diff options
| author | Daniel Vetter <daniel.vetter@ffwll.ch> | 2013-09-24 09:29:24 +0200 |
|---|---|---|
| committer | Daniel Vetter <daniel.vetter@ffwll.ch> | 2013-09-24 09:32:53 +0200 |
| commit | b599c89e8c5cf0c37352e0871be240291f8ce922 (patch) | |
| tree | a03ce75b3a3d6d39852bc201b26bcfab412bf55f /drivers/gpu | |
| parent | ee1452d7458451a7508e0663553ce88d63958157 (diff) | |
| parent | 4a10c2ac2f368583138b774ca41fac4207911983 (diff) | |
Merge tag 'v3.12-rc2' into drm-intel-next
Backmerge Linux 3.12-rc2 to prep for a bunch of -next patches:
- Header cleanup in intel_drv.h, both changed in -fixes and my current
-next pile.
- Cursor handling cleanup for -next which depends upon the cursor
handling fix merged into -rc2.
All just trivial conflicts of the "changed adjacent lines" type:
drivers/gpu/drm/i915/i915_gem.c
drivers/gpu/drm/i915/intel_display.c
drivers/gpu/drm/i915/intel_drv.h
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Diffstat (limited to 'drivers/gpu')
151 files changed, 1991 insertions, 1104 deletions
diff --git a/drivers/gpu/drm/ast/ast_drv.h b/drivers/gpu/drm/ast/ast_drv.h index 796dbb212a4..8492b68e873 100644 --- a/drivers/gpu/drm/ast/ast_drv.h +++ b/drivers/gpu/drm/ast/ast_drv.h @@ -177,7 +177,7 @@ uint8_t ast_get_index_reg_mask(struct ast_private *ast, static inline void ast_open_key(struct ast_private *ast) { - ast_set_index_reg_mask(ast, AST_IO_CRTC_PORT, 0xA1, 0xFF, 0x04); + ast_set_index_reg(ast, AST_IO_CRTC_PORT, 0x80, 0xA8); } #define AST_VIDMEM_SIZE_8M 0x00800000 diff --git a/drivers/gpu/drm/drm_context.c b/drivers/gpu/drm/drm_context.c index b4fb86d8985..224ff965bcf 100644 --- a/drivers/gpu/drm/drm_context.c +++ b/drivers/gpu/drm/drm_context.c @@ -42,6 +42,10 @@ #include <drm/drmP.h> +/******************************************************************/ +/** \name Context bitmap support */ +/*@{*/ + /** * Free a handle from the context bitmap. * @@ -52,48 +56,13 @@ * in drm_device::ctx_idr, while holding the drm_device::struct_mutex * lock. */ -static void drm_ctxbitmap_free(struct drm_device * dev, int ctx_handle) +void drm_ctxbitmap_free(struct drm_device * dev, int ctx_handle) { - if (drm_core_check_feature(dev, DRIVER_MODESET)) - return; - mutex_lock(&dev->struct_mutex); idr_remove(&dev->ctx_idr, ctx_handle); mutex_unlock(&dev->struct_mutex); } -/******************************************************************/ -/** \name Context bitmap support */ -/*@{*/ - -void drm_legacy_ctxbitmap_release(struct drm_device *dev, - struct drm_file *file_priv) -{ - if (drm_core_check_feature(dev, DRIVER_MODESET)) - return; - - mutex_lock(&dev->ctxlist_mutex); - if (!list_empty(&dev->ctxlist)) { - struct drm_ctx_list *pos, *n; - - list_for_each_entry_safe(pos, n, &dev->ctxlist, head) { - if (pos->tag == file_priv && - pos->handle != DRM_KERNEL_CONTEXT) { - if (dev->driver->context_dtor) - dev->driver->context_dtor(dev, - pos->handle); - - drm_ctxbitmap_free(dev, pos->handle); - - list_del(&pos->head); - kfree(pos); - --dev->ctx_count; - } - } - } - mutex_unlock(&dev->ctxlist_mutex); -} - /** * Context bitmap allocation. * @@ -121,12 +90,10 @@ static int drm_ctxbitmap_next(struct drm_device * dev) * * Initialise the drm_device::ctx_idr */ -void drm_legacy_ctxbitmap_init(struct drm_device * dev) +int drm_ctxbitmap_init(struct drm_device * dev) { - if (drm_core_check_feature(dev, DRIVER_MODESET)) - return; - idr_init(&dev->ctx_idr); + return 0; } /** @@ -137,7 +104,7 @@ void drm_legacy_ctxbitmap_init(struct drm_device * dev) * Free all idr members using drm_ctx_sarea_free helper function * while holding the drm_device::struct_mutex lock. */ -void drm_legacy_ctxbitmap_cleanup(struct drm_device * dev) +void drm_ctxbitmap_cleanup(struct drm_device * dev) { mutex_lock(&dev->struct_mutex); idr_destroy(&dev->ctx_idr); @@ -169,9 +136,6 @@ int drm_getsareactx(struct drm_device *dev, void *data, struct drm_local_map *map; struct drm_map_list *_entry; - if (drm_core_check_feature(dev, DRIVER_MODESET)) - return -EINVAL; - mutex_lock(&dev->struct_mutex); map = idr_find(&dev->ctx_idr, request->ctx_id); @@ -216,9 +180,6 @@ int drm_setsareactx(struct drm_device *dev, void *data, struct drm_local_map *map = NULL; struct drm_map_list *r_list = NULL; - if (drm_core_check_feature(dev, DRIVER_MODESET)) - return -EINVAL; - mutex_lock(&dev->struct_mutex); list_for_each_entry(r_list, &dev->maplist, head) { if (r_list->map @@ -319,9 +280,6 @@ int drm_resctx(struct drm_device *dev, void *data, struct drm_ctx ctx; int i; - if (drm_core_check_feature(dev, DRIVER_MODESET)) - return -EINVAL; - if (res->count >= DRM_RESERVED_CONTEXTS) { memset(&ctx, 0, sizeof(ctx)); for (i = 0; i < DRM_RESERVED_CONTEXTS; i++) { @@ -352,9 +310,6 @@ int drm_addctx(struct drm_device *dev, void *data, struct drm_ctx_list *ctx_entry; struct drm_ctx *ctx = data; - if (drm_core_check_feature(dev, DRIVER_MODESET)) - return -EINVAL; - ctx->handle = drm_ctxbitmap_next(dev); if (ctx->handle == DRM_KERNEL_CONTEXT) { /* Skip kernel's context and get a new one. */ @@ -398,9 +353,6 @@ int drm_getctx(struct drm_device *dev, void *data, struct drm_file *file_priv) { struct drm_ctx *ctx = data; - if (drm_core_check_feature(dev, DRIVER_MODESET)) - return -EINVAL; - /* This is 0, because we don't handle any context flags */ ctx->flags = 0; @@ -423,9 +375,6 @@ int drm_switchctx(struct drm_device *dev, void *data, { struct drm_ctx *ctx = data; - if (drm_core_check_feature(dev, DRIVER_MODESET)) - return -EINVAL; - DRM_DEBUG("%d\n", ctx->handle); return drm_context_switch(dev, dev->last_context, ctx->handle); } @@ -446,9 +395,6 @@ int drm_newctx(struct drm_device *dev, void *data, { struct drm_ctx *ctx = data; - if (drm_core_check_feature(dev, DRIVER_MODESET)) - return -EINVAL; - DRM_DEBUG("%d\n", ctx->handle); drm_context_switch_complete(dev, file_priv, ctx->handle); @@ -471,9 +417,6 @@ int drm_rmctx(struct drm_device *dev, void *data, { struct drm_ctx *ctx = data; - if (drm_core_check_feature(dev, DRIVER_MODESET)) - return -EINVAL; - DRM_DEBUG("%d\n", ctx->handle); if (ctx->handle != DRM_KERNEL_CONTEXT) { if (dev->driver->context_dtor) diff --git a/drivers/gpu/drm/drm_fb_helper.c b/drivers/gpu/drm/drm_fb_helper.c index 3d13ca6e257..f6f6cc7fc13 100644 --- a/drivers/gpu/drm/drm_fb_helper.c +++ b/drivers/gpu/drm/drm_fb_helper.c @@ -416,6 +416,14 @@ static void drm_fb_helper_dpms(struct fb_info *info, int dpms_mode) return; /* + * fbdev->blank can be called from irq context in case of a panic. + * Since we already have our own special panic handler which will + * restore the fbdev console mode completely, just bail out early. + */ + if (oops_in_progress) + return; + + /* * For each CRTC in this fb, turn the connectors on/off. */ drm_modeset_lock_all(dev); diff --git a/drivers/gpu/drm/drm_fops.c b/drivers/gpu/drm/drm_fops.c index 4be8e09a32e..3f84277d703 100644 --- a/drivers/gpu/drm/drm_fops.c +++ b/drivers/gpu/drm/drm_fops.c @@ -439,7 +439,26 @@ int drm_release(struct inode *inode, struct file *filp) if (dev->driver->driver_features & DRIVER_GEM) drm_gem_release(dev, file_priv); - drm_legacy_ctxbitmap_release(dev, file_priv); + mutex_lock(&dev->ctxlist_mutex); + if (!list_empty(&dev->ctxlist)) { + struct drm_ctx_list *pos, *n; + + list_for_each_entry_safe(pos, n, &dev->ctxlist, head) { + if (pos->tag == file_priv && + pos->handle != DRM_KERNEL_CONTEXT) { + if (dev->driver->context_dtor) + dev->driver->context_dtor(dev, + pos->handle); + + drm_ctxbitmap_free(dev, pos->handle); + + list_del(&pos->head); + kfree(pos); + --dev->ctx_count; + } + } + } + mutex_unlock(&dev->ctxlist_mutex); mutex_lock(&dev->struct_mutex); diff --git a/drivers/gpu/drm/drm_gem_cma_helper.c b/drivers/gpu/drm/drm_gem_cma_helper.c index 0a4f80574eb..6b51bf90df0 100644 --- a/drivers/gpu/drm/drm_gem_cma_helper.c +++ b/drivers/gpu/drm/drm_gem_cma_helper.c @@ -210,7 +210,7 @@ int drm_gem_cma_dumb_create(struct drm_file *file_priv, cma_obj = drm_gem_cma_create_with_handle(file_priv, dev, args->size, &args->handle); - return PTR_RET(cma_obj); + return PTR_ERR_OR_ZERO(cma_obj); } EXPORT_SYMBOL_GPL(drm_gem_cma_dumb_create); diff --git a/drivers/gpu/drm/drm_stub.c b/drivers/gpu/drm/drm_stub.c index e7eb0276f7f..39d864576be 100644 --- a/drivers/gpu/drm/drm_stub.c +++ b/drivers/gpu/drm/drm_stub.c @@ -292,7 +292,13 @@ int drm_fill_in_dev(struct drm_device *dev, goto error_out_unreg; } - drm_legacy_ctxbitmap_init(dev); + + + retcode = drm_ctxbitmap_init(dev); + if (retcode) { + DRM_ERROR("Cannot allocate memory for context bitmap.\n"); + goto error_out_unreg; + } if (driver->driver_features & DRIVER_GEM) { retcode = drm_gem_init(dev); @@ -446,7 +452,7 @@ void drm_put_dev(struct drm_device *dev) drm_rmmap(dev, r_list->map); drm_ht_remove(&dev->map_hash); - drm_legacy_ctxbitmap_cleanup(dev); + drm_ctxbitmap_cleanup(dev); if (drm_core_check_feature(dev, DRIVER_MODESET)) drm_put_minor(&dev->control); diff --git a/drivers/gpu/drm/exynos/Kconfig b/drivers/gpu/drm/exynos/Kconfig index 772c62a6e2a..45b6ef59596 100644 --- a/drivers/gpu/drm/exynos/Kconfig +++ b/drivers/gpu/drm/exynos/Kconfig @@ -1,11 +1,12 @@ config DRM_EXYNOS tristate "DRM Support for Samsung SoC EXYNOS Series" - depends on DRM && (PLAT_SAMSUNG || ARCH_MULTIPLATFORM) + depends on OF && DRM && (PLAT_SAMSUNG || ARCH_MULTIPLATFORM) select DRM_KMS_HELPER select FB_CFB_FILLRECT select FB_CFB_COPYAREA select FB_CFB_IMAGEBLIT select VT_HW_CONSOLE_BINDING if FRAMEBUFFER_CONSOLE + select VIDEOMODE_HELPERS help Choose this option if you have a Samsung SoC EXYNOS chipset. If M is selected the module will be called exynosdrm. @@ -24,9 +25,8 @@ config DRM_EXYNOS_DMABUF config DRM_EXYNOS_FIMD bool "Exynos DRM FIMD" - depends on OF && DRM_EXYNOS && !FB_S3C && !ARCH_MULTIPLATFORM + depends on DRM_EXYNOS && !FB_S3C && !ARCH_MULTIPLATFORM select FB_MODE_HELPERS - select VIDEOMODE_HELPERS help Choose this option if you want to use Exynos FIMD for DRM. @@ -56,7 +56,7 @@ config DRM_EXYNOS_IPP config DRM_EXYNOS_FIMC bool "Exynos DRM FIMC" - depends on DRM_EXYNOS_IPP && MFD_SYSCON && OF + depends on DRM_EXYNOS_IPP && MFD_SYSCON help Choose this option if you want to use Exynos FIMC for DRM. diff --git a/drivers/gpu/drm/exynos/exynos_ddc.c b/drivers/gpu/drm/exynos/exynos_ddc.c index 30ef41bcd7b..6a8c84e7c83 100644 --- a/drivers/gpu/drm/exynos/exynos_ddc.c +++ b/drivers/gpu/drm/exynos/exynos_ddc.c @@ -15,7 +15,7 @@ #include <linux/kernel.h> #include <linux/i2c.h> - +#include <linux/of.h> #include "exynos_drm_drv.h" #include "exynos_hdmi.h" @@ -41,13 +41,6 @@ static int s5p_ddc_remove(struct i2c_client *client) return 0; } -static struct i2c_device_id ddc_idtable[] = { - {"s5p_ddc", 0}, - {"exynos5-hdmiddc", 0}, - { }, -}; - -#ifdef CONFIG_OF static struct of_device_id hdmiddc_match_types[] = { { .compatible = "samsung,exynos5-hdmiddc", @@ -57,15 +50,13 @@ static struct of_device_id hdmiddc_match_types[] = { /* end node */ } }; -#endif struct i2c_driver ddc_driver = { .driver = { .name = "exynos-hdmiddc", .owner = THIS_MODULE, - .of_match_table = of_match_ptr(hdmiddc_match_types), + .of_match_table = hdmiddc_match_types, }, - .id_table = ddc_idtable, .probe = s5p_ddc_probe, .remove = s5p_ddc_remove, .command = NULL, diff --git a/drivers/gpu/drm/exynos/exynos_drm_buf.c b/drivers/gpu/drm/exynos/exynos_drm_buf.c index b8ac06d92fb..9c8088462c2 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_buf.c +++ b/drivers/gpu/drm/exynos/exynos_drm_buf.c @@ -63,7 +63,8 @@ static int lowlevel_buffer_allocate(struct drm_device *dev, return -ENOMEM; } - buf->kvaddr = dma_alloc_attrs(dev->dev, buf->size, + buf->kvaddr = (void __iomem *)dma_alloc_attrs(dev->dev, + buf->size, &buf->dma_addr, GFP_KERNEL, &buf->dma_attrs); if (!buf->kvaddr) { @@ -90,9 +91,9 @@ static int lowlevel_buffer_allocate(struct drm_device *dev, } buf->sgt = drm_prime_pages_to_sg(buf->pages, nr_pages); - if (!buf->sgt) { + if (IS_ERR(buf->sgt)) { DRM_ERROR("failed to get sg table.\n"); - ret = -ENOMEM; + ret = PTR_ERR(buf->sgt); goto err_free_attrs; } @@ -149,10 +150,8 @@ struct exynos_drm_gem_buf *exynos_drm_init_buf(struct drm_device *dev, DRM_DEBUG_KMS("desired size = 0x%x\n", size); buffer = kzalloc(sizeof(*buffer), GFP_KERNEL); - if (!buffer) { - DRM_ERROR("failed to allocate exynos_drm_gem_buf.\n"); + if (!buffer) return NULL; - } buffer->size = size; return buffer; @@ -161,11 +160,6 @@ struct exynos_drm_gem_buf *exynos_drm_init_buf(struct drm_device *dev, void exynos_drm_fini_buf(struct drm_device *dev, struct exynos_drm_gem_buf *buffer) { - if (!buffer) { - DRM_DEBUG_KMS("buffer is null.\n"); - return; - } - kfree(buffer); buffer = NULL; } diff --git a/drivers/gpu/drm/exynos/exynos_drm_connector.c b/drivers/gpu/drm/exynos/exynos_drm_connector.c index 02a8bc5226c..e082efb2fec 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_connector.c +++ b/drivers/gpu/drm/exynos/exynos_drm_connector.c @@ -17,6 +17,7 @@ #include <drm/exynos_drm.h> #include "exynos_drm_drv.h" #include "exynos_drm_encoder.h" +#include "exynos_drm_connector.h" #define to_exynos_connector(x) container_of(x, struct exynos_drm_connector,\ drm_connector) @@ -28,35 +29,6 @@ struct exynos_drm_connector { uint32_t dpms; }; -/* convert exynos_video_timings to drm_display_mode */ -static inline void -convert_to_display_mode(struct drm_display_mode *mode, - struct exynos_drm_panel_info *panel) -{ - struct fb_videomode *timing = &panel->timing; - - mode->clock = timing->pixclock / 1000; - mode->vrefresh = timing->refresh; - - mode->hdisplay = timing->xres; - mode->hsync_start = mode->hdisplay + timing->right_margin; - mode->hsync_end = mode->hsync_start + timing->hsync_len; - mode->htotal = mode->hsync_end + timing->left_margin; - - mode->vdisplay = timing->yres; - mode->vsync_start = mode->vdisplay + timing->lower_margin; - mode->vsync_end = mode->vsync_start + timing->vsync_len; - mode->vtotal = mode->vsync_end + timing->upper_margin; - mode->width_mm = panel->width_mm; - mode->height_mm = panel->height_mm; - - if (timing->vmode & FB_VMODE_INTERLACED) - mode->flags |= DRM_MODE_FLAG_INTERLACE; - - if (timing->vmode & FB_VMODE_DOUBLE) - mode->flags |= DRM_MODE_FLAG_DBLSCAN; -} - static int exynos_drm_connector_get_modes(struct drm_connector *connector) { struct exynos_drm_connector *exynos_connector = @@ -111,7 +83,9 @@ static int exynos_drm_connector_get_modes(struct drm_connector *connector) return 0; } - convert_to_display_mode(mode, panel); + drm_display_mode_from_videomode(&panel->vm, mode); + mode->width_mm = panel->width_mm; + mode->height_mm = panel->height_mm; connector->display_info.width_mm = mode->width_mm; connector->display_info.height_mm = mode->height_mm; @@ -278,10 +252,8 @@ struct drm_connector *exynos_drm_connector_create(struct drm_device *dev, int err; exynos_connector = kzalloc(sizeof(*exynos_connector), GFP_KERNEL); - if (!exynos_connector) { - DRM_ERROR("failed to allocate connector\n"); + if (!exynos_connector) return NULL; - } connector = &exynos_connector->drm_connector; diff --git a/drivers/gpu/drm/exynos/exynos_drm_crtc.c b/drivers/gpu/drm/exynos/exynos_drm_crtc.c index 14f5c1d3402..ebc01503d50 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_crtc.c +++ b/drivers/gpu/drm/exynos/exynos_drm_crtc.c @@ -15,6 +15,7 @@ #include <drm/drmP.h> #include <drm/drm_crtc_helper.h> +#include "exynos_drm_crtc.h" #include "exynos_drm_drv.h" #include "exynos_drm_encoder.h" #include "exynos_drm_plane.h" @@ -324,10 +325,8 @@ int exynos_drm_crtc_create(struct drm_device *dev, unsigned int nr) struct drm_crtc *crtc; exynos_crtc = kzalloc(sizeof(*exynos_crtc), GFP_KERNEL); - if (!exynos_crtc) { - DRM_ERROR("failed to allocate exynos crtc\n"); + if (!exynos_crtc) return -ENOMEM; - } exynos_crtc->pipe = nr; exynos_crtc->dpms = DRM_MODE_DPMS_OFF; diff --git a/drivers/gpu/drm/exynos/exynos_drm_dmabuf.c b/drivers/gpu/drm/exynos/exynos_drm_dmabuf.c index fd76449cf45..59827cc5e77 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_dmabuf.c +++ b/drivers/gpu/drm/exynos/exynos_drm_dmabuf.c @@ -11,6 +11,7 @@ #include <drm/drmP.h> #include <drm/exynos_drm.h> +#include "exynos_drm_dmabuf.h" #include "exynos_drm_drv.h" #include "exynos_drm_gem.h" @@ -230,7 +231,6 @@ struct drm_gem_object *exynos_dmabuf_prime_import(struct drm_device *drm_dev, buffer = kzalloc(sizeof(*buffer), GFP_KERNEL); if (!buffer) { - DRM_ERROR("failed to allocate exynos_drm_gem_buf.\n"); ret = -ENOMEM; goto err_unmap_attach; } diff --git a/drivers/gpu/drm/exynos/exynos_drm_drv.c b/drivers/gpu/drm/exynos/exynos_drm_drv.c index df81d3c959b..bb82ef78ca8 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_drv.c +++ b/drivers/gpu/drm/exynos/exynos_drm_drv.c @@ -47,10 +47,8 @@ static int exynos_drm_load(struct drm_device *dev, unsigned long flags) int nr; private = kzalloc(sizeof(struct exynos_drm_private), GFP_KERNEL); - if (!private) { - DRM_ERROR("failed to allocate private\n"); + if (!private) return -ENOMEM; - } INIT_LIST_HEAD(&private->pageflip_event_list); dev->dev_private = (void *)private; diff --git a/drivers/gpu/drm/exynos/exynos_drm_encoder.c b/drivers/gpu/drm/exynos/exynos_drm_encoder.c index a99a033793b..06f1b2a09da 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_encoder.c +++ b/drivers/gpu/drm/exynos/exynos_drm_encoder.c @@ -324,10 +324,8 @@ exynos_drm_encoder_create(struct drm_device *dev, return NULL; exynos_encoder = kzalloc(sizeof(*exynos_encoder), GFP_KERNEL); - if (!exynos_encoder) { - DRM_ERROR("failed to allocate encoder\n"); + if (!exynos_encoder) return NULL; - } exynos_encoder->dpms = DRM_MODE_DPMS_OFF; exynos_encoder->manager = manager; diff --git a/drivers/gpu/drm/exynos/exynos_drm_fb.c b/drivers/gpu/drm/exynos/exynos_drm_fb.c index c2d149f0408..ea39e0ef2ae 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_fb.c +++ b/drivers/gpu/drm/exynos/exynos_drm_fb.c @@ -156,10 +156,8 @@ exynos_drm_framebuffer_init(struct drm_device *dev, } exynos_fb = kzalloc(sizeof(*exynos_fb), GFP_KERNEL); - if (!exynos_fb) { - DRM_ERROR("failed to allocate exynos drm framebuffer\n"); + if (!exynos_fb) return ERR_PTR(-ENOMEM); - } drm_helper_mode_fill_fb_struct(&exynos_fb->fb, mode_cmd); exynos_fb->exynos_gem_obj[0] = exynos_gem_obj; @@ -220,10 +218,8 @@ exynos_user_fb_create(struct drm_device *dev, struct drm_file *file_priv, int i, ret; exynos_fb = kzalloc(sizeof(*exynos_fb), GFP_KERNEL); - if (!exynos_fb) { - DRM_ERROR("failed to allocate exynos drm framebuffer\n"); + if (!exynos_fb) return ERR_PTR(-ENOMEM); - } obj = drm_gem_object_lookup(dev, file_priv, mode_cmd->handles[0]); if (!obj) { diff --git a/drivers/gpu/drm/exynos/exynos_drm_fbdev.c b/drivers/gpu/drm/exynos/exynos_drm_fbdev.c index 8e60bd61137..e7c2f2d07f1 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_fbdev.c +++ b/drivers/gpu/drm/exynos/exynos_drm_fbdev.c @@ -16,9 +16,11 @@ #include <drm/drm_crtc.h> #include <drm/drm_fb_helper.h> #include <drm/drm_crtc_helper.h> +#include <drm/exynos_drm.h> #include "exynos_drm_drv.h" #include "exynos_drm_fb.h" +#include "exynos_drm_fbdev.h" #include "exynos_drm_gem.h" #include "exynos_drm_iommu.h" @@ -97,12 +99,13 @@ static int exynos_drm_fbdev_update(struct drm_fb_helper *helper, if (is_drm_iommu_supported(dev)) { unsigned int nr_pages = buffer->size >> PAGE_SHIFT; - buffer->kvaddr = vmap(buffer->pages, nr_pages, VM_MAP, + buffer->kvaddr = (void __iomem *) vmap(buffer->pages, + nr_pages, VM_MAP, pgprot_writecombine(PAGE_KERNEL)); } else { phys_addr_t dma_addr = buffer->dma_addr; if (dma_addr) - buffer->kvaddr = phys_to_virt(dma_addr); + buffer->kvaddr = (void __iomem *)phys_to_virt(dma_addr); else buffer->kvaddr = (void __iomem *)NULL; } @@ -165,8 +168,18 @@ static int exynos_drm_fbdev_create(struct drm_fb_helper *helper, size = mode_cmd.pitches[0] * mode_cmd.height; - /* 0 means to allocate physically continuous memory */ - exynos_gem_obj = exynos_drm_gem_create(dev, 0, size); + exynos_gem_obj = exynos_drm_gem_create(dev, EXYNOS_BO_CONTIG, size); + /* + * If physically contiguous memory allocation fails and if IOMMU is + * supported then try to get buffer from non physically contiguous + * memory area. + */ + if (IS_ERR(exynos_gem_obj) && is_drm_iommu_supported(dev)) { + dev_warn(&pdev->dev, "contiguous FB allocation failed, falling back to non-contiguous\n"); + exynos_gem_obj = exynos_drm_gem_create(dev, EXYNOS_BO_NONCONTIG, + size); + } + if (IS_ERR(exynos_gem_obj)) { ret = PTR_ERR(exynos_gem_obj); goto err_release_framebuffer; @@ -236,10 +249,8 @@ int exynos_drm_fbdev_init(struct drm_device *dev) return 0; fbdev = kzalloc(sizeof(*fbdev), GFP_KERNEL); - if (!fbdev) { - DRM_ERROR("failed to allocate drm fbdev.