From 4c9bc75cbc6f2f447a38a123aa6e0605fab3cb7a Mon Sep 17 00:00:00 2001 From: Dave Airlie Date: Mon, 29 Jun 2009 18:29:12 +1000 Subject: drm/radeon/kms: mmio base/size should be resource_size_t. Unsigned long is incorrect for 64-bit resources on 32-bit hw. Signed-off-by: Dave Airlie --- drivers/gpu/drm/radeon/radeon.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'drivers/gpu/drm/radeon/radeon.h') diff --git a/drivers/gpu/drm/radeon/radeon.h b/drivers/gpu/drm/radeon/radeon.h index d61f2fc61df..e7662ba9abf 100644 --- a/drivers/gpu/drm/radeon/radeon.h +++ b/drivers/gpu/drm/radeon/radeon.h @@ -594,8 +594,8 @@ struct radeon_device { struct radeon_object *fbdev_robj; struct radeon_framebuffer *fbdev_rfb; /* Register mmio */ - unsigned long rmmio_base; - unsigned long rmmio_size; + resource_size_t rmmio_base; + resource_size_t rmmio_size; void *rmmio; radeon_rreg_t mm_rreg; radeon_wreg_t mm_wreg; -- cgit v1.2.3 From ad49f501867cba87e1e45e5ebae0b12435d68bf1 Mon Sep 17 00:00:00 2001 From: Dave Airlie Date: Fri, 10 Jul 2009 22:36:26 +1000 Subject: drm/ttm/radeon: add dma32 support. This add support for using dma32 memory on gpus that really need it. Currently IGPs are left without DMA32 but we might need to change that unless we can fix rs690. Signed-off-by: Dave Airlie --- drivers/gpu/drm/radeon/radeon.h | 1 + 1 file changed, 1 insertion(+) (limited to 'drivers/gpu/drm/radeon/radeon.h') diff --git a/drivers/gpu/drm/radeon/radeon.h b/drivers/gpu/drm/radeon/radeon.h index e7662ba9abf..3060ce14071 100644 --- a/drivers/gpu/drm/radeon/radeon.h +++ b/drivers/gpu/drm/radeon/radeon.h @@ -624,6 +624,7 @@ struct radeon_device { bool gpu_lockup; bool shutdown; bool suspend; + bool need_dma32; }; int radeon_device_init(struct radeon_device *rdev, -- cgit v1.2.3 From 2a0f8918fc34713ecaeb900ffb9afa61df4cb08e Mon Sep 17 00:00:00 2001 From: Dave Airlie Date: Sat, 11 Jul 2009 04:44:47 +1000 Subject: drm/radeon/kms: fix VRAM sizing like DDX does it. Doing this like the DDX seems like the most sure fire way to avoid having to reinvent it slowly and painfully. At the moment we keep getting things wrong with aper vs vram, so we know the DDX does it right. booted on PCI r100, PCIE rv370, IGP rs400. Signed-off-by: Dave Airlie --- drivers/gpu/drm/radeon/radeon.h | 2 ++ 1 file changed, 2 insertions(+) (limited to 'drivers/gpu/drm/radeon/radeon.h') diff --git a/drivers/gpu/drm/radeon/radeon.h b/drivers/gpu/drm/radeon/radeon.h index 3060ce14071..248e3341a98 100644 --- a/drivers/gpu/drm/radeon/radeon.h +++ b/drivers/gpu/drm/radeon/radeon.h @@ -541,6 +541,8 @@ union radeon_asic_config { struct r300_asic r300; }; +/* r100 */ +void r100_vram_init_sizes(struct radeon_device *rdev); /* * IOCTL. -- cgit v1.2.3 From b995e4330de0d8b1b8b9e49ce10cc6dc78e2cbba Mon Sep 17 00:00:00 2001 From: Dave Airlie Date: Tue, 14 Jul 2009 02:02:32 +1000 Subject: drm/radeon/kms: block RN50 from using 3D engine. RN50/ES1000 is a cut-down rv100 chip used in the server market. The 3D engine on these is either not there or unverified so refuse any attempt to configure registers on it. Signed-off-by: Dave Airlie --- drivers/gpu/drm/radeon/radeon.h | 2 ++ 1 file changed, 2 insertions(+) (limited to 'drivers/gpu/drm/radeon/radeon.h') diff --git a/drivers/gpu/drm/radeon/radeon.h b/drivers/gpu/drm/radeon/radeon.h index 248e3341a98..7f007185e7f 100644 --- a/drivers/gpu/drm/radeon/radeon.h +++ b/drivers/gpu/drm/radeon/radeon.h @@ -673,6 +673,8 @@ void r100_pll_errata_after_index(struct radeon_device *rdev); /* * ASICs helpers. */ +#define ASIC_IS_RN50(rdev) ((rdev->pdev->device == 0x515e) || \ + (rdev->pdev->device == 0x5969)) #define ASIC_IS_RV100(rdev) ((rdev->family == CHIP_RV100) || \ (rdev->family == CHIP_RV200) || \ (rdev->family == CHIP_RS100) || \ -- cgit v1.2.3 From e024e11070a0a0dc7163ce1ec2da354a638bdbed Mon Sep 17 00:00:00 2001 From: Dave Airlie Date: Wed, 24 Jun 2009 09:48:08 +1000 Subject: drm/radeon/kms: add initial colortiling support. MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit This adds new set/get tiling interfaces where the pitch and macro/micro tiling enables can be set. Along with a flag to decide if this object should have a surface when mapped. The only thing we need to allocate with a mapped surface should be the frontbuffer. Note rotate scanout shouldn't require one, and back/depth shouldn't either, though mesa needs some fixes. It fixes the TTM interfaces along Thomas's suggestions, and I've tested the surface stealing code with two X servers and not seen any lockdep issues. I've stopped tiling the fbcon frontbuffer, as I don't see there being any advantage other than testing, I've left the testing commands in there, just flip the fb_tiled to true in radeon_fb.c Open: Can we integrate endian swapping in with this? Future features: texture tiling - need to relocate texture registers TXOFFSET* with tiling info. This also merges Michel's cleanup surfaces regs at init time patch even though it makes sense on its own, this patch really relies on it. Some PowerMac firmwares set up a tiling surface at the beginning of VRAM which messes us up otherwise. that patch is: Signed-off-by: Michel Dänzer Signed-off-by: Dave Airlie --- drivers/gpu/drm/radeon/radeon.h | 32 ++++++++++++++++++++++++++++++-- 1 file changed, 30 insertions(+), 2 deletions(-) (limited to 'drivers/gpu/drm/radeon/radeon.h') diff --git a/drivers/gpu/drm/radeon/radeon.h b/drivers/gpu/drm/radeon/radeon.h index 7f007185e7f..af12a2fe322 100644 --- a/drivers/gpu/drm/radeon/radeon.h +++ b/drivers/gpu/drm/radeon/radeon.h @@ -201,6 +201,14 @@ int radeon_fence_wait_last(struct radeon_device *rdev); struct radeon_fence *radeon_fence_ref(struct radeon_fence *fence); void radeon_fence_unref(struct radeon_fence **fence); +/* + * Tiling registers + */ +struct radeon_surface_reg { + struct radeon_object *robj; +}; + +#define RADEON_GEM_MAX_SURFACES 8 /* * Radeon buffer. @@ -213,6 +221,7 @@ struct radeon_object_list { uint64_t gpu_offset; unsigned rdomain; unsigned wdomain; + uint32_t tiling_flags; }; int radeon_object_init(struct radeon_device *rdev); @@ -242,8 +251,15 @@ void radeon_object_list_clean(struct list_head *head); int radeon_object_fbdev_mmap(struct radeon_object *robj, struct vm_area_struct *vma); unsigned long radeon_object_size(struct radeon_object *robj); - - +void radeon_object_clear_surface_reg(struct radeon_object *robj); +int radeon_object_check_tiling(struct radeon_object *robj, bool has_moved, + bool force_drop); +void radeon_object_set_tiling_flags(struct radeon_object *robj, + uint32_t tiling_flags, uint32_t pitch); +void radeon_object_get_tiling_flags(struct radeon_object *robj, uint32_t *tiling_flags, uint32_t *pitch); +void radeon_bo_move_notify(struct ttm_buffer_object *bo, + struct ttm_mem_reg *mem); +void radeon_bo_fault_reserve_notify(struct ttm_buffer_object *bo); /* * GEM objects. */ @@ -535,6 +551,11 @@ struct radeon_asic { void (*set_memory_clock)(struct radeon_device *rdev, uint32_t mem_clock); void (*set_pcie_lanes)(struct radeon_device *rdev, int lanes); void (*set_clock_gating)(struct radeon_device *rdev, int enable); + + int (*set_surface_reg)(struct radeon_device *rdev, int reg, + uint32_t tiling_flags, uint32_t pitch, + uint32_t offset, uint32_t obj_size); + int (*clear_surface_reg)(struct radeon_device *rdev, int reg); }; union radeon_asic_config { @@ -568,6 +589,10 @@ int