe), 0xffff); + + I915_WRITE(HWSTAM, 0xffffffff); + I915_WRITE(PORT_HOTPLUG_EN, 0); + I915_WRITE(PORT_HOTPLUG_STAT, I915_READ(PORT_HOTPLUG_STAT)); + for_each_pipe(pipe) + I915_WRITE(PIPESTAT(pipe), 0xffff); + I915_WRITE(VLV_IIR, 0xffffffff); + I915_WRITE(VLV_IMR, 0xffffffff); + I915_WRITE(VLV_IER, 0x0); + POSTING_READ(VLV_IER); +} + static void ironlake_irq_uninstall(struct drm_device *dev) { drm_i915_private_t *dev_priv = (drm_i915_private_t *) dev->dev_private; @@ -2121,7 +2447,8 @@ void intel_irq_init(struct drm_device *dev) { dev->driver->get_vblank_counter = i915_get_vblank_counter; dev->max_vblank_count = 0xffffff; /* only 24 bits of frame count */ - if (IS_G4X(dev) || IS_GEN5(dev) || IS_GEN6(dev) || IS_IVYBRIDGE(dev)) { + if (IS_G4X(dev) || IS_GEN5(dev) || IS_GEN6(dev) || IS_IVYBRIDGE(dev) || + IS_VALLEYVIEW(dev)) { dev->max_vblank_count = 0xffffffff; /* full 32 bit counter */ dev->driver->get_vblank_counter = gm45_get_vblank_counter; } @@ -2132,7 +2459,14 @@ void intel_irq_init(struct drm_device *dev) dev->driver->get_vblank_timestamp = NULL; dev->driver->get_scanout_position = i915_get_crtc_scanoutpos; - if (IS_IVYBRIDGE(dev)) { + if (IS_VALLEYVIEW(dev)) { + dev->driver->irq_handler = valleyview_irq_handler; + dev->driver->irq_preinstall = valleyview_irq_preinstall; + dev->driver->irq_postinstall = valleyview_irq_postinstall; + dev->driver->irq_uninstall = valleyview_irq_uninstall; + dev->driver->enable_vblank = valleyview_enable_vblank; + dev->driver->disable_vblank = valleyview_disable_vblank; + } else if (IS_IVYBRIDGE(dev)) { /* Share pre & uninstall handlers with ILK/SNB */ dev->driver->irq_handler = ivybridge_irq_handler; dev->driver->irq_preinstall = ironlake_irq_preinstall; diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h index 0bcad74de7b..908550c72a0 100644 --- a/drivers/gpu/drm/i915/i915_reg.h +++ b/drivers/gpu/drm/i915/i915_reg.h @@ -510,6 +510,11 @@ #define IIR 0x020a4 #define IMR 0x020a8 #define ISR 0x020ac +#define VLV_IIR_RW 0x182084 +#define VLV_IER 0x1820a0 +#define VLV_IIR 0x1820a4 +#define VLV_IMR 0x1820a8 +#define VLV_ISR 0x1820ac #define I915_PIPE_CONTROL_NOTIFY_INTERRUPT (1<<18) #define I915_DISPLAY_PORT_INTERRUPT (1<<17) #define I915_RENDER_COMMAND_PARSER_ERROR_INTERRUPT (1<<15) @@ -2533,7 +2538,7 @@ #define PIPEFRAMEPIXEL(pipe) _PIPE(pipe, _PIPEAFRAMEPIXEL, _PIPEBFRAMEPIXEL) #define PIPESTAT(pipe) _PIPE(pipe, _PIPEASTAT, _PIPEBSTAT) -#define DPFLIPSTAT_VLV 0x70028 +#define VLV_DPFLIPSTAT 0x70028 #define PIPEB_LINE_COMPARE_STATUS (1<<29) #define PIPEB_HLINE_INT_EN (1<<28) #define PIPEB_VBLANK_INT_EN (1<<27) diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.c b/drivers/gpu/drm/i915/intel_ringbuffer.c index b7b50a5b047..75081f14639 100644 --- a/drivers/gpu/drm/i915/intel_ringbuffer.c +++ b/drivers/gpu/drm/i915/intel_ringbuffer.c @@ -687,7 +687,7 @@ render_ring_get_irq(struct intel_ring_buffer *ring) spin_lock(&ring->irq_lock); if (ring->irq_refcount++ == 0) { - if (HAS_PCH_SPLIT(dev)) + if (HAS_PCH_SPLIT(dev) || IS_VALLEYVIEW(dev)) ironlake_enable_irq(dev_priv, GT_PIPE_NOTIFY | GT_USER_INTERRUPT); else @@ -706,7 +706,7 @@ render_ring_put_irq(struct intel_ring_buffer *ring) spin_lock(&ring->irq_lock); if (--ring->irq_refcount == 0) { - if (HAS_PCH_SPLIT(dev)) + if (HAS_PCH_SPLIT(dev) || IS_VALLEYVIEW(dev)) ironlake_disable_irq(dev_priv, GT_USER_INTERRUPT | GT_PIPE_NOTIFY); -- cgit v1.2.3-70-g09d2 From e2a1e2f0242c363ed80458282d67039c373fbb1f Mon Sep 17 00:00:00 2001 From: Ben Widawsky Date: Thu, 29 Mar 2012 19:11:26 -0700 Subject: drm/i915: ring irq cleanups - gen6 put/get only need one argument rflags and gflags are always the same (see above explanation) - remove a couple redundantly defined IRQs - reordered some lines to make things go in descending order Every ring has its own interrupts, enables, masks, and status bits that are fed into the main interrupt enable/mask/status registers. At one point in time it seemed like a good idea to make our functions support the notion that each interrupt may have a different bit position in the corresponding register (blitter parser error may be bit n in IMR, but bit m in blitter IMR). It turned out though that the HW designers did us a solid on Gen6+ and this unfortunate situation has been avoided. This allows our interrupt code to be cleaned up a bit. I jammed this into one commit because there should be no functional change with this commit, and staging it into multiple commits was unnecessarily artificial IMO. CC: Chris Wilson CC: Jesse Barnes Signed-off-by: Ben Widawsky Reviewed-by: Chris Wilson [danvet: - fixed up merged conflict with vlv changes. - added GEN6 to GT blitter bit, we only use it on gen6+. - added a comment to both ring irq bits and GT irq bits that on gen6+ these alias. - added comment that GT_BSD_USER_INTERRUPT is ilk-only. - I've got confused a bit that we still use GT_USER_INTERRUPT on ivb for the render ring - but this goes back to ilk where we have only gt interrupt bits and so we be equally confusing if changed.] Signed-off-by: Daniel Vetter --- drivers/gpu/drm/i915/i915_irq.c | 20 +++++++++--------- drivers/gpu/drm/i915/i915_reg.h | 12 +++++++---- drivers/gpu/drm/i915/intel_ringbuffer.c | 36 +++++++++++---------------------- 3 files changed, 30 insertions(+), 38 deletions(-) (limited to 'drivers/gpu/drm/i915/intel_ringbuffer.c') diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c index 01fb6509797..06286a270b0 100644 --- a/drivers/gpu/drm/i915/i915_irq.c +++ b/drivers/gpu/drm/i915/i915_irq.c @@ -462,7 +462,7 @@ static irqreturn_t valleyview_irq_handler(DRM_IRQ_ARGS) notify_ring(dev, &dev_priv->ring[RCS]); if (gt_iir & GT_GEN6_BSD_USER_INTERRUPT) notify_ring(dev, &dev_priv->ring[VCS]); - if (gt_iir & GT_BLT_USER_INTERRUPT) + if (gt_iir & GT_GEN6_BLT_USER_INTERRUPT) notify_ring(dev, &dev_priv->ring[BCS]); if (gt_iir & (GT_GEN6_BLT_CS_ERROR_INTERRUPT | @@ -620,9 +620,9 @@ static irqreturn_t ivybridge_irq_handler(DRM_IRQ_ARGS) if (gt_iir & (GT_USER_INTERRUPT | GT_PIPE_NOTIFY)) notify_ring(dev, &dev_priv->ring[RCS]); - if (gt_iir & GT_GEN6_BSD_USER_INTERRUPT) + if (gt_iir & GEN6_BSD_USER_INTERRUPT) notify_ring(dev, &dev_priv->ring[VCS]); - if (gt_iir & GT_BLT_USER_INTERRUPT) + if (gt_iir & GEN6_BLITTER_USER_INTERRUPT) notify_ring(dev, &dev_priv->ring[BCS]); if (de_iir & DE_GSE_IVB) @@ -688,7 +688,7 @@ static irqreturn_t ironlake_irq_handler(DRM_IRQ_ARGS) atomic_inc(&dev_priv->irq_received); if (IS_GEN6(dev)) - bsd_usr_interrupt = GT_GEN6_BSD_USER_INTERRUPT; + bsd_usr_interrupt = GEN6_BSD_USER_INTERRUPT; /* disable master interrupt before clearing iir */ de_ier = I915_READ(DEIER); @@ -722,7 +722,7 @@ static irqreturn_t ironlake_irq_handler(DRM_IRQ_ARGS) notify_ring(dev, &dev_priv->ring[RCS]); if (gt_iir & bsd_usr_interrupt) notify_ring(dev, &dev_priv->ring[VCS]); - if (gt_iir & GT_BLT_USER_INTERRUPT) + if (gt_iir & GEN6_BLITTER_USER_INTERRUPT) notify_ring(dev, &dev_priv->ring[BCS]); if (de_iir & DE_GSE) @@ -2081,8 +2081,8 @@ static int ironlake_irq_postinstall(struct drm_device *dev) if (IS_GEN6(dev)) render_irqs = GT_USER_INTERRUPT | - GT_GEN6_BSD_USER_INTERRUPT | - GT_BLT_USER_INTERRUPT; + GEN6_BSD_USER_INTERRUPT | + GEN6_BLITTER_USER_INTERRUPT; else render_irqs = GT_USER_INTERRUPT | @@ -2154,8 +2154,8 @@ static int ivybridge_irq_postinstall(struct drm_device *dev) I915_WRITE(GTIIR, I915_READ(GTIIR)); I915_WRITE(GTIMR, dev_priv->gt_irq_mask); - render_irqs = GT_USER_INTERRUPT | GT_GEN6_BSD_USER_INTERRUPT | - GT_BLT_USER_INTERRUPT; + render_irqs = GT_USER_INTERRUPT | GEN6_BSD_USER_INTERRUPT | + GEN6_BLITTER_USER_INTERRUPT; I915_WRITE(GTIER, render_irqs); POSTING_READ(GTIER); @@ -2218,7 +2218,7 @@ static int valleyview_irq_postinstall(struct drm_device *dev) render_irqs = GT_GEN6_BLT_FLUSHDW_NOTIFY_INTERRUPT | GT_GEN6_BLT_CS_ERROR_INTERRUPT | - GT_BLT_USER_INTERRUPT | + GT_GEN6_BLT_USER_INTERRUPT | GT_GEN6_BSD_USER_INTERRUPT | GT_GEN6_BSD_CS_ERROR_INTERRUPT | GT_GEN7_L3_PARITY_ERROR_INTERRUPT | diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h index 0f6f567ae50..56d4db8026e 100644 --- a/drivers/gpu/drm/i915/i915_reg.h +++ b/drivers/gpu/drm/i915/i915_reg.h @@ -643,7 +643,9 @@ #define CACHE_MODE_1 0x7004 /* IVB+ */ #define PIXEL_SUBSPAN_COLLECT_OPT_DISABLE (1<<6) -/* GEN6 interrupt control */ +/* GEN6 interrupt control + * Note that the per-ring interrupt bits do alias with the global interrupt bits + * in GTIMR. */ #define GEN6_RENDER_HWSTAM 0x2098 #define GEN6_RENDER_IMR 0x20a8 #define GEN6_RENDER_CONTEXT_SWITCH_INTERRUPT (1 << 8) @@ -3203,13 +3205,15 @@ #define DEIIR 0x44008 #define DEIER 0x4400c -/* GT interrupt */ +/* GT interrupt. + * Note that for gen6+ the ring-specific interrupt bits do alias with the + * corresponding bits in the per-ring interrupt control registers. */ #define GT_GEN6_BLT_FLUSHDW_NOTIFY_INTERRUPT (1 << 26) #define GT_GEN6_BLT_CS_ERROR_INTERRUPT (1 << 25) -#define GT_BLT_USER_INTERRUPT (1 << 22) +#define GT_GEN6_BLT_USER_INTERRUPT (1 << 22) #define GT_GEN6_BSD_CS_ERROR_INTERRUPT (1 << 15) #define GT_GEN6_BSD_USER_INTERRUPT (1 << 12) -#define GT_BSD_USER_INTERRUPT (1 << 5) +#define GT_BSD_USER_INTERRUPT (1 << 5) /* ilk only */ #define GT_GEN7_L3_PARITY_ERROR_INTERRUPT (1 << 5) #define GT_PIPE_NOTIFY (1 << 4) #define GT_RENDER_CS_ERROR_INTERRUPT (1 << 3) diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.c b/drivers/gpu/drm/i915/intel_ringbuffer.c index 75081f14639..c7eea7fad16 100644 --- a/drivers/gpu/drm/i915/intel_ringbuffer.c +++ b/drivers/gpu/drm/i915/intel_ringbuffer.c @@ -788,7 +788,7 @@ ring_add_request(struct intel_ring_buffer *ring, } static bool -gen6_ring_get_irq(struct intel_ring_buffer *ring, u32 gflag, u32 rflag) +gen6_ring_get_irq(struct intel_ring_buffer *ring, u32 mask) { struct drm_device *dev = ring->dev; drm_i915_private_t *dev_priv = dev->dev_private; @@ -803,9 +803,9 @@ gen6_ring_get_irq(struct intel_ring_buffer *ring, u32 gflag, u32 rflag) spin_lock(&ring->irq_lock); if (ring->irq_refcount++ == 0) { - ring->irq_mask &= ~rflag; + ring->irq_mask &= ~mask; I915_WRITE_IMR(ring, ring->irq_mask); - ironlake_enable_irq(dev_priv, gflag); + ironlake_enable_irq(dev_priv, mask); } spin_unlock(&ring->irq_lock); @@ -813,16 +813,16 @@ gen6_ring_get_irq(struct intel_ring_buffer *ring, u32 gflag, u32 rflag) } static void -gen6_ring_put_irq(struct intel_ring_buffer *ring, u32 gflag, u32 rflag) +gen6_ring_put_irq(struct intel_ring_buffer *ring, u32 mask) { struct drm_device *dev = ring->dev; drm_i915_private_t *dev_priv = dev->dev_private; spin_lock(&ring->irq_lock); if (--ring->irq_refcount == 0) { - ring->irq_mask |= rflag; + ring->irq_mask |= mask; I915_WRITE_IMR(ring, ring->irq_mask); - ironlake_disable_irq(dev_priv, gflag); + ironlake_disable_irq(dev_priv, mask); } spin_unlock(&ring->irq_lock); @@ -1376,33 +1376,25 @@ gen6_ring_dispatch_execbuffer(struct intel_ring_buffer *ring, static bool gen6_render_ring_get_irq(struct intel_ring_buffer *ring) { - return gen6_ring_get_irq(ring, - GT_USER_INTERRUPT, - GEN6_RENDER_USER_INTERRUPT); + return gen6_ring_get_irq(ring, GT_USER_INTERRUPT); } static void gen6_render_ring_put_irq(struct intel_ring_buffer *ring) { - return gen6_ring_put_irq(ring, - GT_USER_INTERRUPT, - GEN6_RENDER_USER_INTERRUPT); + return gen6_ring_put_irq(ring, GT_USER_INTERRUPT); } static bool gen6_bsd_ring_get_irq(struct intel_ring_buffer *ring) { - return gen6_ring_get_irq(ring, - GT_GEN6_BSD_USER_INTERRUPT, - GEN6_BSD_USER_INTERRUPT); + return gen6_ring_get_irq(ring, GEN6_BSD_USER_INTERRUPT); } static void gen6_bsd_ring_put_irq(struct intel_ring_buffer *ring) { - return gen6_ring_put_irq(ring, - GT_GEN6_BSD_USER_INTERRUPT, - GEN6_BSD_USER_INTERRUPT); + return gen6_ring_put_irq(ring, GEN6_BSD_USER_INTERRUPT); } /* ring buffer for Video Codec for Gen6+ */ @@ -1431,17 +1423,13 @@ static const struct intel_ring_buffer gen6_bsd_ring = { static bool blt_ring_get_irq(struct intel_ring_buffer *ring) { - return gen6_ring_get_irq(ring, - GT_BLT_USER_INTERRUPT, - GEN6_BLITTER_USER_INTERRUPT); + return gen6_ring_get_irq(ring, GEN6_BLITTER_USER_INTERRUPT); } static void blt_ring_put_irq(struct intel_ring_buffer *ring) { - gen6_ring_put_irq(ring, - GT_BLT_USER_INTERRUPT, - GEN6_BLITTER_USER_INTERRUPT); + gen6_ring_put_irq(ring, GEN6_BLITTER_USER_INTERRUPT); } static int blt_ring_flush(struct intel_ring_buffer *ring, -- cgit v1.2.3-70-g09d2 From 25c063004a6d0048c4dece74db4da117b9ae623e Mon Sep 17 00:00:00 2001 From: Ben Widawsky Date: Thu, 29 Mar 2012 19:11:27 -0700 Subject: drm/i915: open code gen6+ ring irqs We can now open-code the get/put irq functions as they were just abstracting single register definitions. It would be nice to merge this in with the IRQ handling code... but that is too much work for me at present. In addition I could probably collapse this in to a lot of the Ironlake stuff, but I don't think it's worth the potential regressions. This patch itself should not effect functionality. CC: Jesse Barnes Signed-off-by: Ben Widawsky Reviewed-by: Chris Wilson Signed-off-by: Daniel Vetter --- drivers/gpu/drm/i915/intel_ringbuffer.c | 62 +++++++++------------------------ drivers/gpu/drm/i915/intel_ringbuffer.h | 1 + 2 files changed, 17 insertions(+), 46 deletions(-) (limited to 'drivers/gpu/drm/i915/intel_ringbuffer.c') diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.c b/drivers/gpu/drm/i915/intel_ringbuffer.c index c7eea7fad16..98ac5c0ca37 100644 --- a/drivers/gpu/drm/i915/intel_ringbuffer.c +++ b/drivers/gpu/drm/i915/intel_ringbuffer.c @@ -788,10 +788,11 @@ ring_add_request(struct intel_ring_buffer *ring, } static bool -gen6_ring_get_irq(struct intel_ring_buffer *ring, u32 mask) +gen6_ring_get_irq(struct intel_ring_buffer *ring) { struct drm_device *dev = ring->dev; drm_i915_private_t *dev_priv = dev->dev_private; + u32 mask = ring->irq_enable; if (!dev->irq_enabled) return false; @@ -813,10 +814,11 @@ gen6_ring_get_irq(struct intel_ring_buffer *ring, u32 mask) } static void -gen6_ring_put_irq(struct intel_ring_buffer *ring, u32 mask) +gen6_ring_put_irq(struct intel_ring_buffer *ring) { struct drm_device *dev = ring->dev; drm_i915_private_t *dev_priv = dev->dev_private; + u32 mask = ring->irq_enable; spin_lock(&ring->irq_lock); if (--ring->irq_refcount == 0) { @@ -1373,30 +1375,6 @@ gen6_ring_dispatch_execbuffer(struct intel_ring_buffer *ring, return 0; } -static bool -gen6_render_ring_get_irq(struct intel_ring_buffer *ring) -{ - return gen6_ring_get_irq(ring, GT_USER_INTERRUPT); -} - -static void -gen6_render_ring_put_irq(struct intel_ring_buffer *ring) -{ - return gen6_ring_put_irq(ring, GT_USER_INTERRUPT); -} - -static bool -gen6_bsd_ring_get_irq(struct intel_ring_buffer *ring) -{ - return gen6_ring_get_irq(ring, GEN6_BSD_USER_INTERRUPT); -} - -static void -gen6_bsd_ring_put_irq(struct intel_ring_buffer *ring) -{ - return gen6_ring_put_irq(ring, GEN6_BSD_USER_INTERRUPT); -} - /* ring buffer for Video Codec for Gen6+ */ static const struct intel_ring_buffer gen6_bsd_ring = { .name = "gen6 bsd ring", @@ -1408,8 +1386,9 @@ static const struct intel_ring_buffer gen6_bsd_ring = { .flush = gen6_ring_flush, .add_request = gen6_add_request, .get_seqno = gen6_ring_get_seqno, - .irq_get = gen6_bsd_ring_get_irq, - .irq_put = gen6_bsd_ring_put_irq, + .irq_enable = GEN6_BSD_USER_INTERRUPT, + .irq_get = gen6_ring_get_irq, + .irq_put = gen6_ring_put_irq, .dispatch_execbuffer = gen6_ring_dispatch_execbuffer, .sync_to = gen6_bsd_ring_sync_to, .semaphore_register = {MI_SEMAPHORE_SYNC_VR, @@ -1420,18 +1399,6 @@ static const struct intel_ring_buffer gen6_bsd_ring = { /* Blitter support (SandyBridge+) */ -static bool -blt_ring_get_irq(struct intel_ring_buffer *ring) -{ - return gen6_ring_get_irq(ring, GEN6_BLITTER_USER_INTERRUPT); -} - -static void -blt_ring_put_irq(struct intel_ring_buffer *ring) -{ - gen6_ring_put_irq(ring, GEN6_BLITTER_USER_INTERRUPT); -} - static int blt_ring_flush(struct intel_ring_buffer *ring, u32 invalidate, u32 flush) { @@ -1463,8 +1430,9 @@ static const struct intel_ring_buffer gen6_blt_ring = { .flush = blt_ring_flush, .add_request = gen6_add_request, .get_seqno = gen6_ring_get_seqno, - .irq_get = blt_ring_get_irq, - .irq_put = blt_ring_put_irq, + .irq_get = gen6_ring_get_irq, + .irq_put = gen6_ring_put_irq, + .irq_enable = GEN6_BLITTER_USER_INTERRUPT, .dispatch_execbuffer = gen6_ring_dispatch_execbuffer, .sync_to = gen6_blt_ring_sync_to, .semaphore_register = {MI_SEMAPHORE_SYNC_BR, @@ -1482,8 +1450,9 @@ int intel_init_render_ring_buffer(struct drm_device *dev) if (INTEL_INFO(dev)->gen >= 6) { ring->add_request = gen6_add_request; ring->flush = gen6_render_ring_flush; - ring->irq_get = gen6_render_ring_get_irq; - ring->irq_put = gen6_render_ring_put_irq; + ring->irq_get = gen6_ring_get_irq; + ring->irq_put = gen6_ring_put_irq; + ring->irq_enable = GT_USER_INTERRUPT; ring->get_seqno = gen6_ring_get_seqno; } else if (IS_GEN5(dev)) { ring->add_request = pc_render_add_request; @@ -1506,8 +1475,9 @@ int intel_render_ring_init_dri(struct drm_device *dev, u64 start, u32 size) *ring = render_ring; if (INTEL_INFO(dev)->gen >= 6) { ring->add_request = gen6_add_request; - ring->irq_get = gen6_render_ring_get_irq; - ring->irq_put = gen6_render_ring_put_irq; + ring->irq_get = gen6_ring_get_irq; + ring->irq_put = gen6_ring_put_irq; + ring->irq_enable = GT_USER_INTERRUPT; } else if (IS_GEN5(dev)) { ring->add_request = pc_render_add_request; ring->get_seqno = pc_render_get_seqno; diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.h b/drivers/gpu/drm/i915/intel_ringbuffer.h index bc0365b8fa4..3488a5a127d 100644 --- a/drivers/gpu/drm/i915/intel_ringbuffer.h +++ b/drivers/gpu/drm/i915/intel_ringbuffer.h @@ -59,6 +59,7 @@ struct intel_ring_buffer { spinlock_t irq_lock; u32 irq_refcount; u32 irq_mask; + u32 irq_enable; /* IRQs enabled for this ring */ u32 irq_seqno; /* last seq seem at irq time */ u32 trace_irq_seqno; u32 waiting_seqno; -- cgit v1.2.3-70-g09d2 From 901781b997570e55af950348d7cd5cef64fb6d7c Mon Sep 17 00:00:00 2001 From: Daniel Vetter Date: Fri, 30 Mar 2012 20:24:34 +0200 Subject: drm/i915: use render gen to switch ring irq functions Top-level interrupt bits are usually found in the display block. It therefore makes sense to use HAS_PCH_SPLIT in i915_irq.c But the irq stuff in intel_ring.c only concerns itself with render core/gt-level interrupt sources. It therefore makes more sense to switch based on gpu gen. Kills a vlv special case. Reviewed-by: Jesse Barnes Signed-Off-by: Daniel Vetter --- drivers/gpu/drm/i915/intel_ringbuffer.