\n"); + if (!fbdev) return -ENOMEM; - } private->fb_helper = helper = &fbdev->drm_fb_helper; helper->funcs = &exynos_drm_fb_helper_funcs; diff --git a/drivers/gpu/drm/exynos/exynos_drm_fimc.c b/drivers/gpu/drm/exynos/exynos_drm_fimc.c index 6e047bd53e2..8adfc8f1e08 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_fimc.c +++ b/drivers/gpu/drm/exynos/exynos_drm_fimc.c @@ -17,10 +17,12 @@ #include <linux/regmap.h> #include <linux/clk.h> #include <linux/pm_runtime.h> +#include <linux/of.h> #include <drm/drmP.h> #include <drm/exynos_drm.h> #include "regs-fimc.h" +#include "exynos_drm_drv.h" #include "exynos_drm_ipp.h" #include "exynos_drm_fimc.h" @@ -1343,10 +1345,8 @@ static int fimc_init_prop_list(struct exynos_drm_ippdrv *ippdrv) struct drm_exynos_ipp_prop_list *prop_list; prop_list = devm_kzalloc(ippdrv->dev, sizeof(*prop_list), GFP_KERNEL); - if (!prop_list) { - DRM_ERROR("failed to alloc property list.\n"); + if (!prop_list) return -ENOMEM; - } prop_list->version = 1; prop_list->writeback = 1; diff --git a/drivers/gpu/drm/exynos/exynos_drm_fimd.c b/drivers/gpu/drm/exynos/exynos_drm_fimd.c index 1c263dac3c1..868a14d5299 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_fimd.c +++ b/drivers/gpu/drm/exynos/exynos_drm_fimd.c @@ -16,10 +16,12 @@ #include <linux/kernel.h> #include <linux/platform_device.h> #include <linux/clk.h> +#include <linux/of.h> #include <linux/of_device.h> #include <linux/pm_runtime.h> #include <video/of_display_timing.h> +#include <video/of_videomode.h> #include <video/samsung_fimd.h> #include <drm/exynos_drm.h> @@ -35,6 +37,8 @@ * CPU Interface. */ +#define FIMD_DEFAULT_FRAMERATE 60 + /* position control register for hardware window 0, 2 ~ 4.*/ #define VIDOSD_A(win) (VIDOSD_BASE + 0x00 + (win) * 16) #define VIDOSD_B(win) (VIDOSD_BASE + 0x04 + (win) * 16) @@ -65,11 +69,13 @@ struct fimd_driver_data { unsigned int has_shadowcon:1; unsigned int has_clksel:1; + unsigned int has_limited_fmt:1; }; static struct fimd_driver_data s3c64xx_fimd_driver_data = { .timing_base = 0x0, .has_clksel = 1, + .has_limited_fmt = 1, }; static struct fimd_driver_data exynos4_fimd_driver_data = { @@ -90,6 +96,7 @@ struct fimd_win_data { unsigned int fb_width; unsigned int fb_height; unsigned int bpp; + unsigned int pixel_format; dma_addr_t dma_addr; unsigned int buf_offsize; unsigned int line_size; /* bytes */ @@ -115,11 +122,10 @@ struct fimd_context { wait_queue_head_t wait_vsync_queue; atomic_t wait_vsync_event; - struct exynos_drm_panel_info *panel; + struct exynos_drm_panel_info panel; struct fimd_driver_data *driver_data; }; -#ifdef CONFIG_OF static const struct of_device_id fimd_driver_dt_match[] = { { .compatible = "samsung,s3c6400-fimd", .data = &s3c64xx_fimd_driver_data }, @@ -129,21 +135,14 @@ static const struct of_device_id fimd_driver_dt_match[] = { .data = &exynos5_fimd_driver_data }, {}, }; -#endif static inline struct fimd_driver_data *drm_fimd_get_driver_data( struct platform_device *pdev) { -#ifdef CONFIG_OF const struct of_device_id *of_id = of_match_device(fimd_driver_dt_match, &pdev->dev); - if (of_id) - return (struct fimd_driver_data *)of_id->data; -#endif - - return (struct fimd_driver_data *) - platform_get_device_id(pdev)->driver_data; + return (struct fimd_driver_data *)of_id->data; } static bool fimd_display_is_connected(struct device *dev) @@ -157,7 +156,7 @@ static void *fimd_get_panel(struct device *dev) { struct fimd_context *ctx = get_fimd_context(dev); - return ctx->panel; + return &ctx->panel; } static int fimd_check_mode(struct device *dev, struct drm_display_mode *mode) @@ -237,8 +236,8 @@ static void fimd_apply(struct device *subdrv_dev) static void fimd_commit(struct device *dev) { struct fimd_context *ctx = get_fimd_context(dev); - struct exynos_drm_panel_info *panel = ctx->panel; - struct fb_videomode *timing = &panel->timing; + struct exynos_drm_panel_info *panel = &ctx->panel; + struct videomode *vm = &panel->vm; struct fimd_driver_data *driver_data; u32 val; @@ -250,22 +249,22 @@ static void fimd_commit(struct device *dev) writel(ctx->vidcon1, ctx->regs + driver_data->timing_base + VIDCON1); /* setup vertical timing values. */ - val = VIDTCON0_VBPD(timing->upper_margin - 1) | - VIDTCON0_VFPD(timing->lower_margin - 1) | - VIDTCON0_VSPW(timing->vsync_len - 1); + val = VIDTCON0_VBPD(vm->vback_porch - 1) | + VIDTCON0_VFPD(vm->vfront_porch - 1) | + VIDTCON0_VSPW(vm->vsync_len - 1); writel(val, ctx->regs + driver_data->timing_base + VIDTCON0); /* setup horizontal timing values. */ - val = VIDTCON1_HBPD(timing->left_margin - 1) | - VIDTCON1_HFPD(timing->right_margin - 1) | - VIDTCON1_HSPW(timing->hsync_len - 1); + val = VIDTCON1_HBPD(vm->hback_porch - 1) | + VIDTCON1_HFPD(vm->hfront_porch - 1) | + VIDTCON1_HSPW(vm->hsync_len - 1); writel(val, ctx->regs + driver_data->timing_base + VIDTCON1); /* setup horizontal and vertical display size. */ - val = VIDTCON2_LINEVAL(timing->yres - 1) | - VIDTCON2_HOZVAL(timing->xres - 1) | - VIDTCON2_LINEVAL_E(timing->yres - 1) | - VIDTCON2_HOZVAL_E(timing->xres - 1); + val = VIDTCON2_LINEVAL(vm->vactive - 1) | + VIDTCON2_HOZVAL(vm->hactive - 1) | + VIDTCON2_LINEVAL_E(vm->vactive - 1) | + VIDTCON2_HOZVAL_E(vm->hactive - 1); writel(val, ctx->regs + driver_data->timing_base + VIDTCON2); /* setup clock source, clock divider, enable dma. */ @@ -396,6 +395,7 @@ static void fimd_win_mode_set(struct device *dev, win_data->fb_height = overlay->fb_height; win_data->dma_addr = overlay->dma_addr[0] + offset; win_data->bpp = overlay->bpp; + win_data->pixel_format = overlay->pixel_format; win_data->buf_offsize = (overlay->fb_width - overlay->crtc_width) * (overlay->bpp >> 3); win_data->line_size = overlay->crtc_width * (overlay->bpp >> 3); @@ -417,39 +417,38 @@ static void fimd_win_set_pixfmt(struct device *dev, unsigned int win) val = WINCONx_ENWIN; - switch (win_data->bpp) { - case 1: - val |= WINCON0_BPPMODE_1BPP; - val |= WINCONx_BITSWP; - val |= WINCONx_BURSTLEN_4WORD; - break; - case 2: - val |= WINCON0_BPPMODE_2BPP; - val |= WINCONx_BITSWP; - val |= WINCONx_BURSTLEN_8WORD; - break; - case 4: - val |= WINCON0_BPPMODE_4BPP; - val |= WINCONx_BITSWP; - val |= WINCONx_BURSTLEN_8WORD; - break; - case 8: + /* + * In case of s3c64xx, window 0 doesn't support alpha channel. + * So the request format is ARGB8888 then change it to XRGB8888. + */ + if (ctx->driver_data->has_limited_fmt && !win) { + if (win_data->pixel_format == DRM_FORMAT_ARGB8888) + win_data->pixel_format = DRM_FORMAT_XRGB8888; + } + + switch (win_data->pixel_format) { + case DRM_FORMAT_C8: val |= WINCON0_BPPMODE_8BPP_PALETTE; val |= WINCONx_BURSTLEN_8WORD; val |= WINCONx_BYTSWP; break; - case 16: + case DRM_FORMAT_XRGB1555: + val |= WINCON0_BPPMODE_16BPP_1555; + val |= WINCONx_HAWSWP; + val |= WINCONx_BURSTLEN_16WORD; + break; + case DRM_FORMAT_RGB565: val |= WINCON0_BPPMODE_16BPP_565; val |= WINCONx_HAWSWP; val |= WINCONx_BURSTLEN_16WORD; break; - case 24: + case DRM_FORMAT_XRGB8888: val |= WINCON0_BPPMODE_24BPP_888; val |= WINCONx_WSWP; val |= WINCONx_BURSTLEN_16WORD; break; - case 32: - val |= WINCON1_BPPMODE_28BPP_A4888 + case DRM_FORMAT_ARGB8888: + val |= WINCON1_BPPMODE_25BPP_A1888 | WINCON1_BLD_PIX | WINCON1_ALPHA_SEL; val |= WINCONx_WSWP; val |= WINCONx_BURSTLEN_16WORD; @@ -746,45 +745,54 @@ static void fimd_subdrv_remove(struct drm_device *drm_dev, struct device *dev) drm_iommu_detach_device(drm_dev, dev); } -static int fimd_calc_clkdiv(struct fimd_context *ctx, - struct fb_videomode *timing) +static int fimd_configure_clocks(struct fimd_context *ctx, struct device *dev) { - unsigned long clk = clk_get_rate(ctx->lcd_clk); - u32 retrace; - u32 clkdiv; - u32 best_framerate = 0; - u32 framerate; - - retrace = timing->left_margin + timing->hsync_len + - timing->right_margin + timing->xres; - retrace *= timing->upper_margin + timing->vsync_len + - timing->lower_margin + timing->yres; - - /* default framerate is 60Hz */ - if (!timing->refresh) - timing->refresh = 60; - - clk /= retrace; - - for (clkdiv = 1; clkdiv < 0x100; clkdiv++) { - int tmp; - - /* get best framerate */ - framerate = clk / clkdiv; - tmp = timing->refresh - framerate; - if (tmp < 0) { - best_framerate = framerate; - continue; - } else { - if (!best_framerate) - best_framerate = framerate; - else if (tmp < (best_framerate - framerate)) - best_framerate = framerate; - break; + struct videomode *vm = &ctx->panel.vm; + unsigned long clk; + + ctx->bus_clk = devm_clk_get(dev, "fimd"); + if (IS_ERR(ctx->bus_clk)) { + dev_err(dev, "failed to get bus clock\n"); + return PTR_ERR(ctx->bus_clk); + } + + ctx->lcd_clk = devm_clk_get(dev, "sclk_fimd"); + if (IS_ERR(ctx->lcd_clk)) { + dev_err(dev, "failed to get lcd clock\n"); + return PTR_ERR(ctx->lcd_clk); + } + + clk = clk_get_rate(ctx->lcd_clk); + if (clk == 0) { + dev_err(dev, "error getting sclk_fimd clock rate\n"); + return -EINVAL; + } + + if (vm->pixelclock == 0) { + unsigned long c; + c = vm->hactive + vm->hback_porch + vm->hfront_porch + + vm->hsync_len; + c *= vm->vactive + vm->vback_porch + vm->vfront_porch + + vm->vsync_len; + vm->pixelclock = c * FIMD_DEFAULT_FRAMERATE; + if (vm->pixelclock == 0) { + dev_err(dev, "incorrect display timings\n"); + return -EINVAL; } + dev_warn(dev, "pixel clock recalculated to %luHz (%dHz frame rate)\n", + vm->pixelclock, FIMD_DEFAULT_FRAMERATE); } + ctx->clkdiv = DIV_ROUND_UP(clk, vm->pixelclock); + if (ctx->clkdiv > 256) { + dev_warn(dev, "calculated pixel clock divider too high (%u), lowered to 256\n", + ctx->clkdiv); + ctx->clkdiv = 256; + } + vm->pixelclock = clk / ctx->clkdiv; + DRM_DEBUG_KMS("pixel clock = %lu, clkdiv = %d\n", vm->pixelclock, + ctx->clkdiv); - return clkdiv; + return 0; } static void fimd_clear_win(struct fimd_context *ctx, int win) @@ -876,59 +884,53 @@ static int fimd_activate(struct fimd_context *ctx, bool enable) return 0; } +static int fimd_get_platform_data(struct fimd_context *ctx, struct device *dev) +{ + struct videomode *vm; + int ret; + + vm = &ctx->panel.vm; + ret = of_get_videomode(dev->of_node, vm, OF_USE_NATIVE_MODE); + if (ret) { + DRM_ERROR("failed: of_get_videomode() : %d\n", ret); + return ret; + } + + if (vm->flags & DISPLAY_FLAGS_VSYNC_LOW) + ctx->vidcon1 |= VIDCON1_INV_VSYNC; + if (vm->flags & DISPLAY_FLAGS_HSYNC_LOW) + ctx->vidcon1 |= VIDCON1_INV_HSYNC; + if (vm->flags & DISPLAY_FLAGS_DE_LOW) + ctx->vidcon1 |= VIDCON1_INV_VDEN; + if (vm->flags & DISPLAY_FLAGS_PIXDATA_NEGEDGE) + ctx->vidcon1 |= VIDCON1_INV_VCLK; + + return 0; +} + static int fimd_probe(struct platform_device *pdev) { struct device *dev = &pdev->dev; struct fimd_context *ctx; struct exynos_drm_subdrv *subdrv; - struct exynos_drm_fimd_pdata *pdata; - struct exynos_drm_panel_info *panel; struct resource *res; int win; int ret = -EINVAL; - if (dev->of_node) { - pdata = devm_kzalloc(dev, sizeof(*pdata), GFP_KERNEL); - if (!pdata) { - DRM_ERROR("memory allocation for pdata failed\n"); - return -ENOMEM; - } - - ret = of_get_fb_videomode(dev->of_node, &pdata->panel.timing, - OF_USE_NATIVE_MODE); - if (ret) { - DRM_ERROR("failed: of_get_fb_videomode() : %d\n", ret); - return ret; - } - } else { - pdata = dev->platform_data; - if (!pdata) { - DRM_ERROR("no platform data specified\n"); - return -EINVAL; - } - } - - panel = &pdata->panel; - if (!panel) { - dev_err(dev, "panel is null.\n"); - return -EINVAL; - } + if (!dev->of_node) + return -ENODEV; ctx = devm_kzalloc(dev, sizeof(*ctx), GFP_KERNEL); if (!ctx) return -ENOMEM; - ctx->bus_clk = devm_clk_get(dev, "fimd"); - if (IS_ERR(ctx->bus_clk)) { - dev_err(dev, "failed to get bus clock\n"); - return PTR_ERR(ctx->bus_clk); - } + ret = fimd_get_platform_data(ctx, dev); + if (ret) + return ret; - ctx->lcd_clk = devm_clk_get(dev, "sclk_fimd"); - if (IS_ERR(ctx->lcd_clk)) { - dev_err(dev, "failed to get lcd clock\n"); - return PTR_ERR(ctx->lcd_clk); - } + ret = fimd_configure_clocks(ctx, dev); + if (ret) + return ret; res = platform_get_resource(pdev, IORESOURCE_MEM, 0); @@ -952,10 +954,6 @@ static int fimd_probe(struct platform_device *pdev) } ctx->driver_data = drm_fimd_get_driver_data(pdev); - ctx->vidcon0 = pdata->vidcon0; - ctx->vidcon1 = pdata->vidcon1; - ctx->default_win = pdata->default_win; - ctx->panel = panel; DRM_INIT_WAITQUEUE(&ctx->wait_vsync_queue); atomic_set(&ctx->wait_vsync_event, 0); @@ -973,12 +971,6 @@ static int fimd_probe(struct platform_device *pdev) pm_runtime_enable(dev); pm_runtime_get_sync(dev); - ctx->clkdiv = fimd_calc_clkdiv(ctx, &panel->timing); - panel->timing.pixclock = clk_get_rate(ctx->lcd_clk) / ctx->clkdiv; - - DRM_DEBUG_KMS("pixel clock = %d, clkdiv = %d\n", - panel->timing.pixclock, ctx->clkdiv); - for (win = 0; win < WINDOWS_NR; win++) fimd_clear_win(ctx, win); @@ -1067,20 +1059,6 @@ static int fimd_runtime_resume(struct device *dev) } #endif -static struct platform_device_id fimd_driver_ids[] = { - { - .name = "s3c64xx-fb", - .driver_data = (unsigned long)&s3c64xx_fimd_driver_data, - }, { - .name = "exynos4-fb", - .driver_data = (unsigned long)&exynos4_fimd_driver_data, - }, { - .name = "exynos5-fb", - .driver_data = (unsigned long)&exynos5_fimd_driver_data, - }, - {}, -}; - static const struct dev_pm_ops fimd_pm_ops = { SET_SYSTEM_SLEEP_PM_OPS(fimd_suspend, fimd_resume) SET_RUNTIME_PM_OPS(fimd_runtime_suspend, fimd_runtime_resume, NULL) @@ -1089,11 +1067,10 @@ static const struct dev_pm_ops fimd_pm_ops = { struct platform_driver fimd_driver = { .probe = fimd_probe, .remove = fimd_remove, - .id_table = fimd_driver_ids, .driver = { .name = "exynos4-fb", .owner = THIS_MODULE, .pm = &fimd_pm_ops, - .of_match_table = of_match_ptr(fimd_driver_dt_match), + .of_match_table = fimd_driver_dt_match, }, }; diff --git a/drivers/gpu/drm/exynos/exynos_drm_g2d.c b/drivers/gpu/drm/exynos/exynos_drm_g2d.c index eddea494148..3271fd4b172 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_g2d.c +++ b/drivers/gpu/drm/exynos/exynos_drm_g2d.c @@ -23,6 +23,7 @@ #include <drm/drmP.h> #include <drm/exynos_drm.h> #include "exynos_drm_drv.h" +#include "exynos_drm_g2d.h" #include "exynos_drm_gem.h" #include "exynos_drm_iommu.h" @@ -446,10 +447,8 @@ static dma_addr_t *g2d_userptr_get_dma_addr(struct drm_device *drm_dev, } g2d_userptr = kzalloc(sizeof(*g2d_userptr), GFP_KERNEL); - if (!g2d_userptr) { - DRM_ERROR("failed to allocate g2d_userptr.\n"); + if (!g2d_userptr) return ERR_PTR(-ENOMEM); - } atomic_set(&g2d_userptr->refcount, 1); @@ -499,7 +498,6 @@ static dma_addr_t *g2d_userptr_get_dma_addr(struct drm_device *drm_dev, sgt = kzalloc(sizeof(*sgt), GFP_KERNEL); if (!sgt) { - DRM_ERROR("failed to allocate sg table.\n"); ret = -ENOMEM; goto err_free_userptr; } @@ -808,17 +806,8 @@ static void g2d_dma_start(struct g2d_data *g2d, int ret; ret = pm_runtime_get_sync(g2d->dev); - if (ret < 0) { - dev_warn(g2d->dev, "failed pm power on.\n"); - return; - } - - ret = clk_prepare_enable(g2d->gate_clk); - if (ret < 0) { - dev_warn(g2d->dev, "failed to enable clock.\n"); - pm_runtime_put_sync(g2d->dev); + if (ret < 0) return; - } writel_relaxed(node->dma_addr, g2d->regs + G2D_DMA_SFR_BASE_ADDR); writel_relaxed(G2D_DMA_START, g2d->regs + G2D_DMA_COMMAND); @@ -871,7 +860,6 @@ static void g2d_runqueue_worker(struct work_struct *work) runqueue_work); mutex_lock(&g2d->runqueue_mutex); - clk_disable_unprepare(g2d->gate_clk); pm_runtime_put_sync(g2d->dev); complete(&g2d->runqueue_node->complete); @@ -1096,8 +1084,6 @@ int exynos_g2d_set_cmdlist_ioctl(struct drm_device *drm_dev, void *data, e = kzalloc(sizeof(*node->event), GFP_KERNEL); if (!e) { - dev_err(dev, "failed to allocate event\n"); - spin_lock_irqsave(&drm_dev->event_lock, flags); file->event_space += sizeof(e->event); spin_unlock_irqrestore(&drm_dev->event_lock, flags); @@ -1327,10 +1313,8 @@ static int g2d_open(struct drm_device *drm_dev, struct device *dev, struct exynos_drm_g2d_private *g2d_priv; g2d_priv = kzalloc(sizeof(*g2d_priv), GFP_KERNEL); - if (!g2d_priv) { - dev_err(dev, "failed to allocate g2d private data\n"); + if (!g2d_priv) return -ENOMEM; - } g2d_priv->dev = dev; file_priv->g2d_priv = g2d_priv; @@ -1386,10 +1370,8 @@ static int g2d_probe(struct platform_device *pdev) int ret; g2d = devm_kzalloc(dev, sizeof(*g2d), GFP_KERNEL); - if (!g2d) { - dev_err(dev, "failed to allocate driver data\n"); + if (!g2d) return -ENOMEM; - } g2d->runqueue_slab = kmem_cache_create("g2d_runqueue_slab", sizeof(struct g2d_runqueue_node), 0, 0, NULL); @@ -1524,14 +1506,38 @@ static int g2d_resume(struct device *dev) } #endif -static SIMPLE_DEV_PM_OPS(g2d_pm_ops, g2d_suspend, g2d_resume); +#ifdef CONFIG_PM_RUNTIME +static int g2d_runtime_suspend(struct device *dev) +{ + struct g2d_data *g2d = dev_get_drvdata(dev); + + clk_disable_unprepare(g2d->gate_clk); + + return 0; +} + +static int g2d_runtime_resume(struct device *dev) +{ + struct g2d_data *g2d = dev_get_drvdata(dev); + int ret; + + ret = clk_prepare_enable(g2d->gate_clk); + if (ret < 0) + dev_warn(dev, "failed to enable clock.\n"); + + return ret; +} +#endif + +static const struct dev_pm_ops g2d_pm_ops = { + SET_SYSTEM_SLEEP_PM_OPS(g2d_suspend, g2d_resume) + SET_RUNTIME_PM_OPS(g2d_runtime_suspend, g2d_runtime_resume, NULL) +}; -#ifdef CONFIG_OF static const struct of_device_id exynos_g2d_match[] = { { .compatible = "samsung,exynos5250-g2d" }, {}, }; -#endif struct platform_driver g2d_driver = { .probe = g2d_probe, @@ -1540,6 +1546,6 @@ struct platform_driver g2d_driver = { .name = "s5p-g2d", .owner = THIS_MODULE, .pm = &g2d_pm_ops, - .of_match_table = of_match_ptr(exynos_g2d_match), + .of_match_table = exynos_g2d_match, }, }; diff --git a/drivers/gpu/drm/exynos/exynos_drm_gem.c b/drivers/gpu/drm/exynos/exynos_drm_gem.c index f3c6f40666e..49f9cd23275 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_gem.c +++ b/drivers/gpu/drm/exynos/exynos_drm_gem.c @@ -18,6 +18,7 @@ #include "exynos_drm_drv.h" #include "exynos_drm_gem.h" #include "exynos_drm_buf.h" +#include "exynos_drm_iommu.h" static unsigned int convert_to_vm_err_msg(int msg) { @@ -191,10 +192,8 @@ struct exynos_drm_gem_obj *exynos_drm_gem_init(struct drm_device *dev, int ret; exynos_gem_obj = kzalloc(sizeof(*exynos_gem_obj), GFP_KERNEL); - if (!exynos_gem_obj) { - DRM_ERROR("failed to allocate exynos gem object\n"); + if (!exynos_gem_obj) return NULL; - } exynos_gem_obj->size = size; obj = &exynos_gem_obj->base; @@ -668,6 +667,18 @@ int exynos_drm_gem_dumb_create(struct drm_file *file_priv, exynos_gem_obj = exynos_drm_gem_create(dev, EXYNOS_BO_CONTIG | EXYNOS_BO_WC, args->size); + /* + * If physically contiguous memory allocation fails and if IOMMU is + * supported then try to get buffer from non physically contiguous + * memory area. + */ + if (IS_ERR(exynos_gem_obj) && is_drm_iommu_supported(dev)) { + dev_warn(dev->dev, "contiguous FB allocation failed, falling back to non-contiguous\n"); + exynos_gem_obj = exynos_drm_gem_create(dev, + EXYNOS_BO_NONCONTIG | EXYNOS_BO_WC, + args->size); + } + if (IS_ERR(exynos_gem_obj)) return PTR_ERR(exynos_gem_obj); diff --git a/drivers/gpu/drm/exynos/exynos_drm_gsc.c b/drivers/gpu/drm/exynos/exynos_drm_gsc.c index 90b8a1a5344..cd6aebd53bd 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_gsc.c +++ b/drivers/gpu/drm/exynos/exynos_drm_gsc.c @@ -20,6 +20,7 @@ #include <drm/drmP.h> #include <drm/exynos_drm.h> #include "regs-gsc.h" +#include "exynos_drm_drv.h" #include "exynos_drm_ipp.h" #include "exynos_drm_gsc.h" @@ -1337,10 +1338,8 @@ static int gsc_init_prop_list(struct exynos_drm_ippdrv *ippdrv) struct drm_exynos_ipp_prop_list *prop_list; prop_list = devm_kzalloc(ippdrv->dev, sizeof(*prop_list), GFP_KERNEL); - if (!prop_list) { - DRM_ERROR("failed to alloc property list.