From af447bd358acd92a8e96b61a3bf7a55c61a642f4 Mon Sep 17 00:00:00 2001 From: Alan Cox Date: Wed, 25 Jul 2012 13:49:18 +0100 Subject: vlv: it might be wise if we initialised the flag value... Otherwise our initial behaviour is "randomly save a bogus PLL choice" as far as I can see. Signed-off-by: Alan Cox Signed-off-by: Daniel Vetter --- drivers/gpu/drm/i915/intel_display.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c index e9e476eca89f..3d969ab2ae3f 100644 --- a/drivers/gpu/drm/i915/intel_display.c +++ b/drivers/gpu/drm/i915/intel_display.c @@ -869,6 +869,7 @@ intel_vlv_find_best_pll(const intel_limit_t *limit, struct drm_crtc *crtc, unsigned long bestppm, ppm, absppm; int dotclk, flag; + flag = 0; dotclk = target * 1000; bestppm = 1000000; ppm = absppm = 0; -- cgit v1.2.3-58-ga151 From 0274df3e43be962eee0499ee97c514fdb1c8f3ae Mon Sep 17 00:00:00 2001 From: Alan Cox Date: Wed, 25 Jul 2012 13:51:04 +0100 Subject: i915: fix error path leak in intel_sdvo_write_cmd Signed-off-by: Alan Cox Signed-off-by: Daniel Vetter --- drivers/gpu/drm/i915/intel_sdvo.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/i915/intel_sdvo.c b/drivers/gpu/drm/i915/intel_sdvo.c index 26a6a4d0d078..d172e9873131 100644 --- a/drivers/gpu/drm/i915/intel_sdvo.c +++ b/drivers/gpu/drm/i915/intel_sdvo.c @@ -444,13 +444,16 @@ static bool intel_sdvo_write_cmd(struct intel_sdvo *intel_sdvo, u8 cmd, struct i2c_msg *msgs; int i, ret = true; + /* Would be simpler to allocate both in one go ? */ buf = (u8 *)kzalloc(args_len * 2 + 2, GFP_KERNEL); if (!buf) return false; msgs = kcalloc(args_len + 3, sizeof(*msgs), GFP_KERNEL); - if (!msgs) + if (!msgs) { + kfree(buf); return false; + } intel_sdvo_debug_write(intel_sdvo, cmd, args, args_len); -- cgit v1.2.3-58-ga151 From 9978cf50426bc8d386a0753f74ab895e4d6da554 Mon Sep 17 00:00:00 2001 From: Alan Cox Date: Wed, 25 Jul 2012 13:51:20 +0100 Subject: i915: Remove silly test drv_priv->gmbus is an array. Comparing it with NULL is somewhat less useful than a chocolate teapot. Possibly we should be testing bus != NULL each iteration of the loop instead ? gcc could help by warning too! Signed-off-by: Alan Cox Signed-off-by: Daniel Vetter --- drivers/gpu/drm/i915/intel_i2c.c | 3 --- 1 file changed, 3 deletions(-) diff --git a/drivers/gpu/drm/i915/intel_i2c.c b/drivers/gpu/drm/i915/intel_i2c.c index 1991a4408cf9..d79500bc1ce5 100644 --- a/drivers/gpu/drm/i915/intel_i2c.c +++ b/drivers/gpu/drm/i915/intel_i2c.c @@ -540,9 +540,6 @@ void intel_teardown_gmbus(struct drm_device *dev) struct drm_i915_private *dev_priv = dev->dev_private; int i; - if (dev_priv->gmbus == NULL) - return; - for (i = 0; i < GMBUS_NUM_PORTS; i++) { struct intel_gmbus *bus = &dev_priv->gmbus[i]; i2c_del_adapter(&bus->adapter); -- cgit v1.2.3-58-ga151 From 6af2d180f82151cf3d58952e35a4f96e45bc453a Mon Sep 17 00:00:00 2001 From: Daniel Vetter Date: Thu, 26 Jul 2012 16:24:50 +0200 Subject: drm/i915: fix forcewake related hangs on snb ... by adding seemingly redudant posting reads. This little dragon lair exploded the first time around when we've refactored the code a bit to use the common wait_for_atomic_us in "drm/i915: Group the GT routines together in both code and vtable", which caused QA to file fdo bug #51738. Chris Wilson entertained a few approaches to fixing #51738: Replacing the udelay(1) with the previously-used udelay(10) (or any other "sufficiently larger" delay), adding a posting read, or ditching the delay completely and using cpu_relax. We went with the cpu_relax and "915: Workaround hang with BSD and forcewake on SandyBridge". Which blew up in fdo bug #52424, but adding the posting read while still using cpu_relax seems to also fix that, it looks like the posting read is the important ingriedient to fix these rc6 related hangs on snb. Popular theories as to why this is like it is include: - A herd of pink elephants got royally angered somehow. - The gpu has internally different functional units and judging by the register offsets, the forcewake request register and the forcewake ack registers are _not_ in the same functional unit (or at least aren't reached through the same routes). Hence the posting read syncs up with the wrong block and gets the entire gpu confused. - ... As a minimal ducttape fix for 3.6, let's just put these posting reads into place again. We can try fancier approaches (like adding back the cpu_relax instead of the udelay) in -next. This (re-)fixes a regression introduced in commit 990bbdadabaa51828e475eda86ee5720a4910cc3 Author: Chris Wilson Date: Mon Jul 2 11:51:02 2012 -0300 drm/i915: Group the GT routines together in both code and vtable Cc: Chris Wilson Tested-by: Chris Wilson Tested-by: Du Yan Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=52424 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=51738u Signed-off-by: Daniel Vetter --- drivers/gpu/drm/i915/intel_pm.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/i915/intel_pm.c b/drivers/gpu/drm/i915/intel_pm.c index 94aabcaa3a67..58c07cdafb7e 100644 --- a/drivers/gpu/drm/i915/intel_pm.c +++ b/drivers/gpu/drm/i915/intel_pm.c @@ -3963,6 +3963,7 @@ static void __gen6_gt_force_wake_get(struct drm_i915_private *dev_priv) DRM_ERROR("Force wake wait timed out\n"); I915_WRITE_NOTRACE(FORCEWAKE, 1); + POSTING_READ(FORCEWAKE); if (wait_for_atomic_us((I915_READ_NOTRACE(forcewake_ack) & 1), 500)) DRM_ERROR("Force wake wait timed out\n"); @@ -3983,6 +3984,7 @@ static void __gen6_gt_force_wake_mt_get(struct drm_i915_private *dev_priv) DRM_ERROR("Force wake wait timed out\n"); I915_WRITE_NOTRACE(FORCEWAKE_MT, _MASKED_BIT_ENABLE(1)); + POSTING_READ(FORCEWAKE_MT); if (wait_for_atomic_us((I915_READ_NOTRACE(forcewake_ack) & 1), 500)) DRM_ERROR("Force wake wait timed out\n"); @@ -4018,14 +4020,14 @@ void gen6_gt_check_fifodbg(struct drm_i915_private *dev_priv) static void __gen6_gt_force_wake_put(struct drm_i915_private *dev_priv) { I915_WRITE_NOTRACE(FORCEWAKE, 0); - /* The below doubles as a POSTING_READ */ + POSTING_READ(FORCEWAKE); gen6_gt_check_fifodbg(dev_priv); } static void __gen6_gt_force_wake_mt_put(struct drm_i915_private *dev_priv) { I915_WRITE_NOTRACE(FORCEWAKE_MT, _MASKED_BIT_DISABLE(1)); - /* The below doubles as a POSTING_READ */ + POSTING_READ(FORCEWAKE_MT); gen6_gt_check_fifodbg(dev_priv); } -- cgit v1.2.3-58-ga151 From e844b990b1df9242bb91b7d490552f3198946838 Mon Sep 17 00:00:00 2001 From: Eric Anholt Date: Tue, 31 Jul 2012 15:35:01 -0700 Subject: drm/i915: Don't forget to apply SNB PIPE_CONTROL GTT workaround. If a buffer that was the target of a PIPE_CONTROL from userland was a reused one that hadn't been evicted which had not previously had this workaround applied, then the early return for a correct presumed_offset in this function meant we would not bind it into the GTT and the write would land somewhere else. Fixes reproducible failures with GL_EXT_timer_query usage in apitrace, and I also expect it to fix the intermittent OQ issues on snb that danvet's been working on. Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=48019 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=52932 Signed-off-by: Eric Anholt Reviewed-by: Chris Wilson Reviewed-by: Carl Worth Tested-by: Carl Worth Signed-off-by: Daniel Vetter --- drivers/gpu/drm/i915/i915_gem_execbuffer.c | 20 ++++++++++---------- 1 file changed, 10 insertions(+), 10 deletions(-) diff --git a/drivers/gpu/drm/i915/i915_gem_execbuffer.c b/drivers/gpu/drm/i915/i915_gem_execbuffer.c index 5af631e788c8..ff2819ea0813 100644 --- a/drivers/gpu/drm/i915/i915_gem_execbuffer.c +++ b/drivers/gpu/drm/i915/i915_gem_execbuffer.c @@ -291,6 +291,16 @@ i915_gem_execbuffer_relocate_entry(struct drm_i915_gem_object *obj, target_i915_obj = to_intel_bo(target_obj); target_offset = target_i915_obj->gtt_offset; + /* Sandybridge PPGTT errata: We need a global gtt mapping for MI and + * pipe_control writes because the gpu doesn't properly redirect them + * through the ppgtt for non_secure batchbuffers. */ + if (unlikely(IS_GEN6(dev) && + reloc->write_domain == I915_GEM_DOMAIN_INSTRUCTION && + !target_i915_obj->has_global_gtt_mapping)) { + i915_gem_gtt_bind_object(target_i915_obj, + target_i915_obj->cache_level); + } + /* The target buffer should have appeared before us in the * exec_object list, so it should have a GTT space bound by now. */ @@ -399,16 +409,6 @@ i915_gem_execbuffer_relocate_entry(struct drm_i915_gem_object *obj, io_mapping_unmap_atomic(reloc_page); } - /* Sandybridge PPGTT errata: We need a global gtt mapping for MI and - * pipe_control writes because the gpu doesn't properly redirect them - * through the ppgtt for non_secure batchbuffers. */ - if (unlikely(IS_GEN6(dev) && - reloc->write_domain == I915_GEM_DOMAIN_INSTRUCTION && - !target_i915_obj->has_global_gtt_mapping)) { - i915_gem_gtt_bind_object(target_i915_obj, - target_i915_obj->cache_level); - } - /* and update the user's relocation entry */ reloc->presumed_offset = target_offset; -- cgit v1.2.3-58-ga151 From e47e9ad9186ff265ddf316d25cd4942bab019d57 Mon Sep 17 00:00:00 2001 From: Devendra Naga Date: Sat, 28 Jul 2012 18:46:35 +0545 Subject: drm/i915: remove unused variable MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit the following warning was produced, drivers/gpu/drm/i915/i915_gem_context.c: In function ‘i915_switch_context’: drivers/gpu/drm/i915/i915_gem_context.c:454:6: warning: unused variable ‘ret’ [-Wunused-variable] fix up by removing it Signed-off-by: Devendra Naga Signed-off-by: Daniel Vetter --- drivers/gpu/drm/i915/i915_gem_context.c | 1 - 1 file changed, 1 deletion(-) diff --git a/drivers/gpu/drm/i915/i915_gem_context.c b/drivers/gpu/drm/i915/i915_gem_context.c index da8b01fb1bf8..a9d58d72bb4d 100644 --- a/drivers/gpu/drm/i915/i915_gem_context.c +++ b/drivers/gpu/drm/i915/i915_gem_context.c @@ -451,7 +451,6 @@ int i915_switch_context(struct intel_ring_buffer *ring, struct drm_i915_file_private *file_priv = NULL; struct i915_hw_context *to; struct drm_i915_gem_object *from_obj = ring->last_context_obj; - int ret; if (dev_priv->hw_contexts_disabled) return 0; -- cgit v1.2.3-58-ga151 From f00f979145756e39b5512d7d4e4050534d1c3d7f Mon Sep 17 00:00:00 2001 From: Dave Airlie Date: Tue, 31 Jul 2012 15:58:13 +1000 Subject: i915: don't map imported dma-bufs for dmar. The exporter should have given us pages in the correct place, avoid the prepare object mapping phase on dmar systems. This fixes an oops on a GM45/R600 machine, when running the intel/radeon tests. Signed-off-by: Dave Airlie Signed-off-by: Daniel Vetter --- drivers/gpu/drm/i915/i915_gem_gtt.