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'drivers/gpu/drm/i915/intel_ringbuffer.c') diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.c b/drivers/gpu/drm/i915/intel_ringbuffer.c index 98ac5c0ca37..465a7da3b30 100644 --- a/drivers/gpu/drm/i915/intel_ringbuffer.c +++ b/drivers/gpu/drm/i915/intel_ringbuffer.c @@ -687,7 +687,7 @@ render_ring_get_irq(struct intel_ring_buffer *ring) spin_lock(&ring->irq_lock); if (ring->irq_refcount++ == 0) { - if (HAS_PCH_SPLIT(dev) || IS_VALLEYVIEW(dev)) + if (INTEL_INFO(dev)->gen >= 5) ironlake_enable_irq(dev_priv, GT_PIPE_NOTIFY | GT_USER_INTERRUPT); else @@ -706,7 +706,7 @@ render_ring_put_irq(struct intel_ring_buffer *ring) spin_lock(&ring->irq_lock); if (--ring->irq_refcount == 0) { - if (HAS_PCH_SPLIT(dev) || IS_VALLEYVIEW(dev)) + if (INTEL_INFO(dev)->gen >= 5) ironlake_disable_irq(dev_priv, GT_USER_INTERRUPT | GT_PIPE_NOTIFY); -- cgit v1.2.3-70-g09d2 From 1500f7ea06858819abcf8eec8f952e2f9281c610 Mon Sep 17 00:00:00 2001 From: Ben Widawsky Date: Wed, 11 Apr 2012 11:18:21 -0700 Subject: drm/i915: hide (seqno-1) in ringbuffer code Waiting for seqno-1 in our object synchronization code is an implementation detail given how we've decided to do the waits within the rest of our code. Requested-by: Daniel Vetter Signed-off-by: Ben Widawsky Reviewed-by: Chris Wilson Signed-off-by: Daniel Vetter --- drivers/gpu/drm/i915/i915_gem.c | 2 +- drivers/gpu/drm/i915/intel_ringbuffer.c | 6 ++++++ 2 files changed, 7 insertions(+), 1 deletion(-) (limited to 'drivers/gpu/drm/i915/intel_ringbuffer.c') diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c index 0115b12df57..71934dd0ee4 100644 --- a/drivers/gpu/drm/i915/i915_gem.c +++ b/drivers/gpu/drm/i915/i915_gem.c @@ -2002,7 +2002,7 @@ i915_gem_object_sync(struct drm_i915_gem_object *obj, } - ret = to->sync_to(to, from, seqno - 1); + ret = to->sync_to(to, from, seqno); if (!ret) from->sync_seqno[idx] = seqno; diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.c b/drivers/gpu/drm/i915/intel_ringbuffer.c index dfdb613752c..467b3319e4b 100644 --- a/drivers/gpu/drm/i915/intel_ringbuffer.c +++ b/drivers/gpu/drm/i915/intel_ringbuffer.c @@ -482,6 +482,12 @@ intel_ring_sync(struct intel_ring_buffer *waiter, MI_SEMAPHORE_COMPARE | MI_SEMAPHORE_REGISTER; + /* Throughout all of the GEM code, seqno passed implies our current + * seqno is >= the last seqno executed. However for hardware the + * comparison is strictly greater than. + */ + seqno -= 1; + ret = intel_ring_begin(waiter, 4); if (ret) return ret; -- cgit v1.2.3-70-g09d2 From 6a848ccb800f5330ca368cd804795b7ce644e36c Mon Sep 17 00:00:00 2001 From: Daniel Vetter Date: Wed, 11 Apr 2012 22:12:46 +0200 Subject: drm/i915: rip out ring->irq_mask We only ever enable/disable one interrupt (namely user_interrupts and pipe_notify), so we don't need to track the interrupt masking state. Also rename irq_enable to irq_enable_mask, now that it won't collide - beforehand both a irq_mask and irq_enable_mask would have looked a bit strange. Reviewed-by: Eric Anholt Signed-Off-by: Daniel Vetter --- drivers/gpu/drm/i915/intel_ringbuffer.c | 21 ++++++++------------- drivers/gpu/drm/i915/intel_ringbuffer.h | 3 +-- 2 files changed, 9 insertions(+), 15 deletions(-) (limited to 'drivers/gpu/drm/i915/intel_ringbuffer.c') diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.c b/drivers/gpu/drm/i915/intel_ringbuffer.c index 467b3319e4b..915aa07d5a6 100644 --- a/drivers/gpu/drm/i915/intel_ringbuffer.c +++ b/drivers/gpu/drm/i915/intel_ringbuffer.c @@ -798,7 +798,6 @@ gen6_ring_get_irq(struct intel_ring_buffer *ring) { struct drm_device *dev = ring->dev; drm_i915_private_t *dev_priv = dev->dev_private; - u32 mask = ring->irq_enable; if (!dev->irq_enabled) return false; @@ -810,9 +809,8 @@ gen6_ring_get_irq(struct intel_ring_buffer *ring) spin_lock(&ring->irq_lock); if (ring->irq_refcount++ == 0) { - ring->irq_mask &= ~mask; - I915_WRITE_IMR(ring, ring->irq_mask); - ironlake_enable_irq(dev_priv, mask); + I915_WRITE_IMR(ring, ~ring->irq_enable_mask); + ironlake_enable_irq(dev_priv, ring->irq_enable_mask); } spin_unlock(&ring->irq_lock); @@ -824,13 +822,11 @@ gen6_ring_put_irq(struct intel_ring_buffer *ring) { struct drm_device *dev = ring->dev; drm_i915_private_t *dev_priv = dev->dev_private; - u32 mask = ring->irq_enable; spin_lock(&ring->irq_lock); if (--ring->irq_refcount == 0) { - ring->irq_mask |= mask; - I915_WRITE_IMR(ring, ring->irq_mask); - ironlake_disable_irq(dev_priv, mask); + I915_WRITE_IMR(ring, ~0); + ironlake_disable_irq(dev_priv, ring->irq_enable_mask); } spin_unlock(&ring->irq_lock); @@ -1002,7 +998,6 @@ int intel_init_ring_buffer(struct drm_device *dev, init_waitqueue_head(&ring->irq_queue); spin_lock_init(&ring->irq_lock); - ring->irq_mask = ~0; if (I915_NEED_GFX_HWS(dev)) { ret = init_status_page(ring); @@ -1380,7 +1375,7 @@ static const struct intel_ring_buffer gen6_bsd_ring = { .flush = gen6_ring_flush, .add_request = gen6_add_request, .get_seqno = gen6_ring_get_seqno, - .irq_enable = GEN6_BSD_USER_INTERRUPT, + .irq_enable_mask = GEN6_BSD_USER_INTERRUPT, .irq_get = gen6_ring_get_irq, .irq_put = gen6_ring_put_irq, .dispatch_execbuffer = gen6_ring_dispatch_execbuffer, @@ -1426,7 +1421,7 @@ static const struct intel_ring_buffer gen6_blt_ring = { .get_seqno = gen6_ring_get_seqno, .irq_get = gen6_ring_get_irq, .irq_put = gen6_ring_put_irq, - .irq_enable = GEN6_BLITTER_USER_INTERRUPT, + .irq_enable_mask = GEN6_BLITTER_USER_INTERRUPT, .dispatch_execbuffer = gen6_ring_dispatch_execbuffer, .sync_to = gen6_blt_ring_sync_to, .semaphore_register = {MI_SEMAPHORE_SYNC_BR, @@ -1446,7 +1441,7 @@ int intel_init_render_ring_buffer(struct drm_device *dev) ring->flush = gen6_render_ring_flush; ring->irq_get = gen6_ring_get_irq; ring->irq_put = gen6_ring_put_irq; - ring->irq_enable = GT_USER_INTERRUPT; + ring->irq_enable_mask = GT_USER_INTERRUPT; ring->get_seqno = gen6_ring_get_seqno; } else if (IS_GEN5(dev)) { ring->add_request = pc_render_add_request; @@ -1471,7 +1466,7 @@ int intel_render_ring_init_dri(struct drm_device *dev, u64 start, u32 size) ring->add_request = gen6_add_request; ring->irq_get = gen6_ring_get_irq; ring->irq_put = gen6_ring_put_irq; - ring->irq_enable = GT_USER_INTERRUPT; + ring->irq_enable_mask = GT_USER_INTERRUPT; } else if (IS_GEN5(dev)) { ring->add_request = pc_render_add_request; ring->get_seqno = pc_render_get_seqno; diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.h b/drivers/gpu/drm/i915/intel_ringbuffer.h index 3488a5a127d..06a66adf69c 100644 --- a/drivers/gpu/drm/i915/intel_ringbuffer.h +++ b/drivers/gpu/drm/i915/intel_ringbuffer.h @@ -58,8 +58,7 @@ struct intel_ring_buffer { spinlock_t irq_lock; u32 irq_refcount; - u32 irq_mask; - u32 irq_enable; /* IRQs enabled for this ring */ + u32 irq_enable_mask; /* bitmask to enable ring interrupt */ u32 irq_seqno; /* last seq seem at irq time */ u32 trace_irq_seqno; u32 waiting_seqno; -- cgit v1.2.3-70-g09d2 From dfc9ef2fb0ba01ad8256f4dfa9a8e8fcc6288fc4 Mon Sep 17 00:00:00 2001 From: Daniel Vetter Date: Wed, 11 Apr 2012 22:12:47 +0200 Subject: drm/i915: set ring->size in common ring setup code Eventually we want to scale the ring size depending upon available gtt space. For now just consolidate this instead of replicating it over all ringbuffer templates. Reviewed-by: Eric Anholt Signed-off-by: Daniel Vetter --- drivers/gpu/drm/i915/intel_ringbuffer.c | 5 +---- 1 file changed, 1 insertion(+), 4 deletions(-) (limited to 'drivers/gpu/drm/i915/intel_ringbuffer.c') diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.c b/drivers/gpu/drm/i915/intel_ringbuffer.c index 915aa07d5a6..87e29076123 100644 --- a/drivers/gpu/drm/i915/intel_ringbuffer.c +++ b/drivers/gpu/drm/i915/intel_ringbuffer.c @@ -995,6 +995,7 @@ int intel_init_ring_buffer(struct drm_device *dev, INIT_LIST_HEAD(&ring->active_list); INIT_LIST_HEAD(&ring->request_list); INIT_LIST_HEAD(&ring->gpu_write_list); + ring->size = 32 * PAGE_SIZE; init_waitqueue_head(&ring->irq_queue); spin_lock_init(&ring->irq_lock); @@ -1268,7 +1269,6 @@ static const struct intel_ring_buffer render_ring = { .name = "render ring", .id = RCS, .mmio_base = RENDER_RING_BASE, - .size = 32 * PAGE_SIZE, .init = init_render_ring, .write_tail = ring_write_tail, .flush = render_ring_flush, @@ -1291,7 +1291,6 @@ static const struct intel_ring_buffer bsd_ring = { .name = "bsd ring", .id = VCS, .mmio_base = BSD_RING_BASE, - .size = 32 * PAGE_SIZE, .init = init_ring_common, .write_tail = ring_write_tail, .flush = bsd_ring_flush, @@ -1369,7 +1368,6 @@ static const struct intel_ring_buffer gen6_bsd_ring = { .name = "gen6 bsd ring", .id = VCS, .mmio_base = GEN6_BSD_RING_BASE, - .size = 32 * PAGE_SIZE, .init = init_ring_common, .write_tail = gen6_bsd_ring_write_tail, .flush = gen6_ring_flush, @@ -1413,7 +1411,6 @@ static const struct intel_ring_buffer gen6_blt_ring = { .name = "blt ring", .id = BCS, .mmio_base = BLT_RING_BASE, - .size = 32 * PAGE_SIZE, .init = init_ring_common, .write_tail = ring_write_tail, .flush = blt_ring_flush, -- cgit v1.2.3-70-g09d2 From 59465b5f78db752622350e3e01dbe18a6f19e876 Mon Sep 17 00:00:00 2001 From: Daniel Vetter Date: Wed, 11 Apr 2012 22:12:48 +0200 Subject: drm/i915: dynamically set up the render ring functions and params Our hw is simply not well-designed enough that it neatly fits into boxes. Everywhere else we set up vtables and similar things dynamically using switch statements - it's simply much more flexible. This is prep work to rework the pre-gen6 ring irq stuff - it'll add a few more differences. With the current const struct templates, that would be a mess. This leads to some unfortunate duplication with the old dri1 code, but we can reap that again because gen6 isn't actually supported there. But that's for a separate patch. Reviewed-by: Eric Anholt Signed-Off-by: Daniel Vetter --- drivers/gpu/drm/i915/intel_ringbuffer.c | 71 +++++++++++++++++++++++---------- 1 file changed, 49 insertions(+), 22 deletions(-) (limited to 'drivers/gpu/drm/i915/intel_ringbuffer.c') diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.c b/drivers/gpu/drm/i915/intel_ringbuffer.c index 87e29076123..fd8e27ac207 100644 --- a/drivers/gpu/drm/i915/intel_ringbuffer.c +++ b/drivers/gpu/drm/i915/intel_ringbuffer.c @@ -1265,26 +1265,6 @@ void intel_ring_advance(struct intel_ring_buffer *ring) ring->write_tail(ring, ring->tail); } -static const struct intel_ring_buffer render_ring = { - .name = "render ring", - .id = RCS, - .mmio_base = RENDER_RING_BASE, - .init = init_render_ring, - .write_tail = ring_write_tail, - .flush = render_ring_flush, - .add_request = render_ring_add_request, - .get_seqno = ring_get_seqno, - .irq_get = render_ring_get_irq, - .irq_put = render_ring_put_irq, - .dispatch_execbuffer = render_ring_dispatch_execbuffer, - .cleanup = render_ring_cleanup, - .sync_to = render_ring_sync_to, - .semaphore_register = {MI_SEMAPHORE_SYNC_INVALID, - MI_SEMAPHORE_SYNC_RV, - MI_SEMAPHORE_SYNC_RB}, - .signal_mbox = {GEN6_VRSYNC, GEN6_BRSYNC}, -}; - /* ring buffer for bit-stream decoder */ static const struct intel_ring_buffer bsd_ring = { @@ -1432,7 +1412,10 @@ int intel_init_render_ring_buffer(struct drm_device *dev) drm_i915_private_t *dev_priv = dev->dev_private; struct intel_ring_buffer *ring = &dev_priv->ring[RCS]; - *ring = render_ring; + ring->name = "render ring"; + ring->id = RCS; + ring->mmio_base = RENDER_RING_BASE; + if (INTEL_INFO(dev)->gen >= 6) { ring->add_request = gen6_add_request; ring->flush = gen6_render_ring_flush; @@ -1440,10 +1423,30 @@ int intel_init_render_ring_buffer(struct drm_device *dev) ring->irq_put = gen6_ring_put_irq; ring->irq_enable_mask = GT_USER_INTERRUPT; ring->get_seqno = gen6_ring_get_seqno; + ring->sync_to = render_ring_sync_to; + ring->semaphore_register[0] = MI_SEMAPHORE_SYNC_INVALID; + ring->semaphore_register[1] = MI_SEMAPHORE_SYNC_RV; + ring->semaphore_register[2] = MI_SEMAPHORE_SYNC_RB; + ring->signal_mbox[0] = GEN6_VRSYNC; + ring->signal_mbox[1] = GEN6_BRSYNC; } else if (IS_GEN5(dev)) { ring->add_request = pc_render_add_request; + ring->flush = render_ring_flush; ring->get_seqno = pc_render_get_seqno; + ring->irq_get = render_ring_get_irq; + ring->irq_put = render_ring_put_irq; + } else { + ring->add_request = render_ring_add_request; + ring->flush = render_ring_flush; + ring->get_seqno = ring_get_seqno; + ring->irq_get = render_ring_get_irq; + ring->irq_put = render_ring_put_irq; } + ring->write_tail = ring_write_tail; + ring->dispatch_execbuffer = render_ring_dispatch_execbuffer; + ring->init = init_render_ring; + ring->cleanup = render_ring_cleanup; + if (!I915_NEED_GFX_HWS(dev)) { ring->status_page.page_addr = dev_priv->status_page_dmah->vaddr; @@ -1458,16 +1461,40 @@ int intel_render_ring_init_dri(struct drm_device *dev, u64 start, u32 size) drm_i915_private_t *dev_priv = dev->dev_private; struct intel_ring_buffer *ring = &dev_priv->ring[RCS]; - *ring = render_ring; + ring->name = "render ring"; + ring->id = RCS; + ring->mmio_base = RENDER_RING_BASE; + if (INTEL_INFO(dev)->gen >= 6) { ring->add_request = gen6_add_request; + ring->flush = gen6_render_ring_flush; ring->irq_get = gen6_ring_get_irq; ring->irq_put = gen6_ring_put_irq; ring->irq_enable_mask = GT_USER_INTERRUPT; + ring->get_seqno = gen6_ring_get_seqno; + ring->sync_to = render_ring_sync_to; + ring->semaphore_register[0] = MI_SEMAPHORE_SYNC_INVALID; + ring->semaphore_register[1] = MI_SEMAPHORE_SYNC_RV; + ring->semaphore_register[2] = MI_SEMAPHORE_SYNC_RB; + ring->signal_mbox[0] = GEN6_VRSYNC; + ring->signal_mbox[1] = GEN6_BRSYNC; } else if (IS_GEN5(dev)) { ring->add_request = pc_render_add_request; + ring->flush = render_ring_flush; ring->get_seqno = pc_render_get_seqno; + ring->irq_get = render_ring_get_irq; + ring->irq_put = render_ring_put_irq; + } else { + ring->add_request = render_ring_add_request; + ring->flush = render_ring_flush; + ring->get_seqno = ring_get_seqno; + ring->irq_get = render_ring_get_irq; + ring->irq_put = render_ring_put_irq; } + ring->write_tail = ring_write_tail; + ring->dispatch_execbuffer = render_ring_dispatch_execbuffer; + ring->init = init_render_ring; + ring->cleanup = render_ring_cleanup; if (!I915_NEED_GFX_HWS(dev)) ring->status_page.page_addr = dev_priv->status_page_dmah->vaddr; -- cgit v1.2.3-70-g09d2 From 58fa3835874bc3466e84e3bba6057f43acd6eb18 Mon Sep 17 00:00:00 2001 From: Daniel Vetter Date: Wed, 11 Apr 2012 22:12:49 +0200 Subject: drm/i915: dynamically set up bsd ring functions and params The same treatment for the bsd ring. Again, this will be split up further by the irq rework. Reviewed-by: Eric Anholt Signed-off-by: Daniel Vetter --- drivers/gpu/drm/i915/intel_ringbuffer.c | 72 ++++++++++++++------------------- 1 file changed, 31 insertions(+), 41 deletions(-) (limited to 'drivers/gpu/drm/i915/intel_ringbuffer.c') diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.c b/drivers/gpu/drm/i915/intel_ringbuffer.c index fd8e27ac207..683d1f0fcb3 100644 --- a/drivers/gpu/drm/i915/intel_ringbuffer.c +++ b/drivers/gpu/drm/i915/intel_ringbuffer.c @@ -1265,22 +1265,6 @@ void intel_ring_advance(struct intel_ring_buffer *ring) ring->write_tail(ring, ring->tail); } -/* ring buffer for bit-stream decoder */ - -static const struct intel_ring_buffer bsd_ring = { - .name = "bsd ring", - .id = VCS, - .mmio_base = BSD_RING_BASE, - .init = init_ring_common, - .write_tail = ring_write_tail, - .flush = bsd_ring_flush, - .add_request = ring_add_request, - .get_seqno = ring_get_seqno, - .irq_get = bsd_ring_get_irq, - .irq_put = bsd_ring_put_irq, - .dispatch_execbuffer = ring_dispatch_execbuffer, -}; - static void gen6_bsd_ring_write_tail(struct intel_ring_buffer *ring, u32 value) @@ -1343,27 +1327,6 @@ gen6_ring_dispatch_execbuffer(struct intel_ring_buffer *ring, return 0; } -/* ring buffer for Video Codec for Gen6+ */ -static const struct intel_ring_buffer gen6_bsd_ring = { - .name = "gen6 bsd ring", - .id = VCS, - .mmio_base = GEN6_BSD_RING_BASE, - .init = init_ring_common, - .write_tail = gen6_bsd_ring_write_tail, - .flush = gen6_ring_flush, - .add_request = gen6_add_request, - .get_seqno = gen6_ring_get_seqno, - .irq_enable_mask = GEN6_BSD_USER_INTERRUPT, - .irq_get = gen6_ring_get_irq, - .irq_put = gen6_ring_put_irq, - .dispatch_execbuffer = gen6_ring_dispatch_execbuffer, - .sync_to = gen6_bsd_ring_sync_to, - .semaphore_register = {MI_SEMAPHORE_SYNC_VR, - MI_SEMAPHORE_SYNC_INVALID, - MI_SEMAPHORE_SYNC_VB}, - .signal_mbox = {GEN6_RVSYNC, GEN6_BVSYNC}, -}; - /* Blitter support (SandyBridge+) */ static int blt_ring_flush(struct intel_ring_buffer *ring, @@ -1531,10 +1494,37 @@ int intel_init_bsd_ring_buffer(struct drm_device *dev) drm_i915_private_t *dev_priv = dev->dev_private; struct intel_ring_buffer *ring = &dev_priv->ring[VCS]; - if (IS_GEN6(dev) || IS_GEN7(dev)) - *ring = gen6_bsd_ring; - else - *ring = bsd_ring; + ring->name = "bsd ring"; + ring->id = VCS; + + if (IS_GEN6(dev) || IS_GEN7(dev)) { + ring->mmio_base = GEN6_BSD_RING_BASE; + ring->write_tail = gen6_bsd_ring_write_tail; + ring->flush = gen6_ring_flush; + ring->add_request = gen6_add_request; + ring->get_seqno = gen6_ring_get_seqno; + ring->irq_enable_mask = GEN6_BSD_USER_INTERRUPT; + ring->irq_get = gen6_ring_get_irq; + ring->irq_put = gen6_ring_put_irq; + ring->dispatch_execbuffer = gen6_ring_dispatch_execbuffer; + ring->sync_to = gen6_bsd_ring_sync_to; + ring->semaphore_register[0] = MI_SEMAPHORE_SYNC_VR; + ring->semaphore_register[1] = MI_SEMAPHORE_SYNC_INVALID; + ring->semaphore_register[2] = MI_SEMAPHORE_SYNC_VB; + ring->signal_mbox[0] = GEN6_RVSYNC; + ring->signal_mbox[1] = GEN6_BVSYNC; + } else { + ring->mmio_base = BSD_RING_BASE; + ring->write_tail = ring_write_tail; + ring->flush = bsd_ring_flush; + ring->add_request = ring_add_request; + ring->get_seqno = ring_get_seqno; + ring->irq_get = bsd_ring_get_irq; + ring->irq_put = bsd_ring_put_irq; + ring->dispatch_execbuffer = ring_dispatch_execbuffer; + } + ring->init = init_ring_common; + return intel_init_ring_buffer(dev, ring); } -- cgit v1.2.3-70-g09d2 From 3535d9dd5a19c2f7b88caf67d650bdaa0750b06c Mon Sep 17 00:00:00 2001 From: Daniel Vetter Date: Wed, 11 Apr 2012 22:12:50 +0200 Subject: drm/i915: dynamically set up blt ring functions and parameters Just for consistency. Reviewed-by: Eric Anholt Signed-off-by: Daniel Vetter --- drivers/gpu/drm/i915/intel_ringbuffer.c | 40 ++++++++++++++++----------------- 1 file changed, 19 insertions(+), 21 deletions(-) (limited to 'drivers/gpu/drm/i915/intel_ringbuffer.c') diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.c b/drivers/gpu/drm/i915/intel_ringbuffer.c index 683d1f0fcb3..5b11c53a8ee 100644 --- a/drivers/gpu/drm/i915/intel_ringbuffer.c +++ b/drivers/gpu/drm/i915/intel_ringbuffer.c @@ -1350,26 +1350,6 @@ static int blt_ring_flush(struct intel_ring_buffer *ring, return 0; } -static const struct intel_ring_buffer gen6_blt_ring = { - .name = "blt ring", - .id = BCS, - .mmio_base = BLT_RING_BASE, - .init = init_ring_common, - .write_tail = ring_write_tail, - .flush = blt_ring_flush, - .add_request = gen6_add_request, - .get_seqno = gen6_ring_get_seqno, - .irq_get = gen6_ring_get_irq, - .irq_put = gen6_ring_put_irq, - .irq_enable_mask = GEN6_BLITTER_USER_INTERRUPT, - .dispatch_execbuffer = gen6_ring_dispatch_execbuffer, - .sync_to = gen6_blt_ring_sync_to, - .semaphore_register = {MI_SEMAPHORE_SYNC_BR, - MI_SEMAPHORE_SYNC_BV, - MI_SEMAPHORE_SYNC_INVALID}, - .signal_mbox = {GEN6_RBSYNC, GEN6_VBSYNC}, -}; - int intel_init_render_ring_buffer(struct drm_device *dev) { drm_i915_private_t *dev_priv = dev->dev_private; @@ -1534,7 +1514,25 @@ int intel_init_blt_ring_buffer(struct drm_device *dev) drm_i915_private_t *dev_priv = dev->dev_private; struct intel_ring_buffer *ring = &dev_priv->ring[BCS]; - *ring = gen6_blt_ring; + ring->name = "blitter ring"; + ring->id = BCS; + + ring->mmio_base = BLT_RING_BASE; + ring->write_tail = ring_write_tail; + ring->flush = blt_ring_flush; + ring->add_request = gen6_add_request; + ring->get_seqno = gen6_ring_get_seqno; + ring->irq_enable_mask = GEN6_BLITTER_USER_INTERRUPT; + ring->irq_get = gen6_ring_get_irq; + ring->irq_put = gen6_ring_put_irq; + ring->dispatch_execbuffer = gen6_ring_dispatch_execbuffer; + ring->sync_to = gen6_blt_ring_sync_to; + ring->semaphore_register[0] = MI_SEMAPHORE_SYNC_BR; + ring->semaphore_register[1] = MI_SEMAPHORE_SYNC_BV; + ring->semaphore_register[2] = MI_SEMAPHORE_SYNC_INVALID; + ring->signal_mbox[0] = GEN6_RBSYNC; + ring->signal_mbox[1] = GEN6_VBSYNC; + ring->init = init_ring_common; return intel_init_ring_buffer(dev, ring); } -- cgit v1.2.3-70-g09d2 From b4178f8aaf84ae59d9b2a17a4a21aa19499f577a Mon Sep 17 00:00:00 2001 From: Daniel Vetter Date: Wed, 11 Apr 2012 22:12:51 +0200 Subject: drm/i915: don't set up rings on gen6+ for non-kms It's not supported, and with the patch to refuse loading on gen6+ without kms enabled, there's also no way we can hit this. Reviewed-by: Eric Anholt Signed-Off-by: Daniel Vetter --- drivers/gpu/drm/i915/intel_ringbuffer.c | 14 ++------------ 1 file changed, 2 insertions(+), 12 deletions(-) (limited to 'drivers/gpu/drm/i915/intel_ringbuffer.c') diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.c b/drivers/gpu/drm/i915/intel_ringbuffer.c index 5b11c53a8ee..be405f23dd5 100644 --- a/drivers/gpu/drm/i915/intel_ringbuffer.c +++ b/drivers/gpu/drm/i915/intel_ringbuffer.c @@ -1409,18 +1409,8 @@ int intel_render_ring_init_dri(struct drm_device *dev, u64 start, u32 size) ring->mmio_base = RENDER_RING_BASE; if (INTEL_INFO(dev)->gen >= 6) { - ring->add_request = gen6_add_request; - ring->flush = gen6_render_ring_flush; - ring->irq_get = gen6_ring_get_irq; - ring->irq_put = gen6_ring_put_irq; - ring->irq_enable_mask = GT_USER_INTERRUPT; - ring->get_seqno = gen6_ring_get_seqno; - ring->sync_to = render_ring_sync_to; - ring->semaphore_register[0] = MI_SEMAPHORE_SYNC_INVALID; - ring->semaphore_register[1] = MI_SEMAPHORE_SYNC_RV; - ring->semaphore_register[2] = MI_SEMAPHORE_SYNC_RB; - ring->signal_mbox[0] = GEN6_VRSYNC; - ring->signal_mbox[1] = GEN6_BRSYNC; + /* non-kms not supported on gen6+ */ + return -ENODEV; } else if (IS_GEN5(dev)) { ring->add_request = pc_render_add_request; ring->flush = render_ring_flush; -- cgit v1.2.3-70-g09d2 From 686cb5f9f5dc7db150c9f4e49cc0d45ef952450b Mon Sep 17 00:00:00 2001 From: Daniel Vetter Date: Wed, 11 Apr 2012 22:12:52 +0200 Subject: drm/i915: consolidate ring->sync-to functions The waiter is always the ring itself (otherwise we'd have a decent snafu in a callsite), so we can unify this easily. Also give it the usual gen6_ prefix, in case anyone is foolish enough to implement hw semaphores for gen5. Reviewed-by: Eric Anholt Signed-Off-by: Daniel Vetter --- drivers/gpu/drm/i915/intel_ringbuffer.c | 60 ++++++--------------------------- 1 file changed, 11 insertions(+), 49 deletions(-) (limited to 'drivers/gpu/drm/i915/intel_ringbuffer.c') diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.c b/drivers/gpu/drm/i915/intel_ringbuffer.c index be405f23dd5..bc33f13783b 100644 --- a/drivers/gpu/drm/i915/intel_ringbuffer.c +++ b/drivers/gpu/drm/i915/intel_ringbuffer.c @@ -472,10 +472,9 @@ gen6_add_request(struct intel_ring_buffer *ring, * @seqno - seqno which the waiter will block on */ static int -intel_ring_sync(struct intel_ring_buffer *waiter, - struct intel_ring_buffer *signaller, - int ring, - u32 seqno) +gen6_ring_sync(struct intel_ring_buffer *waiter, + struct intel_ring_buffer *signaller, + u32 seqno) { int ret; u32 dw1 = MI_SEMAPHORE_MBOX | @@ -488,11 +487,15 @@ intel_ring_sync(struct intel_ring_buffer *waiter, */ seqno -= 1; + WARN_ON(signaller->semaphore_register[waiter->id] == + MI_SEMAPHORE_SYNC_INVALID); + ret = intel_ring_begin(waiter, 4); if (ret) return ret; - intel_ring_emit(waiter, dw1 | signaller->semaphore_register[ring]); + intel_ring_emit(waiter, + dw1 | signaller->semaphore_register[waiter->id]); intel_ring_emit(waiter, seqno); intel_ring_emit(waiter, 0); intel_ring_emit(waiter, MI_NOOP); @@ -501,47 +504,6 @@ intel_ring_sync(struct intel_ring_buffer *waiter, return 0; } -/* VCS->RCS (RVSYNC) or BCS->RCS (RBSYNC) */ -int -render_ring_sync_to(struct intel_ring_buffer *waiter, - struct intel_ring_buffer *signaller, - u32 seqno) -{ - WARN_ON(signaller->semaphore_register[RCS] == MI_SEMAPHORE_SYNC_INVALID); - return intel_ring_sync(waiter, - signaller, - RCS, - seqno); -} - -/* RCS->VCS (VRSYNC) or BCS->VCS (VBSYNC) */ -int -gen6_bsd_ring_sync_to(struct intel_ring_buffer *waiter, - struct intel_ring_buffer *signaller, - u32 seqno) -{ - WARN_ON(signaller->semaphore_register[VCS] == MI_SEMAPHORE_SYNC_INVALID); - return intel_ring_sync(waiter, - signaller, - VCS, - seqno); -} - -/* RCS->BCS (BRSYNC) or VCS->BCS (BVSYNC) */ -int -gen6_blt_ring_sync_to(struct intel_ring_buffer *waiter, - struct intel_ring_buffer *signaller, - u32 seqno) -{ - WARN_ON(signaller->semaphore_register[BCS] == MI_SEMAPHORE_SYNC_INVALID); - return intel_ring_sync(waiter, - signaller, - BCS, - seqno); -} - - - #define PIPE_CONTROL_FLUSH(ring__, addr__) \ do { \ intel_ring_emit(ring__, GFX_OP_PIPE_CONTROL(4) | PIPE_CONTROL_QW_WRITE | \ @@ -1366,7 +1328,7 @@ int intel_init_render_ring_buffer(struct drm_device *dev) ring->irq_put = gen6_ring_put_irq; ring->irq_enable_mask = GT_USER_INTERRUPT; ring->get_seqno = gen6_ring_get_seqno; - ring->sync_to = render_ring_sync_to; + ring->sync_to = gen6_ring_sync; ring->semaphore_register[0] = MI_SEMAPHORE_SYNC_INVALID; ring->semaphore_register[1] = MI_SEMAPHORE_SYNC_RV; ring->semaphore_register[2] = MI_SEMAPHORE_SYNC_RB; @@ -1477,7 +1439,7 @@ int intel_init_bsd_ring_buffer(struct drm_device *dev) ring->irq_get = gen6_ring_get_irq; ring->irq_put = gen6_ring_put_irq; ring->dispatch_execbuffer = gen6_ring_dispatch_execbuffer; - ring->sync_to = gen6_bsd_ring_sync_to; + ring->sync_to = gen6_ring_sync; ring->semaphore_register[0] = MI_SEMAPHORE_SYNC_VR; ring->semaphore_register[1] = MI_SEMAPHORE_SYNC_INVALID; ring->semaphore_register[2] = MI_SEMAPHORE_SYNC_VB; @@ -1516,7 +1478,7 @@ int intel_init_blt_ring_buffer(struct drm_device *dev) ring->irq_get = gen6_ring_get_irq; ring->irq_put = gen6_ring_put_irq; ring->dispatch_execbuffer = gen6_ring_dispatch_execbuffer; - ring->sync_to = gen6_blt_ring_sync_to; + ring->sync_to = gen6_ring_sync; ring->semaphore_register[0] = MI_SEMAPHORE_SYNC_BR; ring->semaphore_register[1] = MI_SEMAPHORE_SYNC_BV; ring->semaphore_register[2] = MI_SEMAPHORE_SYNC_INVALID; -- cgit v1.2.3-70-g09d2 From e367031966c3546b213f6699b83669739cb6fb1d Mon Sep 17 00:00:00 2001 From: Daniel Vetter Date: Wed, 11 Apr 2012 22:12:53 +0200 Subject: drm/i915: abstract away ring-specific irq_get/put Inspired by Ben Widawsky's patch for gen6+. Now after restructuring how we set up the ring vtables and parameters, we can do this right. This kills the bsd specific get/put_irq functions, they're now the same. Reviewed-by: Eric Anholt Signed-Off-by: Daniel Vetter --- drivers/gpu/drm/i915/intel_ringbuffer.c | 77 ++++++++++----------------------- 1 file changed, 24 insertions(+), 53 deletions(-) (limited to 'drivers/gpu/drm/i915/intel_ringbuffer.c') diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.c b/drivers/gpu/drm/i915/intel_ringbuffer.c index bc33f13783b..a09e8aa0db8 100644 --- a/drivers/gpu/drm/i915/intel_ringbuffer.c +++ b/drivers/gpu/drm/i915/intel_ringbuffer.c @@ -645,7 +645,7 @@ i915_disable_irq(drm_i915_private_t *dev_priv, u32 mask) } static bool -render_ring_get_irq(struct intel_ring_buffer *ring) +i9xx_ring_get_irq(struct intel_ring_buffer *ring) { struct drm_device *dev = ring->dev; drm_i915_private_t *dev_priv = dev->dev_private; @@ -657,9 +657,9 @@ render_ring_get_irq(struct intel_ring_buffer *ring) if (ring->irq_refcount++ == 0) { if (INTEL_INFO(dev)->gen >= 5) ironlake_enable_irq(dev_priv, - GT_PIPE_NOTIFY | GT_USER_INTERRUPT); + ring->irq_enable_mask); else - i915_enable_irq(dev_priv, I915_USER_INTERRUPT); + i915_enable_irq(dev_priv, ring->irq_enable_mask); } spin_unlock(&ring->irq_lock); @@ -667,7 +667,7 @@ render_ring_get_irq(struct intel_ring_buffer *ring) } static void -render_ring_put_irq(struct intel_ring_buffer *ring) +i9xx_ring_put_irq(struct intel_ring_buffer *ring) { struct drm_device *dev = ring->dev; drm_i915_private_t *dev_priv = dev->dev_private; @@ -676,10 +676,9 @@ render_ring_put_irq(struct intel_ring_buffer *ring) if (--ring->irq_refcount == 0) { if (INTEL_INFO(dev)->gen >= 5) ironlake_disable_irq(dev_priv, - GT_USER_INTERRUPT | - GT_PIPE_NOTIFY); + ring->irq_enable_mask); else - i915_disable_irq(dev_priv, I915_USER_INTERRUPT); + i915_disable_irq(dev_priv, ring->irq_enable_mask); } spin_unlock(&ring->irq_lock); } @@ -795,42 +794,6 @@ gen6_ring_put_irq(struct intel_ring_buffer *ring) gen6_gt_force_wake_put(dev_priv); } -static bool -bsd_ring_get_irq(struct intel_ring_buffer *ring) -{ - struct drm_device *dev = ring->dev; - drm_i915_private_t *dev_priv = dev->dev_private; - - if (!dev->irq_enabled) - return false; - - spin_lock(&ring->irq_lock); - if (ring->irq_refcount++ == 0) { - if (IS_G4X(dev)) - i915_enable_irq(dev_priv, I915_BSD_USER_INTERRUPT); - else - ironlake_enable_irq(dev_priv, GT_BSD_USER_INTERRUPT); - } - spin_unlock(&ring->irq_lock); - - return true; -} -static void -bsd_ring_put_irq(struct intel_ring_buffer *ring) -{ - struct drm_device *dev = ring->dev; - drm_i915_private_t *dev_priv = dev->dev_private; - - spin_lock(&ring->irq_lock); - if (--ring->irq_refcount == 0) { - if (IS_G4X(dev)) - i915_disable_irq(dev_priv, I915_BSD_USER_INTERRUPT); - else - ironlake_disable_irq(dev_priv, GT_BSD_USER_INTERRUPT); - } - spin_unlock(&ring->irq_lock); -} - static int ring_dispatch_execbuffer(struct intel_ring_buffer *ring, u32 offset, u32 length) { @@ -1338,14 +1301,16 @@ int intel_init_render_ring_buffer(struct drm_device *dev) ring->add_request = pc_render_add_request; ring->flush = render_ring_flush; ring->get_seqno = pc_render_get_seqno; - ring->irq_get = render_ring_get_irq; - ring->irq_put = render_ring_put_irq; + ring->irq_get = i9xx_ring_get_irq; + ring->irq_put = i9xx_ring_put_irq; + ring->irq_enable_mask = GT_USER_INTERRUPT | GT_PIPE_NOTIFY; } else { ring->add_request = render_ring_add_request; ring->flush = render_ring_flush; ring->get_seqno = ring_get_seqno; - ring->irq_get = render_ring_get_irq; - ring->irq_put = render_ring_put_irq; + ring->irq_get = i9xx_ring_get_irq; + ring->irq_put = i9xx_ring_put_irq; + ring->irq_enable_mask = I915_USER_INTERRUPT; } ring->write_tail = ring_write_tail; ring->dispatch_execbuffer = render_ring_dispatch_execbuffer; @@ -1377,14 +1342,16 @@ int intel_render_ring_init_dri(struct drm_device *dev, u64 start, u32 size) ring->add_request = pc_render_add_request; ring->flush = render_ring_flush; ring->get_seqno = pc_render_get_seqno; - ring->irq_get = render_ring_get_irq; - ring->irq_put = render_ring_put_irq; + ring->irq_get = i9xx_ring_get_irq; + ring->irq_put = i9xx_ring_put_irq; + ring->irq_enable_mask = GT_USER_INTERRUPT | GT_PIPE_NOTIFY; } else { ring->add_request = render_ring_add_request; ring->flush = render_ring_flush; ring->get_seqno = ring_get_seqno; - ring->irq_get = render_ring_get_irq; - ring->irq_put = render_ring_put_irq; + ring->irq_get = i9xx_ring_get_irq; + ring->irq_put = i9xx_ring_put_irq; + ring->irq_enable_mask = I915_USER_INTERRUPT; } ring->write_tail = ring_write_tail; ring->dispatch_execbuffer = render_ring_dispatch_execbuffer; @@ -1451,8 +1418,12 @@ int intel_init_bsd_ring_buffer(struct drm_device *dev) ring->flush = bsd_ring_flush; ring->add_request = ring_add_request; ring->get_seqno = ring_get_seqno; - ring->irq_get = bsd_ring_get_irq; - ring->irq_put = bsd_ring_put_irq; + ring->irq_get = i9xx_ring_get_irq; + ring->irq_put = i9xx_ring_put_irq; + if (IS_GEN5(dev)) + ring->irq_enable_mask = GT_BSD_USER_INTERRUPT; + else + ring->irq_enable_mask = I915_BSD_USER_INTERRUPT; ring->dispatch_execbuffer = ring_dispatch_execbuffer; } ring->init = init_ring_common; -- cgit v1.2.3-70-g09d2 From e48d86347c602c55159714f6ddcd88969a1b2f21 Mon Sep 17 00:00:00 2001 From: Daniel Vetter Date: Wed, 11 Apr 2012 22:12:54 +0200 Subject: drm/i915: split out the gen5 ring irq get/put functions Now that we have sensibly split up, we can nicely get rid of that ugly is_gen5 check. Reviewed-by: Eric Anholt Signed-Off-by: Daniel Vetter --- drivers/gpu/drm/i915/intel_ringbuffer.c | 66 ++++++++++++++++++++++----------- 1 file changed, 44 insertions(+), 22 deletions(-) (limited to 'drivers/gpu/drm/i915/intel_ringbuffer.c') diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.c b/drivers/gpu/drm/i915/intel_ringbuffer.c index a09e8aa0db8..7e1f2211ea2 100644 --- a/drivers/gpu/drm/i915/intel_ringbuffer.c +++ b/drivers/gpu/drm/i915/intel_ringbuffer.c @@ -644,6 +644,35 @@ i915_disable_irq(drm_i915_private_t *dev_priv, u32 mask) POSTING_READ(IMR); } +static bool +gen5_ring_get_irq(struct intel_ring_buffer *ring) +{ + struct drm_device *dev = ring->dev; + drm_i915_private_t *dev_priv = dev->dev_private; + + if (!dev->irq_enabled) + return false; + + spin_lock(&ring->irq_lock); + if (ring->irq_refcount++ == 0) + ironlake_enable_irq(dev_priv, ring->irq_enable_mask); + spin_unlock(&ring->irq_lock); + + return true; +} + +static void +gen5_ring_put_irq(struct intel_ring_buffer *ring) +{ + struct drm_device *dev = ring->dev; + drm_i915_private_t *dev_priv = dev->dev_private; + + spin_lock(&ring->irq_lock); + if (--ring->irq_refcount == 0) + ironlake_disable_irq(dev_priv, ring->irq_enable_mask); + spin_unlock(&ring->irq_lock); +} + static bool i9xx_ring_get_irq(struct intel_ring_buffer *ring) { @@ -654,13 +683,8 @@ i9xx_ring_get_irq(struct intel_ring_buffer *ring) return false; spin_lock(&ring->irq_lock); - if (ring->irq_refcount++ == 0) { - if (INTEL_INFO(dev)->gen >= 5) - ironlake_enable_irq(dev_priv, - ring->irq_enable_mask); - else - i915_enable_irq(dev_priv, ring->irq_enable_mask); - } + if (ring->irq_refcount++ == 0) + i915_enable_irq(dev_priv, ring->irq_enable_mask); spin_unlock(&ring->irq_lock); return true; @@ -673,13 +697,8 @@ i9xx_ring_put_irq(struct intel_ring_buffer *ring) drm_i915_private_t *dev_priv = dev->dev_private; spin_lock(&ring->irq_lock); - if (--ring->irq_refcount == 0) { - if (INTEL_INFO(dev)->gen >= 5) - ironlake_disable_irq(dev_priv, - ring->irq_enable_mask); - else - i915_disable_irq(dev_priv, ring->irq_enable_mask); - } + if (--ring->irq_refcount == 0) + i915_disable_irq(dev_priv, ring->irq_enable_mask); spin_unlock(&ring->irq_lock); } @@ -1301,8 +1320,8 @@ int intel_init_render_ring_buffer(struct drm_device *dev) ring->add_request = pc_render_add_request; ring->flush = render_ring_flush; ring->get_seqno = pc_render_get_seqno; - ring->irq_get = i9xx_ring_get_irq; - ring->irq_put = i9xx_ring_put_irq; + ring->irq_get = gen5_ring_get_irq; + ring->irq_put = gen5_ring_put_irq; ring->irq_enable_mask = GT_USER_INTERRUPT | GT_PIPE_NOTIFY; } else { ring->add_request = render_ring_add_request; @@ -1342,8 +1361,8 @@ int intel_render_ring_init_dri(struct drm_device *dev, u64 start, u32 size) ring->add_request = pc_render_add_request; ring->flush = render_ring_flush; ring->get_seqno = pc_render_get_seqno; - ring->irq_get = i9xx_ring_get_irq; - ring->irq_put = i9xx_ring_put_irq; + ring->irq_get = gen5_ring_get_irq; + ring->irq_put = gen5_ring_put_irq; ring->irq_enable_mask = GT_USER_INTERRUPT | GT_PIPE_NOTIFY; } else { ring->add_request = render_ring_add_request; @@ -1418,12 +1437,15 @@ int intel_init_bsd_ring_buffer(struct drm_device *dev) ring->flush = bsd_ring_flush; ring->add_request = ring_add_request; ring->get_seqno = ring_get_seqno; - ring->irq_get = i9xx_ring_get_irq; - ring->irq_put = i9xx_ring_put_irq; - if (IS_GEN5(dev)) + if (IS_GEN5(dev)) { ring->irq_enable_mask = GT_BSD_USER_INTERRUPT; - else + ring->irq_get = gen5_ring_get_irq; + ring->irq_put = gen5_ring_put_irq; + } else { ring->irq_enable_mask = I915_BSD_USER_INTERRUPT; + ring->irq_get = i9xx_ring_get_irq; + ring->irq_put = i9xx_ring_put_irq; + } ring->dispatch_execbuffer = ring_dispatch_execbuffer; } ring->init = init_ring_common; -- cgit v1.2.3-70-g09d2 From 0fd2c201482e62492f2d7dc6c2798cf7f66c9570 Mon Sep 17 00:00:00 2001 From: Daniel Vetter Date: Wed, 11 Apr 2012 22:12:55 +0200 Subject: drm/i915: don't enable the gen6 bsd ring tail write enable on gen7 HW engineers have fixed this issue for ivb. Again, a nice cleanup possible thanks to the more flexible ring initialization. Reviewed-by: Eric Anholt Signed-Off-by: Daniel Vetter --- drivers/gpu/drm/i915/intel_ringbuffer.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) (limited to 'drivers/gpu/drm/i915/intel_ringbuffer.c') diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.c b/drivers/gpu/drm/i915/intel_ringbuffer.c index 7e1f2211ea2..68e1255452f 100644 --- a/drivers/gpu/drm/i915/intel_ringbuffer.c +++ b/drivers/gpu/drm/i915/intel_ringbuffer.c @@ -1415,9 +1415,12 @@ int intel_init_bsd_ring_buffer(struct drm_device *dev) ring->name = "bsd ring"; ring->id = VCS; + ring->write_tail = ring_write_tail; if (IS_GEN6(dev) || IS_GEN7(dev)) { ring->mmio_base = GEN6_BSD_RING_BASE; - ring->write_tail = gen6_bsd_ring_write_tail; + /* gen6 bsd needs a special wa for tail updates */ + if (IS_GEN6(dev)) + ring->write_tail = gen6_bsd_ring_write_tail; ring->flush = gen6_ring_flush; ring->add_request = gen6_add_request; ring->get_seqno = gen6_ring_get_seqno; @@ -1433,7 +1436,6 @@ int intel_init_bsd_ring_buffer(struct drm_device *dev) ring->signal_mbox[1] = GEN6_BVSYNC; } else { ring->mmio_base = BSD_RING_BASE; - ring->write_tail = ring_write_tail; ring->flush = bsd_ring_flush; ring->add_request = ring_add_request; ring->get_seqno = ring_get_seqno; -- cgit v1.2.3-70-g09d2 From fb3256da8db2cb70be7d2c7f058b8676a9a85aaf Mon Sep 17 00:00:00 2001 From: Daniel Vetter Date: Wed, 11 Apr 2012 22:12:56 +0200 Subject: drm/i915: split up ring->dispatch_execbuffer functions Now that we can, we should split them up in a way that makes some sense and banishes the IS_ checks into init code. Reviewed-by: Eric Anholt Signed-Off-by: Daniel Vetter --- drivers/gpu/drm/i915/intel_ringbuffer.c | 69 +++++++++++++++++++-------------- 1 file changed, 40 insertions(+), 29 deletions(-) (limited to 'drivers/gpu/drm/i915/intel_ringbuffer.c') diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.c b/drivers/gpu/drm/i915/intel_ringbuffer.c index 68e1255452f..8cffd0b9aaf 100644 --- a/drivers/gpu/drm/i915/intel_ringbuffer.c +++ b/drivers/gpu/drm/i915/intel_ringbuffer.c @@ -814,7 +814,7 @@ gen6_ring_put_irq(struct intel_ring_buffer *ring) } static int -ring_dispatch_execbuffer(struct intel_ring_buffer *ring, u32 offset, u32 length) +i965_dispatch_execbuffer(struct intel_ring_buffer *ring, u32 offset, u32 length) { int ret; @@ -832,37 +832,36 @@ ring_dispatch_execbuffer(struct intel_ring_buffer *ring, u32 offset, u32 length) } static int -render_ring_dispatch_execbuffer(struct intel_ring_buffer *ring, +i830_dispatch_execbuffer(struct intel_ring_buffer *ring, u32 offset, u32 len) { - struct drm_device *dev = ring->dev; int ret; - if (IS_I830(dev) || IS_845G(dev)) { - ret = intel_ring_begin(ring, 4); - if (ret) - return ret; + ret = intel_ring_begin(ring, 4); + if (ret) + return ret; - intel_ring_emit(ring, MI_BATCH_BUFFER); - intel_ring_emit(ring, offset | MI_BATCH_NON_SECURE); - intel_ring_emit(ring, offset + len - 8); - intel_ring_emit(ring, 0); - } else { - ret = intel_ring_begin(ring, 2); - if (ret) - return ret; + intel_ring_emit(ring, MI_BATCH_BUFFER); + intel_ring_emit(ring, offset | MI_BATCH_NON_SECURE); + intel_ring_emit(ring, offset + len - 8); + intel_ring_emit(ring, 0); + intel_ring_advance(ring); - if (INTEL_INFO(dev)->gen >= 4) { - intel_ring_emit(ring, - MI_BATCH_BUFFER_START | (2 << 6) | - MI_BATCH_NON_SECURE_I965); - intel_ring_emit(ring, offset); - } else { - intel_ring_emit(ring, - MI_BATCH_BUFFER_START | (2 << 6)); - intel_ring_emit(ring, offset | MI_BATCH_NON_SECURE); - } - } + return 0; +} + +static int +i915_dispatch_execbuffer(struct intel_ring_buffer *ring, + u32 offset, u32 len) +{ + int ret; + + ret = intel_ring_begin(ring, 2); + if (ret) + return ret; + + intel_ring_emit(ring, MI_BATCH_BUFFER_START | (2 << 6)); + intel_ring_emit(ring, offset | MI_BATCH_NON_SECURE); intel_ring_advance(ring); return 0; @@ -1332,7 +1331,14 @@ int intel_init_render_ring_buffer(struct drm_device *dev) ring->irq_enable_mask = I915_USER_INTERRUPT; } ring->write_tail = ring_write_tail; - ring->dispatch_execbuffer = render_ring_dispatch_execbuffer; + if (INTEL_INFO(dev)->gen >= 6) + ring->dispatch_execbuffer = gen6_ring_dispatch_execbuffer; + else if (INTEL_INFO(dev)->gen >= 4) + ring->dispatch_execbuffer = i965_dispatch_execbuffer; + else if (IS_I830(dev) || IS_845G(dev)) + ring->dispatch_execbuffer = i830_dispatch_execbuffer; + else + ring->dispatch_execbuffer = i915_dispatch_execbuffer; ring->init = init_render_ring; ring->cleanup = render_ring_cleanup; @@ -1373,7 +1379,12 @@ int intel_render_ring_init_dri(struct drm_device *dev, u64 start, u32 size) ring->irq_enable_mask = I915_USER_INTERRUPT; } ring->write_tail = ring_write_tail; - ring->dispatch_execbuffer = render_ring_dispatch_execbuffer; + if (INTEL_INFO(dev)->gen >= 4) + ring->dispatch_execbuffer = i965_dispatch_execbuffer; + else if (IS_I830(dev) || IS_845G(dev)) + ring->dispatch_execbuffer = i830_dispatch_execbuffer; + else + ring->dispatch_execbuffer = i915_dispatch_execbuffer; ring->init = init_render_ring; ring->cleanup = render_ring_cleanup; @@ -1448,7 +1459,7 @@ int intel_init_bsd_ring_buffer(struct drm_device *dev) ring->irq_get = i9xx_ring_get_irq; ring->irq_put = i9xx_ring_put_irq; } - ring->dispatch_execbuffer = ring_dispatch_execbuffer; + ring->dispatch_execbuffer = i965_dispatch_execbuffer; } ring->init = init_ring_common; -- cgit v1.2.3-70-g09d2 From 8620a3a908da34e1a0db9ad457136bc96340911f Mon Sep 17 00:00:00 2001 From: Daniel Vetter Date: Wed, 11 Apr 2012 22:12:57 +0200 Subject: drm/i915: consolidate ring->add_request a bit They're indentical, so just kill one. Also give the other a prefix to distinguish it from the gen6+ functions - this add_request function is not really generic code. v2: Fixup commit message as noted by Ben Widawsky. Reviewed-by: Eric Anholt Signed-Off-by: Daniel Vetter --- drivers/gpu/drm/i915/intel_ringbuffer.c | 29 ++++------------------------- 1 file changed, 4 insertions(+), 25 deletions(-) (limited to 'drivers/gpu/drm/i915/intel_ringbuffer.c') diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.c b/drivers/gpu/drm/i915/intel_ringbuffer.c index 8cffd0b9aaf..ac05c749104 100644 --- a/drivers/gpu/drm/i915/intel_ringbuffer.c +++ b/drivers/gpu/drm/i915/intel_ringbuffer.c @@ -565,27 +565,6 @@ pc_render_add_request(struct intel_ring_buffer *ring, return 0; } -static int -render_ring_add_request(struct intel_ring_buffer *ring, - u32 *result) -{ - u32 seqno = i915_gem_next_request_seqno(ring); - int ret; - - ret = intel_ring_begin(ring, 4); - if (ret) - return ret; - - intel_ring_emit(ring, MI_STORE_DWORD_INDEX); - intel_ring_emit(ring, I915_GEM_HWS_INDEX << MI_STORE_DWORD_INDEX_SHIFT); - intel_ring_emit(ring, seqno); - intel_ring_emit(ring, MI_USER_INTERRUPT); - intel_ring_advance(ring); - - *result = seqno; - return 0; -} - static u32 gen6_ring_get_seqno(struct intel_ring_buffer *ring) { @@ -751,7 +730,7 @@ bsd_ring_flush(struct intel_ring_buffer *ring, } static int -ring_add_request(struct intel_ring_buffer *ring, +i9xx_add_request(struct intel_ring_buffer *ring, u32 *result) { u32 seqno; @@ -1323,7 +1302,7 @@ int intel_init_render_ring_buffer(struct drm_device *dev) ring->irq_put = gen5_ring_put_irq; ring->irq_enable_mask = GT_USER_INTERRUPT | GT_PIPE_NOTIFY; } else { - ring->add_request = render_ring_add_request; + ring->add_request = i9xx_add_request; ring->flush = render_ring_flush; ring->get_seqno = ring_get_seqno; ring->irq_get = i9xx_ring_get_irq; @@ -1371,7 +1350,7 @@ int intel_render_ring_init_dri(struct drm_device *dev, u64 start, u32 size) ring->irq_put = gen5_ring_put_irq; ring->irq_enable_mask = GT_USER_INTERRUPT | GT_PIPE_NOTIFY; } else { - ring->add_request = render_ring_add_request; + ring->add_request = i9xx_add_request; ring->flush = render_ring_flush; ring->get_seqno = ring_get_seqno; ring->irq_get = i9xx_ring_get_irq; @@ -1448,7 +1427,7 @@ int intel_init_bsd_ring_buffer(struct drm_device *dev) } else { ring->mmio_base = BSD_RING_BASE; ring->flush = bsd_ring_flush; - ring->add_request = ring_add_request; + ring->add_request = i9xx_add_request; ring->get_seqno = ring_get_seqno; if (IS_GEN5(dev)) { ring->irq_enable_mask = GT_BSD_USER_INTERRUPT; -- cgit v1.2.3-70-g09d2 From 28f0cbf71f5ed9b080878c04158c754b090a674a Mon Sep 17 00:00:00 2001 From: Daniel Vetter Date: Wed, 11 Apr 2012 22:12:58 +0200 Subject: drm/i915: don't set up gem ring functions on gen5 for !kms We already disallow initialition of gem in this case in the corresponding ioctl, so don't bother setting up the gem support ring functions in the legacy dri render ring init. Reviewed-by: Eric Anholt Signed-Off-by: Daniel Vetter --- drivers/gpu/drm/i915/intel_ringbuffer.c | 24 ++++++++++-------------- 1 file changed, 10 insertions(+), 14 deletions(-) (limited to 'drivers/gpu/drm/i915/intel_ringbuffer.c') diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.c b/drivers/gpu/drm/i915/intel_ringbuffer.c index ac05c749104..6b3ff37e752 100644 --- a/drivers/gpu/drm/i915/intel_ringbuffer.c +++ b/drivers/gpu/drm/i915/intel_ringbuffer.c @@ -1342,21 +1342,17 @@ int intel_render_ring_init_dri(struct drm_device *dev, u64 start, u32 size) if (INTEL_INFO(dev)->gen >= 6) { /* non-kms not supported on gen6+ */ return -ENODEV; - } else if (IS_GEN5(dev)) { - ring->add_request = pc_render_add_request; - ring->flush = render_ring_flush; - ring->get_seqno = pc_render_get_seqno; - ring->irq_get = gen5_ring_get_irq; - ring->irq_put = gen5_ring_put_irq; - ring->irq_enable_mask = GT_USER_INTERRUPT | GT_PIPE_NOTIFY; - } else { - ring->add_request = i9xx_add_request; - ring->flush = render_ring_flush; - ring->get_seqno = ring_get_seqno; - ring->irq_get = i9xx_ring_get_irq; - ring->irq_put = i9xx_ring_put_irq; - ring->irq_enable_mask = I915_USER_INTERRUPT; } + + /* Note: gem is not supported on gen5/ilk without kms (the corresponding + * gem_init ioctl returns with -ENODEV). Hence we do not need to set up + * the special gen5 functions. */ + ring->add_request = i9xx_add_request; + ring->flush = render_ring_flush; + ring->get_seqno = ring_get_seqno; + ring->irq_get = i9xx_ring_get_irq; + ring->irq_put = i9xx_ring_put_irq; + ring->irq_enable_mask = I915_USER_INTERRUPT; ring->write_tail = ring_write_tail; if (INTEL_INFO(dev)->gen >= 4) ring->dispatch_execbuffer = i965_dispatch_execbuffer; -- cgit v1.2.3-70-g09d2 From