\n"); + if (!prop_list) return -ENOMEM; - } prop_list->version = 1; prop_list->writeback = 1; diff --git a/drivers/gpu/drm/exynos/exynos_drm_hdmi.c b/drivers/gpu/drm/exynos/exynos_drm_hdmi.c index 8d3bc01d683..8548b974bd5 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_hdmi.c +++ b/drivers/gpu/drm/exynos/exynos_drm_hdmi.c @@ -403,10 +403,8 @@ static int exynos_drm_hdmi_probe(struct platform_device *pdev) struct drm_hdmi_context *ctx; ctx = devm_kzalloc(dev, sizeof(*ctx), GFP_KERNEL); - if (!ctx) { - DRM_LOG_KMS("failed to alloc common hdmi context.\n"); + if (!ctx) return -ENOMEM; - } subdrv = &ctx->subdrv; diff --git a/drivers/gpu/drm/exynos/exynos_drm_iommu.c b/drivers/gpu/drm/exynos/exynos_drm_iommu.c index 3799d5c2b5d..fb8db037827 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_iommu.c +++ b/drivers/gpu/drm/exynos/exynos_drm_iommu.c @@ -47,10 +47,16 @@ int drm_create_iommu_mapping(struct drm_device *drm_dev) dev->dma_parms = devm_kzalloc(dev, sizeof(*dev->dma_parms), GFP_KERNEL); + if (!dev->dma_parms) + goto error; + dma_set_max_seg_size(dev, 0xffffffffu); dev->archdata.mapping = mapping; return 0; +error: + arm_iommu_release_mapping(mapping); + return -ENOMEM; } /* @@ -91,6 +97,9 @@ int drm_iommu_attach_device(struct drm_device *drm_dev, subdrv_dev->dma_parms = devm_kzalloc(subdrv_dev, sizeof(*subdrv_dev->dma_parms), GFP_KERNEL); + if (!subdrv_dev->dma_parms) + return -ENOMEM; + dma_set_max_seg_size(subdrv_dev, 0xffffffffu); ret = arm_iommu_attach_device(subdrv_dev, dev->archdata.mapping); diff --git a/drivers/gpu/drm/exynos/exynos_drm_ipp.c b/drivers/gpu/drm/exynos/exynos_drm_ipp.c index d2b6ab4def9..824e0705c8d 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_ipp.c +++ b/drivers/gpu/drm/exynos/exynos_drm_ipp.c @@ -408,10 +408,8 @@ static struct drm_exynos_ipp_cmd_work *ipp_create_cmd_work(void) struct drm_exynos_ipp_cmd_work *cmd_work; cmd_work = kzalloc(sizeof(*cmd_work), GFP_KERNEL); - if (!cmd_work) { - DRM_ERROR("failed to alloc cmd_work.\n"); + if (!cmd_work) return ERR_PTR(-ENOMEM); - } INIT_WORK((struct work_struct *)cmd_work, ipp_sched_cmd); @@ -423,10 +421,8 @@ static struct drm_exynos_ipp_event_work *ipp_create_event_work(void) struct drm_exynos_ipp_event_work *event_work; event_work = kzalloc(sizeof(*event_work), GFP_KERNEL); - if (!event_work) { - DRM_ERROR("failed to alloc event_work.\n"); + if (!event_work) return ERR_PTR(-ENOMEM); - } INIT_WORK((struct work_struct *)event_work, ipp_sched_event); @@ -482,10 +478,8 @@ int exynos_drm_ipp_set_property(struct drm_device *drm_dev, void *data, /* allocate command node */ c_node = kzalloc(sizeof(*c_node), GFP_KERNEL); - if (!c_node) { - DRM_ERROR("failed to allocate map node.\n"); + if (!c_node) return -ENOMEM; - } /* create property id */ ret = ipp_create_id(&ctx->prop_idr, &ctx->prop_lock, c_node, @@ -694,10 +688,8 @@ static struct drm_exynos_ipp_mem_node mutex_lock(&c_node->mem_lock); m_node = kzalloc(sizeof(*m_node), GFP_KERNEL); - if (!m_node) { - DRM_ERROR("failed to allocate queue node.\n"); + if (!m_node) goto err_unlock; - } /* clear base address for error handling */ memset(&buf_info, 0x0, sizeof(buf_info)); @@ -798,9 +790,7 @@ static int ipp_get_event(struct drm_device *drm_dev, DRM_DEBUG_KMS("ops_id[%d]buf_id[%d]\n", qbuf->ops_id, qbuf->buf_id); e = kzalloc(sizeof(*e), GFP_KERNEL); - if (!e) { - DRM_ERROR("failed to allocate event.\n"); spin_lock_irqsave(&drm_dev->event_lock, flags); file->event_space += sizeof(e->event); spin_unlock_irqrestore(&drm_dev->event_lock, flags); @@ -1780,10 +1770,8 @@ static int ipp_subdrv_open(struct drm_device *drm_dev, struct device *dev, struct exynos_drm_ipp_private *priv; priv = kzalloc(sizeof(*priv), GFP_KERNEL); - if (!priv) { - DRM_ERROR("failed to allocate priv.\n"); + if (!priv) return -ENOMEM; - } priv->dev = dev; file_priv->ipp_priv = priv; diff --git a/drivers/gpu/drm/exynos/exynos_drm_plane.c b/drivers/gpu/drm/exynos/exynos_drm_plane.c index 6ee55e68e0a..fcb0652e77d 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_plane.c +++ b/drivers/gpu/drm/exynos/exynos_drm_plane.c @@ -16,6 +16,7 @@ #include "exynos_drm_encoder.h" #include "exynos_drm_fb.h" #include "exynos_drm_gem.h" +#include "exynos_drm_plane.h" #define to_exynos_plane(x) container_of(x, struct exynos_plane, base) @@ -264,10 +265,8 @@ struct drm_plane *exynos_plane_init(struct drm_device *dev, int err; exynos_plane = kzalloc(sizeof(struct exynos_plane), GFP_KERNEL); - if (!exynos_plane) { - DRM_ERROR("failed to allocate plane\n"); + if (!exynos_plane) return NULL; - } err = drm_plane_init(dev, &exynos_plane->base, possible_crtcs, &exynos_plane_funcs, formats, ARRAY_SIZE(formats), diff --git a/drivers/gpu/drm/exynos/exynos_drm_rotator.c b/drivers/gpu/drm/exynos/exynos_drm_rotator.c index 49669aa24c4..7b901688def 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_rotator.c +++ b/drivers/gpu/drm/exynos/exynos_drm_rotator.c @@ -21,6 +21,7 @@ #include <drm/exynos_drm.h> #include "regs-rotator.h" #include "exynos_drm.h" +#include "exynos_drm_drv.h" #include "exynos_drm_ipp.h" /* @@ -471,10 +472,8 @@ static int rotator_init_prop_list(struct exynos_drm_ippdrv *ippdrv) struct drm_exynos_ipp_prop_list *prop_list; prop_list = devm_kzalloc(ippdrv->dev, sizeof(*prop_list), GFP_KERNEL); - if (!prop_list) { - DRM_ERROR("failed to alloc property list.\n"); + if (!prop_list) return -ENOMEM; - } prop_list->version = 1; prop_list->flip = (1 << EXYNOS_DRM_FLIP_VERTICAL) | @@ -631,21 +630,96 @@ static int rotator_ippdrv_start(struct device *dev, enum drm_exynos_ipp_cmd cmd) return 0; } +static struct rot_limit_table rot_limit_tbl_4210 = { + .ycbcr420_2p = { + .min_w = 32, + .min_h = 32, + .max_w = SZ_64K, + .max_h = SZ_64K, + .align = 3, + }, + .rgb888 = { + .min_w = 8, + .min_h = 8, + .max_w = SZ_16K, + .max_h = SZ_16K, + .align = 2, + }, +}; + +static struct rot_limit_table rot_limit_tbl_4x12 = { + .ycbcr420_2p = { + .min_w = 32, + .min_h = 32, + .max_w = SZ_32K, + .max_h = SZ_32K, + .align = 3, + }, + .rgb888 = { + .min_w = 8, + .min_h = 8, + .max_w = SZ_8K, + .max_h = SZ_8K, + .align = 2, + }, +}; + +static struct rot_limit_table rot_limit_tbl_5250 = { + .ycbcr420_2p = { + .min_w = 32, + .min_h = 32, + .max_w = SZ_32K, + .max_h = SZ_32K, + .align = 3, + }, + .rgb888 = { + .min_w = 8, + .min_h = 8, + .max_w = SZ_8K, + .max_h = SZ_8K, + .align = 1, + }, +}; + +static const struct of_device_id exynos_rotator_match[] = { + { + .compatible = "samsung,exynos4210-rotator", + .data = &rot_limit_tbl_4210, + }, + { + .compatible = "samsung,exynos4212-rotator", + .data = &rot_limit_tbl_4x12, + }, + { + .compatible = "samsung,exynos5250-rotator", + .data = &rot_limit_tbl_5250, + }, + {}, +}; + static int rotator_probe(struct platform_device *pdev) { struct device *dev = &pdev->dev; struct rot_context *rot; struct exynos_drm_ippdrv *ippdrv; + const struct of_device_id *match; int ret; + if (!dev->of_node) { + dev_err(dev, "cannot find of_node.\n"); + return -ENODEV; + } + rot = devm_kzalloc(dev, sizeof(*rot), GFP_KERNEL); - if (!rot) { - dev_err(dev, "failed to allocate rot\n"); + if (!rot) return -ENOMEM; - } - rot->limit_tbl = (struct rot_limit_table *) - platform_get_device_id(pdev)->driver_data; + match = of_match_node(exynos_rotator_match, dev->of_node); + if (!match) { + dev_err(dev, "failed to match node\n"); + return -ENODEV; + } + rot->limit_tbl = (struct rot_limit_table *)match->data; rot->regs_res = platform_get_resource(pdev, IORESOURCE_MEM, 0); rot->regs = devm_ioremap_resource(dev, rot->regs_res); @@ -717,31 +791,6 @@ static int rotator_remove(struct platform_device *pdev) return 0; } -static struct rot_limit_table rot_limit_tbl = { - .ycbcr420_2p = { - .min_w = 32, - .min_h = 32, - .max_w = SZ_32K, - .max_h = SZ_32K, - .align = 3, - }, - .rgb888 = { - .min_w = 8, - .min_h = 8, - .max_w = SZ_8K, - .max_h = SZ_8K, - .align = 2, - }, -}; - -static struct platform_device_id rotator_driver_ids[] = { - { - .name = "exynos-rot", - .driver_data = (unsigned long)&rot_limit_tbl, - }, - {}, -}; - static int rotator_clk_crtl(struct rot_context *rot, bool enable) { if (enable) { @@ -803,10 +852,10 @@ static const struct dev_pm_ops rotator_pm_ops = { struct platform_driver rotator_driver = { .probe = rotator_probe, .remove = rotator_remove, - .id_table = rotator_driver_ids, .driver = { .name = "exynos-rot", .owner = THIS_MODULE, .pm = &rotator_pm_ops, + .of_match_table = exynos_rotator_match, }, }; diff --git a/drivers/gpu/drm/exynos/exynos_drm_vidi.c b/drivers/gpu/drm/exynos/exynos_drm_vidi.c index c57c56519ad..4400330e444 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_vidi.c +++ b/drivers/gpu/drm/exynos/exynos_drm_vidi.c @@ -23,6 +23,7 @@ #include "exynos_drm_drv.h" #include "exynos_drm_crtc.h" #include "exynos_drm_encoder.h" +#include "exynos_drm_vidi.h" /* vidi has totally three virtual windows. */ #define WINDOWS_NR 3 diff --git a/drivers/gpu/drm/exynos/exynos_hdmi.c b/drivers/gpu/drm/exynos/exynos_hdmi.c index 2f5c6942c96..a0e10aeb0e6 100644 --- a/drivers/gpu/drm/exynos/exynos_hdmi.c +++ b/drivers/gpu/drm/exynos/exynos_hdmi.c @@ -32,6 +32,7 @@ #include <linux/clk.h> #include <linux/regulator/consumer.h> #include <linux/io.h> +#include <linux/of.h> #include <linux/of_gpio.h> #include <drm/exynos_drm.h> @@ -1824,10 +1825,8 @@ static int hdmi_resources_init(struct hdmi_context *hdata) res->regul_bulk = devm_kzalloc(dev, ARRAY_SIZE(supply) * sizeof(res->regul_bulk[0]), GFP_KERNEL); - if (!res->regul_bulk) { - DRM_ERROR("failed to get memory for regulators\n"); + if (!res->regul_bulk) goto fail; - } for (i = 0; i < ARRAY_SIZE(supply); ++i) { res->regul_bulk[i].supply = supply[i]; res->regul_bulk[i].consumer = NULL; @@ -1859,7 +1858,6 @@ void hdmi_attach_hdmiphy_client(struct i2c_client *hdmiphy) hdmi_hdmiphy = hdmiphy; } -#ifdef CONFIG_OF static struct s5p_hdmi_platform_data *drm_hdmi_dt_parse_pdata (struct device *dev) { @@ -1868,10 +1866,8 @@ static struct s5p_hdmi_platform_data *drm_hdmi_dt_parse_pdata u32 value; pd = devm_kzalloc(dev, sizeof(*pd), GFP_KERNEL); - if (!pd) { - DRM_ERROR("memory allocation for pdata failed\n"); + if (!pd) goto err_data; - } if (!of_find_property(np, "hpd-gpio", &value)) { DRM_ERROR("no hpd gpio property found\n"); @@ -1885,33 +1881,7 @@ static struct s5p_hdmi_platform_data *drm_hdmi_dt_parse_pdata err_data: return NULL; } -#else -static struct s5p_hdmi_platform_data *drm_hdmi_dt_parse_pdata - (struct device *dev) -{ - return NULL; -} -#endif - -static struct platform_device_id hdmi_driver_types[] = { - { - .name = "s5pv210-hdmi", - .driver_data = HDMI_TYPE13, - }, { - .name = "exynos4-hdmi", - .driver_data = HDMI_TYPE13, - }, { - .name = "exynos4-hdmi14", - .driver_data = HDMI_TYPE14, - }, { - .name = "exynos5-hdmi", - .driver_data = HDMI_TYPE14, - }, { - /* end node */ - } -}; -#ifdef CONFIG_OF static struct of_device_id hdmi_match_types[] = { { .compatible = "samsung,exynos5-hdmi", @@ -1923,7 +1893,6 @@ static struct of_device_id hdmi_match_types[] = { /* end node */ } }; -#endif static int hdmi_probe(struct platform_device *pdev) { @@ -1932,36 +1901,23 @@ static int hdmi_probe(struct platform_device *pdev) struct hdmi_context *hdata; struct s5p_hdmi_platform_data *pdata; struct resource *res; + const struct of_device_id *match; int ret; - if (dev->of_node) { - pdata = drm_hdmi_dt_parse_pdata(dev); - if (IS_ERR(pdata)) { - DRM_ERROR("failed to parse dt\n"); - return PTR_ERR(pdata); - } - } else { - pdata = dev->platform_data; - } + if (!dev->of_node) + return -ENODEV; - if (!pdata) { - DRM_ERROR("no platform data specified\n"); + pdata = drm_hdmi_dt_parse_pdata(dev); + if (!pdata) return -EINVAL; - } - drm_hdmi_ctx = devm_kzalloc(dev, sizeof(*drm_hdmi_ctx), - GFP_KERNEL); - if (!drm_hdmi_ctx) { - DRM_ERROR("failed to allocate common hdmi context.\n"); + drm_hdmi_ctx = devm_kzalloc(dev, sizeof(*drm_hdmi_ctx), GFP_KERNEL); + if (!drm_hdmi_ctx) return -ENOMEM; - } - hdata = devm_kzalloc(dev, sizeof(struct hdmi_context), - GFP_KERNEL); - if (!hdata) { - DRM_ERROR("out of memory\n"); + hdata = devm_kzalloc(dev, sizeof(struct hdmi_context), GFP_KERNEL); + if (!hdata) return -ENOMEM; - } mutex_init(&hdata->hdmi_mutex); @@ -1970,23 +1926,15 @@ static int hdmi_probe(struct platform_device *pdev) platform_set_drvdata(pdev, drm_hdmi_ctx); - if (dev->of_node) { - const struct of_device_id *match; - match = of_match_node(of_match_ptr(hdmi_match_types), - dev->of_node); - if (match == NULL) - return -ENODEV; - hdata->type = (enum hdmi_type)match->data; - } else { - hdata->type = (enum hdmi_type)platform_get_device_id - (pdev)->driver_data; - } + match = of_match_node(hdmi_match_types, dev->of_node); + if (!match) + return -ENODEV; + hdata->type = (enum hdmi_type)match->data; hdata->hpd_gpio = pdata->hpd_gpio; hdata->dev = dev; ret = hdmi_resources_init(hdata); - if (ret) { DRM_ERROR("hdmi_resources_init failed\n"); return -EINVAL; @@ -2141,11 +2089,10 @@ static const struct dev_pm_ops hdmi_pm_ops = { struct platform_driver hdmi_driver = { .probe = hdmi_probe, .remove = hdmi_remove, - .id_table = hdmi_driver_types, .driver = { .name = "exynos-hdmi", .owner = THIS_MODULE, .pm = &hdmi_pm_ops, - .of_match_table = of_match_ptr(hdmi_match_types), + .of_match_table = hdmi_match_types, }, }; diff --git a/drivers/gpu/drm/exynos/exynos_hdmiphy.c b/drivers/gpu/drm/exynos/exynos_hdmiphy.c index 6e320ae9afe..59abb1494ce 100644 --- a/drivers/gpu/drm/exynos/exynos_hdmiphy.c +++ b/drivers/gpu/drm/exynos/exynos_hdmiphy.c @@ -15,6 +15,7 @@ #include <linux/kernel.h> #include <linux/i2c.h> +#include <linux/of.h> #include "exynos_drm_drv.h" #include "exynos_hdmi.h" @@ -39,13 +40,6 @@ static int hdmiphy_remove(struct i2c_client *client) return 0; } -static const struct i2c_device_id hdmiphy_id[] = { - { "s5p_hdmiphy", 0 }, - { "exynos5-hdmiphy", 0 }, - { }, -}; - -#ifdef CONFIG_OF static struct of_device_id hdmiphy_match_types[] = { { .compatible = "samsung,exynos5-hdmiphy", @@ -57,15 +51,13 @@ static struct of_device_id hdmiphy_match_types[] = { /* end node */ } }; -#endif struct i2c_driver hdmiphy_driver = { .driver = { .name = "exynos-hdmiphy", .owner = THIS_MODULE, - .of_match_table = of_match_ptr(hdmiphy_match_types), + .of_match_table = hdmiphy_match_types, }, - .id_table = hdmiphy_id, .probe = hdmiphy_probe, .remove = hdmiphy_remove, .command = NULL, diff --git a/drivers/gpu/drm/exynos/exynos_mixer.c b/drivers/gpu/drm/exynos/exynos_mixer.c index c9a137caea4..63bc5f92fbb 100644 --- a/drivers/gpu/drm/exynos/exynos_mixer.c +++ b/drivers/gpu/drm/exynos/exynos_mixer.c @@ -30,6 +30,7 @@ #include <linux/pm_runtime.h> #include <linux/clk.h> #include <linux/regulator/consumer.h> +#include <linux/of.h> #include <drm/exynos_drm.h> @@ -1185,16 +1186,12 @@ static int mixer_probe(struct platform_device *pdev) drm_hdmi_ctx = devm_kzalloc(dev, sizeof(*drm_hdmi_ctx), GFP_KERNEL); - if (!drm_hdmi_ctx) { - DRM_ERROR("failed to allocate common hdmi context.\n"); + if (!drm_hdmi_ctx) return -ENOMEM; - } ctx = devm_kzalloc(dev, sizeof(*ctx), GFP_KERNEL); - if (!ctx) { - DRM_ERROR("failed to alloc mixer context.\n"); + if (!ctx) return -ENOMEM; - } mutex_init(&ctx->mixer_mutex); diff --git a/drivers/gpu/drm/i2c/tda998x_drv.c b/drivers/gpu/drm/i2c/tda998x_drv.c index c2bd711e86e..b1f8fc69023 100644 --- a/drivers/gpu/drm/i2c/tda998x_drv.c +++ b/drivers/gpu/drm/i2c/tda998x_drv.c @@ -596,6 +596,10 @@ tda998x_configure_audio(struct drm_encoder *encoder, cts_n = CTS_N_M(3) | CTS_N_K(3); ca_i2s = CA_I2S_CA_I2S(0); break; + + default: + BUG(); + return; } reg_write(encoder, REG_AIP_CLKSEL, clksel_aip); diff --git a/drivers/gpu/drm/i915/i915_dma.c b/drivers/gpu/drm/i915/i915_dma.c index 1eaa0be392d..512645fdaff 100644 --- a/drivers/gpu/drm/i915/i915_dma.c +++ b/drivers/gpu/drm/i915/i915_dma.c @@ -1353,6 +1353,12 @@ static int i915_load_modeset_init(struct drm_device *dev) */ intel_fbdev_initial_config(dev); + /* + * Must do this after fbcon init so that + * vgacon_save_screen() works during the handover. + */ + i915_disable_vga_mem(dev); + /* Only enable hotplug handling once the fbdev is fully set up. */ dev_priv->enable_hotplug_processing = true; @@ -1672,7 +1678,7 @@ int i915_driver_load(struct drm_device *dev, unsigned long flags) return 0; out_gem_unload: - if (dev_priv->mm.inactive_shrinker.shrink) + if (dev_priv->mm.inactive_shrinker.scan_objects) unregister_shrinker(&dev_priv->mm.inactive_shrinker); if (dev->pdev->msi_enabled) @@ -1711,7 +1717,7 @@ int i915_driver_unload(struct drm_device *dev) i915_teardown_sysfs(dev); - if (dev_priv->mm.inactive_shrinker.shrink) + if (dev_priv->mm.inactive_shrinker.scan_objects) unregister_shrinker(&dev_priv->mm.inactive_shrinker); mutex_lock(&dev->struct_mutex); diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c index f1779b352f5..750f35cccf4 100644 --- a/drivers/gpu/drm/i915/i915_gem.c +++ b/drivers/gpu/drm/i915/i915_gem.c @@ -60,10 +60,12 @@ static void i915_gem_object_update_fence(struct drm_i915_gem_object *obj, struct drm_i915_fence_reg *fence, bool enable); -static int i915_gem_inactive_shrink(struct shrinker *shrinker, - struct shrink_control *sc); +static unsigned long i915_gem_inactive_count(struct shrinker *shrinker, + struct shrink_control *sc); +static unsigned long i915_gem_inactive_scan(struct shrinker *shrinker, + struct shrink_control *sc); static long i915_gem_purge(struct drm_i915_private *dev_priv, long target); -static void i915_gem_shrink_all(struct drm_i915_private *dev_priv); +static long i915_gem_shrink_all(struct drm_i915_private *dev_priv); static void i915_gem_object_truncate(struct drm_i915_gem_object *obj); static bool cpu_cache_is_coherent(struct drm_device *dev, @@ -1389,14 +1391,11 @@ out: if (i915_terminally_wedged(&dev_priv->gpu_error)) return VM_FAULT_SIGBUS; case -EAGAIN: - /* Give the error handler a chance to run and move the - * objects off the GPU active list. Next time we service the - * fault, we should be able to transition the page into the - * GTT without touching the GPU (and so avoid further - * EIO/EGAIN). If the GPU is wedged, then there is no issue - * with coherency, just lost writes. + /* + * EAGAIN means the gpu is hung and we'll wait for the error + * handler to reset everything when re-faulting in + * i915_mutex_lock_interruptible. */ - set_need_resched(); case 0: case -ERESTARTSYS: case -EINTR: @@ -1694,6 +1693,7 @@ static long __i915_gem_shrink(struct drm_i915_private *dev_priv, long target, bool purgeable_only) { + struct list_head still_bound_list; struct drm_i915_gem_object *obj, *next; long count = 0; @@ -1708,23 +1708,55 @@ __i915_gem_shrink(struct drm_i915_private *dev_priv, long target, } } - list_for_each_entry_safe(obj, next, &dev_priv->mm.bound_list, - global_list) { + /* + * As we may completely rewrite the bound list whilst unbinding + * (due to retiring requests) we have to strictly process only + * one element of the list at the time, and recheck the list + * on every iteration. + */ + INIT_LIST_HEAD(&still_bound_list); + while (count < target && !list_empty(&dev_priv->mm.bound_list)) { struct i915_vma *vma, *v; + obj = list_first_entry(&dev_priv->mm.bound_list, + typeof(*obj), global_list); + list_move_tail(&obj->global_list, &still_bound_list); + if (!i915_gem_object_is_purgeable(obj) && purgeable_only) continue; + /* + * Hold