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.c b/drivers/gpu/drm/i915/i915_gem_gtt.c index 9fd25a435536..ee9b68f6bc36 100644 --- a/drivers/gpu/drm/i915/i915_gem_gtt.c +++ b/drivers/gpu/drm/i915/i915_gem_gtt.c @@ -361,7 +361,8 @@ int i915_gem_gtt_prepare_object(struct drm_i915_gem_object *obj) struct drm_device *dev = obj->base.dev; struct drm_i915_private *dev_priv = dev->dev_private; - if (dev_priv->mm.gtt->needs_dmar) + /* don't map imported dma buf objects */ + if (dev_priv->mm.gtt->needs_dmar && !obj->sg_table) return intel_gtt_map_memory(obj->pages, obj->base.size >> PAGE_SHIFT, &obj->sg_list, -- cgit v1.2.3-58-ga151 From faea35dd8a9d3605c411c8ecd7f97f0f9ecbaf58 Mon Sep 17 00:00:00 2001 From: Stéphane Marchesin Date: Mon, 30 Jul 2012 13:51:38 -0700 Subject: drm/i915: Make intel_panel_get_backlight static. MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit This function isn't used outside of intel_panel.c, so make it static. Signed-off-by: Stéphane Marchesin Signed-off-by: Daniel Vetter --- drivers/gpu/drm/i915/intel_drv.h | 1 - drivers/gpu/drm/i915/intel_panel.c | 2 +- 2 files changed, 1 insertion(+), 2 deletions(-) diff --git a/drivers/gpu/drm/i915/intel_drv.h b/drivers/gpu/drm/i915/intel_drv.h index 84353559441c..3afe3550e4ae 100644 --- a/drivers/gpu/drm/i915/intel_drv.h +++ b/drivers/gpu/drm/i915/intel_drv.h @@ -380,7 +380,6 @@ extern void intel_pch_panel_fitting(struct drm_device *dev, const struct drm_display_mode *mode, struct drm_display_mode *adjusted_mode); extern u32 intel_panel_get_max_backlight(struct drm_device *dev); -extern u32 intel_panel_get_backlight(struct drm_device *dev); extern void intel_panel_set_backlight(struct drm_device *dev, u32 level); extern int intel_panel_setup_backlight(struct drm_device *dev); extern void intel_panel_enable_backlight(struct drm_device *dev, diff --git a/drivers/gpu/drm/i915/intel_panel.c b/drivers/gpu/drm/i915/intel_panel.c index 10c7d39034e1..9474488db948 100644 --- a/drivers/gpu/drm/i915/intel_panel.c +++ b/drivers/gpu/drm/i915/intel_panel.c @@ -213,7 +213,7 @@ static u32 intel_panel_compute_brightness(struct drm_device *dev, u32 val) return val; } -u32 intel_panel_get_backlight(struct drm_device *dev) +static u32 intel_panel_get_backlight(struct drm_device *dev) { struct drm_i915_private *dev_priv = dev->dev_private; u32 val; -- cgit v1.2.3-58-ga151 From bcf9dcc1e6269fac674e41f25d007ff75f76e840 Mon Sep 17 00:00:00 2001 From: Chris Wilson Date: Sun, 15 Jul 2012 09:42:38 +0100 Subject: drm/i915: Workaround hang with BSD and forcewake on SandyBridge For reasons that are not apparent to anybody, 990bbdadaba (drm/i915: Group the GT routines together in both code and vtable) breaks the use of the BitStream Decoder ring on SandyBridge. The active ingredient of that patch is the conversion from a udelay(10) to a udelay(1) in the busy-wait loop of waiting for the forcewake acknowledge. If we restore that udelay(10) or insert another udelay(1) afterwards (or any wait longer than 250ns) everything works again. An alternative is also to remove any delay from the busy-wait loop. Given that in the atomic sections we want to complete the wait as quick as possible to avoid blocking the CPU for too long, it makes sense to remove the delay altogether and simply spin on the exit condition until it completes. So we replace the udelay(1) with cpu_relax(). Papers over regression from commit 990bbdadabaa51828e475eda86ee5720a4910cc3 Author: Chris Wilson Date: Mon Jul 2 11:51:02 2012 -0300 drm/i915: Group the GT routines together in both code and vtable Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=51738 Signed-off-by: Chris Wilson Signed-off-by: Daniel Vetter --- drivers/gpu/drm/i915/intel_drv.h | 19 ++++++++++--------- 1 file changed, 10 insertions(+), 9 deletions(-) diff --git a/drivers/gpu/drm/i915/intel_drv.h b/drivers/gpu/drm/i915/intel_drv.h index 3afe3550e4ae..132ab511b90c 100644 --- a/drivers/gpu/drm/i915/intel_drv.h +++ b/drivers/gpu/drm/i915/intel_drv.h @@ -46,15 +46,16 @@ }) #define wait_for_atomic_us(COND, US) ({ \ - int i, ret__ = -ETIMEDOUT; \ - for (i = 0; i < (US); i++) { \ - if ((COND)) { \ - ret__ = 0; \ - break; \ - } \ - udelay(1); \ - } \ - ret__; \ + unsigned long timeout__ = jiffies + usecs_to_jiffies(US); \ + int ret__ = 0; \ + while (!(COND)) { \ + if (time_after(jiffies, timeout__)) { \ + ret__ = -ETIMEDOUT; \ + break; \ + } \ + cpu_relax(); \ + } \ + ret__; \ }) #define wait_for(COND, MS) _wait_for(COND, MS, 1) -- cgit v1.2.3-58-ga151 From 5ab3633d6907018b0b830a720e877c3884d679c3 Mon Sep 17 00:00:00 2001 From: Hunt Xu Date: Sun, 1 Jul 2012 03:45:07 +0000 Subject: drm/i915: make rc6 in sysfs functions conditional Commit 0136db586c028f71e7cc21cc183064ff0d5919c8 merges rc6 information into the power group. However, when compiled with CONFIG_PM not set, modprobing i915 would taint since power_group_name is defined as NULL. This patch makes these rc6 in sysfs functions conditional upon the definition of the CONFIG_PM macro to avoid the above-mentioned problem. Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=45181 Cc: stable@vger.kernel.org Tested-by: Kris Karas Signed-off-by: Hunt Xu Signed-off-by: Daniel Vetter --- drivers/gpu/drm/i915/i915_sysfs.c | 12 ++++++++++++ 1 file changed, 12 insertions(+) diff --git a/drivers/gpu/drm/i915/i915_sysfs.c b/drivers/gpu/drm/i915/i915_sysfs.c index 2f5388af8df9..7631807a2788 100644 --- a/drivers/gpu/drm/i915/i915_sysfs.c +++ b/drivers/gpu/drm/i915/i915_sysfs.c @@ -32,6 +32,7 @@ #include "intel_drv.h" #include "i915_drv.h" +#ifdef CONFIG_PM static u32 calc_residency(struct drm_device *dev, const u32 reg) { struct drm_i915_private *dev_priv = dev->dev_private; @@ -224,3 +225,14 @@ void i915_teardown_sysfs(struct drm_device *dev) device_remove_bin_file(&dev->primary->kdev, &dpf_attrs); sysfs_unmerge_group(&dev->primary->kdev.kobj, &rc6_attr_group); } +#else +void i915_setup_sysfs(struct drm_device *dev) +{ + return; +} + +void i915_teardown_sysfs(struct drm_device *dev) +{ + return; +} +#endif /* CONFIG_PM */ -- cgit v1.2.3-58-ga151 From da612d880fbc598ac0efcef579355fb90d4bca4e Mon Sep 17 00:00:00 2001 From: Paulo Zanoni Date: Mon, 6 Aug 2012 18:45:01 -0300 Subject: drm/i915: add more Haswell PCI IDs Also properly indent the HB IDs. Reviewed-by: Rodrigo Vivi Signed-off-by: Paulo Zanoni Signed-off-by: Daniel Vetter --- drivers/char/agp/intel-agp.h | 39 +++++++++++++++++++++++---- drivers/char/agp/intel-gtt.c | 60 ++++++++++++++++++++++++++++++++++++++++- drivers/gpu/drm/i915/i915_drv.c | 31 ++++++++++++++++++++- 3 files changed, 123 insertions(+), 7 deletions(-) diff --git a/drivers/char/agp/intel-agp.h b/drivers/char/agp/intel-agp.h index 57226424690c..6f007b6c240d 100644 --- a/drivers/char/agp/intel-agp.h +++ b/drivers/char/agp/intel-agp.h @@ -239,16 +239,45 @@ #define PCI_DEVICE_ID_INTEL_IVYBRIDGE_S_GT2_IG 0x016A #define PCI_DEVICE_ID_INTEL_VALLEYVIEW_HB 0x0F00 /* VLV1 */ #define PCI_DEVICE_ID_INTEL_VALLEYVIEW_IG 0x0F30 -#define PCI_DEVICE_ID_INTEL_HASWELL_HB 0x0400 /* Desktop */ +#define PCI_DEVICE_ID_INTEL_HASWELL_HB 0x0400 /* Desktop */ #define PCI_DEVICE_ID_INTEL_HASWELL_D_GT1_IG 0x0402 #define PCI_DEVICE_ID_INTEL_HASWELL_D_GT2_IG 0x0412 -#define PCI_DEVICE_ID_INTEL_HASWELL_M_HB 0x0404 /* Mobile */ +#define PCI_DEVICE_ID_INTEL_HASWELL_D_GT2_PLUS_IG 0x0422 +#define PCI_DEVICE_ID_INTEL_HASWELL_M_HB 0x0404 /* Mobile */ #define PCI_DEVICE_ID_INTEL_HASWELL_M_GT1_IG 0x0406 #define PCI_DEVICE_ID_INTEL_HASWELL_M_GT2_IG 0x0416 -#define PCI_DEVICE_ID_INTEL_HASWELL_S_HB 0x0408 /* Server */ +#define PCI_DEVICE_ID_INTEL_HASWELL_M_GT2_PLUS_IG 0x0426 +#define PCI_DEVICE_ID_INTEL_HASWELL_S_HB 0x0408 /* Server */ #define PCI_DEVICE_ID_INTEL_HASWELL_S_GT1_IG 0x040a #define PCI_DEVICE_ID_INTEL_HASWELL_S_GT2_IG 0x041a -#define PCI_DEVICE_ID_INTEL_HASWELL_SDV 0x0c16 /* SDV */ -#define PCI_DEVICE_ID_INTEL_HASWELL_E_HB 0x0c04 +#define PCI_DEVICE_ID_INTEL_HASWELL_S_GT2_PLUS_IG 0x042a +#define PCI_DEVICE_ID_INTEL_HASWELL_E_HB 0x0c04 +#define PCI_DEVICE_ID_INTEL_HASWELL_SDV_D_GT1_IG 0x0C02 +#define PCI_DEVICE_ID_INTEL_HASWELL_SDV_D_GT2_IG 0x0C12 +#define PCI_DEVICE_ID_INTEL_HASWELL_SDV_D_GT2_PLUS_IG 0x0C22 +#define PCI_DEVICE_ID_INTEL_HASWELL_SDV_M_GT1_IG 0x0C06 +#define PCI_DEVICE_ID_INTEL_HASWELL_SDV_M_GT2_IG 0x0C16 +#define PCI_DEVICE_ID_INTEL_HASWELL_SDV_M_GT2_PLUS_IG 0x0C26 +#define PCI_DEVICE_ID_INTEL_HASWELL_SDV_S_GT1_IG 0x0C0A +#define PCI_DEVICE_ID_INTEL_HASWELL_SDV_S_GT2_IG 0x0C1A +#define PCI_DEVICE_ID_INTEL_HASWELL_SDV_S_GT2_PLUS_IG 0x0C2A +#define PCI_DEVICE_ID_INTEL_HASWELL_ULT_D_GT1_IG 0x0A02 +#define PCI_DEVICE_ID_INTEL_HASWELL_ULT_D_GT2_IG 0x0A12 +#define PCI_DEVICE_ID_INTEL_HASWELL_ULT_D_GT2_PLUS_IG 0x0A22 +#define PCI_DEVICE_ID_INTEL_HASWELL_ULT_M_GT1_IG 0x0A06 +#define PCI_DEVICE_ID_INTEL_HASWELL_ULT_M_GT2_IG 0x0A16 +#define PCI_DEVICE_ID_INTEL_HASWELL_ULT_M_GT2_PLUS_IG 0x0A26 +#define PCI_DEVICE_ID_INTEL_HASWELL_ULT_S_GT1_IG 0x0A0A +#define PCI_DEVICE_ID_INTEL_HASWELL_ULT_S_GT2_IG 0x0A1A +#define PCI_DEVICE_ID_INTEL_HASWELL_ULT_S_GT2_PLUS_IG 0x0A2A +#define PCI_DEVICE_ID_INTEL_HASWELL_CRW_D_GT1_IG 0x0D12 +#define PCI_DEVICE_ID_INTEL_HASWELL_CRW_D_GT2_IG 0x0D22 +#define PCI_DEVICE_ID_INTEL_HASWELL_CRW_D_GT2_PLUS_IG 0x0D32 +#define PCI_DEVICE_ID_INTEL_HASWELL_CRW_M_GT1_IG 0x0D16 +#define PCI_DEVICE_ID_INTEL_HASWELL_CRW_M_GT2_IG 0x0D26 +#define PCI_DEVICE_ID_INTEL_HASWELL_CRW_M_GT2_PLUS_IG 0x0D36 +#define PCI_DEVICE_ID_INTEL_HASWELL_CRW_S_GT1_IG 0x0D1A +#define PCI_DEVICE_ID_INTEL_HASWELL_CRW_S_GT2_IG 0x0D2A +#define PCI_DEVICE_ID_INTEL_HASWELL_CRW_S_GT2_PLUS_IG 0x0D3A #endif diff --git a/drivers/char/agp/intel-gtt.c b/drivers/char/agp/intel-gtt.c index 9ed92ef5829b..08fc5cbb13cd 100644 --- a/drivers/char/agp/intel-gtt.c +++ b/drivers/char/agp/intel-gtt.c @@ -1502,15 +1502,73 @@ static const struct intel_gtt_driver_description { "Haswell", &sandybridge_gtt_driver }, { PCI_DEVICE_ID_INTEL_HASWELL_D_GT2_IG, "Haswell", &sandybridge_gtt_driver }, + { PCI_DEVICE_ID_INTEL_HASWELL_D_GT2_PLUS_IG, + "Haswell", &sandybridge_gtt_driver }, { PCI_DEVICE_ID_INTEL_HASWELL_M_GT1_IG, "Haswell", &sandybridge_gtt_driver }, { PCI_DEVICE_ID_INTEL_HASWELL_M_GT2_IG, "Haswell", &sandybridge_gtt_driver }, + { PCI_DEVICE_ID_INTEL_HASWELL_M_GT2_PLUS_IG, + "Haswell", &sandybridge_gtt_driver }, { PCI_DEVICE_ID_INTEL_HASWELL_S_GT1_IG, "Haswell", &sandybridge_gtt_driver }, { PCI_DEVICE_ID_INTEL_HASWELL_S_GT2_IG, "Haswell", &sandybridge_gtt_driver }, - { PCI_DEVICE_ID_INTEL_HASWELL_SDV, + { PCI_DEVICE_ID_INTEL_HASWELL_S_GT2_PLUS_IG, + "Haswell", &sandybridge_gtt_driver }, + { PCI_DEVICE_ID_INTEL_HASWELL_SDV_D_GT1_IG, + "Haswell", &sandybridge_gtt_driver }, + { PCI_DEVICE_ID_INTEL_HASWELL_SDV_D_GT2_IG, + "Haswell", &sandybridge_gtt_driver }, + { PCI_DEVICE_ID_INTEL_HASWELL_SDV_D_GT2_PLUS_IG, + "Haswell", &sandybridge_gtt_driver }, + { PCI_DEVICE_ID_INTEL_HASWELL_SDV_M_GT1_IG, + "Haswell", &sandybridge_gtt_driver }, + { PCI_DEVICE_ID_INTEL_HASWELL_SDV_M_GT2_IG, + "Haswell", &sandybridge_gtt_driver }, + { PCI_DEVICE_ID_INTEL_HASWELL_SDV_M_GT2_PLUS_IG, + "Haswell", &sandybridge_gtt_driver }, + { PCI_DEVICE_ID_INTEL_HASWELL_SDV_S_GT1_IG, + "Haswell", &sandybridge_gtt_driver }, + { PCI_DEVICE_ID_INTEL_HASWELL_SDV_S_GT2_IG, + "Haswell", &sandybridge_gtt_driver }, + { PCI_DEVICE_ID_INTEL_HASWELL_SDV_S_GT2_PLUS_IG, + "Haswell", &sandybridge_gtt_driver }, + { PCI_DEVICE_ID_INTEL_HASWELL_ULT_D_GT1_IG, + "Haswell", &sandybridge_gtt_driver }, + { PCI_DEVICE_ID_INTEL_HASWELL_ULT_D_GT2_IG, + "Haswell", &sandybridge_gtt_driver }, + { PCI_DEVICE_ID_INTEL_HASWELL_ULT_D_GT2_PLUS_IG, + "Haswell", &sandybridge_gtt_driver }, + { PCI_DEVICE_ID_INTEL_HASWELL_ULT_M_GT1_IG, + "Haswell", &sandybridge_gtt_driver }, + { PCI_DEVICE_ID_INTEL_HASWELL_ULT_M_GT2_IG, + "Haswell", &sandybridge_gtt_driver }, + { PCI_DEVICE_ID_INTEL_HASWELL_ULT_M_GT2_PLUS_IG, + "Haswell", &sandybridge_gtt_driver }, + { PCI_DEVICE_ID_INTEL_HASWELL_ULT_S_GT1_IG, + "Haswell", &sandybridge_gtt_driver }, + { PCI_DEVICE_ID_INTEL_HASWELL_ULT_S_GT2_IG, + "Haswell", &sandybridge_gtt_driver }, + { PCI_DEVICE_ID_INTEL_HASWELL_ULT_S_GT2_PLUS_IG, + "Haswell", &sandybridge_gtt_driver }, + { PCI_DEVICE_ID_INTEL_HASWELL_CRW_D_GT1_IG, + "Haswell", &sandybridge_gtt_driver }, + { PCI_DEVICE_ID_INTEL_HASWELL_CRW_D_GT2_IG, + "Haswell", &sandybridge_gtt_driver }, + { PCI_DEVICE_ID_INTEL_HASWELL_CRW_D_GT2_PLUS_IG, + "Haswell", &sandybridge_gtt_driver }, + { PCI_DEVICE_ID_INTEL_HASWELL_CRW_M_GT1_IG, + "Haswell", &sandybridge_gtt_driver }, + { PCI_DEVICE_ID_INTEL_HASWELL_CRW_M_GT2_IG, + "Haswell", &sandybridge_gtt_driver }, + { PCI_DEVICE_ID_INTEL_HASWELL_CRW_M_GT2_PLUS_IG, + "Haswell", &sandybridge_gtt_driver }, + { PCI_DEVICE_ID_INTEL_HASWELL_CRW_S_GT1_IG, + "Haswell", &sandybridge_gtt_driver }, + { PCI_DEVICE_ID_INTEL_HASWELL_CRW_S_GT2_IG, + "Haswell", &sandybridge_gtt_driver }, + { PCI_DEVICE_ID_INTEL_HASWELL_CRW_S_GT2_PLUS_IG, "Haswell", &sandybridge_gtt_driver }, { 0, NULL, NULL } }; diff --git a/drivers/gpu/drm/i915/i915_drv.c b/drivers/gpu/drm/i915/i915_drv.c index ed22612bc847..a24ffbe97c01 100644 --- a/drivers/gpu/drm/i915/i915_drv.c +++ b/drivers/gpu/drm/i915/i915_drv.c @@ -346,11 +346,40 @@ static const struct pci_device_id pciidlist[] = { /* aka */ INTEL_VGA_DEVICE(0x016a, &intel_ivybridge_d_info), /* GT2 server */ INTEL_VGA_DEVICE(0x0402, &intel_haswell_d_info), /* GT1 desktop */ INTEL_VGA_DEVICE(0x0412, &intel_haswell_d_info), /* GT2 desktop */ + INTEL_VGA_DEVICE(0x0422, &intel_haswell_d_info), /* GT2 desktop */ INTEL_VGA_DEVICE(0x040a, &intel_haswell_d_info), /* GT1 server */ INTEL_VGA_DEVICE(0x041a, &intel_haswell_d_info), /* GT2 server */ + INTEL_VGA_DEVICE(0x042a, &intel_haswell_d_info), /* GT2 server */ INTEL_VGA_DEVICE(0x0406, &intel_haswell_m_info), /* GT1 mobile */ INTEL_VGA_DEVICE(0x0416, &intel_haswell_m_info), /* GT2 mobile */ - INTEL_VGA_DEVICE(0x0c16, &intel_haswell_d_info), /* SDV */ + INTEL_VGA_DEVICE(0x0426, &intel_haswell_m_info), /* GT2 mobile */ + INTEL_VGA_DEVICE(0x0C02, &intel_haswell_d_info), /* SDV GT1 desktop */ + INTEL_VGA_DEVICE(0x0C12, &intel_haswell_d_info), /* SDV GT2 desktop */ + INTEL_VGA_DEVICE(0x0C22, &intel_haswell_d_info), /* SDV GT2 desktop */ + INTEL_VGA_DEVICE(0x0C0A, &intel_haswell_d_info), /* SDV GT1 server */ + INTEL_VGA_DEVICE(0x0C1A, &intel_haswell_d_info), /* SDV GT2 server */ + INTEL_VGA_DEVICE(0x0C2A, &intel_haswell_d_info), /* SDV GT2 server */ + INTEL_VGA_DEVICE(0x0C06, &intel_haswell_m_info), /* SDV GT1 mobile */ + INTEL_VGA_DEVICE(0x0C16, &intel_haswell_m_info), /* SDV GT2 mobile */ + INTEL_VGA_DEVICE(0x0C26, &intel_haswell_m_info), /* SDV GT2 mobile */ + INTEL_VGA_DEVICE(0x0A02, &intel_haswell_d_info), /* ULT GT1 desktop */ + INTEL_VGA_DEVICE(0x0A12, &intel_haswell_d_info), /* ULT GT2 desktop */ + INTEL_VGA_DEVICE(0x0A22, &intel_haswell_d_info), /* ULT GT2 desktop */ + INTEL_VGA_DEVICE(0x0A0A, &intel_haswell_d_info), /* ULT GT1 server */ + INTEL_VGA_DEVICE(0x0A1A, &intel_haswell_d_info), /* ULT GT2 server */ + INTEL_VGA_DEVICE(0x0A2A, &intel_haswell_d_info), /* ULT GT2 server */ + INTEL_VGA_DEVICE(0x0A06, &intel_haswell_m_info), /* ULT GT1 mobile */ + INTEL_VGA_DEVICE(0x0A16, &intel_haswell_m_info), /* ULT GT2 mobile */ + INTEL_VGA_DEVICE(0x0A26, &intel_haswell_m_info), /* ULT GT2 mobile */ + INTEL_VGA_DEVICE(0x0D12, &intel_haswell_d_info), /* CRW GT1 desktop */ + INTEL_VGA_DEVICE(0x0D22, &intel_haswell_d_info), /* CRW GT2 desktop */ + INTEL_VGA_DEVICE(0x0D32, &intel_haswell_d_info), /* CRW GT2 desktop */ + INTEL_VGA_DEVICE(0x0D1A, &intel_haswell_d_info), /* CRW GT1 server */ + INTEL_VGA_DEVICE(0x0D2A, &intel_haswell_d_info), /* CRW GT2 server */ + INTEL_VGA_DEVICE(0x0D3A, &intel_haswell_d_info), /* CRW GT2 server */ + INTEL_VGA_DEVICE(0x0D16, &intel_haswell_m_info), /* CRW GT1 mobile */ + INTEL_VGA_DEVICE(0x0D26, &intel_haswell_m_info), /* CRW GT2 mobile */ + INTEL_VGA_DEVICE(0x0D36, &intel_haswell_m_info), /* CRW GT2 mobile */ INTEL_VGA_DEVICE(0x0f30, &intel_valleyview_m_info), INTEL_VGA_DEVICE(0x0157, &intel_valleyview_m_info), INTEL_VGA_DEVICE(0x0155, &intel_valleyview_d_info), -- cgit v1.2.3-58-ga151 From 0d8957c8a90bbb5d34fab9a304459448a5131e06 Mon Sep 17 00:00:00 2001 From: Daniel Vetter Date: Tue, 7 Aug 2012 09:54:14 +0200 Subject: drm/i915: correctly order the ring init sequence We may only start to set up the new register values after having confirmed that the ring is truely off. Otherwise the hw might lose the newly written register values. This is caught later on in the init sequence, when we check whether the register writes have stuck. Cc: stable@vger.kernel.org Reviewed-by: Jani Nikula Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=50522 Tested-by: Yang Guang Signed-off-by: Daniel Vetter --- drivers/gpu/drm/i915/intel_ringbuffer.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.c b/drivers/gpu/drm/i915/intel_ringbuffer.c index bf0195a96d53..414af1e2973b 100644 --- a/drivers/gpu/drm/i915/intel_ringbuffer.c +++ b/drivers/gpu/drm/i915/intel_ringbuffer.c @@ -289,8 +289,6 @@ static int init_ring_common(struct intel_ring_buffer *ring) I915_WRITE_HEAD(ring, 0); ring->write_tail(ring, 0); - /* Initialize the ring. */ - I915_WRITE_START(ring, obj->gtt_offset); head = I915_READ_HEAD(ring) & HEAD_ADDR; /* G45 ring initialization fails to reset head to zero */ @@ -316,6 +314,11 @@ static int init_ring_common(struct intel_ring_buffer *ring) } } + /* Initialize the ring. This must happen _after_ we've cleared the ring + * registers with the above sequence (the readback of the HEAD registers + * also enforces ordering), otherwise the hw might lose the new ring + * register values. */ + I915_WRITE_START(ring, obj->gtt_offset); I915_WRITE_CTL(ring, ((ring->size - PAGE_SIZE) & RING_NR_PAGES) | RING_VALID); -- cgit v1.2.3-58-ga151 From 0f457e488c79922dfd646f94ed058764f7ba758c Mon Sep 17 00:00:00 2001 From: Marek Olšák Date: Sun, 29 Jul 2012 16:24:57 +0200 Subject: drm/radeon/kms: allow "invalid" DB formats as a means to disable DB MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Signed-off-by: Marek Olšák Reviewed-by: Alex Deucher Cc: stable@vger.kernel.org Signed-off-by: Dave Airlie --- drivers/gpu/drm/radeon/evergreen_cs.c | 6 ++++-- drivers/gpu/drm/radeon/evergreend.h | 2 ++ drivers/gpu/drm/radeon/r600_cs.c | 6 ++++-- drivers/gpu/drm/radeon/radeon_drv.c | 3 ++- 4 files changed, 12 insertions(+), 5 deletions(-) diff --git a/drivers/gpu/drm/radeon/evergreen_cs.c b/drivers/gpu/drm/radeon/evergreen_cs.c index c16554122ccd..f2e5c545c973 100644 --- a/drivers/gpu/drm/radeon/evergreen_cs.c +++ b/drivers/gpu/drm/radeon/evergreen_cs.c @@ -961,13 +961,15 @@ static int evergreen_cs_track_check(struct radeon_cs_parser *p) if (track->db_dirty) { /* Check stencil buffer */ - if (G_028800_STENCIL_ENABLE(track->db_depth_control)) { + if (G_028044_FORMAT(track->db_s_info) != V_028044_STENCIL_INVALID && + G_028800_STENCIL_ENABLE(track->db_depth_control)) { r = evergreen_cs_track_validate_stencil(p); if (r) return r; } /* Check depth buffer */ - if (G_028800_Z_ENABLE(track->db_depth_control)) { + if (G_028040_FORMAT(track->db_z_info) != V_028040_Z_INVALID && + G_028800_Z_ENABLE(track->db_depth_control)) { r = evergreen_cs_track_validate_depth(p); if (r) return r; diff --git a/drivers/gpu/drm/radeon/evergreend.h b/drivers/gpu/drm/radeon/evergreend.h index d3bd098e4e19..79347855d9bf 100644 --- a/drivers/gpu/drm/radeon/evergreend.h +++ b/drivers/gpu/drm/radeon/evergreend.h @@ -1277,6 +1277,8 @@ #define S_028044_FORMAT(x) (((x) & 0x1) << 0) #define G_028044_FORMAT(x) (((x) >> 0) & 0x1) #define C_028044_FORMAT 0xFFFFFFFE +#define V_028044_STENCIL_INVALID 0 +#define V_028044_STENCIL_8 1 #define G_028044_TILE_SPLIT(x) (((x) >> 8) & 0x7) #define DB_Z_READ_BASE 0x28048 #define DB_STENCIL_READ_BASE 0x2804c diff --git a/drivers/gpu/drm/radeon/r600_cs.c b/drivers/gpu/drm/radeon/r600_cs.c index ca87f7afaf23..1119e31e5c2f 100644 --- a/drivers/gpu/drm/radeon/r600_cs.c +++ b/drivers/gpu/drm/radeon/r600_cs.c @@ -764,8 +764,10 @@ static int r600_cs_track_check(struct radeon_cs_parser *p) } /* Check depth buffer */ - if (track->db_dirty && (G_028800_STENCIL_ENABLE(track->db_depth_control) || - G_028800_Z_ENABLE(track->db_depth_control))) { + if (track->db_dirty && + G_028010_FORMAT(track->db_depth_info) != V_028010_DEPTH_INVALID && + (G_028800_STENCIL_ENABLE(track->db_depth_control) || + G_028800_Z_ENABLE(track->db_depth_control))) { r = r600_cs_track_validate_db(p); if (r) return r; diff --git a/drivers/gpu/drm/radeon/radeon_drv.c b/drivers/gpu/drm/radeon/radeon_drv.c index dcea6f01ae4e..4b736ecb0aa5 100644 --- a/drivers/gpu/drm/radeon/radeon_drv.c +++ b/drivers/gpu/drm/radeon/radeon_drv.c @@ -59,9 +59,10 @@ * 2.15.0 - add max_pipes query * 2.16.0 - fix evergreen 2D tiled surface calculation * 2.17.0 - add STRMOUT_BASE_UPDATE for r7xx + * 2.18.0 - r600-eg: allow "invalid" DB formats */ #define KMS_DRIVER_MAJOR 2 -#define KMS_DRIVER_MINOR 17 +#define KMS_DRIVER_MINOR 18 #define KMS_DRIVER_PATCHLEVEL 0 int radeon_driver_load_kms(struct drm_device *dev, unsigned long flags); int radeon_driver_unload_kms(struct drm_device *dev); -- cgit v1.2.3-58-ga151 From 959f724728b6a1cb5185230972119dd7bc8b610e Mon Sep 17 00:00:00 2001 From: Thomas Meyer Date: Tue, 7 Aug 2012 06:57:25 +0000 Subject: drm/udl: Use ERR_CAST inlined function instead of ERR_PTR(PTR_ERR(.. [1] The semantic patch that makes this change is available in scripts/coccinelle/api/err_cast.cocci. More information about semantic patching is available at http://coccinelle.lip6.fr/ Signed-off-by: Thomas Meyer Signed-off-by: Dave Airlie --- drivers/gpu/drm/udl/udl_gem.