radeon_gem_busy_ioctl(struct drm_device *dev, void *data, int radeon_gem_wait_idle_ioctl(struct drm_device *dev, void *data, struct drm_file *filp); int radeon_cs_ioctl(struct drm_device *dev, void *data, struct drm_file *filp); +int radeon_gem_set_tiling_ioctl(struct drm_device *dev, void *data, + struct drm_file *filp); +int radeon_gem_get_tiling_ioctl(struct drm_device *dev, void *data, + struct drm_file *filp); /* @@ -627,6 +652,7 @@ struct radeon_device { bool shutdown; bool suspend; bool need_dma32; + struct radeon_surface_reg surface_regs[RADEON_GEM_MAX_SURFACES]; }; int radeon_device_init(struct radeon_device *rdev, @@ -801,5 +827,7 @@ static inline void radeon_ring_write(struct radeon_device *rdev, uint32_t v) #define radeon_set_memory_clock(rdev, e) (rdev)->asic->set_engine_clock((rdev), (e)) #define radeon_set_pcie_lanes(rdev, l) (rdev)->asic->set_pcie_lanes((rdev), (l)) #define radeon_set_clock_gating(rdev, e) (rdev)->asic->set_clock_gating((rdev), (e)) +#define radeon_set_surface_reg(rdev, r, f, p, o, s) ((rdev)->asic->set_surface_reg((rdev), (r), (f), (p), (o), (s))) +#define radeon_clear_surface_reg(rdev, r) ((rdev)->asic->clear_surface_reg((rdev), (r))) #endif -- cgit v1.2.3 From c93bb85b5cba3e3a06f2cad8e9bc5c23d3d10aac Mon Sep 17 00:00:00 2001 From: Jerome Glisse Date: Mon, 13 Jul 2009 21:04:08 +0200 Subject: drm/radeon/kms: fix bandwidth computation on avivo hardware Fix bandwidth computation and crtc priority in memory controller so that crtc memory request are fullfill in time to avoid display artifact. Signed-off-by: Jerome Glisse Signed-off-by: Dave Airlie --- drivers/gpu/drm/radeon/radeon.h | 40 +++++++++++++++++++++++++++++++++++++--- 1 file changed, 37 insertions(+), 3 deletions(-) (limited to 'drivers/gpu/drm/radeon/radeon.h') diff --git a/drivers/gpu/drm/radeon/radeon.h b/drivers/gpu/drm/radeon/radeon.h index af12a2fe322..63a3fe32e58 100644 --- a/drivers/gpu/drm/radeon/radeon.h +++ b/drivers/gpu/drm/radeon/radeon.h @@ -113,6 +113,7 @@ enum radeon_family { CHIP_RV770, CHIP_RV730, CHIP_RV710, + CHIP_RS880, CHIP_LAST, }; @@ -490,6 +491,39 @@ struct radeon_wb { uint64_t gpu_addr; }; +/** + * struct radeon_pm - power management datas + * @max_bandwidth: maximum bandwidth the gpu has (MByte/s) + * @igp_sideport_mclk: sideport memory clock Mhz (rs690,rs740,rs780,rs880) + * @igp_system_mclk: system clock Mhz (rs690,rs740,rs780,rs880) + * @igp_ht_link_clk: ht link clock Mhz (rs690,rs740,rs780,rs880) + * @igp_ht_link_width: ht link width in bits (rs690,rs740,rs780,rs880) + * @k8_bandwidth: k8 bandwidth the gpu has (MByte/s) (IGP) + * @sideport_bandwidth: sideport bandwidth the gpu has (MByte/s) (IGP) + * @ht_bandwidth: ht bandwidth the gpu has (MByte/s) (IGP) + * @core_bandwidth: core GPU bandwidth the gpu has (MByte/s) (IGP) + * @sclk: GPU clock Mhz (core bandwith depends of this clock) + * @needed_bandwidth: current bandwidth needs + * + * It keeps track of various data needed to take powermanagement decision. + * Bandwith need is used to determine minimun clock of the GPU and memory. + * Equation between gpu/memory clock and available bandwidth is hw dependent + * (type of memory, bus size, efficiency, ...) + */ +struct radeon_pm { + fixed20_12 max_bandwidth; + fixed20_12 igp_sideport_mclk; + fixed20_12 igp_system_mclk; + fixed20_12 igp_ht_link_clk; + fixed20_12 igp_ht_link_width; + fixed20_12 k8_bandwidth; + fixed20_12 sideport_bandwidth; + fixed20_12 ht_bandwidth; + fixed20_12 core_bandwidth; + fixed20_12 sclk; + fixed20_12 needed_bandwidth; +}; + /* * Benchmarking @@ -551,19 +585,17 @@ struct radeon_asic { void (*set_memory_clock)(struct radeon_device *rdev, uint32_t mem_clock); void (*set_pcie_lanes)(struct