a reference whilst we unbind this object, as we may + * end up waiting for and retiring requests. This might + * release the final reference (held by the active list) + * and result in the object being freed from under us. + * in this object being freed. + * + * Note 1: Shrinking the bound list is special since only active + * (and hence bound objects) can contain such limbo objects, so + * we don't need special tricks for shrinking the unbound list. + * The only other place where we have to be careful with active + * objects suddenly disappearing due to retiring requests is the + * eviction code. + * + * Note 2: Even though the bound list doesn't hold a reference + * to the object we can safely grab one here: The final object + * unreferencing and the bound_list are both protected by the + * dev->struct_mutex and so we won't ever be able to observe an + * object on the bound_list with a reference count equals 0. + */ + drm_gem_object_reference(&obj->base); + list_for_each_entry_safe(vma, v, &obj->vma_list, vma_link) if (i915_vma_unbind(vma)) break; - if (!i915_gem_object_put_pages(obj)) { + if (i915_gem_object_put_pages(obj) == 0) count += obj->base.size >> PAGE_SHIFT; - if (count >= target) - return count; - } + + drm_gem_object_unreference(&obj->base); } + list_splice(&still_bound_list, &dev_priv->mm.bound_list); return count; } @@ -1735,16 +1767,21 @@ i915_gem_purge(struct drm_i915_private *dev_priv, long target) return __i915_gem_shrink(dev_priv, target, true); } -static void +static long i915_gem_shrink_all(struct drm_i915_private *dev_priv) { struct drm_i915_gem_object *obj, *next; + long freed = 0; i915_gem_evict_everything(dev_priv->dev); list_for_each_entry_safe(obj, next, &dev_priv->mm.unbound_list, - global_list) + global_list) { + if (obj->pages_pin_count == 0) + freed += obj->base.size >> PAGE_SHIFT; i915_gem_object_put_pages(obj); + } + return freed; } static int @@ -4581,7 +4618,8 @@ i915_gem_load(struct drm_device *dev) dev_priv->mm.interruptible = true; - dev_priv->mm.inactive_shrinker.shrink = i915_gem_inactive_shrink; + dev_priv->mm.inactive_shrinker.scan_objects = i915_gem_inactive_scan; + dev_priv->mm.inactive_shrinker.count_objects = i915_gem_inactive_count; dev_priv->mm.inactive_shrinker.seeks = DEFAULT_SEEKS; register_shrinker(&dev_priv->mm.inactive_shrinker); } @@ -4804,8 +4842,8 @@ static bool mutex_is_locked_by(struct mutex *mutex, struct task_struct *task) #endif } -static int -i915_gem_inactive_shrink(struct shrinker *shrinker, struct shrink_control *sc) +static unsigned long +i915_gem_inactive_count(struct shrinker *shrinker, struct shrink_control *sc) { struct drm_i915_private *dev_priv = container_of(shrinker, @@ -4813,45 +4851,35 @@ i915_gem_inactive_shrink(struct shrinker *shrinker, struct shrink_control *sc) mm.inactive_shrinker); struct drm_device *dev = dev_priv->dev; struct drm_i915_gem_object *obj; - int nr_to_scan = sc->nr_to_scan; bool unlock = true; - int cnt; + unsigned long count; if (!mutex_trylock(&dev->struct_mutex)) { if (!mutex_is_locked_by(&dev->struct_mutex, current)) - return 0; + return SHRINK_STOP; if (dev_priv->mm.shrinker_no_lock_stealing) - return 0; + return SHRINK_STOP; unlock = false; } - if (nr_to_scan) { - nr_to_scan -= i915_gem_purge(dev_priv, nr_to_scan); - if (nr_to_scan > 0) - nr_to_scan -= __i915_gem_shrink(dev_priv, nr_to_scan, - false); - if (nr_to_scan > 0) - i915_gem_shrink_all(dev_priv); - } - - cnt = 0; + count = 0; list_for_each_entry(obj, &dev_priv->mm.unbound_list, global_list) if (obj->pages_pin_count == 0) - cnt += obj->base.size >> PAGE_SHIFT; + count += obj->base.size >> PAGE_SHIFT; list_for_each_entry(obj, &dev_priv->mm.bound_list, global_list) { if (obj->active) continue; if (obj->pin_count == 0 && obj->pages_pin_count == 0) - cnt += obj->base.size >> PAGE_SHIFT; + count += obj->base.size >> PAGE_SHIFT; } if (unlock) mutex_unlock(&dev->struct_mutex); - return cnt; + return count; } /* All the new VM stuff */ @@ -4913,3 +4941,37 @@ unsigned long i915_gem_obj_size(struct drm_i915_gem_object *o, return 0; } + +static unsigned long +i915_gem_inactive_scan(struct shrinker *shrinker, struct shrink_control *sc) +{ + struct drm_i915_private *dev_priv = + container_of(shrinker, + struct drm_i915_private, + mm.inactive_shrinker); + struct drm_device *dev = dev_priv->dev; + int nr_to_scan = sc->nr_to_scan; + unsigned long freed; + bool unlock = true; + + if (!mutex_trylock(&dev->struct_mutex)) { + if (!mutex_is_locked_by(&dev->struct_mutex, current)) + return 0; + + if (dev_priv->mm.shrinker_no_lock_stealing) + return 0; + + unlock = false; + } + + freed = i915_gem_purge(dev_priv, nr_to_scan); + if (freed < nr_to_scan) + freed += __i915_gem_shrink(dev_priv, nr_to_scan, + false); + if (freed < nr_to_scan) + freed += i915_gem_shrink_all(dev_priv); + + if (unlock) + mutex_unlock(&dev->struct_mutex); + return freed; +} diff --git a/drivers/gpu/drm/i915/i915_gem_stolen.c b/drivers/gpu/drm/i915/i915_gem_stolen.c index b902f2afc8e..d284d892ed9 100644 --- a/drivers/gpu/drm/i915/i915_gem_stolen.c +++ b/drivers/gpu/drm/i915/i915_gem_stolen.c @@ -201,6 +201,9 @@ int i915_gem_init_stolen(struct drm_device *dev) struct drm_i915_private *dev_priv = dev->dev_private; int bios_reserved = 0; + if (dev_priv->gtt.stolen_size == 0) + return 0; + dev_priv->mm.stolen_base = i915_stolen_to_physical(dev); if (dev_priv->mm.stolen_base == 0) return 0; diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c index a9233e2d48a..b356dc15add 100644 --- a/drivers/gpu/drm/i915/i915_irq.c +++ b/drivers/gpu/drm/i915/i915_irq.c @@ -1483,6 +1483,34 @@ static irqreturn_t ironlake_irq_handler(int irq, void *arg) return ret; } +static void i915_error_wake_up(struct drm_i915_private *dev_priv, + bool reset_completed) +{ + struct intel_ring_buffer *ring; + int i; + + /* + * Notify all waiters for GPU completion events that reset state has + * been changed, and that they need to restart their wait after + * checking for potential errors (and bail out to drop locks if there is + * a gpu reset pending so that i915_error_work_func can acquire them). + */ + + /* Wake up __wait_seqno, potentially holding dev->struct_mutex. */ + for_each_ring(ring, dev_priv, i) + wake_up_all(&ring->irq_queue); + + /* Wake up intel_crtc_wait_for_pending_flips, holding crtc->mutex. */ + wake_up_all(&dev_priv->pending_flip_queue); + + /* + * Signal tasks blocked in i915_gem_wait_for_error that the pending + * reset state is cleared. + */ + if (reset_completed) + wake_up_all(&dev_priv->gpu_error.reset_queue); +} + /** * i915_error_work_func - do process context error handling work * @work: work struct @@ -1497,11 +1525,10 @@ static void i915_error_work_func(struct work_struct *work) drm_i915_private_t *dev_priv = container_of(error, drm_i915_private_t, gpu_error); struct drm_device *dev = dev_priv->dev; - struct intel_ring_buffer *ring; char *error_event[] = { I915_ERROR_UEVENT "=1", NULL }; char *reset_event[] = { I915_RESET_UEVENT "=1", NULL }; char *reset_done_event[] = { I915_ERROR_UEVENT "=0", NULL }; - int i, ret; + int ret; kobject_uevent_env(&dev->primary->kdev.kobj, KOBJ_CHANGE, error_event); @@ -1520,8 +1547,16 @@ static void i915_error_work_func(struct work_struct *work) kobject_uevent_env(&dev->primary->kdev.kobj, KOBJ_CHANGE, reset_event); + /* + * All state reset _must_ be completed before we update the + * reset counter, for otherwise waiters might miss the reset + * pending state and not properly drop locks, resulting in + * deadlocks with the reset work. + */ ret = i915_reset(dev); + intel_display_handle_reset(dev); + if (ret == 0) { /* * After all the gem state is reset, increment the reset @@ -1542,12 +1577,11 @@ static void i915_error_work_func(struct work_struct *work) atomic_set(&error->reset_counter, I915_WEDGED); } - for_each_ring(ring, dev_priv, i) - wake_up_all(&ring->irq_queue); - - intel_display_handle_reset(dev); - - wake_up_all(&dev_priv->gpu_error.reset_queue); + /* + * Note: The wake_up also serves as a memory barrier so that + * waiters see the update value of the reset counter atomic_t. + */ + i915_error_wake_up(dev_priv, true); } } @@ -1656,8 +1690,6 @@ static void i915_report_and_clear_eir(struct drm_device *dev) void i915_handle_error(struct drm_device *dev, bool wedged) { struct drm_i915_private *dev_priv = dev->dev_private; - struct intel_ring_buffer *ring; - int i; i915_capture_error_state(dev); i915_report_and_clear_eir(dev); @@ -1667,14 +1699,28 @@ void i915_handle_error(struct drm_device *dev, bool wedged) &dev_priv->gpu_error.reset_counter); /* - * Wakeup waiting processes so that the reset work item - * doesn't deadlock trying to grab various locks. + * Wakeup waiting processes so that the reset work function + * i915_error_work_func doesn't deadlock trying to grab various + * locks. By bumping the reset counter first, the woken + * processes will see a reset in progress and back off, + * releasing their locks and then wait for the reset completion. + * We must do this for _all_ gpu waiters that might hold locks + * that the reset work needs to acquire. + * + * Note: The wake_up serves as the required memory barrier to + * ensure that the waiters see the updated value of the reset + * counter atomic_t. */ - for_each_ring(ring, dev_priv, i) - wake_up_all(&ring->irq_queue); + i915_error_wake_up(dev_priv, false); } - queue_work(dev_priv->wq, &dev_priv->gpu_error.work); + /* + * Our reset work can grab modeset locks (since it needs to reset the + * state of outstanding pagelips). Hence it must not be run on our own + * dev-priv->wq work queue for otherwise the flush_work in the pageflip + * code will deadlock. + */ + schedule_work(&dev_priv->gpu_error.work); } static void __always_unused i915_pageflip_stall_check(struct drm_device *dev, int pipe) diff --git a/drivers/gpu/drm/i915/intel_ddi.c b/drivers/gpu/drm/i915/intel_ddi.c index 491899539aa..04f68804834 100644 --- a/drivers/gpu/drm/i915/intel_ddi.c +++ b/drivers/gpu/drm/i915/intel_ddi.c @@ -778,7 +778,7 @@ void intel_ddi_enable_transcoder_func(struct drm_crtc *crtc) /* Can only use the always-on power well for eDP when * not using the panel fitter, and when not using motion * blur mitigation (which we don't support). */ - if (intel_crtc->config.pch_pfit.size) + if (intel_crtc->config.pch_pfit.enabled) temp |= TRANS_DDI_EDP_INPUT_A_ONOFF; else temp |= TRANS_DDI_EDP_INPUT_A_ON; diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c index 2334e974301..76d1d32f178 100644 --- a/drivers/gpu/drm/i915/intel_display.c +++ b/drivers/gpu/drm/i915/intel_display.c @@ -2291,7 +2291,7 @@ intel_pipe_set_base(struct drm_crtc *crtc, int x, int y, I915_WRITE(PIPESRC(intel_crtc->pipe), ((crtc->mode.hdisplay - 1) << 16) | (crtc->mode.vdisplay - 1)); - if (!intel_crtc->config.pch_pfit.size && + if (!intel_crtc->config.pch_pfit.enabled && (intel_pipe_has_type(crtc, INTEL_OUTPUT_LVDS) || intel_pipe_has_type(crtc, INTEL_OUTPUT_EDP))) { I915_WRITE(PF_CTL(intel_crtc->pipe), 0); @@ -3246,7 +3246,7 @@ static void ironlake_pfit_enable(struct intel_crtc *crtc) struct drm_i915_private *dev_priv = dev->dev_private; int pipe = crtc->pipe; - if (crtc->config.pch_pfit.size) { + if (crtc->config.pch_pfit.enabled) { /* Force use of hard-coded filter coefficients * as some pre-programmed values are broken, * e.g. x201. @@ -3472,7 +3472,7 @@ static void ironlake_pfit_disable(struct intel_crtc *crtc) /* To avoid upsetting the power well on haswell only disable the pfit if * it's in use. The hw state code will make sure we get this right. */ - if (crtc->config.pch_pfit.size) { + if (crtc->config.pch_pfit.enabled) { I915_WRITE(PF_CTL(pipe), 0); I915_WRITE(PF_WIN_POS(pipe), 0); I915_WRITE(PF_WIN_SZ(pipe), 0); @@ -4933,10 +4933,7 @@ static int i9xx_crtc_mode_set(struct drm_crtc *crtc, } } - /* Ensure that the cursor is valid for the new mode before changing... */ - intel_crtc_update_cursor(crtc, true); - - if (!is_dsi && is_lvds && dev_priv->lvds_downclock_avail) { + if (is_lvds && dev_priv->lvds_downclock_avail) { /* * Ensure we match the reduced clock's P to the target clock. * If the clocks don't match, we can't switch the display clock @@ -5846,9 +5843,6 @@ static int ironlake_crtc_mode_set(struct drm_crtc *crtc, intel_crtc->config.dpll.p2 = clock.p2; } - /* Ensure that the cursor is valid for the new mode before changing... */ - intel_crtc_update_cursor(crtc, true); - /* CPU eDP is the only output that doesn't need a PCH PLL of its own. */ if (intel_crtc->config.has_pch_encoder) { fp = i9xx_dpll_compute_fp(&intel_crtc->config.dpll); @@ -5979,6 +5973,7 @@ static void ironlake_get_pfit_config(struct intel_crtc *crtc, tmp = I915_READ(PF_CTL(crtc->pipe)); if (tmp & PF_ENABLE) { + pipe_config->pch_pfit.enabled = true; pipe_config->pch_pfit.pos = I915_READ(PF_WIN_POS(crtc->pipe)); pipe_config->pch_pfit.size = I915_READ(PF_WIN_SZ(crtc->pipe)); @@ -6381,7 +6376,7 @@ static void haswell_modeset_global_resources(struct drm_device *dev) if (!crtc->base.enabled) continue; - if (crtc->pipe != PIPE_A || crtc->config.pch_pfit.size || + if (crtc->pipe != PIPE_A || crtc->config.pch_pfit.enabled || crtc->config.cpu_transcoder != TRANSCODER_EDP) enable = true; } @@ -6404,9 +6399,6 @@ static int haswell_crtc_mode_set(struct drm_crtc *crtc, if (!intel_ddi_pll_mode_set(crtc)) return -EINVAL; - /* Ensure that the cursor is valid for the new mode before changing... */ - intel_crtc_update_cursor(crtc, true); - if (intel_crtc->config.has_dp_encoder) intel_dp_set_m_n(intel_crtc); @@ -6637,15 +6629,15 @@ static void haswell_write_eld(struct drm_connector *connector, /* Set ELD valid state */ tmp = I915_READ(aud_cntrl_st2); - DRM_DEBUG_DRIVER("HDMI audio: pin eld vld status=0x%8x\n", tmp); + DRM_DEBUG_DRIVER("HDMI audio: pin eld vld status=0x%08x\n", tmp); tmp |= (AUDIO_ELD_VALID_A << (pipe * 4)); I915_WRITE(aud_cntrl_st2, tmp); tmp = I915_READ(aud_cntrl_st2); - DRM_DEBUG_DRIVER("HDMI audio: eld vld status=0x%8x\n", tmp); + DRM_DEBUG_DRIVER("HDMI audio: eld vld status=0x%08x\n", tmp); /* Enable HDMI mode */ tmp = I915_READ(aud_config); - DRM_DEBUG_DRIVER("HDMI audio: audio conf: 0x%8x\n", tmp); + DRM_DEBUG_DRIVER("HDMI audio: audio conf: 0x%08x\n", tmp); /* clear N_programing_enable and N_value_index */ tmp &= ~(AUD_CONFIG_N_VALUE_INDEX | AUD_CONFIG_N_PROG_ENABLE); I915_WRITE(aud_config, tmp); @@ -7081,7 +7073,8 @@ static int intel_crtc_cursor_set(struct drm_crtc *crtc, intel_crtc->cursor_width = width; intel_crtc->cursor_height = height; - intel_crtc_update_cursor(crtc, intel_crtc->cursor_bo != NULL); + if (intel_crtc->active) + intel_crtc_update_cursor(crtc, intel_crtc->cursor_bo != NULL); return 0; fail_unpin: @@ -7100,7 +7093,8 @@ static int intel_crtc_cursor_move(struct drm_crtc *crtc, int x, int y) intel_crtc->cursor_x = x; intel_crtc->cursor_y = y; - intel_crtc_update_cursor(crtc, intel_crtc->cursor_bo != NULL); + if (intel_crtc->active) + intel_crtc_update_cursor(crtc, intel_crtc->cursor_bo != NULL); return 0; } @@ -7997,7 +7991,7 @@ static int intel_gen7_queue_flip(struct drm_device *dev, int len, ret; ring = obj->ring; - if (ring == NULL || ring->id != RCS) + if (IS_VALLEYVIEW(dev) || ring == NULL || ring->id != RCS) ring = &dev_priv->ring[BCS]; ret = intel_pin_and_fence_fb_obj(dev, obj, ring); @@ -8378,9 +8372,10 @@ static void intel_dump_pipe_config(struct intel_crtc *crtc, pipe_config->gmch_pfit.control, pipe_config->gmch_pfit.pgm_ratios, pipe_config->gmch_pfit.lvds_border_bits); - DRM_DEBUG_KMS("pch pfit: pos: 0x%08x, size: 0x%08x\n", + DRM_DEBUG_KMS("pch pfit: pos: 0x%08x, size: 0x%08x, %s\n", pipe_config->pch_pfit.pos, - pipe_config->pch_pfit.size); + pipe_config->pch_pfit.size, + pipe_config->pch_pfit.enabled ? "enabled" : "disabled"); DRM_DEBUG_KMS("ips: %i\n", pipe_config->ips_enabled); DRM_DEBUG_KMS("double wide: %i\n", pipe_config->double_wide); } @@ -8794,8 +8789,11 @@ intel_pipe_config_compare(struct drm_device *dev, if (INTEL_INFO(dev)->gen < 4) PIPE_CONF_CHECK_I(gmch_pfit.pgm_ratios); PIPE_CONF_CHECK_I(gmch_pfit.lvds_border_bits); - PIPE_CONF_CHECK_I(pch_pfit.pos); - PIPE_CONF_CHECK_I(pch_pfit.size); + PIPE_CONF_CHECK_I(pch_pfit.enabled); + if (current_config->pch_pfit.enabled) { + PIPE_CONF_CHECK_I(pch_pfit.pos); + PIPE_CONF_CHECK_I(pch_pfit.size); + } PIPE_CONF_CHECK_I(ips_enabled); @@ -10232,15 +10230,6 @@ static void i915_disable_vga(struct drm_device *dev) outb(SR01, VGA_SR_INDEX); sr1 = inb(VGA_SR_DATA); outb(sr1 | 1<<5, VGA_SR_DATA); - - /* Disable VGA memory on Intel HD */ - if (HAS_PCH_SPLIT(dev)) { - outb(inb(VGA_MSR_READ) & ~VGA_MSR_MEM_EN, VGA_MSR_WRITE); - vga_set_legacy_decoding(dev->pdev, VGA_RSRC_LEGACY_IO | - VGA_RSRC_NORMAL_IO | - VGA_RSRC_NORMAL_MEM); - } - vga_put(dev->pdev, VGA_RSRC_LEGACY_IO); udelay(300); @@ -10248,7 +10237,7 @@ static void i915_disable_vga(struct drm_device *dev) POSTING_READ(vga_reg); } -static void i915_enable_vga(struct drm_device *dev) +static void i915_enable_vga_mem(struct drm_device *dev) { /* Enable VGA memory on Intel HD */ if (HAS_PCH_SPLIT(dev)) { @@ -10262,6 +10251,19 @@ static void i915_enable_vga(struct drm_device *dev) } } +void i915_disable_vga_mem(struct drm_device *dev) +{ + /* Disable VGA memory on Intel HD */ + if (HAS_PCH_SPLIT(dev)) { + vga_get_uninterruptible(dev->pdev, VGA_RSRC_LEGACY_IO); + outb(inb(VGA_MSR_READ) & ~VGA_MSR_MEM_EN, VGA_MSR_WRITE); + vga_set_legacy_decoding(dev->pdev, VGA_RSRC_LEGACY_IO | + VGA_RSRC_NORMAL_IO | + VGA_RSRC_NORMAL_MEM); + vga_put(dev->pdev, VGA_RSRC_LEGACY_IO); + } +} + void intel_modeset_init_hw(struct drm_device *dev) { intel_prepare_ddi(dev); @@ -10538,6 +10540,7 @@ void i915_redisable_vga(struct drm_device *dev) if (I915_READ(vga_reg) != VGA_DISP_DISABLE) { DRM_DEBUG_KMS("Something enabled VGA plane, disabling it\n"); i915_disable_vga(dev); + i915_disable_vga_mem(dev); } } @@ -10742,7 +10745,7 @@ void intel_modeset_cleanup(struct drm_device *dev) intel_disable_fbc(dev); - i915_enable_vga(dev); + i915_enable_vga_mem(dev); intel_disable_gt_powersave(dev); diff --git a/drivers/gpu/drm/i915/intel_drv.h b/drivers/gpu/drm/i915/intel_drv.h index 1f052678035..4b75328462e 100644 --- a/drivers/gpu/drm/i915/intel_drv.h +++ b/drivers/gpu/drm/i915/intel_drv.h @@ -298,6 +298,7 @@ struct intel_crtc_config { struct { u32 pos; u32 size; + bool enabled; } pch_pfit; /* FDI configuration, only valid if has_pch_encoder is set. */ @@ -827,5 +828,6 @@ extern void ironlake_check_encoder_dotclock(const struct intel_crtc_config *pipe int dotclock); extern bool intel_crtc_active(struct drm_crtc *crtc); +extern void i915_disable_vga_mem(struct drm_device *dev); #endif /* __INTEL_DRV_H__ */ diff --git a/drivers/gpu/drm/i915/intel_dvo.c b/drivers/gpu/drm/i915/intel_dvo.c index fe65c7270f0..ff86c366218 100644 --- a/drivers/gpu/drm/i915/intel_dvo.c +++ b/drivers/gpu/drm/i915/intel_dvo.c @@ -265,6 +265,8 @@ static bool intel_dvo_compute_config(struct intel_encoder *encoder, C(vtotal); C(clock); #undef C + + drm_mode_set_crtcinfo(adjusted_mode, 0); } return true; diff --git a/drivers/gpu/drm/i915/intel_opregion.c b/drivers/gpu/drm/i915/intel_opregion.c index fd3062a2a42..2acf5cae20e 100644 --- a/drivers/gpu/drm/i915/intel_opregion.c +++ b/drivers/gpu/drm/i915/intel_opregion.c @@ -406,7 +406,7 @@ static u32 asle_set_backlight(struct drm_device *dev, u32 bclp) return ASLC_BACKLIGHT_FAILED; intel_panel_set_backlight(dev, bclp, 255); - iowrite32((bclp*0x64)/0xff | ASLE_CBLV_VALID, &asle->cblv); + iowrite32(DIV_ROUND_UP(bclp * 100, 255) | ASLE_CBLV_VALID, &asle->cblv); return 0; } diff --git a/drivers/gpu/drm/i915/intel_panel.c b/drivers/gpu/drm/i915/intel_panel.c index 0aaeb8b6d6e..3bc89a6bc3e 100644 --- a/drivers/gpu/drm/i915/intel_panel.c +++ b/drivers/gpu/drm/i915/intel_panel.c @@ -113,6 +113,7 @@ intel_pch_panel_fitting(struct intel_crtc *intel_crtc, done: pipe_config->pch_pfit.pos = (x << 16) | y; pipe_config->pch_pfit.size = (width << 16) | height; + pipe_config->pch_pfit.enabled = pipe_config->pch_pfit.size != 0; } static void diff --git a/drivers/gpu/drm/i915/intel_pm.c b/drivers/gpu/drm/i915/intel_pm.c index 6789dcdb9b1..4692f8cb772 100644 --- a/drivers/gpu/drm/i915/intel_pm.c +++ b/drivers/gpu/drm/i915/intel_pm.c @@ -2110,16 +2110,16 @@ static uint32_t ilk_pipe_pixel_rate(struct drm_device *dev, struct drm_crtc *crtc) { struct intel_crtc *intel_crtc = to_intel_crtc(crtc); - uint32_t pixel_rate, pfit_size; + uint32_t pixel_rate; pixel_rate = intel_crtc->config.adjusted_mode.clock; /* We only use IF-ID interlacing. If we ever use PF-ID we'll need to * adjust the pixel_rate here. */ - pfit_size = intel_crtc->config.pch_pfit.size; - if (pfit_size) { + if (intel_crtc->config.pch_pfit.enabled) { uint64_t pipe_w, pipe_h, pfit_w, pfit_h; + uint32_t pfit_size = intel_crtc->config.pch_pfit.size; pipe_w = intel_crtc->config.pipe_src_w; pipe_h = intel_crtc->config.pipe_src_h; diff --git a/drivers/gpu/drm/i915/intel_sdvo.c b/drivers/gpu/drm/i915/intel_sdvo.c index 8aa7be588f7..91aea9e1ab6 100644 --- a/drivers/gpu/drm/i915/intel_sdvo.c +++ b/drivers/gpu/drm/i915/intel_sdvo.c @@ -788,6 +788,8 @@ static void intel_sdvo_get_dtd_from_mode(struct intel_sdvo_dtd *dtd, uint16_t h_sync_offset, v_sync_offset; int mode_clock; + memset(dtd, 0, sizeof(*dtd)); + width = mode->hdisplay; height = mode->vdisplay; @@ -830,44 +832,51 @@ static void intel_sdvo_get_dtd_from_mode(struct intel_sdvo_dtd *dtd, if (mode->flags & DRM_MODE_FLAG_PVSYNC) dtd->part2.dtd_flags |= DTD_FLAG_VSYNC_POSITIVE; - dtd->part2.sdvo_flags = 0; dtd->part2.v_sync_off_high = v_sync_offset & 0xc0; - dtd->part2.reserved = 0; } -static void intel_sdvo_get_mode_from_dtd(struct drm_display_mode * mode, +static void intel_sdvo_get_mode_from_dtd(struct drm_display_mode *pmode, const struct intel_sdvo_dtd *dtd) { - mode->hdisplay = dtd->part1.h_active; - mode->hdisplay += ((dtd->part1.h_high >> 4) & 0x0f) << 8; - mode->hsync_start = mode->hdisplay + dtd->part2.h_sync_off; - mode->hsync_start += (dtd->part2.sync_off_width_high & 0xc0) << 2; - mode->hsync_end = mode->hsync_start + dtd->part2.h_sync_width; - mode->hsync_end += (dtd->part2.sync_off_width_high & 0x30) << 4; - mode->htotal = mode->hdisplay + dtd->part1.h_blank; - mode->htotal += (dtd->part1.h_high & 0xf) << 8; - - mode->vdisplay = dtd->part1.v_active; - mode->vdisplay += ((dtd->part1.v_high >> 4) & 0x0f) << 8; - mode->vsync_start = mode->vdisplay; - mode->vsync_start += (dtd->part2.v_sync_off_width >> 4) & 0xf; - mode->vsync_start += (dtd->part2.sync_off_width_high & 0x0c) << 2; - mode->vsync_start += dtd->part2.v_sync_off_high & 0xc0; - mode->vsync_end = mode->vsync_start + + struct drm_display_mode mode = {}; + + mode.hdisplay = dtd->part1.h_active; + mode.hdisplay += ((dtd->part1.h_high >> 4) & 0x0f) << 8; + mode.hsync_start = mode.hdisplay + dtd->part2.h_sync_off; + mode.hsync_start += (dtd->part2.sync_off_width_high & 0xc0) << 2; + mode.hsync_end = mode.hsync_start + dtd->part2.h_sync_width; + mode.hsync_end += (dtd->part2.sync_off_width_high & 0x30) << 4; + mode.htotal = mode.hdisplay + dtd->part1.h_blank; + mode.htotal += (dtd->part1.h_high & 0xf) << 8; + + mode.vdisplay = dtd->part1.v_active; + mode.vdisplay += ((dtd->part1.v_high >> 4) & 0x0f) << 8; + mode.vsync_start = mode.vdisplay; + mode.vsync_start += (dtd->part2.v_sync_off_width >> 4) & 0xf; + mode.vsync_start += (dtd->part2.sync_off_width_high & 0x0c) << 2; + mode.vsync_start += dtd->part2.v_sync_off_high & 0xc0; + mode.vsync_end = mode.vsync_start + (dtd->part2.v_sync_off_width & 0xf); - mode->vsync_end += (dtd->part2.sync_off_width_high & 0x3) << 4; - mode->vtotal = mode->vdisplay + dtd->part1.v_blank; - mode->vtotal += (dtd->part1.v_high & 0xf) << 8; + mode.vsync_end += (dtd->part2.sync_off_width_high & 0x3) << 4; + mode.vtotal = mode.vdisplay + dtd->part1.v_blank; + mode.vtotal += (dtd->part1.v_high & 0xf) << 8; - mode->clock = dtd->part1.clock * 10; + mode.clock = dtd->part1.clock * 10; - mode->flags &= ~(DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC); if (dtd->part2.dtd_flags & DTD_FLAG_INTERLACE) - mode->flags |= DRM_MODE_FLAG_INTERLACE; + mode.flags |= DRM_MODE_FLAG_INTERLACE; if (dtd->part2.dtd_flags & DTD_FLAG_HSYNC_POSITIVE) - mode->flags |= DRM_MODE_FLAG_PHSYNC; + mode.flags |= DRM_MODE_FLAG_PHSYNC; + else + mode.flags |= DRM_MODE_FLAG_NHSYNC; if (dtd->part2.dtd_flags & DTD_FLAG_VSYNC_POSITIVE) - mode->flags |= DRM_MODE_FLAG_PVSYNC; + mode.flags |= DRM_MODE_FLAG_PVSYNC; + else + mode.flags |= DRM_MODE_FLAG_NVSYNC; + + drm_mode_set_crtcinfo(&mode, 0); + + drm_mode_copy(pmode, &mode); } static bool intel_sdvo_check_supp_encode(struct intel_sdvo *intel_sdvo) diff --git a/drivers/gpu/drm/msm/adreno/adreno_gpu.c b/drivers/gpu/drm/msm/adreno/adreno_gpu.c index a60584763b6..a0b9d8a95b1 100644 --- a/drivers/gpu/drm/msm/adreno/adreno_gpu.c +++ b/drivers/gpu/drm/msm/adreno/adreno_gpu.c @@ -124,6 +124,8 @@ void adreno_recover(struct msm_gpu *gpu) /* reset completed fence seqno, just discard anything pending: */ adreno_gpu->memptrs->fence = gpu->submitted_fence; + adreno_gpu->memptrs->rptr = 0; + adreno_gpu->memptrs->wptr = 0; gpu->funcs->pm_resume(gpu); ret = gpu->funcs->hw_init(gpu); @@ -229,7 +231,7 @@ void adreno_idle(struct msm_gpu *gpu) return; } while(time_before(jiffies, t)); - DRM_ERROR("timeout waiting for %s to drain ringbuffer!\n", gpu->name); + DRM_ERROR("%s: timeout waiting to drain ringbuffer!\n", gpu->name); /* TODO maybe we need to reset GPU here to recover from hang? */ } @@ -256,11 +258,17 @@ void adreno_wait_ring(struct msm_gpu *gpu, uint32_t ndwords) { struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); uint32_t freedwords; + unsigned long t = jiffies + ADRENO_IDLE_TIMEOUT; do { uint32_t size = gpu->rb->size / 4; uint32_t wptr = get_wptr(gpu->rb); uint32_t rptr = adreno_gpu->memptrs->rptr; freedwords = (rptr + (size - 1) - wptr) % size; + + if (time_after(jiffies, t)) { + DRM_ERROR("%s: timeout waiting for ringbuffer space\n", gpu->name); + break; + } } while(freedwords < ndwords); } diff --git a/drivers/gpu/drm/msm/msm_drv.c b/drivers/gpu/drm/msm/msm_drv.c index 864c9773636..008d772384c 100644 --- a/drivers/gpu/drm/msm/msm_drv.c +++ b/drivers/gpu/drm/msm/msm_drv.c @@ -499,25 +499,41 @@ int msm_wait_fence_interruptable(struct drm_device *dev, uint32_t fence, struct timespec *timeout) { struct msm_drm_private *priv = dev->dev_private; - unsigned long timeout_jiffies = timespec_to_jiffies(timeout); - unsigned long start_jiffies = jiffies; - unsigned long remaining_jiffies; int ret; - if (time_after(start_jiffies, timeout_jiffies)) - remaining_jiffies = 0; - else - remaining_jiffies = timeout_jiffies - start_jiffies; - - ret = wait_event_interruptible_timeout(priv->fence_event, - priv->completed_fence >= fence, - remaining_jiffies); - if (ret == 0) { - DBG("timeout waiting for fence: %u (completed: %u)", - fence, priv->completed_fence); - ret = -ETIMEDOUT; - } else if (ret != -ERESTARTSYS) { - ret = 0; + if (!priv->gpu) + return 0; + + if (fence > priv->gpu->submitted_fence) { + DRM_ERROR("waiting on invalid fence: %u (of %u)\n", + fence, priv->gpu->submitted_fence); + return -EINVAL; + } + + if (!timeout) { + /* no-wait: */ + ret = fence_completed(dev, fence) ? 0 : -EBUSY; + } else { + unsigned long timeout_jiffies = timespec_to_jiffies(timeout); + unsigned long start_jiffies = jiffies; + unsigned long remaining_jiffies; + + if (time_after(start_jiffies, timeout_jiffies)) + remaining_jiffies = 0; + else + remaining_jiffies = timeout_jiffies - start_jiffies; + + ret = wait_event_interruptible_timeout(priv->fence_event, + fence_completed(dev, fence), + remaining_jiffies); + + if (ret == 0) { + DBG("timeout waiting for fence: %u (completed: %u)", + fence, priv->completed_fence); + ret = -ETIMEDOUT; + } else if (ret != -ERESTARTSYS) { + ret = 0; + } } return ret; diff --git a/drivers/gpu/drm/msm/msm_drv.h b/drivers/gpu/drm/msm/msm_drv.h index 80d75094bf0..df8f1d084bc 100644 --- a/drivers/gpu/drm/msm/msm_drv.h +++ b/drivers/gpu/drm/msm/msm_drv.h @@ -153,7 +153,7 @@ void *msm_gem_vaddr(struct drm_gem_object *obj); int msm_gem_queue_inactive_work(struct drm_gem_object *obj, struct work_struct *work); void msm_gem_move_to_active(struct drm_gem_object *obj, - struct msm_gpu *gpu, uint32_t fence); + struct msm_gpu *gpu, bool write, uint32_t fence); void msm_gem_move_to_inactive(struct drm_gem_object *obj); int msm_gem_cpu_prep(struct drm_gem_object *obj, uint32_t op, struct timespec *timeout); @@ -191,6 +191,12 @@ u32 msm_readl(const void __iomem *addr); #define DBG(fmt, ...) DRM_DEBUG(fmt"\n", ##__VA_ARGS__) #define VERB(fmt, ...) if (0) DRM_DEBUG(fmt"\n", ##__VA_ARGS__) +static inline bool fence_completed(struct drm_device *dev, uint32_t fence) +{ + struct msm_drm_private *priv = dev->dev_private; + return priv->completed_fence >= fence; +} + static inline int align_pitch(int width, int bpp) { int bytespp = (bpp + 7) / 8; diff --git a/drivers/gpu/drm/msm/msm_gem.c b/drivers/gpu/drm/msm/msm_gem.c index 6b5a6c8c765..29eacfa29cf 100644 --- a/drivers/gpu/drm/msm/msm_gem.c +++ b/drivers/gpu/drm/msm/msm_gem.c @@ -40,9 +40,9 @@ static struct page **get_pages(struct drm_gem_object *obj) } msm_obj->sgt = drm_prime_pages_to_sg(p, npages); - if (!msm_obj->sgt) { + if (IS_ERR(msm_obj->sgt)) { dev_err(dev->dev, "failed to allocate sgt\n"); - return ERR_PTR(-ENOMEM); + return ERR_CAST(msm_obj->sgt); } msm_obj->pages = p; @@ -159,7 +159,6 @@ out_unlock: out: switch (ret) { case -EAGAIN: - set_need_resched(); case 0: case -ERESTARTSYS: case -EINTR: @@ -393,11 +392,14 @@ int msm_gem_queue_inactive_work(struct drm_gem_object *obj, } void msm_gem_move_to_active(struct drm_gem_object *obj, - struct msm_gpu *gpu, uint32_t fence) + struct msm_gpu *gpu, bool write, uint32_t fence) { struct msm_gem_object *msm_obj = to_msm_bo(obj); msm_obj->gpu = gpu; - msm_obj->fence = fence; + if (write) + msm_obj->write_fence = fence; + else + msm_obj->read_fence = fence; list_del_init(&msm_obj->mm_list); list_add_tail(&msm_obj->mm_list, &gpu->active_list); } @@ -411,7 +413,8 @@ void msm_gem_move_to_inactive(struct drm_gem_object *obj) WARN_ON(!mutex_is_locked(&dev->struct_mutex)); msm_obj->gpu = NULL; - msm_obj->fence = 0; + msm_obj->read_fence = 0; + msm_obj->write_fence = 0; list_del_init(&msm_obj->mm_list); list_add_tail(&msm_obj->mm_list, &priv->inactive_list); @@ -433,8 +436,18 @@ int msm_gem_cpu_prep(struct drm_gem_object *obj, uint32_t op, struct msm_gem_object *msm_obj = to_msm_bo(obj); int ret = 0; - if (is_active(msm_obj) && !(op & MSM_PREP_NOSYNC)) - ret = msm_wait_fence_interruptable(dev, msm_obj->fence, timeout); + if (is_active(msm_obj)) { + uint32_t fence = 0; + + if (op & MSM_PREP_READ) + fence = msm_obj->write_fence; + if (op & MSM_PREP_WRITE) + fence = max(fence, msm_obj->read_fence); + if (op & MSM_PREP_NOSYNC) + timeout = NULL; + + ret = msm_wait_fence_interruptable(dev, fence, timeout); + } /* TODO cache maintenance */ @@ -455,9 +468,10 @@ void msm_gem_describe(struct drm_gem_object *obj, struct seq_file *m) uint64_t off = drm_vma_node_start(&obj->vma_node); WARN_ON(!mutex_is_locked(&dev->struct_mutex)); - seq_printf(m, "%08x: %c(%d) %2d (%2d) %08llx %p %d\n", + seq_printf(m, "%08x: %c(r=%u,w=%u) %2d (%2d) %08llx %p %d\n", msm_obj->flags, is_active(msm_obj) ? 'A' : 'I', - msm_obj->fence, obj->name, obj->refcount.refcount.counter, + msm_obj->read_fence, msm_obj->write_fence, + obj->name, obj->refcount.refcount.counter, off, msm_obj->vaddr, obj->size); } diff --git a/drivers/gpu/drm/msm/msm_gem.h b/drivers/gpu/drm/msm/msm_gem.h index d746f13d283..0676f32e2c6 100644 --- a/drivers/gpu/drm/msm/msm_gem.h +++ b/drivers/gpu/drm/msm/msm_gem.h @@ -36,7 +36,7 @@ struct msm_gem_object { */ struct list_head mm_list; struct msm_gpu *gpu; /* non-null if active */ - uint32_t fence; + uint32_t read_fence, write_fence; /* Transiently in the process of submit ioctl, objects associated * with the submit are on submit->bo_list.. this only lasts for diff --git a/drivers/gpu/drm/msm/msm_gem_submit.c b/drivers/gpu/drm/msm/msm_gem_submit.c index 3e1ef3a00f6..5281d4bc37f 100644 --- a/drivers/gpu/drm/msm/msm_gem_submit.c +++ b/drivers/gpu/drm/msm/msm_gem_submit.c @@ -78,7 +78,7 @@ static int submit_lookup_objects(struct msm_gem_submit *submit, } if (submit_bo.flags & BO_INVALID_FLAGS) { - DBG("invalid flags: %x", submit_bo.flags); + DRM_ERROR("invalid flags: %x\n", submit_bo.flags); ret = -EINVAL; goto out_unlock; } @@ -92,7 +92,7 @@ static int submit_lookup_objects(struct msm_gem_submit *submit, */ obj = idr_find(&file->object_idr, submit_bo.handle); if (!obj) { - DBG("invalid handle %u at index %u", submit_bo.handle, i); + DRM_ERROR("invalid handle %u at index %u\n", submit_bo.handle, i); ret = -EINVAL; goto out_unlock; } @@ -100,7 +100,7 @@ static int submit_lookup_objects(struct msm_gem_submit *submit, msm_obj = to_msm_bo(obj); if (!list_empty(&msm_obj->submit_entry)) { - DBG("handle %u at index %u already on submit list", + DRM_ERROR("handle %u at index %u already on submit list\n", submit_bo.handle, i); ret = -EINVAL; goto out_unlock; @@ -216,8 +216,9 @@ static int submit_bo(struct msm_gem_submit *submit, uint32_t idx, struct msm_gem_object **obj, uint32_t *iova, bool *valid) { if (idx >= submit->nr_bos) { - DBG("invalid buffer index: %u (out of %u)", idx, submit->nr_bos); - return EINVAL; + DRM_ERROR("invalid buffer index: %u (out of %u)\n", + idx, submit->nr_bos); + return -EINVAL; } if (obj) @@ -239,7 +240,7 @@ static int submit_reloc(struct msm_gem_submit *submit, struct msm_gem_object *ob int ret; if (offset % 4) { - DBG("non-aligned cmdstream buffer: %u", offset); + DRM_ERROR("non-aligned cmdstream buffer: %u\n", offset); return -EINVAL; } @@ -266,7 +267,7 @@ static int submit_reloc(struct msm_gem_submit *submit, struct msm_gem_object *ob return -EFAULT; if (submit_reloc.submit_offset % 4) { - DBG("non-aligned reloc offset: %u", + DRM_ERROR("non-aligned reloc offset: %u\n", submit_reloc.submit_offset); return -EINVAL; } @@ -276,7 +277,7 @@ static int submit_reloc(struct msm_gem_submit *submit, struct msm_gem_object *ob if ((off >= (obj->base.size / 4)) || (off < last_offset)) { - DBG("invalid offset %u at reloc %u", off, i); + DRM_ERROR("invalid offset %u at reloc %u\n", off, i); return -EINVAL; } @@ -374,14 +375,15 @@ int msm_ioctl_gem_submit(struct drm_device *dev, void *data, goto out; if (submit_cmd.size % 4) { - DBG("non-aligned cmdstream buffer size: %u", + DRM_ERROR("non-aligned cmdstream buffer size: %u\n", submit_cmd.size); ret = -EINVAL; goto out; } - if (submit_cmd.size >= msm_obj->base.size) { - DBG("invalid cmdstream size: %u", submit_cmd.size); + if ((submit_cmd.size + submit_cmd.submit_offset) >= + msm_obj->base.size) { + DRM_ERROR("invalid cmdstream size: %u\n", submit_cmd.size); ret = -EINVAL; goto out; } diff --git a/drivers/gpu/drm/msm/msm_gpu.c b/drivers/gpu/drm/msm/msm_gpu.c index e1e1ec9321f..3bab937965d 100644 --- a/drivers/gpu/drm/msm/msm_gpu.c +++ b/drivers/gpu/drm/msm/msm_gpu.c @@ -29,13 +29,14 @@ static void bs_init(struct msm_gpu *gpu, struct platform_device *pdev) { struct drm_device *dev = gpu->dev; - struct kgsl_device_platform_data *pdata = pdev->dev.platform_data; + struct kgsl_device_platform_data *pdata; if (!pdev) { dev_err(dev->dev, "could not find dtv pdata\n"); return; } + pdata = pdev->dev.platform_data; if (pdata->bus_scale_table) { gpu->bsc = msm_bus_scale_register_client(pdata->bus_scale_table); DBG("bus scale client: %08x", gpu->bsc); @@ -230,6 +231,8 @@ static void hangcheck_timer_reset(struct msm_gpu *gpu) static void hangcheck_handler(unsigned long data) { struct msm_gpu *gpu = (struct msm_gpu *)data; + struct drm_device *dev = gpu->dev; + struct msm_drm_private *priv = dev->dev_private; uint32_t fence = gpu->funcs->last_fence(gpu); if (fence != gpu->hangcheck_fence) { @@ -237,14 +240,22 @@ static void hangcheck_handler(unsigned long data) gpu->hangcheck_fence = fence; } else if (fence < gpu->submitted_fence) { /* no progress and not done.. hung! */ - struct msm_drm_private *priv = gpu->dev->dev_private; gpu->hangcheck_fence = fence; + dev_err(dev->dev, "%s: hangcheck detected gpu lockup!\n", + gpu->name); + dev_err(dev->dev, "%s: completed fence: %u\n", + gpu->name, fence); + dev_err(dev->dev, "%s: submitted fence: %u\n", + gpu->name, gpu->submitted_fence); queue_work(priv->wq, &gpu->recover_work); } /* if still more pending work, reset the hangcheck timer: */ if (gpu->submitted_fence > gpu->hangcheck_fence) hangcheck_timer_reset(gpu); + + /* workaround for missing irq: */ + queue_work(priv->wq, &gpu->retire_work); } /* @@ -265,7 +276,8 @@ static void retire_worker(struct work_struct *work) obj = list_first_entry(&gpu->active_list, struct msm_gem_object, mm_list); - if (obj->fence <= fence) { + if ((obj->read_fence <= fence) && + (obj->write_fence <= fence)) { /* move to inactive: */ msm_gem_move_to_inactive(&obj->base); msm_gem_put_iova(&obj->base, gpu->id); @@ -321,7 +333,11 @@ int msm_gpu_submit(struct msm_gpu *gpu, struct msm_gem_submit *submit, submit->gpu->id, &iova); } - msm_gem_move_to_active(&msm_obj->base, gpu, submit->fence); + if (submit->bos[i].flags & MSM_SUBMIT_BO_READ) + msm_gem_move_to_active(&msm_obj->base, gpu, false, submit->fence); + + if (submit->bos[i].flags & MSM_SUBMIT_BO_WRITE) + msm_gem_move_to_active(&msm_obj->base, gpu, true, submit->fence); } hangcheck_timer_reset(gpu); mutex_unlock(&dev->struct_mutex); diff --git a/drivers/gpu/drm/nouveau/core/core/ramht.c b/drivers/gpu/drm/nouveau/core/core/ramht.c index 86a64045dd6..f3b9bddc387 100644 --- a/drivers/gpu/drm/nouveau/core/core/ramht.c +++ b/drivers/gpu/drm/nouveau/core/core/ramht.c @@ -22,7 +22,6 @@ #include <core/object.h> #include <core/ramht.h> -#include <core/math.h> #include <subdev/bar.h> @@ -104,6 +103,6 @@ nouveau_ramht_new(struct nouveau_object *parent, struct nouveau_object *pargpu, if (ret) return ret; - ramht->bits = log2i(nv_gpuobj(ramht)->size >> 3); + ramht->bits = order_base_2(nv_gpuobj(ramht)->size >> 3); return 0; } diff --git a/drivers/gpu/drm/nouveau/core/engine/bsp/nv98.c b/drivers/gpu/drm/nouveau/core/engine/bsp/nv98.c index 8bf92b0e6d8..6b089e022fd 100644 --- a/drivers/gpu/drm/nouveau/core/engine/bsp/nv98.c +++ b/drivers/gpu/drm/nouveau/core/engine/bsp/nv98.c @@ -19,16 +19,14 @@ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR * OTHER DEALINGS IN THE SOFTWARE. * - * Authors: Ben Skeggs + * Authors: Ben Skeggs, Maarten Lankhorst, Ilia Mirkin */ -#include <core/engctx.h> -#include <core/class.h> - +#include <engine/falcon.h> #include <engine/bsp.h> struct nv98_bsp_priv { - struct nouveau_engine base; + struct nouveau_falcon base; }; /******************************************************************************* @@ -37,31 +35,49 @@ struct nv98_bsp_priv { static struct nouveau_oclass nv98_bsp_sclass[] = { + { 0x88b1, &nouveau_object_ofuncs }, + { 0x85b1, &nouveau_object_ofuncs }, + { 0x86b1, &nouveau_object_ofuncs }, {}, }; /******************************************************************************* - * BSP context + * PBSP context ******************************************************************************/ static struct nouveau_oclass nv98_bsp_cclass = { .handle = NV_ENGCTX(BSP, 0x98), .ofuncs = &(struct nouveau_ofuncs) { - .ctor = _nouveau_engctx_ctor, - .dtor = _nouveau_engctx_dtor, - .init = _nouveau_engctx_init, - .fini = _nouveau_engctx_fini, - .rd32 = _nouveau_engctx_rd32, - .wr32 = _nouveau_engctx_wr32, + .ctor = _nouveau_falcon_context_ctor, + .dtor = _nouveau_falcon_context_dtor, + .init = _nouveau_falcon_context_init, + .fini = _nouveau_falcon_context_fini, + .rd32 = _nouveau_falcon_context_rd32, + .wr32 = _nouveau_falcon_context_wr32, }, }; /******************************************************************************* - * BSP engine/subdev functions + * PBSP engine/subdev functions ******************************************************************************/ static int +nv98_bsp_init(struct nouveau_object *object) +{ + struct nv98_bsp_priv *priv = (void *)object; + int ret; + + ret = nouveau_falcon_init(&priv->base); + if (ret) + return ret; + + nv_wr32(priv, 0x084010, 0x0000ffd2); + nv_wr32(priv, 0x08401c, 0x0000fff2); + return 0; +} + +static int nv98_bsp_ctor(struct nouveau_object *parent, struct nouveau_object *engine, struct nouveau_oclass *oclass, void *data, u32 size, struct nouveau_object **pobject) @@ -69,7 +85,7 @@ nv98_bsp_ctor(struct nouveau_object *parent, struct nouveau_object *engine, struct nv98_bsp_priv *priv; int ret; - ret = nouveau_engine_create(parent, engine, oclass, true, + ret = nouveau_falcon_create(parent, engine, oclass, 0x084000, true, "PBSP", "bsp", &priv); *pobject = nv_object(priv); if (ret) @@ -86,8 +102,10 @@ nv98_bsp_oclass = { .handle = NV_ENGINE(BSP, 0x98), .ofuncs = &(struct nouveau_ofuncs) { .ctor = nv98_bsp_ctor, - .dtor = _nouveau_engine_dtor, - .init = _nouveau_engine_init, - .fini = _nouveau_engine_fini, + .dtor = _nouveau_falcon_dtor, + .init = nv98_bsp_init, + .fini = _nouveau_falcon_fini, + .rd32 = _nouveau_falcon_rd32, + .wr32 = _nouveau_falcon_wr32, }, }; diff --git a/drivers/gpu/drm/nouveau/core/engine/disp/dacnv50.c b/drivers/gpu/drm/nouveau/core/engine/disp/dacnv50.c index f02fd9f443f..a66b27c0fca 100644 --- a/drivers/gpu/drm/nouveau/core/engine/disp/dacnv50.c +++ b/drivers/gpu/drm/nouveau/core/engine/disp/dacnv50.c @@ -49,18 +49,23 @@ int nv50_dac_sense(struct nv50_disp_priv *priv, int or, u32 loadval) { const u32 doff = (or * 0x800); - int load = -EINVAL; + nv_mask(priv, 0x61a004 + doff, 0x807f0000, 0x80150000); nv_wait(priv, 0x61a004 + doff, 0x80000000, 0x00000000); + nv_wr32(priv, 0x61a00c + doff, 0x00100000 | loadval); mdelay(9); udelay(500); - nv_wr32(priv, 0x61a00c + doff, 0x80000000); - load = (nv_rd32(priv, 0x61a00c + doff) & 0x38000000) >> 27; - nv_wr32(priv, 0x61a00c + doff, 0x00000000); + loadval = nv_mask(priv, 0x61a00c + doff, 0xffffffff, 0x00000000); + nv_mask(priv, 0x61a004 + doff, 0x807f0000, 0x80550000); nv_wait(priv, 0x61a004 + doff, 0x80000000, 0x00000000); - return load; + + nv_debug(priv, "DAC%d sense: 0x%08x\n", or, loadval); + if (!