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/udl/udl_gem.c b/drivers/gpu/drm/udl/udl_gem.c index 7bd65bdd15a8..291ecc145585 100644 --- a/drivers/gpu/drm/udl/udl_gem.c +++ b/drivers/gpu/drm/udl/udl_gem.c @@ -308,7 +308,7 @@ struct drm_gem_object *udl_gem_prime_import(struct drm_device *dev, /* need to attach */ attach = dma_buf_attach(dma_buf, dev->dev); if (IS_ERR(attach)) - return ERR_PTR(PTR_ERR(attach)); + return ERR_CAST(attach); sg = dma_buf_map_attachment(attach, DMA_BIDIRECTIONAL); if (IS_ERR(sg)) { -- cgit v1.2.3-58-ga151 From f7b83b908fbecadefa73c0d76b9c719d09c1d96d Mon Sep 17 00:00:00 2001 From: Alexey Khoroshilov Date: Tue, 7 Aug 2012 12:23:06 +0000 Subject: drm/edid: Fix potential memory leak in edid_load() Do not leak memory by updating pointer with potentially NULL realloc return value. Found by Linux Driver Verification project (linuxtesting.org). Signed-off-by: Alexey Khoroshilov Reviewed-by: Carsten Emde Signed-off-by: Dave Airlie --- drivers/gpu/drm/drm_edid_load.c | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-) diff --git a/drivers/gpu/drm/drm_edid_load.c b/drivers/gpu/drm/drm_edid_load.c index 66d4a28ad5a2..0303935d10e2 100644 --- a/drivers/gpu/drm/drm_edid_load.c +++ b/drivers/gpu/drm/drm_edid_load.c @@ -119,7 +119,7 @@ static int edid_load(struct drm_connector *connector, char *name, { const struct firmware *fw; struct platform_device *pdev; - u8 *fwdata = NULL, *edid; + u8 *fwdata = NULL, *edid, *new_edid; int fwsize, expected; int builtin = 0, err = 0; int i, valid_extensions = 0; @@ -195,12 +195,14 @@ static int edid_load(struct drm_connector *connector, char *name, "\"%s\" for connector \"%s\"\n", valid_extensions, edid[0x7e], name, connector_name); edid[0x7e] = valid_extensions; - edid = krealloc(edid, (valid_extensions + 1) * EDID_LENGTH, + new_edid = krealloc(edid, (valid_extensions + 1) * EDID_LENGTH, GFP_KERNEL); - if (edid == NULL) { + if (new_edid == NULL) { err = -ENOMEM; + kfree(edid); goto relfw_out; } + edid = new_edid; } connector->display_info.raw_edid = edid; -- cgit v1.2.3-58-ga151 From 9830605d4c070b16ec5c24a75503877cc7698409 Mon Sep 17 00:00:00 2001 From: Dave Airlie Date: Thu, 9 Aug 2012 15:00:15 +1000 Subject: drm/mgag200: fix G200ER pll picking algorithm The original code was misported from the X driver, a) an int went to unsigned int, breaking the downward counting testm code b) the port did the vco/computed clock bits completely wrong. This fixes an infinite loop on modprobe on some Dell servers with the G200ER chipset variant. Found in internal testing. Cc: stable@vger.kernel.org Signed-off-by: Dave Airlie --- drivers/gpu/drm/mgag200/mgag200_mode.c | 12 +++++++----- 1 file changed, 7 insertions(+), 5 deletions(-) diff --git a/drivers/gpu/drm/mgag200/mgag200_mode.c b/drivers/gpu/drm/mgag200/mgag200_mode.c index a4d7c500c97b..b69642d5d850 100644 --- a/drivers/gpu/drm/mgag200/mgag200_mode.c +++ b/drivers/gpu/drm/mgag200/mgag200_mode.c @@ -468,10 +468,11 @@ static int mga_g200er_set_plls(struct mga_device *mdev, long clock) { unsigned int vcomax, vcomin, pllreffreq; unsigned int delta, tmpdelta; - unsigned int testr, testn, testm, testo; + int testr, testn, testm, testo; unsigned int p, m, n; - unsigned int computed; + unsigned int computed, vco; int tmp; + const unsigned int m_div_val[] = { 1, 2, 4, 8 }; m = n = p = 0; vcomax = 1488000; @@ -490,12 +491,13 @@ static int mga_g200er_set_plls(struct mga_device *mdev, long clock) if (delta == 0) break; for (testo = 5; testo < 33; testo++) { - computed = pllreffreq * (testn + 1) / + vco = pllreffreq * (testn + 1) / (testr + 1); - if (computed < vcomin) + if (vco < vcomin) continue; - if (computed > vcomax) + if (vco > vcomax) continue; + computed = vco / (m_div_val[testm] * (testo + 1)); if (computed > clock) tmpdelta = computed - clock; else -- cgit v1.2.3-58-ga151 From 3d61bd42135a5801ce6c2bc47f33277fde74e5ea Mon Sep 17 00:00:00 2001 From: Alex Deucher Date: Thu, 19 Jul 2012 20:11:44 -0400 Subject: drm/radeon: fix handling for ddc type 5 on combios When ddc type is 5, need to look up the i2c channel in the i2c table. Signed-off-by: Alex Deucher --- drivers/gpu/drm/radeon/radeon_combios.c | 57 +++++++++++++++++++++------------ 1 file changed, 36 insertions(+), 21 deletions(-) diff --git a/drivers/gpu/drm/radeon/radeon_combios.c b/drivers/gpu/drm/radeon/radeon_combios.c index 576f4f6919f2..f75247d42ffd 100644 --- a/drivers/gpu/drm/radeon/radeon_combios.c +++ b/drivers/gpu/drm/radeon/radeon_combios.c @@ -719,6 +719,34 @@ static struct radeon_i2c_bus_rec combios_setup_i2c_bus(struct radeon_device *rde return i2c; } +static struct radeon_i2c_bus_rec radeon_combios_get_i2c_info_from_table(struct radeon_device *rdev) +{ + struct drm_device *dev = rdev->ddev; + struct radeon_i2c_bus_rec i2c; + u16 offset; + u8 id, blocks, clk, data; + int i; + + i2c.valid = false; + + offset = combios_get_table_offset(dev, COMBIOS_I2C_INFO_TABLE); + if (offset) { + blocks = RBIOS8(offset + 2); + for (i = 0; i < blocks; i++) { + id = RBIOS8(offset + 3 + (i * 5) + 0); + if (id == 136) { + clk = RBIOS8(offset + 3 + (i * 5) + 3); + data = RBIOS8(offset + 3 + (i * 5) + 4); + /* gpiopad */ + i2c = combios_setup_i2c_bus(rdev, DDC_MONID, + (1 << clk), (1 << data)); + break; + } + } + } + return i2c; +} + void radeon_combios_i2c_init(struct radeon_device *rdev) { struct drm_device *dev = rdev->ddev; @@ -755,30 +783,14 @@ void radeon_combios_i2c_init(struct radeon_device *rdev) } else if (rdev->family == CHIP_RS300 || rdev->family == CHIP_RS400 || rdev->family == CHIP_RS480) { - u16 offset; - u8 id, blocks, clk, data; - int i; - /* 0x68 */ i2c = combios_setup_i2c_bus(rdev, DDC_CRT2, 0, 0); rdev->i2c_bus[3] = radeon_i2c_create(dev, &i2c, "MONID"); - offset = combios_get_table_offset(dev, COMBIOS_I2C_INFO_TABLE); - if (offset) { - blocks = RBIOS8(offset + 2); - for (i = 0; i < blocks; i++) { - id = RBIOS8(offset + 3 + (i * 5) + 0); - if (id == 136) { - clk = RBIOS8(offset + 3 + (i * 5) + 3); - data = RBIOS8(offset + 3 + (i * 5) + 4); - /* gpiopad */ - i2c = combios_setup_i2c_bus(rdev, DDC_MONID, - (1 << clk), (1 << data)); - rdev->i2c_bus[4] = radeon_i2c_create(dev, &i2c, "GPIOPAD_MASK"); - break; - } - } - } + /* gpiopad */ + i2c = radeon_combios_get_i2c_info_from_table(rdev); + if (i2c.valid) + rdev->i2c_bus[4] = radeon_i2c_create(dev, &i2c, "GPIOPAD_MASK"); } else if ((rdev->family == CHIP_R200) || (rdev->family >= CHIP_R300)) { /* 0x68 */ @@ -2321,7 +2333,10 @@ bool radeon_get_legacy_connector_info_from_bios(struct drm_device *dev) connector = (tmp >> 12) & 0xf; ddc_type = (tmp >> 8) & 0xf; - ddc_i2c = combios_setup_i2c_bus(rdev, ddc_type, 0, 0); + if (ddc_type == 5) + ddc_i2c = radeon_combios_get_i2c_info_from_table(rdev); + else + ddc_i2c = combios_setup_i2c_bus(rdev, ddc_type, 0, 0); switch (connector) { case CONNECTOR_PROPRIETARY_LEGACY: -- cgit v1.2.3-58-ga151 From f4254a2bb4555e26289a19b71bfa8a4965c5b5ee Mon Sep 17 00:00:00 2001 From: Alex Deucher Date: Tue, 10 Jul 2012 15:20:24 -0400 Subject: drm/radeon/dce4+: set a more reasonable cursor watermark Set a more reasonable default cursor watermark. The recommended default value is 4. This should reduce urgency requests to the MC form the display hw. Signed-off-by: Alex Deucher --- drivers/gpu/drm/radeon/radeon_cursor.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/radeon/radeon_cursor.c b/drivers/gpu/drm/radeon/radeon_cursor.c index 711e95ad39bf..8794744cdf1a 100644 --- a/drivers/gpu/drm/radeon/radeon_cursor.c +++ b/drivers/gpu/drm/radeon/radeon_cursor.c @@ -67,7 +67,8 @@ static void radeon_hide_cursor(struct drm_crtc *crtc) if (ASIC_IS_DCE4(rdev)) { WREG32(RADEON_MM_INDEX, EVERGREEN_CUR_CONTROL + radeon_crtc->crtc_offset); - WREG32(RADEON_MM_DATA, EVERGREEN_CURSOR_MODE(EVERGREEN_CURSOR_24_8_PRE_MULT)); + WREG32(RADEON_MM_DATA, EVERGREEN_CURSOR_MODE(EVERGREEN_CURSOR_24_8_PRE_MULT) | + EVERGREEN_CURSOR_URGENT_CONTROL(EVERGREEN_CURSOR_URGENT_1_2)); } else if (ASIC_IS_AVIVO(rdev)) { WREG32(RADEON_MM_INDEX, AVIVO_D1CUR_CONTROL + radeon_crtc->crtc_offset); WREG32(RADEON_MM_DATA, (AVIVO_D1CURSOR_MODE_24BPP << AVIVO_D1CURSOR_MODE_SHIFT)); @@ -94,7 +95,8 @@ static void radeon_show_cursor(struct drm_crtc *crtc) if (ASIC_IS_DCE4(rdev)) { WREG32(RADEON_MM_INDEX, EVERGREEN_CUR_CONTROL + radeon_crtc->crtc_offset); WREG32(RADEON_MM_DATA, EVERGREEN_CURSOR_EN | - EVERGREEN_CURSOR_MODE(EVERGREEN_CURSOR_24_8_PRE_MULT)); + EVERGREEN_CURSOR_MODE(EVERGREEN_CURSOR_24_8_PRE_MULT) | + EVERGREEN_CURSOR_URGENT_CONTROL(EVERGREEN_CURSOR_URGENT_1_2)); } else if (ASIC_IS_AVIVO(rdev)) { WREG32(RADEON_MM_INDEX, AVIVO_D1CUR_CONTROL + radeon_crtc->crtc_offset); WREG32(RADEON_MM_DATA, AVIVO_D1CURSOR_EN | -- cgit v1.2.3-58-ga151 From 3838f46e363d5a07e61d0352bf11d534dc61f921 Mon Sep 17 00:00:00 2001 From: Alex Deucher Date: Wed, 25 Jul 2012 12:32:59 -0400 Subject: drm/radeon: properly handle SS overrides on TN (v2) The IntegratedSystemInfo table changed versions on TN. Update the SS override lookup to handle it. v2: fix copy-paste typo. Signed-off-by: Alex Deucher --- drivers/gpu/drm/radeon/radeon_atombios.c | 49 ++++++++++++++++++++++++-------- 1 file changed, 37 insertions(+), 12 deletions(-) diff --git a/drivers/gpu/drm/radeon/radeon_atombios.c b/drivers/gpu/drm/radeon/radeon_atombios.c index b1e3820df363..f9c21f9d16bc 100644 --- a/drivers/gpu/drm/radeon/radeon_atombios.c +++ b/drivers/gpu/drm/radeon/radeon_atombios.c @@ -1263,6 +1263,8 @@ bool radeon_atom_get_clock_info(struct drm_device *dev) union igp_info { struct _ATOM_INTEGRATED_SYSTEM_INFO info; struct _ATOM_INTEGRATED_SYSTEM_INFO_V2 info_2; + struct _ATOM_INTEGRATED_SYSTEM_INFO_V6 info_6; + struct _ATOM_INTEGRATED_SYSTEM_INFO_V1_7 info_7; }; bool radeon_atombios_sideport_present(struct radeon_device *rdev) @@ -1390,27 +1392,50 @@ static void radeon_atombios_get_igp_ss_overrides(struct radeon_device *rdev, struct radeon_mode_info *mode_info = &rdev->mode_info; int index = GetIndexIntoMasterTable(DATA, IntegratedSystemInfo); u16 data_offset, size; - struct _ATOM_INTEGRATED_SYSTEM_INFO_V6 *igp_info; + union igp_info *igp_info; u8 frev, crev; u16 percentage = 0, rate = 0; /* get any igp specific overrides */ if (atom_parse_data_header(mode_info->atom_context, index, &size, &frev, &crev, &data_offset)) { - igp_info = (struct _ATOM_INTEGRATED_SYSTEM_INFO_V6 *) + igp_info = (union igp_info *) (mode_info->atom_context->bios + data_offset); - switch (id) { - case ASIC_INTERNAL_SS_ON_TMDS: - percentage = le16_to_cpu(igp_info->usDVISSPercentage); - rate = le16_to_cpu(igp_info->usDVISSpreadRateIn10Hz); + switch (crev) { + case 6: + switch (id) { + case ASIC_INTERNAL_SS_ON_TMDS: + percentage = le16_to_cpu(igp_info->info_6.usDVISSPercentage); + rate = le16_to_cpu(igp_info->info_6.usDVISSpreadRateIn10Hz); + break; + case ASIC_INTERNAL_SS_ON_HDMI: + percentage = le16_to_cpu(igp_info->info_6.usHDMISSPercentage); + rate = le16_to_cpu(igp_info->info_6.usHDMISSpreadRateIn10Hz); + break; + case ASIC_INTERNAL_SS_ON_LVDS: + percentage = le16_to_cpu(igp_info->info_6.usLvdsSSPercentage); + rate = le16_to_cpu(igp_info->info_6.usLvdsSSpreadRateIn10Hz); + break; + } break; - case ASIC_INTERNAL_SS_ON_HDMI: - percentage = le16_to_cpu(igp_info->usHDMISSPercentage); - rate = le16_to_cpu(igp_info->usHDMISSpreadRateIn10Hz); + case 7: + switch (id) { + case ASIC_INTERNAL_SS_ON_TMDS: + percentage = le16_to_cpu(igp_info->info_7.usDVISSPercentage); + rate = le16_to_cpu(igp_info->info_7.usDVISSpreadRateIn10Hz); + break; + case ASIC_INTERNAL_SS_ON_HDMI: + percentage = le16_to_cpu(igp_info->info_7.usHDMISSPercentage); + rate = le16_to_cpu(igp_info->info_7.usHDMISSpreadRateIn10Hz); + break; + case ASIC_INTERNAL_SS_ON_LVDS: + percentage = le16_to_cpu(igp_info->info_7.usLvdsSSPercentage); + rate = le16_to_cpu(igp_info->info_7.usLvdsSSpreadRateIn10Hz); + break; + } break; - case ASIC_INTERNAL_SS_ON_LVDS: - percentage = le16_to_cpu(igp_info->usLvdsSSPercentage); - rate = le16_to_cpu(igp_info->usLvdsSSpreadRateIn10Hz); + default: + DRM_ERROR("Unsupported IGP table: %d %d\n", frev, crev); break; } if (percentage) -- cgit v1.2.3-58-ga151 From 6c0ae2ab85fc4a95cae82047a7db1f688a7737ab Mon Sep 17 00:00:00 2001 From: Alex Deucher Date: Thu, 26 Jul 2012 13:38:52 -0400 Subject: drm/radeon: properly handle crtc powergating Need to make sure the crtc is gated on before modesetting. Explicitly gate the crtc on in prepare() and set a flag so that the dpms functions don't gate it off during mode set. Noticed by sylware on IRC. Signed-off-by: Alex Deucher Cc: stable@vger.kernel.org --- drivers/gpu/drm/radeon/atombios_crtc.c | 14 ++++++++++++-- drivers/gpu/drm/radeon/radeon_legacy_crtc.c | 4 ++++ drivers/gpu/drm/radeon/radeon_mode.h | 1 + 3 files changed, 17 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/radeon/atombios_crtc.c b/drivers/gpu/drm/radeon/atombios_crtc.c index 9e6f76fec527..dc279ebf7afb 100644 --- a/drivers/gpu/drm/radeon/atombios_crtc.c +++ b/drivers/gpu/drm/radeon/atombios_crtc.c @@ -259,7 +259,7 @@ void atombios_crtc_dpms(struct drm_crtc *crtc, int mode) /* adjust pm to dpms changes BEFORE enabling crtcs */ radeon_pm_compute_clocks(rdev); /* disable crtc pair power gating before programming */ - if (ASIC_IS_DCE6(rdev)) + if (ASIC_IS_DCE6(rdev) && !radeon_crtc->in_mode_set) atombios_powergate_crtc(crtc, ATOM_DISABLE); atombios_enable_crtc(crtc, ATOM_ENABLE); if (ASIC_IS_DCE3(rdev) && !ASIC_IS_DCE6(rdev)) @@ -279,7 +279,7 @@ void atombios_crtc_dpms(struct drm_crtc *crtc, int mode) atombios_enable_crtc(crtc, ATOM_DISABLE); radeon_crtc->enabled = false; /* power gating is per-pair */ - if (ASIC_IS_DCE6(rdev)) { + if (ASIC_IS_DCE6(rdev) && !radeon_crtc->in_mode_set) { struct drm_crtc *other_crtc; struct radeon_crtc *other_radeon_crtc; list_for_each_entry(other_crtc, &rdev->ddev->mode_config.crtc_list, head) { @@ -1635,18 +1635,28 @@ static bool atombios_crtc_mode_fixup(struct drm_crtc *crtc, static void atombios_crtc_prepare(struct drm_crtc *crtc) { struct radeon_crtc *radeon_crtc = to_radeon_crtc(crtc); + struct drm_device *dev = crtc->dev; + struct radeon_device *rdev = dev->dev_private; + radeon_crtc->in_mode_set = true; /* pick pll */ radeon_crtc->pll_id = radeon_atom_pick_pll(crtc); + /* disable crtc pair power gating before programming */ + if (ASIC_IS_DCE6(rdev)) + atombios_powergate_crtc(crtc, ATOM_DISABLE); + atombios_lock_crtc(crtc, ATOM_ENABLE); atombios_crtc_dpms(crtc, DRM_MODE_DPMS_OFF); } static void atombios_crtc_commit(struct drm_crtc *crtc) { + struct radeon_crtc *radeon_crtc = to_radeon_crtc(crtc); + atombios_crtc_dpms(crtc, DRM_MODE_DPMS_ON); atombios_lock_crtc(crtc, ATOM_DISABLE); + radeon_crtc->in_mode_set = false; } static void atombios_crtc_disable(struct drm_crtc *crtc) diff --git a/drivers/gpu/drm/radeon/radeon_legacy_crtc.c b/drivers/gpu/drm/radeon/radeon_legacy_crtc.c index d5fd615897ec..94b4a1c12893 100644 --- a/drivers/gpu/drm/radeon/radeon_legacy_crtc.c +++ b/drivers/gpu/drm/radeon/radeon_legacy_crtc.c @@ -1025,9 +1025,11 @@ static int radeon_crtc_mode_set(struct drm_crtc *crtc, static void radeon_crtc_prepare(struct drm_crtc *crtc) { + struct radeon_crtc *radeon_crtc = to_radeon_crtc(crtc); struct drm_device *dev = crtc->dev; struct drm_crtc *crtci; + radeon_crtc->in_mode_set = true; /* * The hardware wedges sometimes if you reconfigure one CRTC * whilst another is running (see fdo bug #24611). @@ -1038,6 +1040,7 @@ static void radeon_crtc_prepare(struct drm_crtc *crtc) static void radeon_crtc_commit(struct drm_crtc *crtc) { + struct radeon_crtc *radeon_crtc = to_radeon_crtc(crtc); struct drm_device *dev = crtc->dev; struct drm_crtc *crtci; @@ -1048,6 +1051,7 @@ static void radeon_crtc_commit(struct drm_crtc *crtc) if (crtci->enabled) radeon_crtc_dpms(crtci, DRM_MODE_DPMS_ON); } + radeon_crtc->in_mode_set = false; } static const struct drm_crtc_helper_funcs legacy_helper_funcs = { diff --git a/drivers/gpu/drm/radeon/radeon_mode.h b/drivers/gpu/drm/radeon/radeon_mode.h index f380d59c5763..d56978949f34 100644 --- a/drivers/gpu/drm/radeon/radeon_mode.h +++ b/drivers/gpu/drm/radeon/radeon_mode.h @@ -275,6 +275,7 @@ struct radeon_crtc { u16 lut_r[256], lut_g[256], lut_b[256]; bool enabled; bool can_tile; + bool in_mode_set; uint32_t crtc_offset; struct drm_gem_object *cursor_bo; uint64_t cursor_addr; -- cgit v1.2.3-58-ga151 From dca571a6a4edda1f61ba7ecb47431a22245490a3 Mon Sep 17 00:00:00 2001 From: Christian König Date: Tue, 31 Jul 2012 13:48:51 +0200 Subject: drm/radeon: fix bank tiling parameters on SI MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The sixteen bank case wasn't handled here, leading to GPU crashes because of userspace miscalculation. Signed-off-by: Christian König Cc: stable@vger.kernel.org Signed-off-by: Alex Deucher --- drivers/gpu/drm/radeon/si.c | 16 ++++++++++++---- 1 file changed, 12 insertions(+), 4 deletions(-) diff --git a/drivers/gpu/drm/radeon/si.c b/drivers/gpu/drm/radeon/si.c index c053f8193771..c153a7f359c8 100644 --- a/drivers/gpu/drm/radeon/si.c +++ b/drivers/gpu/drm/radeon/si.c @@ -1639,11 +1639,19 @@ static void si_gpu_init(struct radeon_device *rdev) /* XXX what about 12? */ rdev->config.si.tile_config |= (3 << 0); break; - } - if ((mc_arb_ramcfg & NOOFBANK_MASK) >> NOOFBANK_SHIFT) - rdev->config.si.tile_config |= 1 << 4; - else + } + switch ((mc_arb_ramcfg & NOOFBANK_MASK) >> NOOFBANK_SHIFT) { + case 0: /* four banks */ rdev->config.si.tile_config |= 0 << 4; + break; + case 1: /* eight banks */ + rdev->config.si.tile_config |= 1 << 4; + break; + case 2: /* sixteen banks */ + default: + rdev->config.si.tile_config |= 2 << 4; + break; + } rdev->config.si.tile_config |= ((gb_addr_config & PIPE_INTERLEAVE_SIZE_MASK) >> PIPE_INTERLEAVE_SIZE_SHIFT) << 8; rdev->config.si.tile_config |= -- cgit v1.2.3-58-ga151 From c8d15edc17d836686d1f071e564800e1a2724fa6 Mon Sep 17 00:00:00 2001 From: Alex Deucher Date: Tue, 31 Jul 2012 11:01:10 -0400 Subject: drm/radeon: fix bank tiling parameters on evergreen Handle the 16 bank case. Signed-off-by: Alex Deucher Cc: stable@vger.kernel.org --- drivers/gpu/drm/radeon/evergreen.c | 14 +++++++++++--- 1 file changed, 11 insertions(+), 3 deletions(-) diff --git a/drivers/gpu/drm/radeon/evergreen.c b/drivers/gpu/drm/radeon/evergreen.c index e585a3b947eb..f4ef24f11419 100644 --- a/drivers/gpu/drm/radeon/evergreen.c +++ b/drivers/gpu/drm/radeon/evergreen.c @@ -1986,10 +1986,18 @@ static void evergreen_gpu_init(struct radeon_device *rdev) if (rdev->flags & RADEON_IS_IGP) rdev->config.evergreen.tile_config |= 1 << 4; else { - if ((mc_arb_ramcfg & NOOFBANK_MASK) >> NOOFBANK_SHIFT) - rdev->config.evergreen.tile_config |= 1 << 4; - else + switch ((mc_arb_ramcfg & NOOFBANK_MASK) >> NOOFBANK_SHIFT) { + case 0: /* four banks */ rdev->config.evergreen.tile_config |= 0 << 4; + break; + case 1: /* eight banks */ + rdev->config.evergreen.tile_config |= 1 << 4; + break; + case 2: /* sixteen banks */ + default: + rdev->config.evergreen.tile_config |= 2 << 4; + break; + } } rdev->config.evergreen.tile_config |= 0 << 8; rdev->config.evergreen.tile_config |= -- cgit v1.2.3-58-ga151 From 5b23c9045a8b61352986270b2d109edf5085e113 Mon Sep 17 00:00:00 2001 From: Alex Deucher Date: Tue, 31 Jul 2012 11:05:11 -0400 Subject: drm/radeon: fix bank tiling parameters on cayman Handle the 16 bank case. Signed-off-by: Alex Deucher Cc: stable@vger.kernel.org --- drivers/gpu/drm/radeon/ni.c | 14 +++++++++++--- 1 file changed, 11 insertions(+), 3 deletions(-) diff --git a/drivers/gpu/drm/radeon/ni.c b/drivers/gpu/drm/radeon/ni.c index 9945d86d9001..853800e8582f 100644 --- a/drivers/gpu/drm/radeon/ni.c +++ b/drivers/gpu/drm/radeon/ni.c @@ -574,10 +574,18 @@ static void cayman_gpu_init(struct radeon_device *rdev) if (rdev->flags & RADEON_IS_IGP) rdev->config.cayman.tile_config |= 1 << 4; else { - if ((mc_arb_ramcfg & NOOFBANK_MASK) >> NOOFBANK_SHIFT) - rdev->config.cayman.tile_config |= 1 << 4; - else + switch ((mc_arb_ramcfg & NOOFBANK_MASK) >> NOOFBANK_SHIFT) { + case 0: /* four banks */ rdev->config.cayman.tile_config |= 0 << 4; + break; + case 1: /* eight banks */ + rdev->config.cayman.tile_config |= 1 << 4; + break; + case 2: /* sixteen banks */ + default: + rdev->config.cayman.tile_config |= 2 << 4; + break; + } } rdev->config.cayman.tile_config |= ((gb_addr_config & PIPE_INTERLEAVE_SIZE_MASK) >> PIPE_INTERLEAVE_SIZE_SHIFT) << 8; -- cgit v1.2.3-58-ga151 From 81ee8fb6b52ec69eeed37fe7943446af1dccecc5 Mon Sep 17 00:00:00 2001 From: Jerome Glisse Date: Fri, 27 Jul 2012 16:32:24 -0400 Subject: drm/radeon: do not reenable crtc after moving vram start address It seems we can not update the crtc scanout address. After disabling crtc, update to base address do not take effect after crtc being reenable leading to at least frame being scanout from the old crtc base address. Disabling crtc display request lead to same behavior. So after changing the vram address if we don't keep crtc disabled we will have the GPU trying to read some random system memory address with some iommu this will broke the crtc engine and will lead to broken display and iommu error message. So to avoid this, disable crtc. For flicker less boot we will need to avoid moving the vram start address. This patch should also fix : https://bugs.freedesktop.org/show_bug.cgi?id=42373 Cc: Signed-off-by: Jerome Glisse --- drivers/gpu/drm/radeon/evergreen.c | 57 ------------------------------------ drivers/gpu/drm/radeon/radeon_asic.h | 8 ++--- drivers/gpu/drm/radeon/rv515.c | 13 -------- 3 files changed, 2 insertions(+), 76 deletions(-) diff --git a/drivers/gpu/drm/radeon/evergreen.c b/drivers/gpu/drm/radeon/evergreen.c index f4ef24f11419..e93b80a6d4e9 100644 --- a/drivers/gpu/drm/radeon/evergreen.c +++ b/drivers/gpu/drm/radeon/evergreen.c @@ -1229,24 +1229,8 @@ void evergreen_agp_enable(struct radeon_device *rdev) void evergreen_mc_stop(struct radeon_device *rdev, struct evergreen_mc_save *save) { - save->vga_control[0] = RREG32(D1VGA_CONTROL); - save->vga_control[1] = RREG32(D2VGA_CONTROL); save->vga_render_control = RREG32(VGA_RENDER_CONTROL); save->vga_hdp_control = RREG32(VGA_HDP_CONTROL); - save->crtc_control[0] = RREG32(EVERGREEN_CRTC_CONTROL + EVERGREEN_CRTC0_REGISTER_OFFSET); - save->crtc_control[1] = RREG32(EVERGREEN_CRTC_CONTROL + EVERGREEN_CRTC1_REGISTER_OFFSET); - if (rdev->num_crtc >= 4) { - save->vga_control[2] = RREG32(EVERGREEN_D3VGA_CONTROL); - save->vga_control[3] = RREG32(EVERGREEN_D4VGA_CONTROL); - save->crtc_control[2] = RREG32(EVERGREEN_CRTC_CONTROL + EVERGREEN_CRTC2_REGISTER_OFFSET); - save->crtc_control[3] = RREG32(EVERGREEN_CRTC_CONTROL + EVERGREEN_CRTC3_REGISTER_OFFSET); - } - if (rdev->num_crtc >= 6) { - save->vga_control[4] = RREG32(EVERGREEN_D5VGA_CONTROL); - save->vga_control[5] = RREG32(EVERGREEN_D6VGA_CONTROL); - save->crtc_control[4] = RREG32(EVERGREEN_CRTC_CONTROL + EVERGREEN_CRTC4_REGISTER_OFFSET); - save->crtc_control[5] = RREG32(EVERGREEN_CRTC_CONTROL + EVERGREEN_CRTC5_REGISTER_OFFSET); - } /* Stop all video */ WREG32(VGA_RENDER_CONTROL, 0); @@ -1357,47 +1341,6 @@ void evergreen_mc_resume(struct radeon_device *rdev, struct evergreen_mc_save *s /* Unlock host access */ WREG32(VGA_HDP_CONTROL, save->vga_hdp_control); mdelay(1); - /* Restore video state */ - WREG32(D1VGA_CONTROL, save->vga_control[0]); - WREG32(D2VGA_CONTROL, save->vga_control[1]); - if (rdev->num_crtc >= 4) { - WREG32(EVERGREEN_D3VGA_CONTROL, save->vga_control[2]); - WREG32(EVERGREEN_D4VGA_CONTROL, save->vga_control[3]); - } - if (rdev->num_crtc >= 6) { - WREG32(EVERGREEN_D5VGA_CONTROL, save->vga_control[4]); - WREG32(EVERGREEN_D6VGA_CONTROL, save->vga_control[5]); - } - WREG32(EVERGREEN_CRTC_UPDATE_LOCK + EVERGREEN_CRTC0_REGISTER_OFFSET, 1); - WREG32(EVERGREEN_CRTC_UPDATE_LOCK + EVERGREEN_CRTC1_REGISTER_OFFSET, 1); - if (rdev->num_crtc >= 4) { - WREG32(EVERGREEN_CRTC_UPDATE_LOCK + EVERGREEN_CRTC2_REGISTER_OFFSET, 1); - WREG32(EVERGREEN_CRTC_UPDATE_LOCK + EVERGREEN_CRTC3_REGISTER_OFFSET, 1); - } - if (rdev->num_crtc >= 6) { - WREG32(EVERGREEN_CRTC_UPDATE_LOCK + EVERGREEN_CRTC4_REGISTER_OFFSET, 1); - WREG32(EVERGREEN_CRTC_UPDATE_LOCK + EVERGREEN_CRTC5_REGISTER_OFFSET, 1); - } - WREG32(EVERGREEN_CRTC_CONTROL + EVERGREEN_CRTC0_REGISTER_OFFSET, save->crtc_control[0]); - WREG32(EVERGREEN_CRTC_CONTROL + EVERGREEN_CRTC1_REGISTER_OFFSET, save->crtc_control[1]); - if (rdev->num_crtc >= 4) { - WREG32(EVERGREEN_CRTC_CONTROL + EVERGREEN_CRTC2_REGISTER_OFFSET, save->crtc_control[2]); - WREG32(EVERGREEN_CRTC_CONTROL + EVERGREEN_CRTC3_REGISTER_OFFSET, save->crtc_control[3]); - } - if (rdev->num_crtc >= 6) { - WREG32(EVERGREEN_CRTC_CONTROL + EVERGREEN_CRTC4_REGISTER_OFFSET, save->crtc_control[4]); - WREG32(EVERGREEN_CRTC_CONTROL + EVERGREEN_CRTC5_REGISTER_OFFSET, save->crtc_control[5]); - } - WREG32(EVERGREEN_CRTC_UPDATE_LOCK + EVERGREEN_CRTC0_REGISTER_OFFSET, 0); - WREG32(EVERGREEN_CRTC_UPDATE_LOCK + EVERGREEN_CRTC1_REGISTER_OFFSET, 0); - if (rdev->num_crtc >= 4) { - WREG32(EVERGREEN_CRTC_UPDATE_LOCK + EVERGREEN_CRTC2_REGISTER_OFFSET, 0); - WREG32(EVERGREEN_CRTC_UPDATE_LOCK + EVERGREEN_CRTC3_REGISTER_OFFSET, 0); - } - if (rdev->num_crtc >= 6) { - WREG32(EVERGREEN_CRTC_UPDATE_LOCK + EVERGREEN_CRTC4_REGISTER_OFFSET, 0); - WREG32(EVERGREEN_CRTC_UPDATE_LOCK + EVERGREEN_CRTC5_REGISTER_OFFSET, 0); - } WREG32(VGA_RENDER_CONTROL, save->vga_render_control); } diff --git a/drivers/gpu/drm/radeon/radeon_asic.h b/drivers/gpu/drm/radeon/radeon_asic.h index f4af24310438..0d445e7d00d1 100644 --- a/drivers/gpu/drm/radeon/radeon_asic.h +++ b/drivers/gpu/drm/radeon/radeon_asic.h @@ -255,13 +255,10 @@ extern int rs690_mc_wait_for_idle(struct radeon_device *rdev); * rv515 */ struct rv515_mc_save { - u32 d1vga_control; - u32 d2vga_control; u32 vga_render_control; u32 vga_hdp_control; - u32 d1crtc_control; - u32 d2crtc_control; }; + int rv515_init(struct radeon_device *rdev); void rv515_fini(struct radeon_device *rdev); uint32_t rv515_mc_rreg(struct radeon_device *rdev, uint32_t reg); @@ -389,11 +386,10 @@ void r700_cp_fini(struct radeon_device *rdev); * evergreen */ struct evergreen_mc_save { - u32 vga_control[6]; u32 vga_render_control; u32 vga_hdp_control; - u32 crtc_control[6]; }; + void evergreen_pcie_gart_tlb_flush(struct radeon_device *rdev); int evergreen_init(struct radeon_device *rdev); void evergreen_fini(struct radeon_device *rdev); diff --git a/drivers/gpu/drm/radeon/rv515.c b/drivers/gpu/drm/radeon/rv515.c index a12fbcc8ccb6..aa8ef491ef3c 100644 --- a/drivers/gpu/drm/radeon/rv515.c +++ b/drivers/gpu/drm/radeon/rv515.c @@ -281,12 +281,8 @@ int rv515_debugfs_ga_info_init(struct radeon_device *rdev) void rv515_mc_stop(struct radeon_device *rdev, struct rv515_mc_save *save) { - save->d1vga_control = RREG32(R_000330_D1VGA_CONTROL); - save->d2vga_control = RREG32(R_000338_D2VGA_CONTROL); save->vga_render_control = RREG32(R_000300_VGA_RENDER_CONTROL); save->vga_hdp_control = RREG32(R_000328_VGA_HDP_CONTROL); - save->d1crtc_control = RREG32(R_006080_D1CRTC_CONTROL); - save->d2crtc_control = RREG32(R_006880_D2CRTC_CONTROL); /* Stop all video */ WREG32(R_0068E8_D2CRTC_UPDATE_LOCK, 0); @@ -311,15 +307,6 @@ void rv515_mc_resume(struct radeon_device *rdev, struct rv515_mc_save *save) /* Unlock host access */ WREG32(R_000328_VGA_HDP_CONTROL, save->vga_hdp_control); mdelay(1); - /* Restore video state */ - WREG32(R_000330_D1VGA_CONTROL, save->d1vga_control); - WREG32(R_000338_D2VGA_CONTROL, save->d2vga_control); - WREG32(R_0060E8_D1CRTC_UPDATE_LOCK, 1); - WREG32(R_0068E8_D2CRTC_UPDATE_LOCK, 1); - WREG32(R_006080_D1CRTC_CONTROL, save->d1crtc_control); - WREG32(R_006880_D2CRTC_CONTROL, save->d2crtc_control); - WREG32(R_0060E8_D1CRTC_UPDATE_LOCK, 0); - WREG32(R_0068E8_D2CRTC_UPDATE_LOCK, 0); WREG32(R_000300_VGA_RENDER_CONTROL, save->vga_render_control); } -- cgit v1.2.3-58-ga151 From ecd67955fd4c8e66e4df312098989d5fa7da624c Mon Sep 17 00:00:00 2001 From: Alex Deucher Date: Mon, 6 Aug 2012 17:06:03 -0400 Subject: drm/radeon: fix ordering in pll picking on dce4+ No functional change, but re-order the cases so they evaluate properly due to the way the DCE macros work. Noticed by kallisti5 on IRC. Signed-off-by: Alex Deucher --- drivers/gpu/drm/radeon/atombios_crtc.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/drivers/gpu/drm/radeon/atombios_crtc.c b/drivers/gpu/drm/radeon/atombios_crtc.c index dc279ebf7afb..c6fcb5b86a45 100644 --- a/drivers/gpu/drm/radeon/atombios_crtc.c +++ b/drivers/gpu/drm/radeon/atombios_crtc.c @@ -1531,12 +1531,12 @@ static int radeon_atom_pick_pll(struct drm_crtc *crtc) * crtc virtual pixel clock. */ if (ENCODER_MODE_IS_DP(atombios_get_encoder_mode(test_encoder))) { - if (ASIC_IS_DCE5(rdev)) - return ATOM_DCPLL; + if (rdev->clock.dp_extclk) + return ATOM_PPLL_INVALID; else if (ASIC_IS_DCE6(rdev)) return ATOM_PPLL0; - else if (rdev->clock.dp_extclk) - return ATOM_PPLL_INVALID; + else if (ASIC_IS_DCE5(rdev)) + return ATOM_DCPLL; } } } -- cgit v1.2.3-58-ga151 From 2f292004dd1fb005788dc0a9cdd5559812ed866e Mon Sep 17 00:00:00 2001 From: Alex Deucher Date: Mon, 6 Aug 2012 10:03:59 -0400 Subject: drm/radeon: add some new SI pci ids Signed-off-by: Alex Deucher Cc: stable@vger.kernel.org --- include/drm/drm_pciids.h | 3 +++ 1 file changed, 3 insertions(+) diff --git a/include/drm/drm_pciids.h b/include/drm/drm_pciids.h index 7ff5c99b1638..c78bb997e2c6 100644 --- a/include/drm/drm_pciids.h +++ b/include/drm/drm_pciids.h @@ -213,9 +213,12 @@ {0x1002, 0x6800, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_PITCAIRN|RADEON_IS_MOBILITY|RADEON_NEW_MEMMAP}, \ {0x1002, 0x6801, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_PITCAIRN|RADEON_IS_MOBILITY|RADEON_NEW_MEMMAP}, \ {0x1002, 0x6802, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_PITCAIRN|RADEON_IS_MOBILITY|RADEON_NEW_MEMMAP}, \ + {0x1002, 0x6806, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_PITCAIRN|RADEON_NEW_MEMMAP}, \ {0x1002, 0x6808, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_PITCAIRN|RADEON_NEW_MEMMAP}, \ {0x1002, 0x6809, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_PITCAIRN|RADEON_NEW_MEMMAP}, \ {0x1002, 0x6810, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_PITCAIRN|RADEON_NEW_MEMMAP}, \ + {0x1002, 0x6816, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_PITCAIRN|RADEON_NEW_MEMMAP}, \ + {0x1002, 0x6817, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_PITCAIRN|RADEON_NEW_MEMMAP}, \ {0x1002, 0x6818, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_PITCAIRN|RADEON_NEW_MEMMAP}, \ {0x1002, 0x6819, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_PITCAIRN|RADEON_NEW_MEMMAP}, \ {0x1002, 0x6820, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_VERDE|RADEON_IS_MOBILITY|RADEON_NEW_MEMMAP}, \ -- cgit v1.2.3-58-ga151 From 69b62ad8a488727d5aab931c238aa399e6c3550c Mon Sep 17 00:00:00 2001 From: Alex Deucher Date: Fri, 3 Aug 2012 11:50:54 -0400 Subject: drm/radeon: fix some missing parens in asic macros Better safe than sorry. Signed-off-by: Alex Deucher Reviewed-by: Jerome Glisse --- drivers/gpu/drm/radeon/radeon.h | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) diff --git a/drivers/gpu/drm/radeon/radeon.h b/drivers/gpu/drm/radeon/radeon.h index 5431af292408..583e0f19e1c7 100644 --- a/drivers/gpu/drm/radeon/radeon.h +++ b/drivers/gpu/drm/radeon/radeon.h @@ -1733,11 +1733,11 @@ void radeon_ring_write(struct radeon_ring *ring, uint32_t v); #define radeon_pm_finish(rdev) (rdev)->asic->pm.finish((rdev)) #define radeon_pm_init_profile(rdev) (rdev)->asic->pm.init_profile((rdev)) #define radeon_pm_get_dynpm_state(rdev) (rdev)->asic->pm.get_dynpm_state((rdev)) -#define radeon_pre_page_flip(rdev, crtc) rdev->asic->pflip.pre_page_flip((rdev), (crtc)) -#define radeon_page_flip(rdev, crtc, base) rdev->asic->pflip.page_flip((rdev), (crtc), (base)) -#define radeon_post_page_flip(rdev, crtc) rdev->asic->pflip.post_page_flip((rdev), (crtc)) -#define radeon_wait_for_vblank(rdev, crtc) rdev->asic->display.wait_for_vblank((rdev), (crtc)) -#define radeon_mc_wait_for_idle(rdev) rdev->asic->mc_wait_for_idle((rdev)) +#define radeon_pre_page_flip(rdev, crtc) (rdev)->asic->pflip.pre_page_flip((rdev), (crtc)) +#define radeon_page_flip(rdev, crtc, base) (rdev)->asic->pflip.page_flip((rdev), (crtc), (base)) +#define radeon_post_page_flip(rdev, crtc) (rdev)->asic->pflip.post_page_flip((rdev), (crtc)) +#define radeon_wait_for_vblank(rdev, crtc) (rdev)->asic->display.wait_for_vblank((rdev), (crtc)) +#define radeon_mc_wait_for_idle(rdev) (rdev)->asic->mc_wait_for_idle((rdev)) /* Common functions */ /* AGP */ -- cgit v1.2.3-58-ga151 From e43b5ec05afdc232be25aa481315035c1888d389 Mon Sep 17 00:00:00 2001 From: Jerome Glisse Date: Mon, 6 Aug 2012 12:32:21 -0400 Subject: drm/radeon: fence virtual address and free it once idle v4 Virtual address need to be fenced to know when we can safely remove it. This patch also properly clear the pagetable. Previously it was serouisly broken. Kernel 3.5/3.4 need a similar patch but adapted for difference in mutex locking. v2: For to update pagetable when unbinding bo (don't bailout if bo_va->valid is true). v3: Add kernel 3.5/3.4 comment. v4: Fix compilation warnings. Signed-off-by: Jerome Glisse Reviewed-by: Alex Deucher --- drivers/gpu/drm/radeon/radeon.h | 1 + drivers/gpu/drm/radeon/radeon_cs.c | 32 +++++++++++++++++++++++++++++--- drivers/gpu/drm/radeon/radeon_gart.c | 24 ++++++++++++++++++++++-- drivers/gpu/drm/radeon/radeon_gem.c | 13 ++----------- drivers/gpu/drm/radeon/radeon_object.c | 6 +----- 5 files changed, 55 insertions(+), 21 deletions(-) diff --git a/drivers/gpu/drm/radeon/radeon.h b/drivers/gpu/drm/radeon/radeon.h index 583e0f19e1c7..b237a29142d1 100644 --- a/drivers/gpu/drm/radeon/radeon.h +++ b/drivers/gpu/drm/radeon/radeon.h @@ -300,6 +300,7 @@ struct radeon_bo_va { uint64_t soffset; uint64_t eoffset; uint32_t flags; + struct