f637fde434c9e3687798730c7ddd367e93666013 Mon Sep 17 00:00:00 2001 From: Daniel Vetter Date: Wed, 11 Apr 2012 22:12:59 +0200 Subject: drm/i915: inline enable/disable_irq into ring->get/put_irq Now that these are properly refactored this additional indirection doesn't really buy us anything but confusion. Hence inline them. This duplicates the ironlake gt enable/disable code snippet, but we've already separate ilk from gen6+ gt irq in i915_irq.c, so I think this makes more sense. Reviewed-by: Eric Anholt Signed-Off-by: Daniel Vetter --- drivers/gpu/drm/i915/intel_ringbuffer.c | 68 +++++++++++++-------------------- 1 file changed, 26 insertions(+), 42 deletions(-) (limited to 'drivers/gpu/drm/i915/intel_ringbuffer.c') diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.c b/drivers/gpu/drm/i915/intel_ringbuffer.c index 6b3ff37e752..6c144574a5b 100644 --- a/drivers/gpu/drm/i915/intel_ringbuffer.c +++ b/drivers/gpu/drm/i915/intel_ringbuffer.c @@ -591,38 +591,6 @@ pc_render_get_seqno(struct intel_ring_buffer *ring) return pc->cpu_page[0]; } -static void -ironlake_enable_irq(drm_i915_private_t *dev_priv, u32 mask) -{ - dev_priv->gt_irq_mask &= ~mask; - I915_WRITE(GTIMR, dev_priv->gt_irq_mask); - POSTING_READ(GTIMR); -} - -static void -ironlake_disable_irq(drm_i915_private_t *dev_priv, u32 mask) -{ - dev_priv->gt_irq_mask |= mask; - I915_WRITE(GTIMR, dev_priv->gt_irq_mask); - POSTING_READ(GTIMR); -} - -static void -i915_enable_irq(drm_i915_private_t *dev_priv, u32 mask) -{ - dev_priv->irq_mask &= ~mask; - I915_WRITE(IMR, dev_priv->irq_mask); - POSTING_READ(IMR); -} - -static void -i915_disable_irq(drm_i915_private_t *dev_priv, u32 mask) -{ - dev_priv->irq_mask |= mask; - I915_WRITE(IMR, dev_priv->irq_mask); - POSTING_READ(IMR); -} - static bool gen5_ring_get_irq(struct intel_ring_buffer *ring) { @@ -633,8 +601,11 @@ gen5_ring_get_irq(struct intel_ring_buffer *ring) return false; spin_lock(&ring->irq_lock); - if (ring->irq_refcount++ == 0) - ironlake_enable_irq(dev_priv, ring->irq_enable_mask); + if (ring->irq_refcount++ == 0) { + dev_priv->gt_irq_mask &= ~ring->irq_enable_mask; + I915_WRITE(GTIMR, dev_priv->gt_irq_mask); + POSTING_READ(GTIMR); + } spin_unlock(&ring->irq_lock); return true; @@ -647,8 +618,11 @@ gen5_ring_put_irq(struct intel_ring_buffer *ring) drm_i915_private_t *dev_priv = dev->dev_private; spin_lock(&ring->irq_lock); - if (--ring->irq_refcount == 0) - ironlake_disable_irq(dev_priv, ring->irq_enable_mask); + if (--ring->irq_refcount == 0) { + dev_priv->gt_irq_mask |= ring->irq_enable_mask; + I915_WRITE(GTIMR, dev_priv->gt_irq_mask); + POSTING_READ(GTIMR); + } spin_unlock(&ring->irq_lock); } @@ -662,8 +636,11 @@ i9xx_ring_get_irq(struct intel_ring_buffer *ring) return false; spin_lock(&ring->irq_lock); - if (ring->irq_refcount++ == 0) - i915_enable_irq(dev_priv, ring->irq_enable_mask); + if (ring->irq_refcount++ == 0) { + dev_priv->irq_mask &= ~ring->irq_enable_mask; + I915_WRITE(IMR, dev_priv->irq_mask); + POSTING_READ(IMR); + } spin_unlock(&ring->irq_lock); return true; @@ -676,8 +653,11 @@ i9xx_ring_put_irq(struct intel_ring_buffer *ring) drm_i915_private_t *dev_priv = dev->dev_private; spin_lock(&ring->irq_lock); - if (--ring->irq_refcount == 0) - i915_disable_irq(dev_priv, ring->irq_enable_mask); + if (--ring->irq_refcount == 0) { + dev_priv->irq_mask |= ring->irq_enable_mask; + I915_WRITE(IMR, dev_priv->irq_mask); + POSTING_READ(IMR); + } spin_unlock(&ring->irq_lock); } @@ -769,7 +749,9 @@ gen6_ring_get_irq(struct intel_ring_buffer *ring) spin_lock(&ring->irq_lock); if (ring->irq_refcount++ == 0) { I915_WRITE_IMR(ring, ~ring->irq_enable_mask); - ironlake_enable_irq(dev_priv, ring->irq_enable_mask); + dev_priv->gt_irq_mask &= ~ring->irq_enable_mask; + I915_WRITE(GTIMR, dev_priv->gt_irq_mask); + POSTING_READ(GTIMR); } spin_unlock(&ring->irq_lock); @@ -785,7 +767,9 @@ gen6_ring_put_irq(struct intel_ring_buffer *ring) spin_lock(&ring->irq_lock); if (--ring->irq_refcount == 0) { I915_WRITE_IMR(ring, ~0); - ironlake_disable_irq(dev_priv, ring->irq_enable_mask); + dev_priv->gt_irq_mask |= ring->irq_enable_mask; + I915_WRITE(GTIMR, dev_priv->gt_irq_mask); + POSTING_READ(GTIMR); } spin_unlock(&ring->irq_lock); -- cgit v1.2.3-70-g09d2 From c43b5634037ff00c83cde4ab7fc5e46831c846ef Mon Sep 17 00:00:00 2001 From: Ben Widawsky Date: Mon, 16 Apr 2012 14:07:40 -0700 Subject: drm/i915: [sparse] trivial sparse fixes This should contain all the changes which require no thought to make sparse happy. Signed-off-by: Ben Widawsky Reviewed-by: Chris Wilson Signed-off-by: Daniel Vetter --- drivers/gpu/drm/i915/i915_debugfs.c | 2 +- drivers/gpu/drm/i915/i915_drv.h | 2 ++ drivers/gpu/drm/i915/i915_ioc32.c | 5 ++++- drivers/gpu/drm/i915/i915_trace_points.c | 2 ++ drivers/gpu/drm/i915/intel_acpi.c | 1 + drivers/gpu/drm/i915/intel_display.c | 2 +- drivers/gpu/drm/i915/intel_fb.c | 2 +- drivers/gpu/drm/i915/intel_ringbuffer.c | 4 ++-- drivers/gpu/drm/i915/intel_sdvo.c | 2 +- 9 files changed, 15 insertions(+), 7 deletions(-) (limited to 'drivers/gpu/drm/i915/intel_ringbuffer.c') diff --git a/drivers/gpu/drm/i915/i915_debugfs.c b/drivers/gpu/drm/i915/i915_debugfs.c index 97c963082d4..35462df7cef 100644 --- a/drivers/gpu/drm/i915/i915_debugfs.c +++ b/drivers/gpu/drm/i915/i915_debugfs.c @@ -1832,7 +1832,7 @@ static int i915_forcewake_open(struct inode *inode, struct file *file) return 0; } -int i915_forcewake_release(struct inode *inode, struct file *file) +static int i915_forcewake_release(struct inode *inode, struct file *file) { struct drm_device *dev = inode->i_private; struct drm_i915_private *dev_priv = dev->dev_private; diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h index 422f424deb4..303cee71b49 100644 --- a/drivers/gpu/drm/i915/i915_drv.h +++ b/drivers/gpu/drm/i915/i915_drv.h @@ -1128,8 +1128,10 @@ extern void i915_driver_preclose(struct drm_device *dev, extern void i915_driver_postclose(struct drm_device *dev, struct drm_file *file_priv); extern int i915_driver_device_is_agp(struct drm_device * dev); +#ifdef CONFIG_COMPAT extern long i915_compat_ioctl(struct file *filp, unsigned int cmd, unsigned long arg); +#endif extern int i915_emit_box(struct drm_device *dev, struct drm_clip_rect *box, int DR1, int DR4); diff --git a/drivers/gpu/drm/i915/i915_ioc32.c b/drivers/gpu/drm/i915/i915_ioc32.c index 13b028994b2..0e72abb9f70 100644 --- a/drivers/gpu/drm/i915/i915_ioc32.c +++ b/drivers/gpu/drm/i915/i915_ioc32.c @@ -34,6 +34,7 @@ #include "drmP.h" #include "drm.h" #include "i915_drm.h" +#include "i915_drv.h" typedef struct _drm_i915_batchbuffer32 { int start; /* agp offset */ @@ -181,7 +182,7 @@ static int compat_i915_alloc(struct file *file, unsigned int cmd, (unsigned long)request); } -drm_ioctl_compat_t *i915_compat_ioctls[] = { +static drm_ioctl_compat_t *i915_compat_ioctls[] = { [DRM_I915_BATCHBUFFER] = compat_i915_batchbuffer, [DRM_I915_CMDBUFFER] = compat_i915_cmdbuffer, [DRM_I915_GETPARAM] = compat_i915_getparam, @@ -189,6 +190,7 @@ drm_ioctl_compat_t *i915_compat_ioctls[] = { [DRM_I915_ALLOC] = compat_i915_alloc }; +#ifdef CONFIG_COMPAT /** * Called whenever a 32-bit process running under a 64-bit kernel * performs an ioctl on /dev/dri/card. @@ -217,3 +219,4 @@ long i915_compat_ioctl(struct file *filp, unsigned int cmd, unsigned long arg) return ret; } +#endif diff --git a/drivers/gpu/drm/i915/i915_trace_points.c b/drivers/gpu/drm/i915/i915_trace_points.c index ead876eb6ea..f1df2bd4ecf 100644 --- a/drivers/gpu/drm/i915/i915_trace_points.c +++ b/drivers/gpu/drm/i915/i915_trace_points.c @@ -7,5 +7,7 @@ #include "i915_drv.h" +#ifndef __CHECKER__ #define CREATE_TRACE_POINTS #include "i915_trace.h" +#endif diff --git a/drivers/gpu/drm/i915/intel_acpi.c b/drivers/gpu/drm/i915/intel_acpi.c index f152b2a7fc5..f413899475e 100644 --- a/drivers/gpu/drm/i915/intel_acpi.c +++ b/drivers/gpu/drm/i915/intel_acpi.c @@ -9,6 +9,7 @@ #include #include "drmP.h" +#include "i915_drv.h" #define INTEL_DSM_REVISION_ID 1 /* For Calpella anyway... */ diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c index eb7ebf49f97..78179e03a90 100644 --- a/drivers/gpu/drm/i915/intel_display.c +++ b/drivers/gpu/drm/i915/intel_display.c @@ -9503,7 +9503,7 @@ struct intel_quirk { void (*hook)(struct drm_device *dev); }; -struct intel_quirk intel_quirks[] = { +static struct intel_quirk intel_quirks[] = { /* HP Mini needs pipe A force quirk (LP: #322104) */ { 0x27ae, 0x103c, 0x361a, quirk_pipea_force }, diff --git a/drivers/gpu/drm/i915/intel_fb.c b/drivers/gpu/drm/i915/intel_fb.c index 19ecd78b8a2..71ef2896be9 100644 --- a/drivers/gpu/drm/i915/intel_fb.c +++ b/drivers/gpu/drm/i915/intel_fb.c @@ -94,7 +94,7 @@ static int intelfb_create(struct intel_fbdev *ifbdev, mutex_lock(&dev->struct_mutex); /* Flush everything out, we'll be doing GTT only from now on */ - ret = intel_pin_and_fence_fb_obj(dev, obj, false); + ret = intel_pin_and_fence_fb_obj(dev, obj, NULL); if (ret) { DRM_ERROR("failed to pin fb: %d\n", ret); goto out_unref; diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.c b/drivers/gpu/drm/i915/intel_ringbuffer.c index 9d4e5f06edc..492812db537 100644 --- a/drivers/gpu/drm/i915/intel_ringbuffer.c +++ b/drivers/gpu/drm/i915/intel_ringbuffer.c @@ -891,8 +891,8 @@ err: return ret; } -int intel_init_ring_buffer(struct drm_device *dev, - struct intel_ring_buffer *ring) +static int intel_init_ring_buffer(struct drm_device *dev, + struct intel_ring_buffer *ring) { struct drm_i915_gem_object *obj; int ret; diff --git a/drivers/gpu/drm/i915/intel_sdvo.c b/drivers/gpu/drm/i915/intel_sdvo.c index 6898145b44c..aa1a0d5f0a5 100644 --- a/drivers/gpu/drm/i915/intel_sdvo.c +++ b/drivers/gpu/drm/i915/intel_sdvo.c @@ -1258,7 +1258,7 @@ intel_sdvo_get_analog_edid(struct drm_connector *connector) dev_priv->crt_ddc_pin)); } -enum drm_connector_status +static enum drm_connector_status intel_sdvo_tmds_sink_detect(struct drm_connector *connector) { struct intel_sdvo *intel_sdvo = intel_attached_sdvo(connector); -- cgit v1.2.3-70-g09d2 From 65f5687603ea6ede1cb01b3d6c16a8c1fac88541 Mon Sep 17 00:00:00 2001 From: Chris Wilson Date: Tue, 17 Apr 2012 16:38:12 +0100 Subject: drm/i915: Replace open coded MI_BATCH_GTT The (2<<6) virtual memory space selector harks back to gen3 and is mandatory given our use of GTT space for batchbuffers. On gen4+, use of the GTT became mandatory and bit6 marked reserved. However the code must now explicitly set (1<<7), which conveniently is also (2<<6). To clarify the meaning for future readers, replace the open coded (2<<6) with MI_BATCH_GTT. Signed-off-by: Chris Wilson Signed-off-by: Daniel Vetter --- drivers/gpu/drm/i915/i915_reg.h | 1 + drivers/gpu/drm/i915/intel_ringbuffer.c | 5 +++-- 2 files changed, 4 insertions(+), 2 deletions(-) (limited to 'drivers/gpu/drm/i915/intel_ringbuffer.c') diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h index d093dba8224..0d3b97f0169 100644 --- a/drivers/gpu/drm/i915/i915_reg.h +++ b/drivers/gpu/drm/i915/i915_reg.h @@ -231,6 +231,7 @@ #define MI_BATCH_NON_SECURE (1) #define MI_BATCH_NON_SECURE_I965 (1<<8) #define MI_BATCH_BUFFER_START MI_INSTR(0x31, 0) +#define MI_BATCH_GTT (2<<6) /* aliased with (1<<7) on gen4 */ #define MI_SEMAPHORE_MBOX MI_INSTR(0x16, 1) /* gen6+ */ #define MI_SEMAPHORE_GLOBAL_GTT (1<<22) #define MI_SEMAPHORE_UPDATE (1<<21) diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.c b/drivers/gpu/drm/i915/intel_ringbuffer.c index 492812db537..4ae651bb1c9 100644 --- a/drivers/gpu/drm/i915/intel_ringbuffer.c +++ b/drivers/gpu/drm/i915/intel_ringbuffer.c @@ -786,7 +786,8 @@ i965_dispatch_execbuffer(struct intel_ring_buffer *ring, u32 offset, u32 length) return ret; intel_ring_emit(ring, - MI_BATCH_BUFFER_START | (2 << 6) | + MI_BATCH_BUFFER_START | + MI_BATCH_GTT | MI_BATCH_NON_SECURE_I965); intel_ring_emit(ring, offset); intel_ring_advance(ring); @@ -823,7 +824,7 @@ i915_dispatch_execbuffer(struct intel_ring_buffer *ring, if (ret) return ret; - intel_ring_emit(ring, MI_BATCH_BUFFER_START | (2 << 6)); + intel_ring_emit(ring, MI_BATCH_BUFFER_START | MI_BATCH_GTT); intel_ring_emit(ring, offset | MI_BATCH_NON_SECURE); intel_ring_advance(ring); -- cgit v1.2.3-70-g09d2 From 46f0f8d120c4afae53a5670bf3ac80a928340ff3 Mon Sep 17 00:00:00 2001 From: Chris Wilson Date: Wed, 18 Apr 2012 11:12:11 +0100 Subject: drm/i915: Don't set a MBZ bit in gen2/3 MI_FLUSH On gen2 MI_EXE_FLUSH is actually an AGP flush bit and on gen3 marked as reserved. On both it is documented as being must-be-zero. So obey the documentation, and separate the gen2 flush into its own little routine and share with gen3. This means that we can rename the existing render_ring_flush() to reflect the generation from which it first applies and remove the code for handling earlier generations from it. v2: Applies to gen3 as well v3: Make it compile and improve the commit message. Signed-off-by: Chris Wilson Signed-off-by: Daniel Vetter --- drivers/gpu/drm/i915/intel_ringbuffer.c | 55 +++++++++++++++++++++++---------- 1 file changed, 39 insertions(+), 16 deletions(-) (limited to 'drivers/gpu/drm/i915/intel_ringbuffer.c') diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.c b/drivers/gpu/drm/i915/intel_ringbuffer.c index 4ae651bb1c9..6f610f28b1b 100644 --- a/drivers/gpu/drm/i915/intel_ringbuffer.c +++ b/drivers/gpu/drm/i915/intel_ringbuffer.c @@ -53,9 +53,35 @@ static inline int ring_space(struct intel_ring_buffer *ring) } static int -render_ring_flush(struct intel_ring_buffer *ring, - u32 invalidate_domains, - u32 flush_domains) +gen2_render_ring_flush(struct intel_ring_buffer *ring, + u32 invalidate_domains, + u32 flush_domains) +{ + u32 cmd; + int ret; + + cmd = MI_FLUSH; + if ((flush_domains & I915_GEM_DOMAIN_RENDER) == 0) + cmd |= MI_NO_WRITE_FLUSH; + + if (invalidate_domains & I915_GEM_DOMAIN_SAMPLER) + cmd |= MI_READ_FLUSH; + + ret = intel_ring_begin(ring, 2); + if (ret) + return ret; + + intel_ring_emit(ring, cmd); + intel_ring_emit(ring, MI_NOOP); + intel_ring_advance(ring); + + return 0; +} + +static int +gen4_render_ring_flush(struct intel_ring_buffer *ring, + u32 invalidate_domains, + u32 flush_domains) { struct drm_device *dev = ring->dev; u32 cmd; @@ -90,17 +116,8 @@ render_ring_flush(struct intel_ring_buffer *ring, */ cmd = MI_FLUSH | MI_NO_WRITE_FLUSH; - if ((invalidate_domains|flush_domains) & - I915_GEM_DOMAIN_RENDER) + if ((invalidate_domains|flush_domains) & I915_GEM_DOMAIN_RENDER) cmd &= ~MI_NO_WRITE_FLUSH; - if (INTEL_INFO(dev)->gen < 4) { - /* - * On the 965, the sampler cache always gets flushed - * and this bit is reserved. - */ - if (invalidate_domains & I915_GEM_DOMAIN_SAMPLER) - cmd |= MI_READ_FLUSH; - } if (invalidate_domains & I915_GEM_DOMAIN_INSTRUCTION) cmd |= MI_EXE_FLUSH; @@ -1281,14 +1298,17 @@ int intel_init_render_ring_buffer(struct drm_device *dev) ring->signal_mbox[1] = GEN6_BRSYNC; } else if (IS_GEN5(dev)) { ring->add_request = pc_render_add_request; - ring->flush = render_ring_flush; + ring->flush = gen4_render_ring_flush; ring->get_seqno = pc_render_get_seqno; ring->irq_get = gen5_ring_get_irq; ring->irq_put = gen5_ring_put_irq; ring->irq_enable_mask = GT_USER_INTERRUPT | GT_PIPE_NOTIFY; } else { ring->add_request = i9xx_add_request; - ring->flush = render_ring_flush; + if (INTEL_INFO(dev)->gen < 4) + ring->flush = gen2_render_ring_flush; + else + ring->flush = gen4_render_ring_flush; ring->get_seqno = ring_get_seqno; ring->irq_get = i9xx_ring_get_irq; ring->irq_put = i9xx_ring_put_irq; @@ -1333,7 +1353,10 @@ int intel_render_ring_init_dri(struct drm_device *dev, u64 start, u32 size) * gem_init ioctl returns with -ENODEV). Hence we do not need to set up * the special gen5 functions. */ ring->add_request = i9xx_add_request; - ring->flush = render_ring_flush; + if (INTEL_INFO(dev)->gen < 4) + ring->flush = gen2_render_ring_flush; + else + ring->flush = gen4_render_ring_flush; ring->get_seqno = ring_get_seqno; ring->irq_get = i9xx_ring_get_irq; ring->irq_put = i9xx_ring_put_irq; -- cgit v1.2.3-70-g09d2 From 31b14c9fc596a68a2a982d7e3c136922f81a2e25 Mon Sep 17 00:00:00 2001 From: Daniel Vetter Date: Thu, 19 Apr 2012 16:45:22 +0200 Subject: drm/i915: invalidate render cache on gen2 It looks like we also need to flush the render cache when we just invalidate it. This fixes a regression in i-g-t/gem_tiled_blits on my i855gm. I guess the render cache there is virtually indexed, so we need to clean it when changing gtt mappings. This regression has been introduce in commit 46f0f8d120c4afae53a5670bf3ac80a928340ff3 Author: Chris Wilson Date: Wed Apr 18 11:12:11 2012 +0100 drm/i915: Don't set a MBZ bit in gen2/3 MI_FLUSH Reviewed-by: Chris Wilson Signed-Off-by: Daniel Vetter --- drivers/gpu/drm/i915/intel_ringbuffer.