radeon_device *rdev, int lanes); void (*set_clock_gating)(struct radeon_device *rdev, int enable); - int (*set_surface_reg)(struct radeon_device *rdev, int reg, uint32_t tiling_flags, uint32_t pitch, uint32_t offset, uint32_t obj_size); int (*clear_surface_reg)(struct radeon_device *rdev, int reg); + void (*bandwidth_update)(struct radeon_device *rdev); }; union radeon_asic_config { struct r300_asic r300; }; -/* r100 */ -void r100_vram_init_sizes(struct radeon_device *rdev); /* * IOCTL. @@ -646,6 +678,7 @@ struct radeon_device { struct radeon_irq irq; struct radeon_asic *asic; struct radeon_gem gem; + struct radeon_pm pm; struct mutex cs_mutex; struct radeon_wb wb; bool gpu_lockup; @@ -829,5 +862,6 @@ static inline void radeon_ring_write(struct radeon_device *rdev, uint32_t v) #define radeon_set_clock_gating(rdev, e) (rdev)->asic->set_clock_gating((rdev), (e)) #define radeon_set_surface_reg(rdev, r, f, p, o, s) ((rdev)->asic->set_surface_reg((rdev), (r), (f), (p), (o), (s))) #define radeon_clear_surface_reg(rdev, r) ((rdev)->asic->clear_surface_reg((rdev), (r))) +#define radeon_bandwidth_update(rdev) (rdev)->asic->bandwidth_update((rdev)) #endif -- cgit v1.2.3 From 7a50f01a4ab89d5c05eb2cf62e206ac0bfc61d2c Mon Sep 17 00:00:00 2001 From: Dave Airlie Date: Tue, 21 Jul 2009 20:39:30 +1000 Subject: drm/radeon/kms: vram sizing on certain r100 chips needs workaround. If an rn50/r100/m6/m7 GPU has < 64MB RAM, i.e. 8/16/32, the aperture used to calculate the MC_FB_LOCATION needs to be worked out from the CONFIG_APER_SIZE register, and not the actual vram size. TTM VRAM size was also being initialised wrong, use actual vram size to initialise it. Signed-off-by: Dave Airlie --- drivers/gpu/drm/radeon/radeon.h | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) (limited to 'drivers/gpu/drm/radeon/radeon.h') diff --git a/drivers/gpu/drm/radeon/radeon.h b/drivers/gpu/drm/radeon/radeon.h index 63a3fe32e58..045b33b3bf2 100644 --- a/drivers/gpu/drm/radeon/radeon.h +++ b/drivers/gpu/drm/radeon/radeon.h @@ -332,8 +332,11 @@ struct radeon_mc { unsigned gtt_location; unsigned gtt_size; unsigned vram_location; - unsigned vram_size; + /* for some chips with <= 32MB we need to lie + * about vram size near mc fb location */ + unsigned mc_vram_size; unsigned vram_width; + unsigned real_vram_size; int vram_mtrr; bool vram_is_ddr; }; -- cgit v1.2.3 From ecc0b32645bf19a3a240e72be3022ab3b46ad3d0 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Michel=20D=C3=A4nzer?= Date: Tue, 21 Jul 2009 11:23:57 +0200 Subject: drm/radeon: Add radeon.test parameter for running BO GPU copy tests. MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit If enabled, during initialization BO GTT->VRAM and VRAM->GTT GPU copies are tested across the whole GTT aperture. This has helped uncover the benchmark copy size bug and verify the maximum aperture size supported by the AGP bridge in my PowerBook. Signed-off-by: Michel Dänzer Signed-off-by: Dave Airlie --- drivers/gpu/drm/radeon/radeon.h | 7 +++++++ 1 file changed, 7 insertions(+) (limited to 'drivers/gpu/drm/radeon/radeon.h') diff --git a/drivers/gpu/drm/radeon/radeon.h b/drivers/gpu/drm/radeon/radeon.h index 045b33b3bf2..b1d945b8ed6 100644 --- a/drivers/gpu/drm/radeon/radeon.h +++ b/drivers/gpu/drm/radeon/radeon.h @@ -64,6 +64,7 @@ extern int radeon_agpmode; extern int radeon_vram_limit; extern int radeon_gart_size; extern int radeon_benchmarking; +extern int radeon_testing; extern int radeon_connector_table; /* @@ -534,6 +535,12 @@ struct radeon_pm { void radeon_benchmark(struct radeon_device *rdev); +/* + * Testing + */ +void radeon_test_moves(struct radeon_device *rdev); + + /* * Debugfs */ -- cgit v1.2.3