radeon_fence *fence; bool valid; }; diff --git a/drivers/gpu/drm/radeon/radeon_cs.c b/drivers/gpu/drm/radeon/radeon_cs.c index 8a4c49ef0cc4..b4a0db24f4dd 100644 --- a/drivers/gpu/drm/radeon/radeon_cs.c +++ b/drivers/gpu/drm/radeon/radeon_cs.c @@ -278,6 +278,30 @@ int radeon_cs_parser_init(struct radeon_cs_parser *p, void *data) return 0; } +static void radeon_bo_vm_fence_va(struct radeon_cs_parser *parser, + struct radeon_fence *fence) +{ + struct radeon_fpriv *fpriv = parser->filp->driver_priv; + struct radeon_vm *vm = &fpriv->vm; + struct radeon_bo_list *lobj; + + if (parser->chunk_ib_idx == -1) { + return; + } + if ((parser->cs_flags & RADEON_CS_USE_VM) == 0) { + return; + } + + list_for_each_entry(lobj, &parser->validated, tv.head) { + struct radeon_bo_va *bo_va; + struct radeon_bo *rbo = lobj->bo; + + bo_va = radeon_bo_va(rbo, vm); + radeon_fence_unref(&bo_va->fence); + bo_va->fence = radeon_fence_ref(fence); + } +} + /** * cs_parser_fini() - clean parser states * @parser: parser structure holding parsing context. @@ -290,11 +314,14 @@ static void radeon_cs_parser_fini(struct radeon_cs_parser *parser, int error) { unsigned i; - if (!error) + if (!error) { + /* fence all bo va before ttm_eu_fence_buffer_objects so bo are still reserved */ + radeon_bo_vm_fence_va(parser, parser->ib.fence); ttm_eu_fence_buffer_objects(&parser->validated, parser->ib.fence); - else + } else { ttm_eu_backoff_reservation(&parser->validated); + } if (parser->relocs != NULL) { for (i = 0; i < parser->nrelocs; i++) { @@ -388,7 +415,6 @@ static int radeon_cs_ib_vm_chunk(struct radeon_device *rdev, if (parser->chunk_ib_idx == -1) return 0; - if ((parser->cs_flags & RADEON_CS_USE_VM) == 0) return 0; diff --git a/drivers/gpu/drm/radeon/radeon_gart.c b/drivers/gpu/drm/radeon/radeon_gart.c index b3720054614d..9912182682ec 100644 --- a/drivers/gpu/drm/radeon/radeon_gart.c +++ b/drivers/gpu/drm/radeon/radeon_gart.c @@ -814,7 +814,7 @@ int radeon_vm_bo_update_pte(struct radeon_device *rdev, return -EINVAL; } - if (bo_va->valid) + if (bo_va->valid && mem) return 0; ngpu_pages = radeon_bo_ngpu_pages(bo); @@ -859,11 +859,27 @@ int radeon_vm_bo_rmv(struct radeon_device *rdev, struct radeon_bo *bo) { struct radeon_bo_va *bo_va; + int r; bo_va = radeon_bo_va(bo, vm); if (bo_va == NULL) return 0; + /* wait for va use to end */ + while (bo_va->fence) { + r = radeon_fence_wait(bo_va->fence, false); + if (r) { + DRM_ERROR("error while waiting for fence: %d\n", r); + } + if (r == -EDEADLK) { + r = radeon_gpu_reset(rdev); + if (!r) + continue; + } + break; + } + radeon_fence_unref(&bo_va->fence); + mutex_lock(&rdev->vm_manager.lock); mutex_lock(&vm->mutex); radeon_vm_bo_update_pte(rdev, vm, bo, NULL); @@ -952,12 +968,15 @@ void radeon_vm_fini(struct radeon_device *rdev, struct radeon_vm *vm) radeon_vm_unbind_locked(rdev, vm); mutex_unlock(&rdev->vm_manager.lock); - /* remove all bo */ + /* remove all bo at this point non are busy any more because unbind + * waited for the last vm fence to signal + */ r = radeon_bo_reserve(rdev->ring_tmp_bo.bo, false); if (!r) { bo_va = radeon_bo_va(rdev->ring_tmp_bo.bo, vm); list_del_init(&bo_va->bo_list); list_del_init(&bo_va->vm_list); + radeon_fence_unref(&bo_va->fence); radeon_bo_unreserve(rdev->ring_tmp_bo.bo); kfree(bo_va); } @@ -969,6 +988,7 @@ void radeon_vm_fini(struct radeon_device *rdev, struct radeon_vm *vm) r = radeon_bo_reserve(bo_va->bo, false); if (!r) { list_del_init(&bo_va->bo_list); + radeon_fence_unref(&bo_va->fence); radeon_bo_unreserve(bo_va->bo); kfree(bo_va); } diff --git a/drivers/gpu/drm/radeon/radeon_gem.c b/drivers/gpu/drm/radeon/radeon_gem.c index 84d045245739..1b57b0058ad6 100644 --- a/drivers/gpu/drm/radeon/radeon_gem.c +++ b/drivers/gpu/drm/radeon/radeon_gem.c @@ -134,25 +134,16 @@ void radeon_gem_object_close(struct drm_gem_object *obj, struct radeon_device *rdev = rbo->rdev; struct radeon_fpriv *fpriv = file_priv->driver_priv; struct radeon_vm *vm = &fpriv->vm; - struct radeon_bo_va *bo_va, *tmp; if (rdev->family < CHIP_CAYMAN) { return; } if (radeon_bo_reserve(rbo, false)) { + dev_err(rdev->dev, "leaking bo va because we fail to reserve bo\n"); return; } - list_for_each_entry_safe(bo_va, tmp, &rbo->va, bo_list) { - if (bo_va->vm == vm) { - /* remove from this vm address space */ - mutex_lock(&vm->mutex); - list_del(&bo_va->vm_list); - mutex_unlock(&vm->mutex); - list_del(&bo_va->bo_list); - kfree(bo_va); - } - } + radeon_vm_bo_rmv(rdev, vm, rbo); radeon_bo_unreserve(rbo); } diff --git a/drivers/gpu/drm/radeon/radeon_object.c b/drivers/gpu/drm/radeon/radeon_object.c index 1f1a4c803c1d..1cb014b571ab 100644 --- a/drivers/gpu/drm/radeon/radeon_object.c +++ b/drivers/gpu/drm/radeon/radeon_object.c @@ -52,11 +52,7 @@ void radeon_bo_clear_va(struct radeon_bo *bo) list_for_each_entry_safe(bo_va, tmp, &bo->va, bo_list) { /* remove from all vm address space */ - mutex_lock(&bo_va->vm->mutex); - list_del(&bo_va->vm_list); - mutex_unlock(&bo_va->vm->mutex); - list_del(&bo_va->bo_list); - kfree(bo_va); + radeon_vm_bo_rmv(bo->rdev, bo_va->vm, bo); } } -- cgit v1.2.3-58-ga151 From f00245f182fab57c5eed28a50764600b2bf1ccdb Mon Sep 17 00:00:00 2001 From: Marek Olšák Date: Thu, 9 Aug 2012 16:34:15 +0200 Subject: drm/radeon/kms: reorder code in r600_check_texture_resource MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Signed-off-by: Marek Olšák Reviewed-by: Jerome Glisse Signed-off-by: Alex Deucher --- drivers/gpu/drm/radeon/r600_cs.c | 52 ++++++++++++++++++++-------------------- 1 file changed, 26 insertions(+), 26 deletions(-) diff --git a/drivers/gpu/drm/radeon/r600_cs.c b/drivers/gpu/drm/radeon/r600_cs.c index 1119e31e5c2f..ff61402ef795 100644 --- a/drivers/gpu/drm/radeon/r600_cs.c +++ b/drivers/gpu/drm/radeon/r600_cs.c @@ -1559,13 +1559,14 @@ static int r600_check_texture_resource(struct radeon_cs_parser *p, u32 idx, u32 tiling_flags) { struct r600_cs_track *track = p->track; - u32 nfaces, llevel, blevel, w0, h0, d0; - u32 word0, word1, l0_size, mipmap_size, word2, word3; + u32 dim, nfaces, llevel, blevel, w0, h0, d0; + u32 word0, word1, l0_size, mipmap_size, word2, word3, word4, word5; u32 height_align, pitch, pitch_align, depth_align; - u32 array, barray, larray; + u32 barray, larray; u64 base_align; struct array_mode_checker array_check; u32 format; + bool is_array; /* on legacy kernel we don't perform advanced check */ if (p->rdev == NULL) @@ -1583,12 +1584,28 @@ static int r600_check_texture_resource(struct radeon_cs_parser *p, u32 idx, word0 |= S_038000_TILE_MODE(V_038000_ARRAY_1D_TILED_THIN1); } word1 = radeon_get_ib_value(p, idx + 1); + word2 = radeon_get_ib_value(p, idx + 2) << 8; + word3 = radeon_get_ib_value(p, idx + 3) << 8; + word4 = radeon_get_ib_value(p, idx + 4); + word5 = radeon_get_ib_value(p, idx + 5); + dim = G_038000_DIM(word0); w0 = G_038000_TEX_WIDTH(word0) + 1; + pitch = (G_038000_PITCH(word0) + 1) * 8; h0 = G_038004_TEX_HEIGHT(word1) + 1; d0 = G_038004_TEX_DEPTH(word1); + format = G_038004_DATA_FORMAT(word1); + blevel = G_038010_BASE_LEVEL(word4); + llevel = G_038014_LAST_LEVEL(word5); + /* pitch in texels */ + array_check.array_mode = G_038000_TILE_MODE(word0); + array_check.group_size = track->group_size; + array_check.nbanks = track->nbanks; + array_check.npipes = track->npipes; + array_check.nsamples = 1; + array_check.blocksize = r600_fmt_get_blocksize(format); nfaces = 1; - array = 0; - switch (G_038000_DIM(word0)) { + is_array = false; + switch (dim) { case V_038000_SQ_TEX_DIM_1D: case V_038000_SQ_TEX_DIM_2D: case V_038000_SQ_TEX_DIM_3D: @@ -1601,7 +1618,7 @@ static int r600_check_texture_resource(struct radeon_cs_parser *p, u32 idx, break; case V_038000_SQ_TEX_DIM_1D_ARRAY: case V_038000_SQ_TEX_DIM_2D_ARRAY: - array = 1; + is_array = true; break; case V_038000_SQ_TEX_DIM_2D_MSAA: case V_038000_SQ_TEX_DIM_2D_ARRAY_MSAA: @@ -1609,21 +1626,12 @@ static int r600_check_texture_resource(struct radeon_cs_parser *p, u32 idx, dev_warn(p->dev, "this kernel doesn't support %d texture dim\n", G_038000_DIM(word0)); return -EINVAL; } - format = G_038004_DATA_FORMAT(word1); if (!r600_fmt_is_valid_texture(format, p->family)) { dev_warn(p->dev, "%s:%d texture invalid format %d\n", __func__, __LINE__, format); return -EINVAL; } - /* pitch in texels */ - pitch = (G_038000_PITCH(word0) + 1) * 8; - array_check.array_mode = G_038000_TILE_MODE(word0); - array_check.group_size = track->group_size; - array_check.nbanks = track->nbanks; - array_check.npipes = track->npipes; - array_check.nsamples = 1; - array_check.blocksize = r600_fmt_get_blocksize(format); if (r600_get_array_mode_alignment(&array_check, &pitch_align, &height_align, &depth_align, &base_align)) { dev_warn(p->dev, "%s:%d tex array mode (%d) invalid\n", @@ -1649,20 +1657,13 @@ static int r600_check_texture_resource(struct radeon_cs_parser *p, u32 idx, return -EINVAL; } - word2 = radeon_get_ib_value(p, idx + 2) << 8; - word3 = radeon_get_ib_value(p, idx + 3) << 8; - - word0 = radeon_get_ib_value(p, idx + 4); - word1 = radeon_get_ib_value(p, idx + 5); - blevel = G_038010_BASE_LEVEL(word0); - llevel = G_038014_LAST_LEVEL(word1); if (blevel > llevel) { dev_warn(p->dev, "texture blevel %d > llevel %d\n", blevel, llevel); } - if (array == 1) { - barray = G_038014_BASE_ARRAY(word1); - larray = G_038014_LAST_ARRAY(word1); + if (is_array) { + barray = G_038014_BASE_ARRAY(word5); + larray = G_038014_LAST_ARRAY(word5); nfaces = larray - barray + 1; } @@ -1679,7 +1680,6 @@ static int r600_check_texture_resource(struct radeon_cs_parser *p, u32 idx, return -EINVAL; } /* using get ib will give us the offset into the mipmap bo */ - word3 = radeon_get_ib_value(p, idx + 3) << 8; if ((mipmap_size + word3) > radeon_bo_size(mipmap)) { /*dev_warn(p->dev, "mipmap bo too small (%d %d %d %d %d %d -> %d have %ld)\n", w0, h0, format, blevel, nlevels, word3, mipmap_size, radeon_bo_size(texture));*/ -- cgit v1.2.3-58-ga151 From b51ad12a36234c1f6707c9c54a414cfca23f3cdb Mon Sep 17 00:00:00 2001 From: Marek Olšák Date: Thu, 9 Aug 2012 16:34:16 +0200 Subject: drm/radeon/kms: add MSAA texture support for r600-evergreen MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Most of the checking seems to be in place already. As you can see, log2(number of samples) resides in LAST_LEVEL. This is required for MSAA support (namely for depth-stencil resolve and blitting between MSAA resources). Signed-off-by: Marek Olšák Reviewed-by: Jerome Glisse Signed-off-by: Alex Deucher --- drivers/gpu/drm/radeon/evergreen_cs.c | 7 +++++++ drivers/gpu/drm/radeon/r600_cs.c | 7 ++++++- drivers/gpu/drm/radeon/radeon_drv.c | 3 ++- 3 files changed, 15 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/radeon/evergreen_cs.c b/drivers/gpu/drm/radeon/evergreen_cs.c index f2e5c545c973..e44a62a07fe3 100644 --- a/drivers/gpu/drm/radeon/evergreen_cs.c +++ b/drivers/gpu/drm/radeon/evergreen_cs.c @@ -788,6 +788,13 @@ static int evergreen_cs_track_validate_texture(struct radeon_cs_parser *p, case V_030000_SQ_TEX_DIM_1D_ARRAY: case