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'drivers/gpu/drm/i915/intel_ringbuffer.c') diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.c b/drivers/gpu/drm/i915/intel_ringbuffer.c index 6f610f28b1b..12d9bc789df 100644 --- a/drivers/gpu/drm/i915/intel_ringbuffer.c +++ b/drivers/gpu/drm/i915/intel_ringbuffer.c @@ -61,7 +61,7 @@ gen2_render_ring_flush(struct intel_ring_buffer *ring, int ret; cmd = MI_FLUSH; - if ((flush_domains & I915_GEM_DOMAIN_RENDER) == 0) + if (((invalidate_domains|flush_domains) & I915_GEM_DOMAIN_RENDER) == 0) cmd |= MI_NO_WRITE_FLUSH; if (invalidate_domains & I915_GEM_DOMAIN_SAMPLER) -- cgit v1.2.3-70-g09d2 From c2798b19bac2538393fc932bfbe59807a4734b3e Mon Sep 17 00:00:00 2001 From: Chris Wilson Date: Sun, 22 Apr 2012 21:13:57 +0100 Subject: drm/i915: i8xx interrupt handler gen2 hardware has some significant differences from the other interrupt routines that were glossed over and then forgotten about in the transition to KMS. Such as - 16bit IIR - PendingFlip status bit This patch reintroduces a handler specifically for gen2 for the purpose of handling pageflips correctly, simplifying code in the process. v2: Also fixup ring get/put irq to only access 16bit registers (Daniel) Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=24202 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=41793 Signed-off-by: Chris Wilson [danvet: use posting_read16 in intel_ringbuffer.c and kill _driver from the function names.] Signed-off-by: Daniel Vetter --- drivers/gpu/drm/i915/i915_irq.c | 161 +++++++++++++++++++++++++++++++- drivers/gpu/drm/i915/intel_ringbuffer.c | 53 ++++++++++- 2 files changed, 206 insertions(+), 8 deletions(-) (limited to 'drivers/gpu/drm/i915/intel_ringbuffer.c') diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c index ab023ca73b4..d45b43a35f1 100644 --- a/drivers/gpu/drm/i915/i915_irq.c +++ b/drivers/gpu/drm/i915/i915_irq.c @@ -2446,6 +2446,152 @@ static void i915_driver_irq_uninstall(struct drm_device * dev) I915_WRITE(IIR, I915_READ(IIR)); } +static void i8xx_irq_preinstall(struct drm_device * dev) +{ + drm_i915_private_t *dev_priv = (drm_i915_private_t *) dev->dev_private; + int pipe; + + atomic_set(&dev_priv->irq_received, 0); + + for_each_pipe(pipe) + I915_WRITE(PIPESTAT(pipe), 0); + I915_WRITE16(IMR, 0xffff); + I915_WRITE16(IER, 0x0); + POSTING_READ16(IER); +} + +static int i8xx_irq_postinstall(struct drm_device *dev) +{ + drm_i915_private_t *dev_priv = (drm_i915_private_t *) dev->dev_private; + + dev_priv->vblank_pipe = DRM_I915_VBLANK_PIPE_A | DRM_I915_VBLANK_PIPE_B; + + dev_priv->pipestat[0] = 0; + dev_priv->pipestat[1] = 0; + + I915_WRITE16(EMR, + ~(I915_ERROR_PAGE_TABLE | I915_ERROR_MEMORY_REFRESH)); + + /* Unmask the interrupts that we always want on. */ + dev_priv->irq_mask = + ~(I915_DISPLAY_PIPE_A_EVENT_INTERRUPT | + I915_DISPLAY_PIPE_B_EVENT_INTERRUPT | + I915_DISPLAY_PLANE_A_FLIP_PENDING_INTERRUPT | + I915_DISPLAY_PLANE_B_FLIP_PENDING_INTERRUPT | + I915_RENDER_COMMAND_PARSER_ERROR_INTERRUPT); + I915_WRITE16(IMR, dev_priv->irq_mask); + + I915_WRITE16(IER, + I915_DISPLAY_PIPE_A_EVENT_INTERRUPT | + I915_DISPLAY_PIPE_B_EVENT_INTERRUPT | + I915_RENDER_COMMAND_PARSER_ERROR_INTERRUPT | + I915_USER_INTERRUPT); + POSTING_READ16(IER); + + return 0; +} + +static irqreturn_t i8xx_irq_handler(DRM_IRQ_ARGS) +{ + struct drm_device *dev = (struct drm_device *) arg; + drm_i915_private_t *dev_priv = (drm_i915_private_t *) dev->dev_private; + struct drm_i915_master_private *master_priv; + u16 iir, new_iir; + u32 pipe_stats[2]; + unsigned long irqflags; + int irq_received; + int pipe; + u16 flip_mask = + I915_DISPLAY_PLANE_A_FLIP_PENDING_INTERRUPT | + I915_DISPLAY_PLANE_B_FLIP_PENDING_INTERRUPT; + + atomic_inc(&dev_priv->irq_received); + + iir = I915_READ16(IIR); + if (iir == 0) + return IRQ_NONE; + + while (iir & ~flip_mask) { + /* Can't rely on pipestat interrupt bit in iir as it might + * have been cleared after the pipestat interrupt was received. + * It doesn't set the bit in iir again, but it still produces + * interrupts (for non-MSI). + */ + spin_lock_irqsave(&dev_priv->irq_lock, irqflags); + if (iir & I915_RENDER_COMMAND_PARSER_ERROR_INTERRUPT) + i915_handle_error(dev, false); + + for_each_pipe(pipe) { + int reg = PIPESTAT(pipe); + pipe_stats[pipe] = I915_READ(reg); + + /* + * Clear the PIPE*STAT regs before the IIR + */ + if (pipe_stats[pipe] & 0x8000ffff) { + if (pipe_stats[pipe] & PIPE_FIFO_UNDERRUN_STATUS) + DRM_DEBUG_DRIVER("pipe %c underrun\n", + pipe_name(pipe)); + I915_WRITE(reg, pipe_stats[pipe]); + irq_received = 1; + } + } + spin_unlock_irqrestore(&dev_priv->irq_lock, irqflags); + + I915_WRITE16(IIR, iir & ~flip_mask); + new_iir = I915_READ16(IIR); /* Flush posted writes */ + + if (dev->primary->master) { + master_priv = dev->primary->master->driver_priv; + if (master_priv->sarea_priv) + master_priv->sarea_priv->last_dispatch = + READ_BREADCRUMB(dev_priv); + } + + if (iir & I915_USER_INTERRUPT) + notify_ring(dev, &dev_priv->ring[RCS]); + + if (pipe_stats[0] & PIPE_VBLANK_INTERRUPT_STATUS && + drm_handle_vblank(dev, 0)) { + if (iir & I915_DISPLAY_PLANE_A_FLIP_PENDING_INTERRUPT) { + intel_prepare_page_flip(dev, 0); + intel_finish_page_flip(dev, 0); + flip_mask &= ~I915_DISPLAY_PLANE_A_FLIP_PENDING_INTERRUPT; + } + } + + if (pipe_stats[1] & PIPE_VBLANK_INTERRUPT_STATUS && + drm_handle_vblank(dev, 1)) { + if (iir & I915_DISPLAY_PLANE_B_FLIP_PENDING_INTERRUPT) { + intel_prepare_page_flip(dev, 1); + intel_finish_page_flip(dev, 1); + flip_mask &= ~I915_DISPLAY_PLANE_B_FLIP_PENDING_INTERRUPT; + } + } + + iir = new_iir; + } + + return IRQ_HANDLED; +} + +static void i8xx_irq_uninstall(struct drm_device * dev) +{ + drm_i915_private_t *dev_priv = (drm_i915_private_t *) dev->dev_private; + int pipe; + + dev_priv->vblank_pipe = 0; + + for_each_pipe(pipe) { + /* Clear enable bits; then clear status bits */ + I915_WRITE(PIPESTAT(pipe), 0); + I915_WRITE(PIPESTAT(pipe), I915_READ(PIPESTAT(pipe))); + } + I915_WRITE16(IMR, 0xffff); + I915_WRITE16(IER, 0x0); + I915_WRITE16(IIR, I915_READ16(IIR)); +} + void intel_irq_init(struct drm_device *dev) { dev->driver->get_vblank_counter = i915_get_vblank_counter; @@ -2485,10 +2631,17 @@ void intel_irq_init(struct drm_device *dev) dev->driver->enable_vblank = ironlake_enable_vblank; dev->driver->disable_vblank = ironlake_disable_vblank; } else { - dev->driver->irq_preinstall = i915_driver_irq_preinstall; - dev->driver->irq_postinstall = i915_driver_irq_postinstall; - dev->driver->irq_uninstall = i915_driver_irq_uninstall; - dev->driver->irq_handler = i915_driver_irq_handler; + if (INTEL_INFO(dev)->gen == 2) { + dev->driver->irq_preinstall = i8xx_irq_preinstall; + dev->driver->irq_postinstall = i8xx_irq_postinstall; + dev->driver->irq_handler = i8xx_irq_handler; + dev->driver->irq_uninstall = i8xx_irq_uninstall; + } else { + dev->driver->irq_preinstall = i915_driver_irq_preinstall; + dev->driver->irq_postinstall = i915_driver_irq_postinstall; + dev->driver->irq_uninstall = i915_driver_irq_uninstall; + dev->driver->irq_handler = i915_driver_irq_handler; + } dev->driver->enable_vblank = i915_enable_vblank; dev->driver->disable_vblank = i915_disable_vblank; } diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.c b/drivers/gpu/drm/i915/intel_ringbuffer.c index 12d9bc789df..6249a7fa9ac 100644 --- a/drivers/gpu/drm/i915/intel_ringbuffer.c +++ b/drivers/gpu/drm/i915/intel_ringbuffer.c @@ -678,6 +678,41 @@ i9xx_ring_put_irq(struct intel_ring_buffer *ring) spin_unlock(&ring->irq_lock); } +static bool +i8xx_ring_get_irq(struct intel_ring_buffer *ring) +{ + struct drm_device *dev = ring->dev; + drm_i915_private_t *dev_priv = dev->dev_private; + + if (!dev->irq_enabled) + return false; + + spin_lock(&ring->irq_lock); + if (ring->irq_refcount++ == 0) { + dev_priv->irq_mask &= ~ring->irq_enable_mask; + I915_WRITE16(IMR, dev_priv->irq_mask); + POSTING_READ16(IMR); + } + spin_unlock(&ring->irq_lock); + + return true; +} + +static void +i8xx_ring_put_irq(struct intel_ring_buffer *ring) +{ + struct drm_device *dev = ring->dev; + drm_i915_private_t *dev_priv = dev->dev_private; + + spin_lock(&ring->irq_lock); + if (--ring->irq_refcount == 0) { + dev_priv->irq_mask |= ring->irq_enable_mask; + I915_WRITE16(IMR, dev_priv->irq_mask); + POSTING_READ16(IMR); + } + spin_unlock(&ring->irq_lock); +} + void intel_ring_setup_status_page(struct intel_ring_buffer *ring) { struct drm_device *dev = ring->dev; @@ -1310,8 +1345,13 @@ int intel_init_render_ring_buffer(struct drm_device *dev) else ring->flush = gen4_render_ring_flush; ring->get_seqno = ring_get_seqno; - ring->irq_get = i9xx_ring_get_irq; - ring->irq_put = i9xx_ring_put_irq; + if (IS_GEN2(dev)) { + ring->irq_get = i8xx_ring_get_irq; + ring->irq_put = i8xx_ring_put_irq; + } else { + ring->irq_get = i9xx_ring_get_irq; + ring->irq_put = i9xx_ring_put_irq; + } ring->irq_enable_mask = I915_USER_INTERRUPT; } ring->write_tail = ring_write_tail; @@ -1358,8 +1398,13 @@ int intel_render_ring_init_dri(struct drm_device *dev, u64 start, u32 size) else ring->flush = gen4_render_ring_flush; ring->get_seqno = ring_get_seqno; - ring->irq_get = i9xx_ring_get_irq; - ring->irq_put = i9xx_ring_put_irq; + if (IS_GEN2(dev)) { + ring->irq_get = i8xx_ring_get_irq; + ring->irq_put = i8xx_ring_put_irq; + } else { + ring->irq_get = i9xx_ring_get_irq; + ring->irq_put = i9xx_ring_put_irq; + } ring->irq_enable_mask = I915_USER_INTERRUPT; ring->write_tail = ring_write_tail; if (INTEL_INFO(dev)->gen >= 4) -- cgit v1.2.3-70-g09d2 From 6b26c86d615747f67fede6ca9609b0f8b2037d11 Mon Sep 17 00:00:00 2001 From: Daniel Vetter Date: Tue, 24 Apr 2012 14:04:12 +0200 Subject: drm/i915: create macros to handle masked bits ... and put them to so good use. Note that there's functional change in vlv clock gating code, we now no longer spuriously read back the current value of the bit. According to Bspec the high bits should always read zero, so ORing this in should have no effect. Reviewed-by: Jesse Barnes Reviewed-by: Chris Wilson Reviewed-by: Eugeni Dodonov Signed-Off-by: Daniel Vetter --- drivers/gpu/drm/i915/i915_drv.c | 4 ++-- drivers/gpu/drm/i915/i915_gem.c | 8 ++++---- drivers/gpu/drm/i915/i915_irq.c | 5 ++--- drivers/gpu/drm/i915/i915_reg.h | 8 +++----- drivers/gpu/drm/i915/intel_pm.c | 5 ++--- drivers/gpu/drm/i915/intel_ringbuffer.c | 13 +++++-------- 6 files changed, 18 insertions(+), 25 deletions(-) (limited to 'drivers/gpu/drm/i915/intel_ringbuffer.c') diff --git a/drivers/gpu/drm/i915/i915_drv.c b/drivers/gpu/drm/i915/i915_drv.c index 95ccdffb5de..8a98f9a1641 100644 --- a/drivers/gpu/drm/i915/i915_drv.c +++ b/drivers/gpu/drm/i915/i915_drv.c @@ -438,7 +438,7 @@ void __gen6_gt_force_wake_mt_get(struct drm_i915_private *dev_priv) while (count++ < 50 && (I915_READ_NOTRACE(FORCEWAKE_MT_ACK) & 1)) udelay(10); - I915_WRITE_NOTRACE(FORCEWAKE_MT, (1<<16) | 1); + I915_WRITE_NOTRACE(FORCEWAKE_MT, _MASKED_BIT_ENABLE(1)); POSTING_READ(FORCEWAKE_MT); count = 0; @@ -480,7 +480,7 @@ void __gen6_gt_force_wake_put(struct drm_i915_private *dev_priv) void __gen6_gt_force_wake_mt_put(struct drm_i915_private *dev_priv) { - I915_WRITE_NOTRACE(FORCEWAKE_MT, (1<<16) | 0); + I915_WRITE_NOTRACE(FORCEWAKE_MT, _MASKED_BIT_DISABLE(1)); /* The below doubles as a POSTING_READ */ gen6_gt_check_fifodbg(dev_priv); } diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c index 38490cdf2d9..cfbcf7ef567 100644 --- a/drivers/gpu/drm/i915/i915_gem.c +++ b/drivers/gpu/drm/i915/i915_gem.c @@ -3494,9 +3494,9 @@ void i915_gem_init_swizzling(struct drm_device *dev) I915_WRITE(TILECTL, I915_READ(TILECTL) | TILECTL_SWZCTL); if (IS_GEN6(dev)) - I915_WRITE(ARB_MODE, ARB_MODE_ENABLE(ARB_MODE_SWIZZLE_SNB)); + I915_WRITE(ARB_MODE, _MASKED_BIT_ENABLE(ARB_MODE_SWIZZLE_SNB)); else - I915_WRITE(ARB_MODE, ARB_MODE_ENABLE(ARB_MODE_SWIZZLE_IVB)); + I915_WRITE(ARB_MODE, _MASKED_BIT_ENABLE(ARB_MODE_SWIZZLE_IVB)); } void i915_gem_init_ppgtt(struct drm_device *dev) @@ -3545,7 +3545,7 @@ void i915_gem_init_ppgtt(struct drm_device *dev) ecochk = I915_READ(GAM_ECOCHK); I915_WRITE(GAM_ECOCHK, ecochk | ECOCHK_SNB_BIT | ECOCHK_PPGTT_CACHE64B); - I915_WRITE(GFX_MODE, GFX_MODE_ENABLE(GFX_PPGTT_ENABLE)); + I915_WRITE(GFX_MODE, _MASKED_BIT_ENABLE(GFX_PPGTT_ENABLE)); } else if (INTEL_INFO(dev)->gen >= 7) { I915_WRITE(GAM_ECOCHK, ECOCHK_PPGTT_CACHE64B); /* GFX_MODE is per-ring on gen7+ */ @@ -3556,7 +3556,7 @@ void i915_gem_init_ppgtt(struct drm_device *dev) if (INTEL_INFO(dev)->gen >= 7) I915_WRITE(RING_MODE_GEN7(ring), - GFX_MODE_ENABLE(GFX_PPGTT_ENABLE)); + _MASKED_BIT_ENABLE(GFX_PPGTT_ENABLE)); I915_WRITE(RING_PP_DIR_DCLV(ring), PP_DIR_DCLV_2G); I915_WRITE(RING_PP_DIR_BASE(ring), pd_offset); diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c index d45b43a35f1..26172eef978 100644 --- a/drivers/gpu/drm/i915/i915_irq.c +++ b/drivers/gpu/drm/i915/i915_irq.c @@ -1648,7 +1648,7 @@ static int i915_enable_vblank(struct drm_device *dev, int pipe) /* maintain vblank delivery even in deep C-states */ if (dev_priv->info->gen == 3) - I915_WRITE(INSTPM, INSTPM_AGPBUSY_DIS << 16); + I915_WRITE(INSTPM, _MASKED_BIT_DISABLE(INSTPM_AGPBUSY_DIS)); spin_unlock_irqrestore(&dev_priv->irq_lock, irqflags); return 0; @@ -1722,8 +1722,7 @@ static void i915_disable_vblank(struct drm_device *dev, int pipe) spin_lock_irqsave(&dev_priv->irq_lock, irqflags); if (dev_priv->info->gen == 3) - I915_WRITE(INSTPM, - INSTPM_AGPBUSY_DIS << 16 | INSTPM_AGPBUSY_DIS); + I915_WRITE(INSTPM, _MASKED_BIT_ENABLE(INSTPM_AGPBUSY_DIS)); i915_disable_pipestat(dev_priv, pipe, PIPE_VBLANK_INTERRUPT_ENABLE | diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h index 91f1d1cd007..f1f4d8f1df6 100644 --- a/drivers/gpu/drm/i915/i915_reg.h +++ b/drivers/gpu/drm/i915/i915_reg.h @@ -29,6 +29,9 @@ #define _PORT(port, a, b) ((a) + (port)*((b)-(a))) +#define _MASKED_BIT_ENABLE(a) (((a) << 16) | (a)) +#define _MASKED_BIT_DISABLE(a) ((a) << 16) + /* * The Bridge device's PCI config space has information about the * fb aperture size and the amount of pre-reserved memory. @@ -425,8 +428,6 @@ #define ARB_MODE 0x04030 #define ARB_MODE_SWIZZLE_SNB (1<<4) #define ARB_MODE_SWIZZLE_IVB (1<<5) -#define ARB_MODE_ENABLE(x) GFX_MODE_ENABLE(x) -#define ARB_MODE_DISABLE(x) GFX_MODE_DISABLE(x) #define RENDER_HWS_PGA_GEN7 (0x04080) #define RING_FAULT_REG(ring) (0x4094 + 0x100*(ring)->id) #define DONE_REG 0x40b0 @@ -514,9 +515,6 @@ #define GFX_PSMI_GRANULARITY (1<<10) #define GFX_PPGTT_ENABLE (1<<9) -#define GFX_MODE_ENABLE(bit) (((bit) << 16) | (bit)) -#define GFX_MODE_DISABLE(bit) (((bit) << 16) | (0)) - #define SCPD0 0x0209c /* 915+ only */ #define IER 0x020a0 #define IIR 0x020a4 diff --git a/drivers/gpu/drm/i915/intel_pm.c b/drivers/gpu/drm/i915/intel_pm.c index a2d2ce474d0..a26bf49c464 100644 --- a/drivers/gpu/drm/i915/intel_pm.c +++ b/drivers/gpu/drm/i915/intel_pm.c @@ -2821,9 +2821,8 @@ static void valleyview_init_clock_gating(struct drm_device *dev) intel_flush_display_plane(dev_priv, pipe); } - I915_WRITE(CACHE_MODE_1, I915_READ(CACHE_MODE_1) | - (PIXEL_SUBSPAN_COLLECT_OPT_DISABLE << 16) | - PIXEL_SUBSPAN_COLLECT_OPT_DISABLE); + I915_WRITE(CACHE_MODE_1, + _MASKED_BIT_ENABLE(PIXEL_SUBSPAN_COLLECT_OPT_DISABLE)); } static void g4x_init_clock_gating(struct drm_device *dev) diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.c b/drivers/gpu/drm/i915/intel_ringbuffer.c index 6249a7fa9ac..f797613e6c4 100644 --- a/drivers/gpu/drm/i915/intel_ringbuffer.c +++ b/drivers/gpu/drm/i915/intel_ringbuffer.c @@ -401,12 +401,11 @@ static int init_render_ring(struct intel_ring_buffer *ring) int ret = init_ring_common(ring); if (INTEL_INFO(dev)->gen > 3) { - int mode = VS_TIMER_DISPATCH << 16 | VS_TIMER_DISPATCH; - I915_WRITE(MI_MODE, mode); + I915_WRITE(MI_MODE, _MASKED_BIT_ENABLE(VS_TIMER_DISPATCH)); if (IS_GEN7(dev)) I915_WRITE(GFX_MODE_GEN7, - GFX_MODE_DISABLE(GFX_TLB_INVALIDATE_ALWAYS) | - GFX_MODE_ENABLE(GFX_REPLAY_MODE)); + _MASKED_BIT_DISABLE(GFX_TLB_INVALIDATE_ALWAYS) | + _MASKED_BIT_ENABLE(GFX_REPLAY_MODE)); } if (INTEL_INFO(dev)->gen >= 5) { @@ -415,10 +414,8 @@ static int init_render_ring(struct intel_ring_buffer *ring) return ret; } - if (INTEL_INFO(dev)->gen >= 6) { - I915_WRITE(INSTPM, - INSTPM_FORCE_ORDERING << 16 | INSTPM_FORCE_ORDERING); - } + if (INTEL_INFO(dev)->gen >= 6) + I915_WRITE(INSTPM, _MASKED_BIT_ENABLE(INSTPM_FORCE_ORDERING)); return ret; } -- cgit v1.2.3-70-g09d2 From 7338aefa5cad08eb6794aa8e9a5853dedb990aa0 Mon Sep 17 00:00:00 2001 From: Chris Wilson Date: Tue, 24 Apr 2012 21:48:47 +0100 Subject: drm/i915: Use a global lock for modifying global irq flags We were attempting to use a per-ring spinlock whilst modifying global IRQ flags. A recipe for rare missed interrupts. Signed-off-by: Chris Wilson Acked-by: Ben Widawsky Signed-off-by: Daniel Vetter --- drivers/gpu/drm/i915/intel_ringbuffer.c | 41 +++++++++++++++++++-------------- drivers/gpu/drm/i915/intel_ringbuffer.h | 3 +-- 2 files changed, 25 insertions(+), 19 deletions(-) (limited to 'drivers/gpu/drm/i915/intel_ringbuffer.c') diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.c b/drivers/gpu/drm/i915/intel_ringbuffer.c index f797613e6c4..b7f446ee28d 100644 --- a/drivers/gpu/drm/i915/intel_ringbuffer.c +++ b/drivers/gpu/drm/i915/intel_ringbuffer.c @@ -610,17 +610,18 @@ gen5_ring_get_irq(struct intel_ring_buffer *ring) { struct drm_device *dev = ring->dev; drm_i915_private_t *dev_priv = dev->dev_private; + unsigned long flags; if (!dev->irq_enabled) return false; - spin_lock(&ring->irq_lock); + spin_lock_irqsave(&dev_priv->irq_lock, flags); if (ring->irq_refcount++ == 0) { dev_priv->gt_irq_mask &= ~ring->irq_enable_mask; I915_WRITE(GTIMR, dev_priv->gt_irq_mask); POSTING_READ(GTIMR); } - spin_unlock(&ring->irq_lock); + spin_unlock_irqrestore(&dev_priv->irq_lock, flags); return true; } @@ -630,14 +631,15 @@ gen5_ring_put_irq(struct intel_ring_buffer *ring) { struct drm_device *dev = ring->dev; drm_i915_private_t *dev_priv = dev->dev_private; + unsigned long flags; - spin_lock(&ring->irq_lock); + spin_lock_irqsave(&dev_priv->irq_lock, flags); if (--ring->irq_refcount == 0) { dev_priv->gt_irq_mask |= ring->irq_enable_mask; I915_WRITE(GTIMR, dev_priv->gt_irq_mask); POSTING_READ(GTIMR); } - spin_unlock(&ring->irq_lock); + spin_unlock_irqrestore(&dev_priv->irq_lock, flags); } static bool @@ -645,17 +647,18 @@ i9xx_ring_get_irq(struct intel_ring_buffer *ring) { struct drm_device *dev = ring->dev; drm_i915_private_t *dev_priv = dev->dev_private; + unsigned long flags; if (!dev->irq_enabled) return false; - spin_lock(&ring->irq_lock); + spin_lock_irqsave(&dev_priv->irq_lock, flags); if (ring->irq_refcount++ == 0) { dev_priv->irq_mask &= ~ring->irq_enable_mask; I915_WRITE(IMR, dev_priv->irq_mask); POSTING_READ(IMR); } - spin_unlock(&ring->irq_lock); + spin_unlock_irqrestore(&dev_priv->irq_lock, flags); return true; } @@ -665,14 +668,15 @@ i9xx_ring_put_irq(struct intel_ring_buffer *ring) { struct drm_device *dev = ring->dev; drm_i915_private_t *dev_priv = dev->dev_private; + unsigned long flags; - spin_lock(&ring->irq_lock); + spin_lock_irqsave(&dev_priv->irq_lock, flags); if (--ring->irq_refcount == 0) { dev_priv->irq_mask |= ring->irq_enable_mask; I915_WRITE(IMR, dev_priv->irq_mask); POSTING_READ(IMR); } - spin_unlock(&ring->irq_lock); + spin_unlock_irqrestore(&dev_priv->irq_lock, flags); } static bool @@ -680,17 +684,18 @@ i8xx_ring_get_irq(struct intel_ring_buffer *ring) { struct drm_device *dev = ring->dev; drm_i915_private_t *dev_priv = dev->dev_private; + unsigned long flags; if (!dev->irq_enabled) return false; - spin_lock(&ring->irq_lock); + spin_lock_irqsave(&dev_priv->irq_lock, flags); if (ring->irq_refcount++ == 0) { dev_priv->irq_mask &= ~ring->irq_enable_mask; I915_WRITE16(IMR, dev_priv->irq_mask); POSTING_READ16(IMR); } - spin_unlock(&ring->irq_lock); + spin_unlock_irqrestore(&dev_priv->irq_lock, flags); return true; } @@ -700,14 +705,15 @@ i8xx_ring_put_irq(struct intel_ring_buffer *ring) { struct drm_device *dev = ring->dev; drm_i915_private_t *dev_priv = dev->dev_private; + unsigned long flags; - spin_lock(&ring->irq_lock); + spin_lock_irqsave(&dev_priv->irq_lock, flags); if (--ring->irq_refcount == 0) { dev_priv->irq_mask |= ring->irq_enable_mask; I915_WRITE16(IMR, dev_priv->irq_mask); POSTING_READ16(IMR); } - spin_unlock(&ring->irq_lock); + spin_unlock_irqrestore(&dev_priv->irq_lock, flags); } void intel_ring_setup_status_page(struct intel_ring_buffer *ring) @@ -786,6 +792,7 @@ gen6_ring_get_irq(struct intel_ring_buffer *ring) { struct drm_device *dev = ring->dev; drm_i915_private_t *dev_priv = dev->dev_private; + unsigned long flags; if (!dev->irq_enabled) return false; @@ -795,14 +802,14 @@ gen6_ring_get_irq(struct intel_ring_buffer *ring) * blt/bsd rings on ivb. */ gen6_gt_force_wake_get(dev_priv); - spin_lock(&ring->irq_lock); + spin_lock_irqsave(&dev_priv->irq_lock, flags); if (ring->irq_refcount++ == 0) { I915_WRITE_IMR(ring, ~ring->irq_enable_mask); dev_priv->gt_irq_mask &= ~ring->irq_enable_mask; I915_WRITE(GTIMR, dev_priv->gt_irq_mask); POSTING_READ(GTIMR); } - spin_unlock(&ring->irq_lock); + spin_unlock_irqrestore(&dev_priv->irq_lock, flags); return true; } @@ -812,15 +819,16 @@ gen6_ring_put_irq(struct intel_ring_buffer *ring) { struct drm_device *dev = ring->dev; drm_i915_private_t *dev_priv = dev->dev_private; + unsigned long flags; - spin_lock(&ring->irq_lock); + spin_lock_irqsave(&dev_priv->irq_lock, flags); if (--ring->irq_refcount == 0) { I915_WRITE_IMR(ring, ~0); dev_priv->gt_irq_mask |= ring->irq_enable_mask; I915_WRITE(GTIMR, dev_priv->gt_irq_mask); POSTING_READ(GTIMR); } - spin_unlock(&ring->irq_lock); + spin_unlock_irqrestore(&dev_priv->irq_lock, flags); gen6_gt_force_wake_put(dev_priv); } @@ -954,7 +962,6 @@ static int intel_init_ring_buffer(struct drm_device *dev, ring->size = 32 * PAGE_SIZE; init_waitqueue_head(&ring->irq_queue); - spin_lock_init(&ring->irq_lock); if (I915_NEED_GFX_HWS(dev)) { ret = init_status_page(ring); diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.h b/drivers/gpu/drm/i915/intel_ringbuffer.h index 06a66adf69c..e0b25bb2fb5 100644 --- a/drivers/gpu/drm/i915/intel_ringbuffer.h +++ b/drivers/gpu/drm/i915/intel_ringbuffer.h @@ -56,8 +56,7 @@ struct intel_ring_buffer { */ u32 last_retired_head; - spinlock_t irq_lock; - u32 irq_refcount; + u32 irq_refcount; /* protected by dev_priv->irq_lock */ u32 irq_enable_mask; /* bitmask to enable ring interrupt */ u32 irq_seqno; /* last seq seem at irq time */ u32 trace_irq_seqno; -- cgit v1.2.3-70-g09d2 From 63ed2cb2d15019edf8af324479028996046c8631 Mon Sep 17 00:00:00 2001 From: Daniel Vetter Date: Mon, 23 Apr 2012 16:50:50 +0200 Subject: drm/i915: rip out GEM drm feature checks We always set it so there's no point in checking. We could instead add a bit that tells us whether gem is actually initialized (i.e. either kms or gem_init_ioctl called), but that's imho not worth it. So just rip it out. There's a little change in the wait_ring timeout, but we've never run with anything else than the 60 second timeout, even on dri1 userspace. Reviewed-by: Chris Wilson Signed-off-by: Daniel Vetter --- drivers/gpu/drm/i915/i915_gem.c | 19 +------------------ drivers/gpu/drm/i915/intel_ringbuffer.c | 15 ++++++--------- 2 files changed, 7 insertions(+), 27 deletions(-) (limited to 'drivers/gpu/drm/i915/intel_ringbuffer.c') diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c index 1b808d04194..0d53eacd293 100644 --- a/drivers/gpu/drm/i915/i915_gem.c +++ b/drivers/gpu/drm/i915/i915_gem.c @@ -168,9 +168,6 @@ i915_gem_get_aperture_ioctl(struct drm_device *dev, void *data, struct drm_i915_gem_object *obj; size_t pinned; - if (!(dev->driver->driver_features & DRIVER_GEM)) - return -ENODEV; - pinned = 0; mutex_lock(&dev->struct_mutex); list_for_each_entry(obj, &dev_priv->mm.gtt_list, gtt_list) @@ -246,6 +243,7 @@ i915_gem_create_ioctl(struct drm_device *dev, void *data, struct drm_file *file) { struct drm_i915_gem_create *args = data; + return i915_gem_create(file, dev, args->size, &args->handle); } @@ -931,9 +929,6 @@ i915_gem_set_domain_ioctl(struct drm_device *dev, void *data, uint32_t write_domain = args->write_domain; int ret; - if (!(dev->driver->driver_features & DRIVER_GEM)) - return -ENODEV; - /* Only handle setting domains to types used by the CPU. */ if (write_domain & I915_GEM_GPU_DOMAINS) return -EINVAL; @@ -987,9 +982,6 @@ i915_gem_sw_finish_ioctl(struct drm_device *dev, void *data, struct drm_i915_gem_object *obj; int ret = 0; - if (!(dev->driver->driver_features & DRIVER_GEM)) - return -ENODEV; - ret = i915_mutex_lock_interruptible(dev); if (ret) return ret; @@ -1025,9 +1017,6 @@ i915_gem_mmap_ioctl(struct drm_device *dev, void *data, struct drm_gem_object *obj; unsigned long addr; - if (!(dev->driver->driver_features & DRIVER_GEM)) - return -ENODEV; - obj = drm_gem_object_lookup(dev, file, args->handle); if (obj == NULL) return -ENOENT; @@ -1255,9 +1244,6 @@ i915_gem_mmap_gtt(struct drm_file *file, struct drm_i915_gem_object *obj; int ret; - if (!(dev->driver->driver_features & DRIVER_GEM)) - return -ENODEV; - ret = i915_mutex_lock_interruptible(dev); if (ret) return ret; @@ -1315,9 +1301,6 @@ i915_gem_mmap_gtt_ioctl(struct drm_device *dev, void *data, { struct drm_i915_gem_mmap_gtt *args = data; - if (!(dev->driver->driver_features & DRIVER_GEM)) - return -ENODEV; - return i915_gem_mmap_gtt(file, dev, args->handle, &args->offset); } diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.c b/drivers/gpu/drm/i915/intel_ringbuffer.c index b7f446ee28d..427b7c55ffe 100644 --- a/drivers/gpu/drm/i915/intel_ringbuffer.c +++ b/drivers/gpu/drm/i915/intel_ringbuffer.c @@ -1164,15 +1164,12 @@ int intel_wait_ring_buffer(struct intel_ring_buffer *ring, int n) return ret; trace_i915_ring_wait_begin(ring); - if (drm_core_check_feature(dev, DRIVER_GEM)) - /* With GEM the hangcheck timer should kick us out of the loop, - * leaving it early runs the risk of corrupting GEM state (due - * to running on almost untested codepaths). But on resume - * timers don't work yet, so prevent a complete hang in that - * case by choosing an insanely large timeout. */ - end = jiffies + 60 * HZ; - else - end = jiffies + 3 * HZ; + /* With GEM the hangcheck timer should kick us out of the loop, + * leaving it early runs the risk of corrupting GEM state (due + * to running on almost untested codepaths). But on resume + * timers don't work yet, so prevent a complete hang in that + * case by choosing an insanely large timeout. */ + end = jiffies + 60 * HZ; do { ring->head = I915_READ_HEAD(ring); -- cgit v1.2.3-70-g09d2 From b2da9fe5d5994a104bbae154590070d698279919 Mon Sep 17 00:00:00 2001 From: Ben Widawsky Date: Thu, 26 Apr 2012 16:02:58 -0700 Subject: drm/i915: remove do_retire from i915_wait_request This originates from a hack by me to quickly fix a bug in an earlier patch where we needed control over whether or not waiting on a seqno actually did any retire list processing. Since the two operations aren't clearly related, we should pull the parameter out of the wait function, and make the caller responsible for retiring if the action is desired. The only function call site which did not get an explicit retire_request call (on purpose) is i915_gem_inactive_shrink(). That code was already calling retire_request a second time. v2: don't modify any behavior excepit i915_gem_inactive_shrink(Daniel) Signed-off-by: Ben Widawsky Signed-off-by: Daniel Vetter --- drivers/gpu/drm/i915/i915_dma.c | 3 ++- drivers/gpu/drm/i915/i915_drv.h | 5 ++--- drivers/gpu/drm/i915/i915_gem.c | 33 ++++++++++-------------------- drivers/gpu/drm/i915/i915_gem_evict.c | 15 +++++++++++--- drivers/gpu/drm/i915/i915_gem_execbuffer.c | 3 ++- drivers/gpu/drm/i915/i915_gem_gtt.c | 2 +- drivers/gpu/drm/i915/intel_overlay.c | 8 ++++---- drivers/gpu/drm/i915/intel_ringbuffer.c | 4 +++- 8 files changed, 37 insertions(+), 36 deletions(-) (limited to 'drivers/gpu/drm/i915/intel_ringbuffer.c') diff --git a/drivers/gpu/drm/i915/i915_dma.c b/drivers/gpu/drm/i915/i915_dma.c index 7a55abbddc9..17081dad3fa 100644 --- a/drivers/gpu/drm/i915/i915_dma.c +++ b/drivers/gpu/drm/i915/i915_dma.c @@ -2009,9 +2009,10 @@ int i915_driver_unload(struct drm_device *dev) unregister_shrinker(&dev_priv->mm.inactive_shrinker); mutex_lock(&dev->struct_mutex); - ret = i915_gpu_idle(dev, true); + ret = i915_gpu_idle(dev); if (ret) DRM_ERROR("failed to idle hardware: %d\n", ret); + i915_gem_retire_requests(dev); mutex_unlock(&dev->struct_mutex); /* Cancel the retire work handler, which should be idle now. */ diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h index 0095c8d1256..2113a1ab962 100644 --- a/drivers/gpu/drm/i915/i915_drv.h +++ b/drivers/gpu/drm/i915/i915_drv.h @@ -1297,14 +1297,13 @@ int __must_check i915_gem_init_hw(struct drm_device *dev); void i915_gem_init_swizzling(struct drm_device *dev); void i915_gem_init_ppgtt(struct drm_device *dev); void i915_gem_cleanup_ringbuffer(struct drm_device *dev); -int __must_check i915_gpu_idle(struct drm_device *dev, bool do_retire); +int __must_check i915_gpu_idle(struct drm_device *dev); int __must_check i915_gem_idle(struct drm_device *dev); int __must_check i915_add_request(struct intel_ring_buffer *ring, struct drm_file *file, struct drm_i915_gem_request *request); int __must_check i915_wait_request(struct intel_ring_buffer *ring, - uint32_t seqno, - bool do_retire); + uint32_t seqno); int i915_gem_fault(struct vm_area_struct *vma, struct vm_fault *vmf); int __must_check i915_gem_object_set_to_gtt_domain(struct drm_i915_gem_object *obj, diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c index b46a3fd1774..e378204970f 100644 --- a/drivers/gpu/drm/i915/i915_gem.c +++ b/drivers/gpu/drm/i915/i915_gem.c @@ -1825,8 +1825,7 @@ i915_gem_retire_work_handler(struct work_struct *work) */ int i915_wait_request(struct intel_ring_buffer *ring, - uint32_t seqno, - bool do_retire) + uint32_t seqno) { drm_i915_private_t *dev_priv = ring->dev->dev_private; u32 ier; @@ -1902,14 +1901,6 @@ i915_wait_request(struct intel_ring_buffer *ring, if (atomic_read(&dev_priv->mm.wedged)) ret = -EAGAIN; - /* Directly dispatch request retiring. While we have the work queue - * to handle this, the waiter on a request often wants an associated - * buffer to have made it to the inactive list, and we would need - * a separate wait queue to handle that. - */ - if (ret == 0 && do_retire) - i915_gem_retire_requests_ring(ring); - return ret; } @@ -1931,10 +1922,10 @@ i915_gem_object_wait_rendering(struct drm_i915_gem_object *obj) * it. */ if (obj->active) { - ret = i915_wait_request(obj->ring, obj->last_rendering_seqno, - true); + ret = i915_wait_request(obj->ring, obj->last_rendering_seqno); if (ret) return ret; + i915_gem_retire_requests_ring(obj->ring); } return 0; @@ -2117,7 +2108,7 @@ i915_gem_flush_ring(struct intel_ring_buffer *ring, return 0; } -static int i915_ring_idle(struct intel_ring_buffer *ring, bool do_retire) +static int i915_ring_idle(struct intel_ring_buffer *ring) { int ret; @@ -2131,18 +2122,17 @@ static int i915_ring_idle(struct intel_ring_buffer *ring, bool do_retire) return ret; } - return i915_wait_request(ring, i915_gem_next_request_seqno(ring), - do_retire); + return i915_wait_request(ring, i915_gem_next_request_seqno(ring)); } -int i915_gpu_idle(struct drm_device *dev, bool do_retire) +int i915_gpu_idle(struct drm_device *dev) { drm_i915_private_t *dev_priv = dev->dev_private; int ret, i; /* Flush everything onto the inactive list. */ for (i = 0; i < I915_NUM_RINGS; i++) { - ret = i915_ring_idle(&dev_priv->ring[i], do_retire); + ret = i915_ring_idle(&dev_priv->ring[i]); if (ret) return ret; } @@ -2331,9 +2321,7 @@ i915_gem_object_flush_fence(struct drm_i915_gem_object *obj) } if (obj->last_fenced_seqno) { - ret = i915_wait_request(obj->ring, - obj->last_fenced_seqno, - false); + ret = i915_wait_request(obj->ring, obj->last_fenced_seqno); if (ret) return ret; @@ -3394,11 +3382,12 @@ i915_gem_idle(struct drm_device *dev) return 0; } - ret = i915_gpu_idle(dev, true); + ret = i915_gpu_idle(dev); if (ret) { mutex_unlock(&dev->struct_mutex); return ret; } + i915_gem_retire_requests(dev); /* Under UMS, be paranoid and evict. */ if (!drm_core_check_feature(dev, DRIVER_MODESET)) @@ -4025,7 +4014,7 @@ rescan: * This has a dramatic impact to reduce the number of * OOM-killer events whilst running the GPU aggressively. */ - if (i915_gpu_idle(dev, true) == 0) + if (i915_gpu_idle(dev) == 0) goto rescan; } mutex_unlock(&dev->struct_mutex); diff --git a/drivers/gpu/drm/i915/i915_gem_evict.c b/drivers/gpu/drm/i915/i915_gem_evict.c index 91ebb94d7c8..3bcf0451d07 100644 --- a/drivers/gpu/drm/i915/i915_gem_evict.c +++ b/drivers/gpu/drm/i915/i915_gem_evict.c @@ -168,7 +168,7 @@ i915_gem_evict_everything(struct drm_device *dev, bool purgeable_only) drm_i915_private_t *dev_priv = dev->dev_private; struct drm_i915_gem_object *obj, *next; bool lists_empty; - int ret; + int ret,i; lists_empty = (list_empty(&dev_priv->mm.inactive_list) && list_empty(&dev_priv->mm.flushing_list) && @@ -178,11 +178,20 @@ i915_gem_evict_everything(struct drm_device *dev, bool purgeable_only) trace_i915_gem_evict_everything(dev, purgeable_only); - /* Flush everything (on to the inactive lists) and evict */ - ret = i915_gpu_idle(dev, true); + ret = i915_gpu_idle(dev); if (ret) return ret; + /* The gpu_idle will flush everything in the write domain to the + * active list. Then we must move everything off the active list + * with retire requests. + */ + for (i = 0; i < I915_NUM_RINGS; i++) + if (WARN_ON(!list_empty(&dev_priv->ring[i].gpu_write_list))) + return -EBUSY; + + i915_gem_retire_requests(dev); + BUG_ON(!list_empty(&dev_priv->mm.flushing_list)); /* Having flushed everything, unbind() should never raise an error */ diff --git a/drivers/gpu/drm/i915/i915_gem_execbuffer.c b/drivers/gpu/drm/i915/i915_gem_execbuffer.c index 68ec0130a62..cbba0aa6104 100644 --- a/drivers/gpu/drm/i915/i915_gem_execbuffer.c +++ b/drivers/gpu/drm/i915/i915_gem_execbuffer.c @@ -1220,9 +1220,10 @@ i915_gem_do_execbuffer(struct drm_device *dev, void *data, * so every billion or so execbuffers, we need to stall * the GPU in order to reset the counters. */ - ret = i915_gpu_idle(dev, true); + ret = i915_gpu_idle(dev); if (ret) goto err; + i915_gem_retire_requests(dev); BUG_ON(ring->sync_seqno[i]); } diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.c b/drivers/gpu/drm/i915/i915_gem_gtt.c index 25c8bf9d1d4..29d573c27b3 100644 --- a/drivers/gpu/drm/i915/i915_gem_gtt.c +++ b/drivers/gpu/drm/i915/i915_gem_gtt.c @@ -317,7 +317,7 @@ static bool do_idling(struct drm_i915_private *dev_priv) if (unlikely(dev_priv->mm.gtt->do_idle_maps)) { dev_priv->mm.interruptible = false; - if (i915_gpu_idle(dev_priv->dev, false)) { + if (i915_gpu_idle(dev_priv->dev)) { DRM_ERROR("Couldn't idle GPU\n"); /* Wait a bit, in hopes it avoids the hang */ udelay(10); diff --git a/drivers/gpu/drm/i915/intel_overlay.c b/drivers/gpu/drm/i915/intel_overlay.c index a0b5053c5a3..e06e46a3075 100644 --- a/drivers/gpu/drm/i915/intel_overlay.c +++ b/drivers/gpu/drm/i915/intel_overlay.c @@ -225,10 +225,10 @@ static int intel_overlay_do_wait_request(struct intel_overlay *overlay, } overlay->last_flip_req = request->seqno; overlay->flip_tail = tail; - ret = i915_wait_request(LP_RING(dev_priv), overlay->last_flip_req, - true); + ret = i915_wait_request(LP_RING(dev_priv), overlay->last_flip_req); if (ret) return ret; + i915_gem_retire_requests(dev); overlay->last_flip_req = 0; return 0; @@ -447,10 +447,10 @@ static int intel_overlay_recover_from_interrupt(struct intel_overlay *overlay) if (overlay->last_flip_req == 0) return 0; - ret = i915_wait_request(LP_RING(dev_priv), overlay->last_flip_req, - true); + ret = i915_wait_request(LP_RING(dev_priv), overlay->last_flip_req); if (ret) return ret; + i915_gem_retire_requests(dev); if (overlay->flip_tail) overlay->flip_tail(overlay); diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.c b/drivers/gpu/drm/i915/intel_ringbuffer.c index 427b7c55ffe..a7d97d17b28 100644 --- a/drivers/gpu/drm/i915/intel_ringbuffer.c +++ b/drivers/gpu/drm/i915/intel_ringbuffer.c @@ -1088,9 +1088,11 @@ static int intel_ring_wait_seqno(struct intel_ring_buffer *ring, u32 seqno) was_interruptible = dev_priv->mm.interruptible; dev_priv->mm.interruptible = false; - ret = i915_wait_request(ring, seqno, true); + ret = i915_wait_request(ring, seqno); dev_priv->mm.interruptible = was_interruptible; + if (!ret) + i915_gem_retire_requests_ring(ring); return ret; } -- cgit v1.2.3-70-g09d2 From 627965ad3e2f8aa9311f8b238060ddb648268f6b Mon Sep 17 00:00:00 2001 From: Daniel Vetter Date: Thu, 26 Apr 2012 23:28:14 +0200 Subject: drm/i915: kill pointless clearing of dev_priv->hws_map We kzalloc dev_priv, and we never use hws_map in intel_ringbuffer.c. Reviewed-by: Chris Wilson Acked-by: Jesse Barnes Signed-Off-by: Daniel Vetter --- drivers/gpu/drm/i915/intel_ringbuffer.c | 5 ----- 1 file changed, 5 deletions(-) (limited to 'drivers/gpu/drm/i915/intel_ringbuffer.c') diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.c b/drivers/gpu/drm/i915/intel_ringbuffer.c index a7d97d17b28..df3a770d60f 100644 --- a/drivers/gpu/drm/i915/intel_ringbuffer.c +++ b/drivers/gpu/drm/i915/intel_ringbuffer.c @@ -890,7 +890,6 @@ i915_dispatch_execbuffer(struct intel_ring_buffer *ring, static void cleanup_status_page(struct intel_ring_buffer *ring) { - drm_i915_private_t *dev_priv = ring->dev->dev_private; struct drm_i915_gem_object *obj; obj = ring->status_page.obj; @@ -901,14 +900,11 @@ static void cleanup_status_page(struct intel_ring_buffer *ring) i915_gem_object_unpin(obj); drm_gem_object_unreference(&obj->base); ring->status_page.obj = NULL; - - memset(&dev_priv->hws_map, 0, sizeof(dev_priv->hws_map)); } static int init_status_page(struct intel_ring_buffer *ring) { struct drm_device *dev = ring->dev; - drm_i915_private_t *dev_priv = dev->dev_private; struct drm_i915_gem_object *obj; int ret; @@ -929,7 +925,6 @@ static int init_status_page(struct intel_ring_buffer *ring) ring->status_page.gfx_addr = obj->gtt_offset; ring->status_page.page_addr = kmap(obj->pages[0]); if (ring->status_page.page_addr == NULL) { - memset(&dev_priv->hws_map, 0, sizeof(dev_priv->hws_map)); goto err_unpin; } ring->status_page.obj = obj; -- cgit v1.2.3-70-g09d2 From 4225d0f219d22440e33a5686bf806356cb25bcf5 Mon Sep 17 00:00:00 2001 From: Daniel Vetter Date: Thu, 26 Apr 2012 23:28:16 +0200 Subject: drm/i915: fixup __iomem mixups in ringbuffer.c Two things: - ring->virtual start is an __iomem pointer, treat it accordingly. - dev_priv->status_page.page_addr is now always a cpu addr, no pointer casting needed for that. Take the opportunity to remove the unnecessary drm indirection when setting up the ringbuffer iomapping. v2: Add a compiler barrier before reading the hw status page. Acked-by: Jesse Barnes Reviewed-by: Chris Wilson Signed-off-by: Daniel Vetter --- drivers/gpu/drm/i915/i915_dma.c | 2 +- drivers/gpu/drm/i915/intel_ringbuffer.c | 39 ++++++++++----------------------- drivers/gpu/drm/i915/intel_ringbuffer.h | 7 +++--- 3 files changed, 17 insertions(+), 31 deletions(-) (limited to 'drivers/gpu/drm/i915/intel_ringbuffer.c') diff --git a/drivers/gpu/drm/i915/i915_dma.c b/drivers/gpu/drm/i915/i915_dma.c index f0c0a7ed90e..b5a1a72d332 100644 --- a/drivers/gpu/drm/i915/i915_dma.c +++ b/drivers/gpu/drm/i915/i915_dma.c @@ -256,7 +256,7 @@ static int i915_dma_resume(struct drm_device * dev) DRM_DEBUG_DRIVER("%s\n", __func__); - if (ring->map.handle == NULL) { + if (ring->virtual_start == NULL) { DRM_ERROR("can not ioremap virtual address for" " ring buffer\n"); return -ENOMEM; diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.c b/drivers/gpu/drm/i915/intel_ringbuffer.c index df3a770d60f..38096080a3d 100644 --- a/drivers/gpu/drm/i915/intel_ringbuffer.c +++ b/drivers/gpu/drm/i915/intel_ringbuffer.c @@ -977,20 +977,14 @@ static int intel_init_ring_buffer(struct drm_device *dev, if (ret) goto err_unref; - ring->map.size = ring->size; - ring->map.offset = dev->agp->base + obj->gtt_offset; - ring->map.type = 0; - ring->map.flags = 0; - ring->map.mtrr = 0; - - drm_core_ioremap_wc(&ring->map, dev); - if (ring->map.handle == NULL) { + ring->virtual_start = ioremap_wc(dev->agp->base + obj->gtt_offset, + ring->size); + if (ring->virtual_start == NULL) { DRM_ERROR("Failed to map ringbuffer.\n"); ret = -EINVAL; goto err_unpin; } - ring->virtual_start = ring->map.handle; ret = ring->init(ring); if (ret) goto err_unmap; @@ -1006,7 +1000,7 @@ static int intel_init_ring_buffer(struct drm_device *dev, return 0; err_unmap: - drm_core_ioremapfree(&ring->map, dev); + iounmap(ring->virtual_start); err_unpin: i915_gem_object_unpin(obj); err_unref: @@ -1034,7 +1028,7 @@ void intel_cleanup_ring_buffer(struct intel_ring_buffer *ring) I915_WRITE_CTL(ring, 0); - drm_core_ioremapfree(&ring->map, ring->dev); + iounmap(ring->virtual_start); i915_gem_object_unpin(ring->obj); drm_gem_object_unreference(&ring->obj->base); @@ -1048,7 +1042,7 @@ void intel_cleanup_ring_buffer(struct intel_ring_buffer *ring) static int intel_wrap_ring_buffer(struct intel_ring_buffer *ring) { - unsigned int *virt; + uint32_t __iomem *virt; int rem = ring->size - ring->tail; if (ring->space < rem) { @@ -1057,12 +1051,10 @@ static int intel_wrap_ring_buffer(struct intel_ring_buffer *ring) return ret; } - virt = (unsigned int *)(ring->virtual_start + ring->tail); - rem /= 8; - while (rem--) { - *virt++ = MI_NOOP; - *virt++ = MI_NOOP; - } + virt = ring->virtual_start + ring->tail; + rem /= 4; + while (rem--) + iowrite32(MI_NOOP, virt++); ring->tail = 0; ring->space = ring_space(ring); @@ -1427,20 +1419,13 @@ int intel_render_ring_init_dri(struct drm_device *dev, u64 start, u32 size) if (IS_I830(ring->dev)) ring->effective_size -= 128; - ring->map.offset = start; - ring->map.size = size; - ring->map.type = 0; - ring->map.flags = 0; - ring->map.mtrr = 0; - - drm_core_ioremap_wc(&ring->map, dev); - if (ring->map.handle == NULL) { + ring->virtual_start = ioremap_wc(start, size); + if (ring->virtual_start == NULL) { DRM_ERROR("can not ioremap virtual address for" " ring buffer\n"); return -ENOMEM; } - ring->virtual_start = (void __force __iomem *)ring->map.handle; return 0; } diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.h b/drivers/gpu/drm/i915/intel_ringbuffer.h index 7b879926969..baba7571457 100644 --- a/drivers/gpu/drm/i915/intel_ringbuffer.h +++ b/drivers/gpu/drm/i915/intel_ringbuffer.h @@ -2,7 +2,7 @@ #define _INTEL_RINGBUFFER_H_ struct intel_hw_status_page { - u32 __iomem *page_addr; + u32 *page_addr; unsigned int gfx_addr; struct drm_i915_gem_object *obj; }; @@ -115,7 +115,6 @@ struct intel_ring_buffer { u32 outstanding_lazy_request; wait_queue_head_t irq_queue; - drm_local_map_t map; void *private; }; @@ -149,7 +148,9 @@ static inline u32 intel_read_status_page(struct intel_ring_buffer *ring, int reg) { - return ioread32(ring->status_page.page_addr + reg); + /* Ensure that the compiler doesn't optimize away the load. */ + barrier(); + return ring->status_page.page_addr[reg]; } /** -- cgit v1.2.3-70-g09d2 From e5eb3d63c6182d3f21fbfc836ded748d49d521f9 Mon Sep 17 00:00:00 2001 From: Daniel Vetter Date: Thu, 3 May 2012 14:48:16 +0200 Subject: drm/i915: add interface to simulate gpu hangs gpu reset is a very important piece of our infrastructure. Unfortunately we only really it test by actually hanging the gpu, which often has bad side-effects for the entire system. And the gpu hang handling code is one of the rather complicated pieces of code we have, consisting of - hang detection - error capture - actual gpu reset - reset of all the gem bookkeeping - reinitialition of the entire gpu This patch adds a debugfs to selectively stopping rings by ceasing to update the hw tail pointer, which will result in the gpu no longer updating it's head pointer and eventually to the hangcheck firing. This way we can exercise the gpu hang code under controlled conditions without a dying gpu taking down the entire systems. Patch motivated by me forgetting to properly reinitialize ppgtt after a gpu reset. Usage: echo $((1 << $ringnum)) > i915_ring_stop # stops one ring echo 0xffffffff > i915_ring_stop # stops all, future-proof version then run whatever testload is desired. i915_ring_stop automatically resets after a gpu hang is detected to avoid hanging the gpu to fast and declaring it wedged. v2: Incorporate feedback from Chris Wilson. v3: Add the missing cleanup. v4: Fix up inconsistent size of ring_stop_read vs _write, noticed by Eugeni Dodonov. Reviewed-by: Chris Wilson Reviewed-by: Eugeni Dodonov Signed-Off-by: Daniel Vetter --- drivers/gpu/drm/i915/i915_debugfs.c | 65 +++++++++++++++++++++++++++++++++ drivers/gpu/drm/i915/i915_drv.c | 2 + drivers/gpu/drm/i915/i915_drv.h | 2 + drivers/gpu/drm/i915/intel_ringbuffer.c | 4 ++ 4 files changed, 73 insertions(+) (limited to 'drivers/gpu/drm/i915/intel_ringbuffer.c') diff --git a/drivers/gpu/drm/i915/i915_debugfs.c b/drivers/gpu/drm/i915/i915_debugfs.c index ae68ac1c488..192b27e9046 100644 --- a/drivers/gpu/drm/i915/i915_debugfs.c +++ b/drivers/gpu/drm/i915/i915_debugfs.c @@ -1585,6 +1585,64 @@ static const struct file_operations i915_wedged_fops = { .llseek = default_llseek, }; +static ssize_t +i915_ring_stop_read(struct file *filp, + char __user *ubuf, + size_t max, + loff_t *ppos) +{ + struct drm_device *dev = filp->private_data; + drm_i915_private_t *dev_priv = dev->dev_private; + char buf[20]; + int len; + + len = snprintf(buf, sizeof(buf), + "0x%08x\n", dev_priv->stop_rings); + + if (len > sizeof(buf)) + len = sizeof(buf); + + return simple_read_from_buffer(ubuf, max, ppos, buf, len); +} + +static ssize_t +i915_ring_stop_write(struct file *filp, + const char __user *ubuf, + size_t cnt, + loff_t *ppos) +{ + struct drm_device *dev = filp->private_data; + struct drm_i915_private *dev_priv = dev->dev_private; + char buf[20]; + int val = 0; + + if (cnt > 0) { + if (cnt > sizeof(buf) - 1) + return -EINVAL; + + if (copy_from_user(buf, ubuf, cnt)) + return -EFAULT; + buf[cnt] = 0; + + val = simple_strtoul(buf, NULL, 0); + } + + DRM_DEBUG_DRIVER("Stopping rings 0x%08x\n", val); + + mutex_lock(&dev->struct_mutex); + dev_priv->stop_rings = val; + mutex_unlock(&dev->struct_mutex); + + return cnt; +} + +static const struct file_operations i915_ring_stop_fops = { + .owner = THIS_MODULE, + .open = simple_open, + .read = i915_ring_stop_read, + .write = i915_ring_stop_write, + .llseek = default_llseek, +}; static ssize_t i915_max_freq_read(struct file *filp, char __user *ubuf, @@ -1885,6 +1943,11 @@ int i915_debugfs_init(struct drm_minor *minor) &i915_cache_sharing_fops); if (ret) return ret; + ret = i915_debugfs_create(minor->debugfs_root, minor, + "i915_ring_stop", + &i915_ring_stop_fops); + if (ret) + return ret; return drm_debugfs_create_files(i915_debugfs_list, I915_DEBUGFS_ENTRIES, @@ -1903,6 +1966,8 @@ void i915_debugfs_cleanup(struct drm_minor *minor) 1, minor); drm_debugfs_remove_files((struct drm_info_list *) &i915_cache_sharing_fops, 1, minor); + drm_debugfs_remove_files((struct drm_info_list *) &i915_ring_stop_fops, + 1, minor); } #endif /* CONFIG_DEBUG_FS */ diff --git a/drivers/gpu/drm/i915/i915_drv.c b/drivers/gpu/drm/i915/i915_drv.c index 8a98f9a1641..90a84f9de8e 100644 --- a/drivers/gpu/drm/i915/i915_drv.c +++ b/drivers/gpu/drm/i915/i915_drv.c @@ -800,6 +800,8 @@ int i915_reset(struct drm_device *dev, u8 flags) if (!mutex_trylock(&dev->struct_mutex)) return -EBUSY; + dev_priv->stop_rings = 0; + i915_gem_reset(dev); ret = -ENODEV; diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h index 00a20e985d2..090ec20293f 100644 --- a/drivers/gpu/drm/i915/i915_drv.h +++ b/drivers/gpu/drm/i915/i915_drv.h @@ -412,6 +412,8 @@ typedef struct drm_i915_private { uint32_t last_instdone; uint32_t last_instdone1; + unsigned int stop_rings; + unsigned long cfb_size; unsigned int cfb_fb; enum plane cfb_plane; diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.c b/drivers/gpu/drm/i915/intel_ringbuffer.c index 38096080a3d..3aabe8dfe5c 100644 --- a/drivers/gpu/drm/i915/intel_ringbuffer.c +++ b/drivers/gpu/drm/i915/intel_ringbuffer.c @@ -1210,7 +1210,11 @@ int intel_ring_begin(struct intel_ring_buffer *ring, void intel_ring_advance(struct intel_ring_buffer *ring) { + struct drm_i915_private *dev_priv = ring->dev->dev_private; + ring->tail &= ring->size - 1; + if (dev_priv->stop_rings & intel_ring_flag(ring)) + return; ring->write_tail(ring, ring->tail); } -- cgit v1.2.3-70-g09d2