V_030000_SQ_TEX_DIM_2D_ARRAY: depth = 1; + break; + case V_030000_SQ_TEX_DIM_2D_MSAA: + case V_030000_SQ_TEX_DIM_2D_ARRAY_MSAA: + surf.nsamples = 1 << llevel; + llevel = 0; + depth = 1; + break; case V_030000_SQ_TEX_DIM_3D: break; default: diff --git a/drivers/gpu/drm/radeon/r600_cs.c b/drivers/gpu/drm/radeon/r600_cs.c index ff61402ef795..3dab49cb1d4a 100644 --- a/drivers/gpu/drm/radeon/r600_cs.c +++ b/drivers/gpu/drm/radeon/r600_cs.c @@ -1620,8 +1620,13 @@ static int r600_check_texture_resource(struct radeon_cs_parser *p, u32 idx, case V_038000_SQ_TEX_DIM_2D_ARRAY: is_array = true; break; - case V_038000_SQ_TEX_DIM_2D_MSAA: case V_038000_SQ_TEX_DIM_2D_ARRAY_MSAA: + is_array = true; + /* fall through */ + case V_038000_SQ_TEX_DIM_2D_MSAA: + array_check.nsamples = 1 << llevel; + llevel = 0; + break; default: dev_warn(p->dev, "this kernel doesn't support %d texture dim\n", G_038000_DIM(word0)); return -EINVAL; diff --git a/drivers/gpu/drm/radeon/radeon_drv.c b/drivers/gpu/drm/radeon/radeon_drv.c index 4b736ecb0aa5..a7f8ac0d8f03 100644 --- a/drivers/gpu/drm/radeon/radeon_drv.c +++ b/drivers/gpu/drm/radeon/radeon_drv.c @@ -60,9 +60,10 @@ * 2.16.0 - fix evergreen 2D tiled surface calculation * 2.17.0 - add STRMOUT_BASE_UPDATE for r7xx * 2.18.0 - r600-eg: allow "invalid" DB formats + * 2.19.0 - r600-eg: MSAA textures */ #define KMS_DRIVER_MAJOR 2 -#define KMS_DRIVER_MINOR 18 +#define KMS_DRIVER_MINOR 19 #define KMS_DRIVER_PATCHLEVEL 0 int radeon_driver_load_kms(struct drm_device *dev, unsigned long flags); int radeon_driver_unload_kms(struct drm_device *dev); -- cgit v1.2.3-58-ga151 From 6759a0a7a0496dbbd4fb062c6a76d61c55d0fbd9 Mon Sep 17 00:00:00 2001 From: Marek Olšák Date: Thu, 9 Aug 2012 16:34:17 +0200 Subject: drm/radeon/kms: implement timestamp userspace query (v2) MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Returns a snapshot of the GPU clock counter. Needed for certain OpenGL extensions. v2: agd5f - address Jerome's comments - add function documentation Signed-off-by: Marek Olšák Reviewed-by: Jerome Glisse Signed-off-by: Alex Deucher --- drivers/gpu/drm/radeon/r600.c | 20 +++++++++++++++++++ drivers/gpu/drm/radeon/r600d.h | 3 +++ drivers/gpu/drm/radeon/radeon.h | 1 + drivers/gpu/drm/radeon/radeon_asic.h | 2 ++ drivers/gpu/drm/radeon/radeon_device.c | 1 + drivers/gpu/drm/radeon/radeon_drv.c | 3 ++- drivers/gpu/drm/radeon/radeon_kms.c | 35 ++++++++++++++++++++++++++++------ drivers/gpu/drm/radeon/si.c | 19 ++++++++++++++++++ drivers/gpu/drm/radeon/sid.h | 3 +++ include/drm/radeon_drm.h | 2 ++ 10 files changed, 82 insertions(+), 7 deletions(-) diff --git a/drivers/gpu/drm/radeon/r600.c b/drivers/gpu/drm/radeon/r600.c index 637280f541a3..d79c639ae739 100644 --- a/drivers/gpu/drm/radeon/r600.c +++ b/drivers/gpu/drm/radeon/r600.c @@ -3789,3 +3789,23 @@ static void r600_pcie_gen2_enable(struct radeon_device *rdev) WREG32_PCIE_P(PCIE_LC_LINK_WIDTH_CNTL, link_width_cntl); } } + +/** + * r600_get_gpu_clock - return GPU clock counter snapshot + * + * @rdev: radeon_device pointer + * + * Fetches a GPU clock counter snapshot (R6xx-cayman). + * Returns the 64 bit clock counter snapshot. + */ +uint64_t r600_get_gpu_clock(struct radeon_device *rdev) +{ + uint64_t clock; + + mutex_lock(&rdev->gpu_clock_mutex); + WREG32(RLC_CAPTURE_GPU_CLOCK_COUNT, 1); + clock = (uint64_t)RREG32(RLC_GPU_CLOCK_COUNT_LSB) | + ((uint64_t)RREG32(RLC_GPU_CLOCK_COUNT_MSB) << 32ULL); + mutex_unlock(&rdev->gpu_clock_mutex); + return clock; +} diff --git a/drivers/gpu/drm/radeon/r600d.h b/drivers/gpu/drm/radeon/r600d.h index 4b116ae75fc2..fd328f4c3ea8 100644 --- a/drivers/gpu/drm/radeon/r600d.h +++ b/drivers/gpu/drm/radeon/r600d.h @@ -602,6 +602,9 @@ #define RLC_HB_WPTR 0x3f1c #define RLC_HB_WPTR_LSB_ADDR 0x3f14 #define RLC_HB_WPTR_MSB_ADDR 0x3f18 +#define RLC_GPU_CLOCK_COUNT_LSB 0x3f38 +#define RLC_GPU_CLOCK_COUNT_MSB 0x3f3c +#define RLC_CAPTURE_GPU_CLOCK_COUNT 0x3f40 #define RLC_MC_CNTL 0x3f44 #define RLC_UCODE_CNTL 0x3f48 #define RLC_UCODE_ADDR 0x3f2c diff --git a/drivers/gpu/drm/radeon/radeon.h b/drivers/gpu/drm/radeon/radeon.h index b237a29142d1..99304194a65c 100644 --- a/drivers/gpu/drm/radeon/radeon.h +++ b/drivers/gpu/drm/radeon/radeon.h @@ -1534,6 +1534,7 @@ struct radeon_device { unsigned debugfs_count; /* virtual memory */ struct radeon_vm_manager vm_manager; + struct mutex gpu_clock_mutex; }; int radeon_device_init(struct radeon_device *rdev, diff --git a/drivers/gpu/drm/radeon/radeon_asic.h b/drivers/gpu/drm/radeon/radeon_asic.h index 0d445e7d00d1..18c38d14c8cd 100644 --- a/drivers/gpu/drm/radeon/radeon_asic.h +++ b/drivers/gpu/drm/radeon/radeon_asic.h @@ -368,6 +368,7 @@ void r600_kms_blit_copy(struct radeon_device *rdev, unsigned num_gpu_pages, struct radeon_sa_bo *vb); int r600_mc_wait_for_idle(struct radeon_device *rdev); +uint64_t r600_get_gpu_clock(struct radeon_device *rdev); /* * rv770,rv730,rv710,rv740 @@ -468,5 +469,6 @@ int si_vm_bind(struct radeon_device *rdev, struct radeon_vm *vm, int id); void si_vm_unbind(struct radeon_device *rdev, struct radeon_vm *vm); void si_vm_tlb_flush(struct radeon_device *rdev, struct radeon_vm *vm); int si_ib_parse(struct radeon_device *rdev, struct radeon_ib *ib); +uint64_t si_get_gpu_clock(struct radeon_device *rdev); #endif diff --git a/drivers/gpu/drm/radeon/radeon_device.c b/drivers/gpu/drm/radeon/radeon_device.c index 742af8244e89..d2e243867ac6 100644 --- a/drivers/gpu/drm/radeon/radeon_device.c +++ b/drivers/gpu/drm/radeon/radeon_device.c @@ -1009,6 +1009,7 @@ int radeon_device_init(struct radeon_device *rdev, atomic_set(&rdev->ih.lock, 0); mutex_init(&rdev->gem.mutex); mutex_init(&rdev->pm.mutex); + mutex_init(&rdev->gpu_clock_mutex); init_rwsem(&rdev->pm.mclk_lock); init_rwsem(&rdev->exclusive_lock); init_waitqueue_head(&rdev->irq.vblank_queue); diff --git a/drivers/gpu/drm/radeon/radeon_drv.c b/drivers/gpu/drm/radeon/radeon_drv.c index a7f8ac0d8f03..d7269f48d37c 100644 --- a/drivers/gpu/drm/radeon/radeon_drv.c +++ b/drivers/gpu/drm/radeon/radeon_drv.c @@ -61,9 +61,10 @@ * 2.17.0 - add STRMOUT_BASE_UPDATE for r7xx * 2.18.0 - r600-eg: allow "invalid" DB formats * 2.19.0 - r600-eg: MSAA textures + * 2.20.0 - r600-si: RADEON_INFO_TIMESTAMP query */ #define KMS_DRIVER_MAJOR 2 -#define KMS_DRIVER_MINOR 19 +#define KMS_DRIVER_MINOR 20 #define KMS_DRIVER_PATCHLEVEL 0 int radeon_driver_load_kms(struct drm_device *dev, unsigned long flags); int radeon_driver_unload_kms(struct drm_device *dev); diff --git a/drivers/gpu/drm/radeon/radeon_kms.c b/drivers/gpu/drm/radeon/radeon_kms.c index 1d73f16b5d97..414b4acf6947 100644 --- a/drivers/gpu/drm/radeon/radeon_kms.c +++ b/drivers/gpu/drm/radeon/radeon_kms.c @@ -29,6 +29,7 @@ #include "drm_sarea.h" #include "radeon.h" #include "radeon_drm.h" +#include "radeon_asic.h" #include #include @@ -167,17 +168,39 @@ static void radeon_set_filp_rights(struct drm_device *dev, int radeon_info_ioctl(struct drm_device *dev, void *data, struct drm_file *filp) { struct radeon_device *rdev = dev->dev_private; - struct drm_radeon_info *info; + struct drm_radeon_info *info = data; struct radeon_mode_info *minfo = &rdev->mode_info; - uint32_t *value_ptr; - uint32_t value; + uint32_t value, *value_ptr; + uint64_t value64, *value_ptr64; struct drm_crtc *crtc; int i, found; - info = data; + /* TIMESTAMP is a 64-bit value, needs special handling. */ + if (info->request == RADEON_INFO_TIMESTAMP) { + if (rdev->family >= CHIP_R600) { + value_ptr64 = (uint64_t*)((unsigned long)info->value); + if (rdev->family >= CHIP_TAHITI) { + value64 = si_get_gpu_clock(rdev); + } else { + value64 = r600_get_gpu_clock(rdev); + } + + if (DRM_COPY_TO_USER(value_ptr64, &value64, sizeof(value64))) { + DRM_ERROR("copy_to_user %s:%u\n", __func__, __LINE__); + return -EFAULT; + } + return 0; + } else { + DRM_DEBUG_KMS("timestamp is r6xx+ only!\n"); + return -EINVAL; + } + } + value_ptr = (uint32_t *)((unsigned long)info->value); - if (DRM_COPY_FROM_USER(&value, value_ptr, sizeof(value))) + if (DRM_COPY_FROM_USER(&value, value_ptr, sizeof(value))) { + DRM_ERROR("copy_from_user %s:%u\n", __func__, __LINE__); return -EFAULT; + } switch (info->request) { case RADEON_INFO_DEVICE_ID: @@ -337,7 +360,7 @@ int radeon_info_ioctl(struct drm_device *dev, void *data, struct drm_file *filp) return -EINVAL; } if (DRM_COPY_TO_USER(value_ptr, &value, sizeof(uint32_t))) { - DRM_ERROR("copy_to_user\n"); + DRM_ERROR("copy_to_user %s:%u\n", __func__, __LINE__); return -EFAULT; } return 0; diff --git a/drivers/gpu/drm/radeon/si.c b/drivers/gpu/drm/radeon/si.c index c153a7f359c8..0139e227e3c7 100644 --- a/drivers/gpu/drm/radeon/si.c +++ b/drivers/gpu/drm/radeon/si.c @@ -3968,3 +3968,22 @@ void si_fini(struct radeon_device *rdev) rdev->bios = NULL; } +/** + * si_get_gpu_clock - return GPU clock counter snapshot + * + * @rdev: radeon_device pointer + * + * Fetches a GPU clock counter snapshot (SI). + * Returns the 64 bit clock counter snapshot. + */ +uint64_t si_get_gpu_clock(struct radeon_device *rdev) +{ + uint64_t clock; + + mutex_lock(&rdev->gpu_clock_mutex); + WREG32(RLC_CAPTURE_GPU_CLOCK_COUNT, 1); + clock = (uint64_t)RREG32(RLC_GPU_CLOCK_COUNT_LSB) | + ((uint64_t)RREG32(RLC_GPU_CLOCK_COUNT_MSB) << 32ULL); + mutex_unlock(&rdev->gpu_clock_mutex); + return clock; +} diff --git a/drivers/gpu/drm/radeon/sid.h b/drivers/gpu/drm/radeon/sid.h index 7869089e8761..ef4815c27b1c 100644 --- a/drivers/gpu/drm/radeon/sid.h +++ b/drivers/gpu/drm/radeon/sid.h @@ -698,6 +698,9 @@ #define RLC_UCODE_ADDR 0xC32C #define RLC_UCODE_DATA 0xC330 +#define RLC_GPU_CLOCK_COUNT_LSB 0xC338 +#define RLC_GPU_CLOCK_COUNT_MSB 0xC33C +#define RLC_CAPTURE_GPU_CLOCK_COUNT 0xC340 #define RLC_MC_CNTL 0xC344 #define RLC_UCODE_CNTL 0xC348 diff --git a/include/drm/radeon_drm.h b/include/drm/radeon_drm.h index 58056865b8e9..dc3a8cd7db8a 100644 --- a/include/drm/radeon_drm.h +++ b/include/drm/radeon_drm.h @@ -964,6 +964,8 @@ struct drm_radeon_cs { #define RADEON_INFO_IB_VM_MAX_SIZE 0x0f /* max pipes - needed for compute shaders */ #define RADEON_INFO_MAX_PIPES 0x10 +/* timestamp for GL_ARB_timer_query (OpenGL), returns the current GPU clock */ +#define RADEON_INFO_TIMESTAMP 0x11 struct drm_radeon_info { uint32_t request; -- cgit v1.2.3-58-ga151 From f59abbf28696389c91c2697c7db31f20cfa91d8a Mon Sep 17 00:00:00 2001 From: Dmitrii Cherkasov Date: Mon, 13 Aug 2012 10:53:29 -0400 Subject: drm/radeon: fix typo in function header comment Signed-off-by: Dmitrii Cherkasov Signed-off-by: Alex Deucher --- drivers/gpu/drm/radeon/radeon_gart.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/radeon/radeon_gart.c b/drivers/gpu/drm/radeon/radeon_gart.c index 9912182682ec..bb3b7fe05ccd 100644 --- a/drivers/gpu/drm/radeon/radeon_gart.c +++ b/drivers/gpu/drm/radeon/radeon_gart.c @@ -950,7 +950,7 @@ int radeon_vm_init(struct radeon_device *rdev, struct radeon_vm *vm) } /** - * radeon_vm_init - tear down a vm instance + * radeon_vm_fini - tear down a vm instance * * @rdev: radeon_device pointer * @vm: requested vm -- cgit v1.2.3-58-ga151