(loadval & 0x80000000)) + return -ETIMEDOUT; + + return (loadval & 0x38000000) >> 27; } int diff --git a/drivers/gpu/drm/nouveau/core/engine/disp/nv50.c b/drivers/gpu/drm/nouveau/core/engine/disp/nv50.c index 7ffe2f309f1..c168ae3eaa9 100644 --- a/drivers/gpu/drm/nouveau/core/engine/disp/nv50.c +++ b/drivers/gpu/drm/nouveau/core/engine/disp/nv50.c @@ -628,7 +628,7 @@ nv50_disp_base_init(struct nouveau_object *object) } /* ... PIOR caps */ - for (i = 0; i < 3; i++) { + for (i = 0; i < priv->pior.nr; i++) { tmp = nv_rd32(priv, 0x61e000 + (i * 0x800)); nv_wr32(priv, 0x6101f0 + (i * 0x04), tmp); } @@ -834,10 +834,11 @@ exec_script(struct nv50_disp_priv *priv, int head, int id) u8 ver, hdr, cnt, len; u16 data; u32 ctrl = 0x00000000; + u32 reg; int i; /* DAC */ - for (i = 0; !(ctrl & (1 << head)) && i < 3; i++) + for (i = 0; !(ctrl & (1 << head)) && i < priv->dac.nr; i++) ctrl = nv_rd32(priv, 0x610b5c + (i * 8)); /* SOR */ @@ -845,19 +846,18 @@ exec_script(struct nv50_disp_priv *priv, int head, int id) if (nv_device(priv)->chipset < 0x90 || nv_device(priv)->chipset == 0x92 || nv_device(priv)->chipset == 0xa0) { - for (i = 0; !(ctrl & (1 << head)) && i < 2; i++) - ctrl = nv_rd32(priv, 0x610b74 + (i * 8)); - i += 4; + reg = 0x610b74; } else { - for (i = 0; !(ctrl & (1 << head)) && i < 4; i++) - ctrl = nv_rd32(priv, 0x610798 + (i * 8)); - i += 4; + reg = 0x610798; } + for (i = 0; !(ctrl & (1 << head)) && i < priv->sor.nr; i++) + ctrl = nv_rd32(priv, reg + (i * 8)); + i += 4; } /* PIOR */ if (!(ctrl & (1 << head))) { - for (i = 0; !(ctrl & (1 << head)) && i < 3; i++) + for (i = 0; !(ctrl & (1 << head)) && i < priv->pior.nr; i++) ctrl = nv_rd32(priv, 0x610b84 + (i * 8)); i += 8; } @@ -893,10 +893,11 @@ exec_clkcmp(struct nv50_disp_priv *priv, int head, int id, u32 pclk, u8 ver, hdr, cnt, len; u32 ctrl = 0x00000000; u32 data, conf = ~0; + u32 reg; int i; /* DAC */ - for (i = 0; !(ctrl & (1 << head)) && i < 3; i++) + for (i = 0; !(ctrl & (1 << head)) && i < priv->dac.nr; i++) ctrl = nv_rd32(priv, 0x610b58 + (i * 8)); /* SOR */ @@ -904,19 +905,18 @@ exec_clkcmp(struct nv50_disp_priv *priv, int head, int id, u32 pclk, if (nv_device(priv)->chipset < 0x90 || nv_device(priv)->chipset == 0x92 || nv_device(priv)->chipset == 0xa0) { - for (i = 0; !(ctrl & (1 << head)) && i < 2; i++) - ctrl = nv_rd32(priv, 0x610b70 + (i * 8)); - i += 4; + reg = 0x610b70; } else { - for (i = 0; !(ctrl & (1 << head)) && i < 4; i++) - ctrl = nv_rd32(priv, 0x610794 + (i * 8)); - i += 4; + reg = 0x610794; } + for (i = 0; !(ctrl & (1 << head)) && i < priv->sor.nr; i++) + ctrl = nv_rd32(priv, reg + (i * 8)); + i += 4; } /* PIOR */ if (!(ctrl & (1 << head))) { - for (i = 0; !(ctrl & (1 << head)) && i < 3; i++) + for (i = 0; !(ctrl & (1 << head)) && i < priv->pior.nr; i++) ctrl = nv_rd32(priv, 0x610b80 + (i * 8)); i += 8; } diff --git a/drivers/gpu/drm/nouveau/core/engine/fifo/nv50.c b/drivers/gpu/drm/nouveau/core/engine/fifo/nv50.c index e9b8217d007..7e5dff51d3c 100644 --- a/drivers/gpu/drm/nouveau/core/engine/fifo/nv50.c +++ b/drivers/gpu/drm/nouveau/core/engine/fifo/nv50.c @@ -26,7 +26,6 @@ #include <core/engctx.h> #include <core/ramht.h> #include <core/class.h> -#include <core/math.h> #include <subdev/timer.h> #include <subdev/bar.h> @@ -278,7 +277,7 @@ nv50_fifo_chan_ctor_ind(struct nouveau_object *parent, return ret; ioffset = args->ioffset; - ilength = log2i(args->ilength / 8); + ilength = order_base_2(args->ilength / 8); nv_wo32(base->ramfc, 0x3c, 0x403f6078); nv_wo32(base->ramfc, 0x44, 0x01003fff); diff --git a/drivers/gpu/drm/nouveau/core/engine/fifo/nv84.c b/drivers/gpu/drm/nouveau/core/engine/fifo/nv84.c index 7f53196cff5..91a87cd7195 100644 --- a/drivers/gpu/drm/nouveau/core/engine/fifo/nv84.c +++ b/drivers/gpu/drm/nouveau/core/engine/fifo/nv84.c @@ -28,7 +28,6 @@ #include <core/ramht.h> #include <core/event.h> #include <core/class.h> -#include <core/math.h> #include <subdev/timer.h> #include <subdev/bar.h> @@ -57,6 +56,7 @@ nv84_fifo_context_attach(struct nouveau_object *parent, case NVDEV_ENGINE_SW : return 0; case NVDEV_ENGINE_GR : addr = 0x0020; break; case NVDEV_ENGINE_VP : addr = 0x0040; break; + case NVDEV_ENGINE_PPP : case NVDEV_ENGINE_MPEG : addr = 0x0060; break; case NVDEV_ENGINE_BSP : addr = 0x0080; break; case NVDEV_ENGINE_CRYPT: addr = 0x00a0; break; @@ -92,6 +92,7 @@ nv84_fifo_context_detach(struct nouveau_object *parent, bool suspend, case NVDEV_ENGINE_SW : return 0; case NVDEV_ENGINE_GR : engn = 0; addr = 0x0020; break; case NVDEV_ENGINE_VP : engn = 3; addr = 0x0040; break; + case NVDEV_ENGINE_PPP : case NVDEV_ENGINE_MPEG : engn = 1; addr = 0x0060; break; case NVDEV_ENGINE_BSP : engn = 5; addr = 0x0080; break; case NVDEV_ENGINE_CRYPT: engn = 4; addr = 0x00a0; break; @@ -258,7 +259,7 @@ nv84_fifo_chan_ctor_ind(struct nouveau_object *parent, nv_parent(chan)->object_detach = nv50_fifo_object_detach; ioffset = args->ioffset; - ilength = log2i(args->ilength / 8); + ilength = order_base_2(args->ilength / 8); nv_wo32(base->ramfc, 0x3c, 0x403f6078); nv_wo32(base->ramfc, 0x44, 0x01003fff); diff --git a/drivers/gpu/drm/nouveau/core/engine/fifo/nvc0.c b/drivers/gpu/drm/nouveau/core/engine/fifo/nvc0.c index 46dfa68c47b..ce92f289e75 100644 --- a/drivers/gpu/drm/nouveau/core/engine/fifo/nvc0.c +++ b/drivers/gpu/drm/nouveau/core/engine/fifo/nvc0.c @@ -29,7 +29,6 @@ #include <core/engctx.h> #include <core/event.h> #include <core/class.h> -#include <core/math.h> #include <core/enum.h> #include <subdev/timer.h> @@ -200,7 +199,7 @@ nvc0_fifo_chan_ctor(struct nouveau_object *parent, usermem = chan->base.chid * 0x1000; ioffset = args->ioffset; - ilength = log2i(args->ilength / 8); + ilength = order_base_2(args->ilength / 8); for (i = 0; i < 0x1000; i += 4) nv_wo32(priv->user.mem, usermem + i, 0x00000000); diff --git a/drivers/gpu/drm/nouveau/core/engine/fifo/nve0.c b/drivers/gpu/drm/nouveau/core/engine/fifo/nve0.c index 09644fa9602..8e8121abe31 100644 --- a/drivers/gpu/drm/nouveau/core/engine/fifo/nve0.c +++ b/drivers/gpu/drm/nouveau/core/engine/fifo/nve0.c @@ -29,7 +29,6 @@ #include <core/engctx.h> #include <core/event.h> #include <core/class.h> -#include <core/math.h> #include <core/enum.h> #include <subdev/timer.h> @@ -240,7 +239,7 @@ nve0_fifo_chan_ctor(struct nouveau_object *parent, usermem = chan->base.chid * 0x200; ioffset = args->ioffset; - ilength = log2i(args->ilength / 8); + ilength = order_base_2(args->ilength / 8); for (i = 0; i < 0x200; i += 4) nv_wo32(priv->user.mem, usermem + i, 0x00000000); diff --git a/drivers/gpu/drm/nouveau/core/engine/graph/nv40.h b/drivers/gpu/drm/nouveau/core/engine/graph/nv40.h index 7da35a4e797..ad820937752 100644 --- a/drivers/gpu/drm/nouveau/core/engine/graph/nv40.h +++ b/drivers/gpu/drm/nouveau/core/engine/graph/nv40.h @@ -1,6 +1,9 @@ #ifndef __NV40_GRAPH_H__ #define __NV40_GRAPH_H__ +#include <core/device.h> +#include <core/gpuobj.h> + /* returns 1 if device is one of the nv4x using the 0x4497 object class, * helpful to determine a number of other hardware features */ diff --git a/drivers/gpu/drm/nouveau/core/engine/ppp/nv98.c b/drivers/gpu/drm/nouveau/core/engine/ppp/nv98.c index 5a5b2a773ed..13bf31c40aa 100644 --- a/drivers/gpu/drm/nouveau/core/engine/ppp/nv98.c +++ b/drivers/gpu/drm/nouveau/core/engine/ppp/nv98.c @@ -19,21 +19,14 @@ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR * OTHER DEALINGS IN THE SOFTWARE. * - * Authors: Ben Skeggs + * Authors: Ben Skeggs, Maarten Lankhorst, Ilia Mirkin */ -#include <core/engine.h> -#include <core/engctx.h> -#include <core/class.h> - +#include <engine/falcon.h> #include <engine/ppp.h> struct nv98_ppp_priv { - struct nouveau_engine base; -}; - -struct nv98_ppp_chan { - struct nouveau_engctx base; + struct nouveau_falcon base; }; /******************************************************************************* @@ -42,6 +35,8 @@ struct nv98_ppp_chan { static struct nouveau_oclass nv98_ppp_sclass[] = { + { 0x88b3, &nouveau_object_ofuncs }, + { 0x85b3, &nouveau_object_ofuncs }, {}, }; @@ -53,12 +48,12 @@ static struct nouveau_oclass nv98_ppp_cclass = { .handle = NV_ENGCTX(PPP, 0x98), .ofuncs = &(struct nouveau_ofuncs) { - .ctor = _nouveau_engctx_ctor, - .dtor = _nouveau_engctx_dtor, - .init = _nouveau_engctx_init, - .fini = _nouveau_engctx_fini, - .rd32 = _nouveau_engctx_rd32, - .wr32 = _nouveau_engctx_wr32, + .ctor = _nouveau_falcon_context_ctor, + .dtor = _nouveau_falcon_context_dtor, + .init = _nouveau_falcon_context_init, + .fini = _nouveau_falcon_context_fini, + .rd32 = _nouveau_falcon_context_rd32, + .wr32 = _nouveau_falcon_context_wr32, }, }; @@ -67,6 +62,21 @@ nv98_ppp_cclass = { ******************************************************************************/ static int +nv98_ppp_init(struct nouveau_object *object) +{ + struct nv98_ppp_priv *priv = (void *)object; + int ret; + + ret = nouveau_falcon_init(&priv->base); + if (ret) + return ret; + + nv_wr32(priv, 0x086010, 0x0000ffd2); + nv_wr32(priv, 0x08601c, 0x0000fff2); + return 0; +} + +static int nv98_ppp_ctor(struct nouveau_object *parent, struct nouveau_object *engine, struct nouveau_oclass *oclass, void *data, u32 size, struct nouveau_object **pobject) @@ -74,7 +84,7 @@ nv98_ppp_ctor(struct nouveau_object *parent, struct nouveau_object *engine, struct nv98_ppp_priv *priv; int ret; - ret = nouveau_engine_create(parent, engine, oclass, true, + ret = nouveau_falcon_create(parent, engine, oclass, 0x086000, true, "PPPP", "ppp", &priv); *pobject = nv_object(priv); if (ret) @@ -91,8 +101,10 @@ nv98_ppp_oclass = { .handle = NV_ENGINE(PPP, 0x98), .ofuncs = &(struct nouveau_ofuncs) { .ctor = nv98_ppp_ctor, - .dtor = _nouveau_engine_dtor, - .init = _nouveau_engine_init, - .fini = _nouveau_engine_fini, + .dtor = _nouveau_falcon_dtor, + .init = nv98_ppp_init, + .fini = _nouveau_falcon_fini, + .rd32 = _nouveau_falcon_rd32, + .wr32 = _nouveau_falcon_wr32, }, }; diff --git a/drivers/gpu/drm/nouveau/core/engine/vp/nv98.c b/drivers/gpu/drm/nouveau/core/engine/vp/nv98.c index 8a8236bc84d..fc9ae0ff1ef 100644 --- a/drivers/gpu/drm/nouveau/core/engine/vp/nv98.c +++ b/drivers/gpu/drm/nouveau/core/engine/vp/nv98.c @@ -19,16 +19,14 @@ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR * OTHER DEALINGS IN THE SOFTWARE. * - * Authors: Ben Skeggs + * Authors: Ben Skeggs, Maarten Lankhorst, Ilia Mirkin */ -#include <core/engctx.h> -#include <core/class.h> - +#include <engine/falcon.h> #include <engine/vp.h> struct nv98_vp_priv { - struct nouveau_engine base; + struct nouveau_falcon base; }; /******************************************************************************* @@ -37,6 +35,8 @@ struct nv98_vp_priv { static struct nouveau_oclass nv98_vp_sclass[] = { + { 0x88b2, &nouveau_object_ofuncs }, + { 0x85b2, &nouveau_object_ofuncs }, {}, }; @@ -48,12 +48,12 @@ static struct nouveau_oclass nv98_vp_cclass = { .handle = NV_ENGCTX(VP, 0x98), .ofuncs = &(struct nouveau_ofuncs) { - .ctor = _nouveau_engctx_ctor, - .dtor = _nouveau_engctx_dtor, - .init = _nouveau_engctx_init, - .fini = _nouveau_engctx_fini, - .rd32 = _nouveau_engctx_rd32, - .wr32 = _nouveau_engctx_wr32, + .ctor = _nouveau_falcon_context_ctor, + .dtor = _nouveau_falcon_context_dtor, + .init = _nouveau_falcon_context_init, + .fini = _nouveau_falcon_context_fini, + .rd32 = _nouveau_falcon_context_rd32, + .wr32 = _nouveau_falcon_context_wr32, }, }; @@ -62,6 +62,21 @@ nv98_vp_cclass = { ******************************************************************************/ static int +nv98_vp_init(struct nouveau_object *object) +{ + struct nv98_vp_priv *priv = (void *)object; + int ret; + + ret = nouveau_falcon_init(&priv->base); + if (ret) + return ret; + + nv_wr32(priv, 0x085010, 0x0000ffd2); + nv_wr32(priv, 0x08501c, 0x0000fff2); + return 0; +} + +static int nv98_vp_ctor(struct nouveau_object *parent, struct nouveau_object *engine, struct nouveau_oclass *oclass, void *data, u32 size, struct nouveau_object **pobject) @@ -69,7 +84,7 @@ nv98_vp_ctor(struct nouveau_object *parent, struct nouveau_object *engine, struct nv98_vp_priv *priv; int ret; - ret = nouveau_engine_create(parent, engine, oclass, true, + ret = nouveau_falcon_create(parent, engine, oclass, 0x085000, true, "PVP", "vp", &priv); *pobject = nv_object(priv); if (ret) @@ -86,8 +101,10 @@ nv98_vp_oclass = { .handle = NV_ENGINE(VP, 0x98), .ofuncs = &(struct nouveau_ofuncs) { .ctor = nv98_vp_ctor, - .dtor = _nouveau_engine_dtor, - .init = _nouveau_engine_init, - .fini = _nouveau_engine_fini, + .dtor = _nouveau_falcon_dtor, + .init = nv98_vp_init, + .fini = _nouveau_falcon_fini, + .rd32 = _nouveau_falcon_rd32, + .wr32 = _nouveau_falcon_wr32, }, }; diff --git a/drivers/gpu/drm/nouveau/core/include/core/math.h b/drivers/gpu/drm/nouveau/core/include/core/math.h deleted file mode 100644 index f808131c5cd..00000000000 --- a/drivers/gpu/drm/nouveau/core/include/core/math.h +++ /dev/null @@ -1,16 +0,0 @@ -#ifndef __NOUVEAU_MATH_H__ -#define __NOUVEAU_MATH_H__ - -static inline int -log2i(u64 base) -{ - u64 temp = base >> 1; - int log2; - - for (log2 = 0; temp; log2++, temp >>= 1) { - } - - return (base & (base - 1)) ? log2 + 1: log2; -} - -#endif diff --git a/drivers/gpu/drm/nouveau/core/include/subdev/i2c.h b/drivers/gpu/drm/nouveau/core/include/subdev/i2c.h index 888384c0bed..7e4e2775f24 100644 --- a/drivers/gpu/drm/nouveau/core/include/subdev/i2c.h +++ b/drivers/gpu/drm/nouveau/core/include/subdev/i2c.h @@ -39,8 +39,8 @@ struct nouveau_i2c_func { int (*drv_ctl)(struct nouveau_i2c_port *, int lane, int sw, int pe); }; -#define nouveau_i2c_port_create(p,e,o,i,a,d) \ - nouveau_i2c_port_create_((p), (e), (o), (i), (a), \ +#define nouveau_i2c_port_create(p,e,o,i,a,f,d) \ + nouveau_i2c_port_create_((p), (e), (o), (i), (a), (f), \ sizeof(**d), (void **)d) #define nouveau_i2c_port_destroy(p) ({ \ struct nouveau_i2c_port *port = (p); \ @@ -53,7 +53,9 @@ struct nouveau_i2c_func { int nouveau_i2c_port_create_(struct nouveau_object *, struct nouveau_object *, struct nouveau_oclass *, u8, - const struct i2c_algorithm *, int, void **); + const struct i2c_algorithm *, + const struct nouveau_i2c_func *, + int, void **); void _nouveau_i2c_port_dtor(struct nouveau_object *); #define _nouveau_i2c_port_init nouveau_object_init #define _nouveau_i2c_port_fini nouveau_object_fini diff --git a/drivers/gpu/drm/nouveau/core/include/subdev/mc.h b/drivers/gpu/drm/nouveau/core/include/subdev/mc.h index 9d2cd200625..ce6569f365a 100644 --- a/drivers/gpu/drm/nouveau/core/include/subdev/mc.h +++ b/drivers/gpu/drm/nouveau/core/include/subdev/mc.h @@ -12,6 +12,7 @@ struct nouveau_mc_intr { struct nouveau_mc { struct nouveau_subdev base; const struct nouveau_mc_intr *intr_map; + bool use_msi; }; static inline struct nouveau_mc * diff --git a/drivers/gpu/drm/nouveau/core/include/subdev/timer.h b/drivers/gpu/drm/nouveau/core/include/subdev/timer.h index e465d158d35..9ab70dfe5b0 100644 --- a/drivers/gpu/drm/nouveau/core/include/subdev/timer.h +++ b/drivers/gpu/drm/nouveau/core/include/subdev/timer.h @@ -22,6 +22,7 @@ bool nouveau_timer_wait_eq(void *, u64 nsec, u32 addr, u32 mask, u32 data); bool nouveau_timer_wait_ne(void *, u64 nsec, u32 addr, u32 mask, u32 data); bool nouveau_timer_wait_cb(void *, u64 nsec, bool (*func)(void *), void *data); void nouveau_timer_alarm(void *, u32 nsec, struct nouveau_alarm *); +void nouveau_timer_alarm_cancel(void *, struct nouveau_alarm *); #define NV_WAIT_DEFAULT 2000000000ULL #define nv_wait(o,a,m,v) \ @@ -35,6 +36,7 @@ struct nouveau_timer { struct nouveau_subdev base; u64 (*read)(struct nouveau_timer *); void (*alarm)(struct nouveau_timer *, u64 time, struct nouveau_alarm *); + void (*alarm_cancel)(struct nouveau_timer *, struct nouveau_alarm *); }; static inline struct nouveau_timer * diff --git a/drivers/gpu/drm/nouveau/core/os.h b/drivers/gpu/drm/nouveau/core/os.h index 3bd9be2ab37..191e739f30d 100644 --- a/drivers/gpu/drm/nouveau/core/os.h +++ b/drivers/gpu/drm/nouveau/core/os.h @@ -13,11 +13,13 @@ #include <linux/i2c-algo-bit.h> #include <linux/delay.h> #include <linux/io-mapping.h> -#include <linux/vmalloc.h> #include <linux/acpi.h> +#include <linux/vmalloc.h> #include <linux/dmi.h> #include <linux/reboot.h> #include <linux/interrupt.h> +#include <linux/log2.h> +#include <linux/pm_runtime.h> #include <asm/unaligned.h> diff --git a/drivers/gpu/drm/nouveau/core/subdev/bios/init.c b/drivers/gpu/drm/nouveau/core/subdev/bios/init.c index 2e11ea02cf8..57cda2a1437 100644 --- a/drivers/gpu/drm/nouveau/core/subdev/bios/init.c +++ b/drivers/gpu/drm/nouveau/core/subdev/bios/init.c @@ -579,8 +579,22 @@ static void init_reserved(struct nvbios_init *init) { u8 opcode = nv_ro08(init->bios, init->offset); - trace("RESERVED\t0x%02x\n", opcode); - init->offset += 1; + u8 length, i; + + switch (opcode) { + case 0xaa: + length = 4; + break; + default: + length = 1; + break; + } + + trace("RESERVED 0x%02x\t", opcode); + for (i = 1; i < length; i++) + cont(" 0x%02x", nv_ro08(init->bios, init->offset + i)); + cont("\n"); + init->offset += length; } /** @@ -1437,7 +1451,7 @@ init_configure_mem(struct nvbios_init *init) data = init_rdvgai(init, 0x03c4, 0x01); init_wrvgai(init, 0x03c4, 0x01, data | 0x20); - while ((addr = nv_ro32(bios, sdata)) != 0xffffffff) { + for (; (addr = nv_ro32(bios, sdata)) != 0xffffffff; sdata += 4) { switch (addr) { case 0x10021c: /* CKE_NORMAL */ case 0x1002d0: /* CMD_REFRESH */ @@ -2135,6 +2149,7 @@ static struct nvbios_init_opcode { [0x99] = { init_zm_auxch }, [0x9a] = { init_i2c_long_if }, [0xa9] = { init_gpio_ne }, + [0xaa] = { init_reserved }, }; #define init_opcode_nr (sizeof(init_opcode) / sizeof(init_opcode[0])) diff --git a/drivers/gpu/drm/nouveau/core/subdev/bios/therm.c b/drivers/gpu/drm/nouveau/core/subdev/bios/therm.c index 22a20573ed1..22ac6dbd6c8 100644 --- a/drivers/gpu/drm/nouveau/core/subdev/bios/therm.c +++ b/drivers/gpu/drm/nouveau/core/subdev/bios/therm.c @@ -184,7 +184,8 @@ nvbios_therm_fan_parse(struct nouveau_bios *bios, cur_trip->fan_duty = value; break; case 0x26: - fan->pwm_freq = value; + if (!fan->pwm_freq) + fan->pwm_freq = value; break; case 0x3b: fan->bump_period = value; diff --git a/drivers/gpu/drm/nouveau/core/subdev/i2c/anx9805.c b/drivers/gpu/drm/nouveau/core/subdev/i2c/anx9805.c index dec94e9d776..4b195ac4da6 100644 --- a/drivers/gpu/drm/nouveau/core/subdev/i2c/anx9805.c +++ b/drivers/gpu/drm/nouveau/core/subdev/i2c/anx9805.c @@ -118,7 +118,8 @@ anx9805_aux_chan_ctor(struct nouveau_object *parent, int ret; ret = nouveau_i2c_port_create(parent, engine, oclass, index, - &nouveau_i2c_aux_algo, &chan); + &nouveau_i2c_aux_algo, &anx9805_aux_func, + &chan); *pobject = nv_object(chan); if (ret) return ret; @@ -140,8 +141,6 @@ anx9805_aux_chan_ctor(struct nouveau_object *parent, struct i2c_algo_bit_data *algo = mast->adapter.algo_data; algo->udelay = max(algo->udelay, 40); } - - chan->base.func = &anx9805_aux_func; return 0; } @@ -234,7 +233,8 @@ anx9805_ddc_port_ctor(struct nouveau_object *parent, int ret; ret = nouveau_i2c_port_create(parent, engine, oclass, index, - &anx9805_i2c_algo, &port); + &anx9805_i2c_algo, &anx9805_i2c_func, + &port); *pobject = nv_object(port); if (ret) return ret; @@ -256,8 +256,6 @@ anx9805_ddc_port_ctor(struct nouveau_object *parent, struct i2c_algo_bit_data *algo = mast->adapter.algo_data; algo->udelay = max(algo->udelay, 40); } - - port->base.func = &anx9805_i2c_func; return 0; } diff --git a/drivers/gpu/drm/nouveau/core/subdev/i2c/base.c b/drivers/gpu/drm/nouveau/core/subdev/i2c/base.c index 8ae2625415e..2895c19bb15 100644 --- a/drivers/gpu/drm/nouveau/core/subdev/i2c/base.c +++ b/drivers/gpu/drm/nouveau/core/subdev/i2c/base.c @@ -95,6 +95,7 @@ nouveau_i2c_port_create_(struct nouveau_object *parent, struct nouveau_object *engine, struct nouveau_oclass *oclass, u8 index, const struct i2c_algorithm *algo, + const struct nouveau_i2c_func *func, int size, void **pobject) { struct nouveau_device *device = nv_device(parent); @@ -112,6 +113,7 @@ nouveau_i2c_port_create_(struct nouveau_object *parent, port->adapter.owner = THIS_MODULE; port->adapter.dev.parent = &device->pdev->dev; port->index = index; + port->func = func; i2c_set_adapdata(&port->adapter, i2c); if ( algo == &nouveau_i2c_bit_algo && diff --git a/drivers/gpu/drm/nouveau/core/subdev/i2c/nv04.c b/drivers/gpu/drm/nouveau/core/subdev/i2c/nv04.c index 2ad18840fe6..860d5d2365d 100644 --- a/drivers/gpu/drm/nouveau/core/subdev/i2c/nv04.c +++ b/drivers/gpu/drm/nouveau/core/subdev/i2c/nv04.c @@ -91,12 +91,12 @@ nv04_i2c_port_ctor(struct nouveau_object *parent, struct nouveau_object *engine, int ret; ret = nouveau_i2c_port_create(parent, engine, oclass, index, - &nouveau_i2c_bit_algo, &port); + &nouveau_i2c_bit_algo, &nv04_i2c_func, + &port); *pobject = nv_object(port); if (ret) return ret; - port->base.func = &nv04_i2c_func; port->drive = info->drive; port->sense = info->sense; return 0; diff --git a/drivers/gpu/drm/nouveau/core/subdev/i2c/nv4e.c b/drivers/gpu/drm/nouveau/core/subdev/i2c/nv4e.c index f501ae25dbb..0c2655a03bb 100644 --- a/drivers/gpu/drm/nouveau/core/subdev/i2c/nv4e.c +++ b/drivers/gpu/drm/nouveau/core/subdev/i2c/nv4e.c @@ -84,12 +84,12 @@ nv4e_i2c_port_ctor(struct nouveau_object *parent, struct nouveau_object *engine, int ret; ret = nouveau_i2c_port_create(parent, engine, oclass, index, - &nouveau_i2c_bit_algo, &port); + &nouveau_i2c_bit_algo, &nv4e_i2c_func, + &port); *pobject = nv_object(port); if (ret) return ret; - port->base.func = &nv4e_i2c_func; port->addr = 0x600800 + info->drive; return 0; } diff --git a/drivers/gpu/drm/nouveau/core/subdev/i2c/nv50.c b/drivers/gpu/drm/nouveau/core/subdev/i2c/nv50.c index 378dfa324e5..a8d67a28770 100644 --- a/drivers/gpu/drm/nouveau/core/subdev/i2c/nv50.c +++ b/drivers/gpu/drm/nouveau/core/subdev/i2c/nv50.c @@ -85,7 +85,8 @@ nv50_i2c_port_ctor(struct nouveau_object *parent, struct nouveau_object *engine, int ret; ret = nouveau_i2c_port_create(parent, engine, oclass, index, - &nouveau_i2c_bit_algo, &port); + &nouveau_i2c_bit_algo, &nv50_i2c_func, + &port); *pobject = nv_object(port); if (ret) return ret; @@ -93,7 +94,6 @@ nv50_i2c_port_ctor(struct nouveau_object *parent, struct nouveau_object *engine, if (info->drive >= nv50_i2c_addr_nr) return -EINVAL; - port->base.func = &nv50_i2c_func; port->state = 0x00000007; port->addr = nv50_i2c_addr[info->drive]; return 0; diff --git a/drivers/gpu/drm/nouveau/core/subdev/i2c/nv94.c b/drivers/gpu/drm/nouveau/core/subdev/i2c/nv94.c index 61b771670bf..df6d3e4b68b 100644 --- a/drivers/gpu/drm/nouveau/core/subdev/i2c/nv94.c +++ b/drivers/gpu/drm/nouveau/core/subdev/i2c/nv94.c @@ -186,7 +186,8 @@ nv94_i2c_port_ctor(struct nouveau_object *parent, struct nouveau_object *engine, int ret; ret = nouveau_i2c_port_create(parent, engine, oclass, index, - &nouveau_i2c_bit_algo, &port); + &nouveau_i2c_bit_algo, &nv94_i2c_func, + &port); *pobject = nv_object(port); if (ret) return ret; @@ -194,7 +195,6 @@ nv94_i2c_port_ctor(struct nouveau_object *parent, struct nouveau_object *engine, if (info->drive >= nv50_i2c_addr_nr) return -EINVAL; - port->base.func = &nv94_i2c_func; port->state = 7; port->addr = nv50_i2c_addr[info->drive]; if (info->share != DCB_I2C_UNUSED) { @@ -221,12 +221,12 @@ nv94_aux_port_ctor(struct nouveau_object *parent, struct nouveau_object *engine, int ret; ret = nouveau_i2c_port_create(parent, engine, oclass, index, - &nouveau_i2c_aux_algo, &port); + &nouveau_i2c_aux_algo, &nv94_aux_func, + &port); *pobject = nv_object(port); if (ret) return ret; - port->base.func = &nv94_aux_func; port->addr = info->drive; if (info->share != DCB_I2C_UNUSED) { port->ctrl = 0x00e500 + (info->drive * 0x50); diff --git a/drivers/gpu/drm/nouveau/core/subdev/i2c/nvd0.c b/drivers/gpu/drm/nouveau/core/subdev/i2c/nvd0.c index f761b8a610f..29967d30f97 100644 --- a/drivers/gpu/drm/nouveau/core/subdev/i2c/nvd0.c +++ b/drivers/gpu/drm/nouveau/core/subdev/i2c/nvd0.c @@ -60,12 +60,12 @@ nvd0_i2c_port_ctor(struct nouveau_object *parent, struct nouveau_object *engine, int ret; ret = nouveau_i2c_port_create(parent, engine, oclass, index, - &nouveau_i2c_bit_algo, &port); + &nouveau_i2c_bit_algo, &nvd0_i2c_func, + &port); *pobject = nv_object(port); if (ret) return ret; - port->base.func = &nvd0_i2c_func; port->state = 0x00000007; port->addr = 0x00d014 + (info->drive * 0x20); if (info->share != DCB_I2C_UNUSED) { diff --git a/drivers/gpu/drm/nouveau/core/subdev/instmem/nv40.c b/drivers/gpu/drm/nouveau/core/subdev/instmem/nv40.c index 716bf41bc3c..b10a143787a 100644 --- a/drivers/gpu/drm/nouveau/core/subdev/instmem/nv40.c +++ b/drivers/gpu/drm/nouveau/core/subdev/instmem/nv40.c @@ -22,15 +22,9 @@ * Authors: Ben Skeggs */ -#include "nv04.h" +#include <engine/graph/nv40.h> -static inline int -nv44_graph_class(struct nv04_instmem_priv *priv) -{ - if ((nv_device(priv)->chipset & 0xf0) == 0x60) - return 1; - return !(0x0baf & (1 << (nv_device(priv)->chipset & 0x0f))); -} +#include "nv04.h" static int nv40_instmem_ctor(struct nouveau_object *parent, struct nouveau_object *engine, diff --git a/drivers/gpu/drm/nouveau/core/subdev/mc/base.c b/drivers/gpu/drm/nouveau/core/subdev/mc/base.c index 20f9a538746..37712a6df92 100644 --- a/drivers/gpu/drm/nouveau/core/subdev/mc/base.c +++ b/drivers/gpu/drm/nouveau/core/subdev/mc/base.c @@ -23,7 +23,7 @@ */ #include <subdev/mc.h> -#include <linux/pm_runtime.h> +#include <core/option.h> static irqreturn_t nouveau_mc_intr(int irq, void *arg) @@ -47,6 +47,9 @@ nouveau_mc_intr(int irq, void *arg) map++; } + if (pmc->use_msi) + nv_wr08(pmc->base.base.parent, 0x00088068, 0xff); + if (intr) { nv_error(pmc, "unknown intr 0x%08x\n", stat); } @@ -81,6 +84,8 @@ _nouveau_mc_dtor(struct nouveau_object *object) struct nouveau_device *device = nv_device(object); struct nouveau_mc *pmc = (void *)object; free_irq(device->pdev->irq, pmc); + if (pmc->use_msi) + pci_disable_msi(device->pdev); nouveau_subdev_destroy(&pmc->base); } @@ -102,6 +107,23 @@ nouveau_mc_create_(struct nouveau_object *parent, struct nouveau_object *engine, pmc->intr_map = intr_map; + switch (device->pdev->device & 0x0ff0) { + case 0x00f0: /* BR02? */ + case 0x02e0: /* BR02? */ + pmc->use_msi = false; + break; + default: + pmc->use_msi = nouveau_boolopt(device->cfgopt, "NvMSI", true); + if (pmc->use_msi) { + pmc->use_msi = pci_enable_msi(device->pdev) == 0; + if (pmc->use_msi) { + nv_info(pmc, "MSI interrupts enabled\n"); + nv_wr08(device, 0x00088068, 0xff); + } + } + break; + } + ret = request_irq(device->pdev->irq, nouveau_mc_intr, IRQF_SHARED, "nouveau", pmc); if (ret < 0) diff --git a/drivers/gpu/drm/nouveau/core/subdev/mc/nv98.c b/drivers/gpu/drm/nouveau/core/subdev/mc/nv98.c index 0d57b4d3e00..06710419a59 100644 --- a/drivers/gpu/drm/nouveau/core/subdev/mc/nv98.c +++ b/drivers/gpu/drm/nouveau/core/subdev/mc/nv98.c @@ -35,6 +35,7 @@ nv98_mc_intr[] = { { 0x00001000, NVDEV_ENGINE_GR }, { 0x00004000, NVDEV_ENGINE_CRYPT }, /* NV84:NVA3 */ { 0x00008000, NVDEV_ENGINE_BSP }, + { 0x00020000, NVDEV_ENGINE_VP }, { 0x00080000, NVDEV_SUBDEV_THERM }, /* NVA3:NVC0 */ { 0x00100000, NVDEV_SUBDEV_TIMER }, { 0x00200000, NVDEV_SUBDEV_GPIO }, @@ -42,7 +43,7 @@ nv98_mc_intr[] = { { 0x04000000, NVDEV_ENGINE_DISP }, { 0x10000000, NVDEV_SUBDEV_BUS }, { 0x80000000, NVDEV_ENGINE_SW }, - { 0x0040d101, NVDEV_SUBDEV_FB }, + { 0x0042d101, NVDEV_SUBDEV_FB }, {}, }; diff --git a/drivers/gpu/drm/nouveau/core/subdev/therm/base.c b/drivers/gpu/drm/nouveau/core/subdev/therm/base.c index a00a5a76e2d..f1de7a9c572 100644 --- a/drivers/gpu/drm/nouveau/core/subdev/therm/base.c +++ b/drivers/gpu/drm/nouveau/core/subdev/therm/base.c @@ -95,12 +95,14 @@ nouveau_therm_update(struct nouveau_therm *therm, int mode) int duty; spin_lock_irqsave(&priv->lock, flags); + nv_debug(therm, "FAN speed check\n"); if (mode < 0) mode = priv->mode; priv->mode = mode; switch (mode) { case NOUVEAU_THERM_CTRL_MANUAL: + ptimer->alarm_cancel(ptimer, &priv->alarm); duty = nouveau_therm_fan_get(therm); if (duty < 0) duty = 100; @@ -113,6 +115,7 @@ nouveau_therm_update(struct nouveau_therm *therm, int mode) break; case NOUVEAU_THERM_CTRL_NONE: default: + ptimer->alarm_cancel(ptimer, &priv->alarm); goto done; } @@ -122,6 +125,8 @@ nouveau_therm_update(struct nouveau_therm *therm, int mode) done: if (list_empty(&priv->alarm.head) && (mode == NOUVEAU_THERM_CTRL_AUTO)) ptimer->alarm(ptimer, 1000000000ULL, &priv->alarm); + else if (!list_empty(&priv->alarm.head)) + nv_debug(therm, "therm fan alarm list is not empty\n"); spin_unlock_irqrestore(&priv->lock, flags); } @@ -267,9 +272,15 @@ _nouveau_therm_init(struct nouveau_object *object) if (ret) return ret; - if (priv->suspend >= 0) - nouveau_therm_fan_mode(therm, priv->mode); - priv->sensor.program_alarms(therm); + if (priv->suspend >= 0) { + /* restore the pwm value only when on manual or auto mode */ + if (priv->suspend > 0) + nouveau_therm_fan_set(therm, true, priv->fan->percent); + + nouveau_therm_fan_mode(therm, priv->suspend); + } + nouveau_therm_sensor_init(therm); + nouveau_therm_fan_init(therm); return 0; } @@ -279,6 +290,8 @@ _nouveau_therm_fini(struct nouveau_object *object, bool suspend) struct nouveau_therm *therm = (void *)object; struct nouveau_therm_priv *priv = (void *)therm; + nouveau_therm_fan_fini(therm, suspend); + nouveau_therm_sensor_fini(therm, suspend); if (suspend) { priv->suspend = priv->mode; priv->mode = NOUVEAU_THERM_CTRL_NONE; diff --git a/drivers/gpu/drm/nouveau/core/subdev/therm/fan.c b/drivers/gpu/drm/nouveau/core/subdev/therm/fan.c index c728380d3d6..39f47b950ad 100644 --- a/drivers/gpu/drm/nouveau/core/subdev/therm/fan.c +++ b/drivers/gpu/drm/nouveau/core/subdev/therm/fan.c @@ -204,6 +204,23 @@ nouveau_therm_fan_safety_checks(struct nouveau_therm *therm) } int +nouveau_therm_fan_init(struct nouveau_therm *therm) +{ + return 0; +} + +int +nouveau_therm_fan_fini(struct nouveau_therm *therm, bool suspend) +{ + struct nouveau_therm_priv *priv = (void *)therm; + struct nouveau_timer *ptimer = nouveau_timer(therm); + + if (suspend) + ptimer->alarm_cancel(ptimer, &priv->fan->alarm); + return 0; +} + +int nouveau_therm_fan_ctor(struct nouveau_therm *therm) { struct nouveau_therm_priv *priv = (void *)therm; @@ -234,6 +251,9 @@ nouveau_therm_fan_ctor(struct nouveau_therm *therm) nv_info(therm, "FAN control: %s\n", priv->fan->type); + /* read the current speed, it is useful when resuming */ + priv->fan->percent = nouveau_therm_fan_get(therm); + /* attempt to detect a tachometer connection */ ret = gpio->find(gpio, 0, DCB_GPIO_FAN_SENSE, 0xff, &priv->fan->tach); if (ret) diff --git a/drivers/gpu/drm/nouveau/core/subdev/therm/priv.h b/drivers/gpu/drm/nouveau/core/subdev/therm/priv.h index 15ca64e481f..dd38529262f 100644 --- a/drivers/gpu/drm/nouveau/core/subdev/therm/priv.h +++ b/drivers/gpu/drm/nouveau/core/subdev/therm/priv.h @@ -113,6 +113,8 @@ void nouveau_therm_ic_ctor(struct nouveau_therm *therm); int nouveau_therm_sensor_ctor(struct nouveau_therm *therm); int nouveau_therm_fan_ctor(struct nouveau_therm *therm); +int nouveau_therm_fan_init(struct nouveau_therm *therm); +int nouveau_therm_fan_fini(struct nouveau_therm *therm, bool suspend); int nouveau_therm_fan_get(struct nouveau_therm *therm); int nouveau_therm_fan_set(struct nouveau_therm *therm, bool now, int percent); int nouveau_therm_fan_user_get(struct nouveau_therm *therm); @@ -122,6 +124,8 @@ int nouveau_therm_fan_sense(struct nouveau_therm *therm); int nouveau_therm_preinit(struct nouveau_therm *); +int nouveau_therm_sensor_init(struct nouveau_therm *therm); +int nouveau_therm_sensor_fini(struct nouveau_therm *therm, bool suspend); void nouveau_therm_sensor_preinit(struct nouveau_therm *); void nouveau_therm_sensor_set_threshold_state(struct nouveau_therm *therm, enum nouveau_therm_thrs thrs, diff --git a/drivers/gpu/drm/nouveau/core/subdev/therm/temp.c b/drivers/gpu/drm/nouveau/core/subdev/therm/temp.c index dde746c78c8..b80a33011b9 100644 --- a/drivers/gpu/drm/nouveau/core/subdev/therm/temp.c +++ b/drivers/gpu/drm/nouveau/core/subdev/therm/temp.c @@ -180,6 +180,8 @@ alarm_timer_callback(struct nouveau_alarm *alarm) spin_lock_irqsave(&priv->sensor.alarm_program_lock, flags); + nv_debug(therm, "polling the internal temperature\n"); + nouveau_therm_threshold_hyst_polling(therm, &sensor->thrs_fan_boost, NOUVEAU_THERM_THRS_FANBOOST); @@ -216,6 +218,25 @@ nouveau_therm_program_alarms_polling(struct nouveau_therm *therm) alarm_timer_callback(&priv->sensor.therm_poll_alarm); } +int +nouveau_therm_sensor_init(struct nouveau_therm *therm) +{ + struct nouveau_therm_priv *priv = (void *)therm; + priv->sensor.program_alarms(therm); + return 0; +} + +int +nouveau_therm_sensor_fini(struct nouveau_therm *therm, bool suspend) +{ + struct nouveau_therm_priv *priv = (void *)therm; + struct nouveau_timer *ptimer = nouveau_timer(therm); + + if (suspend) + ptimer->alarm_cancel(ptimer, &priv->sensor.therm_poll_alarm); + return 0; +} + void nouveau_therm_sensor_preinit(struct nouveau_therm *therm) { diff --git a/drivers/gpu/drm/nouveau/core/subdev/timer/base.c b/drivers/gpu/drm/nouveau/core/subdev/timer/base.c index 5d417cc9949..cf8a0e0f8ee 100644 --- a/drivers/gpu/drm/nouveau/core/subdev/timer/base.c +++ b/drivers/gpu/drm/nouveau/core/subdev/timer/base.c @@ -85,3 +85,10 @@ nouveau_timer_alarm(void *obj, u32 nsec, struct nouveau_alarm *alarm) struct nouveau_timer *ptimer = nouveau_timer(obj); ptimer->alarm(ptimer, nsec, alarm); } + +void +nouveau_timer_alarm_cancel(void *obj, struct nouveau_alarm *alarm) +{ + struct nouveau_timer *ptimer = nouveau_timer(obj); + ptimer->alarm_cancel(ptimer, alarm); +} diff --git a/drivers/gpu/drm/nouveau/core/subdev/timer/nv04.c b/drivers/gpu/drm/nouveau/core/subdev/timer/nv04.c index 9469b827567..57711ecb566 100644 --- a/drivers/gpu/drm/nouveau/core/subdev/timer/nv04.c +++ b/drivers/gpu/drm/nouveau/core/subdev/timer/nv04.c @@ -36,6 +36,7 @@ struct nv04_timer_priv { struct nouveau_timer base; struct list_head alarms; spinlock_t lock; + u64 suspend_time; }; static u64 @@ -113,6 +114,25 @@ nv04_timer_alarm(struct nouveau_timer *ptimer, u64 time, } static void +nv04_timer_alarm_cancel(struct nouveau_timer *ptimer, + struct nouveau_alarm *alarm) +{ + struct nv04_timer_priv *priv = (void *)ptimer; + unsigned long flags; + + /* avoid deleting an entry while the alarm intr is running */ + spin_lock_irqsave(&priv->lock, flags); + + /* delete the alarm from the list */ + list_del(&alarm->head); + + /* reset the head so as list_empty returns 1 */ + INIT_LIST_HEAD(&alarm->head); + + spin_unlock_irqrestore(&priv->lock, flags); +} + +static void nv04_timer_intr(struct nouveau_subdev *subdev) { struct nv04_timer_priv *priv = (void *)subdev; @@ -146,6 +166,8 @@ nv04_timer_ctor(struct nouveau_object *parent, struct nouveau_object *engine, priv->base.base.intr = nv04_timer_intr; priv->base.read = nv04_timer_read; priv->base.alarm = nv04_timer_alarm; + priv->base.alarm_cancel = nv04_timer_alarm_cancel; + priv->suspend_time = 0; INIT_LIST_HEAD(&priv->alarms); spin_lock_init(&priv->lock); @@ -164,7 +186,7 @@ nv04_timer_init(struct nouveau_object *object) { struct nouveau_device *device = nv_device(object); struct nv04_timer_priv *priv = (void *)object; - u32 m = 1, f, n, d; + u32 m = 1, f, n, d, lo, hi; int ret; ret = nouveau_timer_init(&priv->base); @@ -221,16 +243,25 @@ nv04_timer_init(struct nouveau_object *object) d >>= 1; } + /* restore the time before suspend */ + lo = priv->suspend_time; + hi = (priv->suspend_time >> 32); + nv_debug(priv, "input frequency : %dHz\n", f); nv_debug(priv, "input multiplier: %d\n", m); nv_debug(priv, "numerator : 0x%08x\n", n); nv_debug(priv, "denominator : 0x%08x\n", d); nv_debug(priv, "timer frequency : %dHz\n", (f * m) * d / n); + nv_debug(priv, "time low : 0x%08x\n", lo); + nv_debug(priv, "time high : 0x%08x\n", hi); nv_wr32(priv, NV04_PTIMER_NUMERATOR, n); nv_wr32(priv, NV04_PTIMER_DENOMINATOR, d); nv_wr32(priv, NV04_PTIMER_INTR_0, 0xffffffff); nv_wr32(priv, NV04_PTIMER_INTR_EN_0, 0x00000000); + nv_wr32(priv, NV04_PTIMER_TIME_1, hi); + nv_wr32(priv, NV04_PTIMER_TIME_0, lo); + return 0; } @@ -238,6 +269,8 @@ static int nv04_timer_fini(struct nouveau_object *object, bool suspend) { struct nv04_timer_priv *priv = (void *)object; + if (suspend) + priv->suspend_time = nv04_timer_read(&priv->base); nv_wr32(priv, NV04_PTIMER_INTR_EN_0, 0x00000000); return nouveau_timer_fini(&priv->base, suspend); } diff --git a/drivers/gpu/drm/nouveau/core/subdev/vm/nv50.c b/drivers/gpu/drm/nouveau/core/subdev/vm/nv50.c index 07dd1fe2d6f..a4aa81a2173 100644 --- a/drivers/gpu/drm/nouveau/core/subdev/vm/nv50.c +++ b/drivers/gpu/drm/nouveau/core/subdev/vm/nv50.c @@ -174,6 +174,7 @@ nv50_vm_flush(struct nouveau_vm *vm) case NVDEV_ENGINE_GR : vme = 0x00; break; case NVDEV_ENGINE_VP : vme = 0x01; break; case NVDEV_SUBDEV_BAR : vme = 0x06; break; + case NVDEV_ENGINE_PPP : case NVDEV_ENGINE_MPEG : vme = 0x08; break; case NVDEV_ENGINE_BSP : vme = 0x09; break; case NVDEV_ENGINE_CRYPT: vme = 0x0a; break; diff --git a/drivers/gpu/drm/nouveau/nouveau_display.c b/drivers/gpu/drm/nouveau/nouveau_display.c index 77ffded6883..7848590f556 100644 --- a/drivers/gpu/drm/nouveau/nouveau_display.c +++ b/drivers/gpu/drm/nouveau/nouveau_display.c @@ -107,6 +107,11 @@ nouveau_framebuffer_init(struct drm_device *dev, return -EINVAL; } + if (nvbo->tile_flags & NOUVEAU_GEM_TILE_NONCONTIG) { + NV_ERROR(drm, "framebuffer requires contiguous bo\n"); + return -EINVAL; + } + if (nv_device(drm->device)->chipset == 0x50) nv_fb->r_format |= (tile_flags << 8); @@ -273,7 +278,6 @@ nouveau_display_create(struct drm_device *dev) { struct nouveau_drm *drm = nouveau_drm(dev); struct nouveau_display *disp; - u32 pclass = dev->pdev->class >> 8; int ret, gen; disp = drm->display = kzalloc(sizeof(*disp), GFP_KERNEL); @@ -335,29 +339,25 @@ nouveau_display_create(struct drm_device *dev) drm_kms_helper_poll_init(dev); drm_kms_helper_poll_disable(dev); - if (nouveau_modeset == 1 || - (nouveau_modeset < 0 && pclass == PCI_CLASS_DISPLAY_VGA)) { - if (drm->vbios.dcb.entries) { - if (nv_device(drm->device)->card_type < NV_50) - ret = nv04_display_create(dev); - else - ret = nv50_display_create(dev); - } else { - ret = 0; - } - - if (ret) - goto disp_create_err; + if (drm->vbios.dcb.entries) { + if (nv_device(drm->device)->card_type < NV_50) + ret = nv04_display_create(dev); + else + ret = nv50_display_create(dev); + } else { + ret = 0; + } - if (dev->mode_config.num_crtc) { - ret = drm_vblank_init(dev, dev->mode_config.num_crtc); - if (ret) - goto vblank_err; - } + if (ret) + goto disp_create_err; - nouveau_backlight_init(dev); + if (dev->mode_config.num_crtc) { + ret = drm_vblank_init(dev, dev->mode_config.num_crtc); + if (ret) + goto vblank_err; } + nouveau_backlight_init(dev); return 0; vblank_err: diff --git a/drivers/gpu/drm/nouveau/nouveau_drm.c b/drivers/gpu/drm/nouveau/nouveau_drm.c index 8863644024b..e893c536240 100644 --- a/drivers/gpu/drm/nouveau/nouveau_drm.c +++ b/drivers/gpu/drm/nouveau/nouveau_drm.c @@ -636,7 +636,8 @@ int nouveau_pmops_resume(struct device *dev) nouveau_fbcon_set_suspend(drm_dev, 0); nouveau_fbcon_zfill_all(drm_dev); - nouveau_display_resume(drm_dev); + if (drm_dev->mode_config.num_crtc) + nouveau_display_resume(drm_dev); nv_suspend_set_printk_level(NV_DBG_DEBUG); return 0; } @@ -671,7 +672,8 @@ static int nouveau_pmops_thaw(struct device *dev) if (drm_dev->mode_config.num_crtc) nouveau_fbcon_set_suspend(drm_dev, 0); nouveau_fbcon_zfill_all(drm_dev); - nouveau_display_resume(drm_dev); + if (drm_dev->mode_config.num_crtc) + nouveau_display_resume(drm_dev); nv_suspend_set_printk_level(NV_DBG_DEBUG); return 0; } @@ -906,7 +908,8 @@ static int nouveau_pmops_runtime_resume(struct device *dev) pci_set_master(pdev); ret = nouveau_do_resume(drm_dev); - nouveau_display_resume(drm_dev); + if (drm_dev->mode_config.num_crtc) + nouveau_display_resume(drm_dev); drm_kms_helper_poll_enable(drm_dev); /* do magic */ nv_mask(device, 0x88488, (1 << 25), (1 << 25)); diff --git a/drivers/gpu/drm/nouveau/nouveau_fbcon.c b/drivers/gpu/drm/nouveau/nouveau_fbcon.c index 8f6d63d7edd..a86ecf65c16 100644 --- a/drivers/gpu/drm/nouveau/nouveau_fbcon.c +++ b/drivers/gpu/drm/nouveau/nouveau_fbcon.c @@ -454,7 +454,8 @@ nouveau_fbcon_init(struct drm_device *dev) int preferred_bpp; int ret; - if (!dev->mode_config.num_crtc) + if (!dev->mode_config.num_crtc || + (dev->pdev->class >> 8) != PCI_CLASS_DISPLAY_VGA) return 0; fbcon = kzalloc(sizeof(struct nouveau_fbdev), GFP_KERNEL); diff --git a/drivers/gpu/drm/nouveau/nouveau_gem.c b/drivers/gpu/drm/nouveau/nouveau_gem.c index 487242fb3fd..f32b71238c0 100644 --- a/drivers/gpu/drm/nouveau/nouveau_gem.c +++ b/drivers/gpu/drm/nouveau/nouveau_gem.c @@ -579,18 +579,31 @@ nouveau_gem_pushbuf_validate(struct nouveau_channel *chan, return 0; } +static inline void +u_free(void *addr) +{ + if (!is_vmalloc_addr(addr)) + kfree(addr); + else + vfree(addr); +} + static inline void * u_memcpya(uint64_t user, unsigned nmemb, unsigned size) { void *mem; void __user *userptr = (void __force __user *)(uintptr_t)user; - mem = kmalloc(nmemb * size, GFP_KERNEL); + size *= nmemb; + + mem = kmalloc(size, GFP_KERNEL | __GFP_NOWARN); + if (!mem) + mem = vmalloc(size); if (!mem) return ERR_PTR(-ENOMEM); - if (DRM_COPY_FROM_USER(mem, userptr, nmemb * size)) { - kfree(mem); + if (DRM_COPY_FROM_USER(mem, userptr, size)) { + u_free(mem); return ERR_PTR(-EFAULT); } @@ -676,7 +689,7 @@ nouveau_gem_pushbuf_reloc_apply(struct nouveau_cli *cli, nouveau_bo_wr32(nvbo, r->reloc_bo_offset >> 2, data); } - kfree(reloc); + u_free(reloc); return ret; } @@ -738,7 +751,7 @@ nouveau_gem_ioctl_pushbuf(struct drm_device *dev, void *data, bo = u_memcpya(req->buffers, req->nr_buffers, sizeof(*bo)); if (IS_ERR(bo)) { - kfree(push); + u_free(push); return nouveau_abi16_put(abi16, PTR_ERR(bo)); } @@ -849,8 +862,8 @@ out: nouveau_fence_unref(&fence); out_prevalid: - kfree(bo); - kfree(push); + u_free(bo); + u_free(push); out_next: if (chan->dma.ib_max) { diff --git a/drivers/gpu/drm/nouveau/nouveau_sgdma.c b/drivers/gpu/drm/nouveau/nouveau_sgdma.c index ca5492ac2da..0843ebc910d 100644 --- a/drivers/gpu/drm/nouveau/nouveau_sgdma.c +++ b/drivers/gpu/drm/nouveau/nouveau_sgdma.c @@ -104,9 +104,7 @@ nouveau_sgdma_create_ttm(struct ttm_bo_device *bdev, else nvbe->ttm.ttm.func = &nv50_sgdma_backend; - if (ttm_dma_tt_init(&nvbe->ttm, bdev, size, page_flags, dummy_read_page)) { - kfree(nvbe); + if (ttm_dma_tt_init(&nvbe->ttm, bdev, size, page_flags, dummy_read_page)) return NULL; - } return &nvbe->ttm.ttm; } diff --git a/drivers/gpu/drm/nouveau/nv50_display.c b/drivers/gpu/drm/nouveau/nv50_display.c index 9d2092a5ed3..f8e66c08b11 100644 --- a/drivers/gpu/drm/nouveau/nv50_display.c +++ b/drivers/gpu/drm/nouveau/nv50_display.c @@ -1583,7 +1583,7 @@ nv50_dac_detect(struct drm_encoder *encoder, struct drm_connector *connector) load = 340; ret = nv_exec(disp->core, NV50_DISP_DAC_LOAD + or, &load, sizeof(load)); - if (ret || load != 7) + if (ret || !load) return connector_status_disconnected; return connector_status_connected; diff --git a/drivers/gpu/drm/omapdrm/omap_encoder.c b/drivers/gpu/drm/omapdrm/omap_encoder.c index c29451ba65d..6a12e899235 100644 --- a/drivers/gpu/drm/omapdrm/omap_encoder.c +++ b/drivers/gpu/drm/omapdrm/omap_encoder.c @@ -133,7 +133,7 @@ int omap_encoder_update(struct drm_encoder *encoder, struct omap_dss_driver *dssdrv = dssdev->driver; int ret; - dssdev->output->manager = mgr; + dssdev->src->manager = mgr; if (dssdrv->check_timings) { ret = dssdrv->check_timings(dssdev, timings); diff --git a/drivers/gpu/drm/radeon/atombios_encoders.c b/drivers/gpu/drm/radeon/atombios_encoders.c index dfac7965ea2..32923d2f600 100644 --- a/drivers/gpu/drm/radeon/atombios_encoders.c +++ b/drivers/gpu/drm/radeon/atombios_encoders.c @@ -707,8 +707,9 @@ atombios_get_encoder_mode(struct drm_encoder *encoder) switch (connector->connector_type) { case DRM_MODE_CONNECTOR_DVII: case DRM_MODE_CONNECTOR_HDMIB: /* HDMI-B is basically DL-DVI; analog works fine */ - if (drm_detect_hdmi_monitor(radeon_connector->edid) && - radeon_audio) + if ((radeon_connector->audio == RADEON_AUDIO_ENABLE) || + (drm_detect_hdmi_monitor(radeon_connector->edid) && + (radeon_connector->audio == RADEON_AUDIO_AUTO))) return ATOM_ENCODER_MODE_HDMI; else if (radeon_connector->use_digital) return ATOM_ENCODER_MODE_DVI; @@ -718,8 +719,9 @@ atombios_get_encoder_mode(struct drm_encoder *encoder) case DRM_MODE_CONNECTOR_DVID: case DRM_MODE_CONNECTOR_HDMIA: default: - if (drm_detect_hdmi_monitor(radeon_connector->edid) && - radeon_audio) + if ((radeon_connector->audio == RADEON_AUDIO_ENABLE) || + (drm_detect_hdmi_monitor(radeon_connector->edid) && + (radeon_connector->audio == RADEON_AUDIO_AUTO))) return ATOM_ENCODER_MODE_HDMI; else return ATOM_ENCODER_MODE_DVI; @@ -732,8 +734,9 @@ atombios_get_encoder_mode(struct drm_encoder *encoder) if ((dig_connector->dp_sink_type == CONNECTOR_OBJECT_ID_DISPLAYPORT) || (dig_connector->dp_sink_type == CONNECTOR_OBJECT_ID_eDP)) return ATOM_ENCODER_MODE_DP; - else if (drm_detect_hdmi_monitor(radeon_connector->edid) && - radeon_audio) + else if ((radeon_connector->audio == RADEON_AUDIO_ENABLE) || + (drm_detect_hdmi_monitor(radeon_connector->edid) && + (radeon_connector->audio == RADEON_AUDIO_AUTO))) return ATOM_ENCODER_MODE_HDMI; else return ATOM_ENCODER_MODE_DVI; @@ -1647,8 +1650,12 @@ radeon_atom_encoder_dpms_dig(struct drm_encoder *encoder, int mode) atombios_dig_encoder_setup(encoder, ATOM_ENABLE, 0); atombios_dig_transmitter_setup(encoder, ATOM_TRANSMITTER_ACTION_SETUP, 0, 0); atombios_dig_transmitter_setup(encoder, ATOM_TRANSMITTER_ACTION_ENABLE, 0, 0); - /* some early dce3.2 boards have a bug in their transmitter control table */ - if ((rdev->family != CHIP_RV710) && (rdev->family != CHIP_RV730)) + /* some dce3.x boards have a bug in their transmitter control table. + * ACTION_ENABLE_OUTPUT can probably be dropped since ACTION_ENABLE + * does the same thing and more. + */ + if ((rdev->family != CHIP_RV710) && (rdev->family != CHIP_RV730) && + (rdev->family != CHIP_RS880)) atombios_dig_transmitter_setup(encoder, ATOM_TRANSMITTER_ACTION_ENABLE_OUTPUT, 0, 0); } if (ENCODER_MODE_IS_DP(atombios_get_encoder_mode(encoder)) && connector) { diff --git a/drivers/gpu/drm/radeon/btc_dpm.c b/drivers/gpu/drm/radeon/btc_dpm.c index 084e69414fd..05ff315e8e9 100644 --- a/drivers/gpu/drm/radeon/btc_dpm.c +++ b/drivers/gpu/drm/radeon/btc_dpm.c @@ -2340,12 +2340,6 @@ int btc_dpm_set_power_state(struct radeon_device *rdev) return ret; } - ret = rv770_dpm_force_performance_level(rdev, RADEON_DPM_FORCED_LEVEL_AUTO); - if (ret) { - DRM_ERROR("rv770_dpm_force_performance_level failed\n"); - return ret; - } - return 0; } diff --git a/drivers/gpu/drm/radeon/ci_dpm.c b/drivers/gpu/drm/radeon/ci_dpm.c index 916630fdc79..89962744303 100644 --- a/drivers/gpu/drm/radeon/ci_dpm.c +++ b/drivers/gpu/drm/radeon/ci_dpm.c @@ -4208,6 +4208,7 @@ static void ci_request_link_speed_change_before_state_change(struct radeon_devic pi->pspp_notify_required = false; if (target_link_speed > current_link_speed) { switch (target_link_speed) { +#ifdef CONFIG_ACPI case RADEON_PCIE_GEN3: if (radeon_acpi_pcie_performance_request(rdev, PCIE_PERF_REQ_PECI_GEN3, false) == 0) break; @@ -4217,6 +4218,7 @@ static void ci_request_link_speed_change_before_state_change(struct radeon_devic case RADEON_PCIE_GEN2: if (radeon_acpi_pcie_performance_request(rdev, PCIE_PERF_REQ_PECI_GEN2, false) == 0) break; +#endif default: pi->force_pcie_gen = ci_get_current_pcie_speed(rdev); break; @@ -4248,7 +4250,9 @@ static void ci_notify_link_speed_change_after_state_change(struct radeon_device (ci_get_current_pcie_speed(rdev) > 0)) return; +#ifdef CONFIG_ACPI radeon_acpi_pcie_performance_request(rdev, request, false); +#endif } } @@ -4744,12 +4748,6 @@ int ci_dpm_set_power_state(struct radeon_device *rdev) if (pi->pcie_performance_request) ci_notify_link_speed_change_after_state_change(rdev, new_ps, old_ps); - ret = ci_dpm_force_performance_level(rdev, RADEON_DPM_FORCED_LEVEL_AUTO); - if (ret) { - DRM_ERROR("ci_dpm_force_performance_level failed\n"); - return ret; - } - cik_update_cg(rdev, (RADEON_CG_BLOCK_GFX | RADEON_CG_BLOCK_MC | RADEON_CG_BLOCK_SDMA | diff --git a/drivers/gpu/drm/radeon/ci_smc.c b/drivers/gpu/drm/radeon/ci_smc.c index 53b43dd3cf1..252e10a41cf 100644 --- a/drivers/gpu/drm/radeon/ci_smc.c +++ b/drivers/gpu/drm/radeon/ci_smc.c @@ -47,10 +47,11 @@ int ci_copy_bytes_to_smc(struct radeon_device *rdev, u32 smc_start_address, const u8 *src, u32 byte_count, u32 limit) { + unsigned long flags; u32 data, original_data; u32 addr; u32 extra_shift; - int ret; + int ret = 0; if (smc_start_address & 3) return -EINVAL; @@ -59,13 +60,14 @@ int ci_copy_bytes_to_smc(struct radeon_device *rdev, addr = smc_start_address; + spin_lock_irqsave(&rdev->smc_idx_lock, flags); while (byte_count >= 4) { /* SMC address space is BE */ data = (src[0] << 24) | (src[1] << 16) | (src[2] << 8) | src[3]; ret = ci_set_smc_sram_address(rdev, addr, limit); if (ret) - return ret; + goto done; WREG32(SMC_IND_DATA_0, data); @@ -80,7 +82,7 @@ int ci_copy_bytes_to_smc(struct radeon_device *rdev, ret = ci_set_smc_sram_address(rdev, addr, limit); if (ret) - return ret; + goto done; original_data = RREG32(SMC_IND_DATA_0); @@ -97,11 +99,15 @@ int ci_copy_bytes_to_smc(struct radeon_device *rdev, ret = ci_set_smc_sram_address(rdev, addr, limit); if (ret) - return ret; + goto done; WREG32(SMC_IND_DATA_0, data); } - return 0; + +done: + spin_unlock_irqrestore(&rdev->smc_idx_lock, flags); + + return ret; } void ci_start_smc(struct radeon_device *rdev) @@ -197,6 +203,7 @@ PPSMC_Result ci_wait_for_smc_inactive(struct radeon_device *rdev) int ci_load_smc_ucode(struct radeon_device *rdev, u32 limit) { + unsigned long flags; u32 ucode_start_address; u32 ucode_size; const u8 *src; @@ -219,6 +226,7 @@ int ci_load_smc_ucode(struct radeon_device *rdev, u32 limit) return -EINVAL; src = (const u8 *)rdev->smc_fw->data; + spin_lock_irqsave(&rdev->smc_idx_lock, flags); WREG32(SMC_IND_INDEX_0, ucode_start_address); WREG32_P(SMC_IND_ACCESS_CNTL, AUTO_INCREMENT_IND_0, ~AUTO_INCREMENT_IND_0); while (ucode_size >= 4) { @@ -231,6 +239,7 @@ int ci_load_smc_ucode(struct radeon_device *rdev, u32 limit) ucode_size -= 4; } WREG32_P(SMC_IND_ACCESS_CNTL, 0, ~AUTO_INCREMENT_IND_0); + spin_unlock_irqrestore(&rdev->smc_idx_lock, flags); return 0; } @@ -238,25 +247,29 @@ int ci_load_smc_ucode(struct radeon_device *rdev, u32 limit) int ci_read_smc_sram_dword(struct radeon_device *rdev, u32 smc_address, u32 *value, u32 limit) { + unsigned long flags; int ret; + spin_lock_irqsave(&rdev->smc_idx_lock, flags); ret = ci_set_smc_sram_address(rdev, smc_address, limit); - if (ret) - return ret; + if (ret == 0) + *value = RREG32(SMC_IND_DATA_0); + spin_unlock_irqrestore(&rdev->smc_idx_lock, flags); - *value = RREG32(SMC_IND_DATA_0); - return 0; + return ret; } int ci_write_smc_sram_dword(struct radeon_device *rdev, u32 smc_address, u32 value, u32 limit) { + unsigned long flags; int ret; + spin_lock_irqsave(&rdev->smc_idx_lock, flags); ret = ci_set_smc_sram_address(rdev, smc_address, limit); - if (ret) - return ret; + if (ret == 0) + WREG32(SMC_IND_DATA_0, value); + spin_unlock_irqrestore(&rdev->smc_idx_lock, flags); - WREG32(SMC_IND_DATA_0, value); - return 0; + return ret; } diff --git a/drivers/gpu/drm/radeon/cik.c b/drivers/gpu/drm/radeon/cik.c index a3bba058727..adbdb6503b0 100644 --- a/drivers/gpu/drm/radeon/cik.c +++ b/drivers/gpu/drm/radeon/cik.c @@ -77,6 +77,8 @@ static void cik_pcie_gen3_enable(struct radeon_device *rdev); static void cik_program_aspm(struct radeon_device *rdev); static void cik_init_pg(struct radeon_device *rdev); static void cik_init_cg(struct radeon_device *rdev); +static void cik_enable_gui_idle_interrupt(struct radeon_device *rdev, + bool enable); /* get temperature in millidegrees */ int ci_get_temp(struct radeon_device *rdev) @@ -120,20 +122,27 @@ int kv_get_temp(struct radeon_device *rdev) */ u32 cik_pciep_rreg(struct radeon_device *rdev, u32 reg) { + unsigned long flags; u32 r; + spin_lock_irqsave(&rdev->pciep_idx_lock, flags); WREG32(PCIE_INDEX, reg); (void)RREG32(PCIE_INDEX); r = RREG32(PCIE_DATA); + spin_unlock_irqrestore(&rdev->pciep_idx_lock, flags); return r; } void cik_pciep_wreg(struct radeon_device *rdev, u32 reg, u32 v) { + unsigned long flags; + + spin_lock_irqsave(&rdev->pciep_idx_lock, flags); WREG32(PCIE_INDEX, reg); (void)RREG32(PCIE_INDEX); WREG32(PCIE_DATA, v); (void)RREG32(PCIE_DATA); + spin_unlock_irqrestore(&rdev->pciep_idx_lock, flags); } static const u32 spectre_rlc_save_restore_register_list[] = @@ -2722,7 +2731,8 @@ static void cik_gpu_init(struct radeon_device *rdev) } else if ((rdev->pdev->device == 0x1309) || (rdev->pdev->device == 0x130A) || (rdev->pdev->device == 0x130D) || - (rdev->pdev->device == 0x1313)) { + (rdev->pdev->device == 0x1313) || + (rdev->pdev->device == 0x131D)) { rdev->config.cik.max_cu_per_sh = 6; rdev->config.cik.max_backends_per_se = 2; } else if ((rdev->pdev->device == 0x1306) || @@ -4013,6 +4023,8 @@ static int cik_cp_resume(struct radeon_device *rdev) { int r; + cik_enable_gui_idle_interrupt(rdev, false); + r = cik_cp_load_microcode(rdev); if (r) return r; @@ -4024,6 +4036,8 @@ static int cik_cp_resume(struct radeon_device *rdev) if (r) return r; + cik_enable_gui_idle_interrupt(rdev, true); + return 0; } @@ -5376,7 +5390,9 @@ static void cik_enable_hdp_ls(struct radeon_device *rdev, void cik_update_cg(struct radeon_device *rdev, u32 block, bool enable) { + if (block & RADEON_CG_BLOCK_GFX) { + cik_enable_gui_idle_interrupt(rdev, false); /* order matters! */ if (enable) { cik_enable_mgcg(rdev, true); @@ -5385,6 +5401,7 @@ void cik_update_cg(struct radeon_device *rdev, cik_enable_cgcg(rdev, false); cik_enable_mgcg(rdev, false); } + cik_enable_gui_idle_interrupt(rdev, true); } if (block & RADEON_CG_BLOCK_MC) { @@ -5541,7 +5558,7 @@ static void cik_enable_gfx_cgpg(struct radeon_device *rdev, { u32 data, orig; - if (enable && (rdev->pg_flags & RADEON_PG_SUPPORT_GFX_CG)) { + if (enable && (rdev->pg_flags & RADEON_PG_SUPPORT_GFX_PG)) { orig = data = RREG32(RLC_PG_CNTL); data |= GFX_PG_ENABLE; if (orig != data) @@ -5805,7 +5822,7 @@ static void cik_init_pg(struct radeon_device *rdev) if (rdev->pg_flags) { cik_enable_sck_slowdown_on_pu(rdev, true); cik_enable_sck_slowdown_on_pd(rdev, true); - if (rdev->pg_flags & RADEON_PG_SUPPORT_GFX_CG) { + if (rdev->pg_flags & RADEON_PG_SUPPORT_GFX_PG) { cik_init_gfx_cgpg(rdev); cik_enable_cp_pg(rdev, true); cik_enable_gds_pg(rdev, true); @@ -5819,7 +5836,7 @@ static void cik_fini_pg(struct radeon_device *rdev) { if (rdev->pg_flags) { cik_update_gfx_pg(rdev, false); - if (rdev->pg_flags & RADEON_PG_SUPPORT_GFX_CG) { + if (rdev->pg_flags & RADEON_PG_SUPPORT_GFX_PG) { cik_enable_cp_pg(rdev, false); cik_enable_gds_pg(rdev, false); } @@ -5895,7 +5912,9 @@ static void cik_disable_interrupt_state(struct radeon_device *rdev) u32 tmp; /* gfx ring */ - WREG32(CP_INT_CNTL_RING0, CNTX_BUSY_INT_ENABLE | CNTX_EMPTY_INT_ENABLE); + tmp = RREG32(CP_INT_CNTL_RING0) & + (CNTX_BUSY_INT_ENABLE | CNTX_EMPTY_INT_ENABLE); + WREG32(CP_INT_CNTL_RING0, tmp); /* sdma */ tmp = RREG32(SDMA0_CNTL + SDMA0_REGISTER_OFFSET) & ~TRAP_ENABLE; WREG32(SDMA0_CNTL + SDMA0_REGISTER_OFFSET, tmp); @@ -6036,8 +6055,7 @@ static int cik_irq_init(struct radeon_device *rdev) */ int cik_irq_set(struct radeon_device *rdev) { - u32 cp_int_cntl = CNTX_BUSY_INT_ENABLE | CNTX_EMPTY_INT_ENABLE | - PRIV_INSTR_INT_ENABLE | PRIV_REG_INT_ENABLE; + u32 cp_int_cntl; u32 cp_m1p0, cp_m1p1, cp_m1p2, cp_m1p3; u32 cp_m2p0, cp_m2p1, cp_m2p2, cp_m2p3; u32 crtc1 = 0, crtc2 = 0, crtc3 = 0, crtc4 = 0, crtc5 = 0, crtc6 = 0; @@ -6058,6 +6076,10 @@ int cik_irq_set(struct radeon_device *rdev) return 0; } + cp_int_cntl = RREG32(CP_INT_CNTL_RING0) & + (CNTX_BUSY_INT_ENABLE | CNTX_EMPTY_INT_ENABLE); + cp_int_cntl |= PRIV_INSTR_INT_ENABLE | PRIV_REG_INT_ENABLE; + hpd1 = RREG32(DC_HPD1_INT_CONTROL) & ~DC_HPDx_INT_EN; hpd2 = RREG32(DC_HPD2_INT_CONTROL) & ~DC_HPDx_INT_EN; hpd3 = RREG32(DC_HPD3_INT_CONTROL) & ~DC_HPDx_INT_EN; diff --git a/drivers/gpu/drm/radeon/cypress_dpm.c b/drivers/gpu/drm/radeon/cypress_dpm.c index 95a66db08d9..91bb470de0a 100644 --- a/drivers/gpu/drm/radeon/cypress_dpm.c +++ b/drivers/gpu/drm/radeon/cypress_dpm.c @@ -2014,12 +2014,6 @@ int cypress_dpm_set_power_state(struct radeon_device *rdev) if (eg_pi->pcie_performance_request) cypress_notify_link_speed_change_after_state_change(rdev, new_ps, old_ps); - ret = rv770_dpm_force_performance_level(rdev, RADEON_DPM_FORCED_LEVEL_AUTO); - if (ret) { - DRM_ERROR("rv770_dpm_force_performance_level failed\n"); - return ret; - } - return 0; } diff --git a/drivers/gpu/drm/radeon/dce6_afmt.c b/drivers/gpu/drm/radeon/dce6_afmt.c index 8953255e894..85a69d2ea3d 100644 --- a/drivers/gpu/drm/radeon/dce6_afmt.c +++ b/drivers/gpu/drm/radeon/dce6_afmt.c @@ -28,22 +28,30 @@ static u32 dce6_endpoint_rreg(struct radeon_device *rdev, u32 block_offset, u32 reg) { + unsigned long flags; u32 r; + spin_lock_irqsave(&rdev->end_idx_lock, flags); WREG32(AZ_F0_CODEC_ENDPOINT_INDEX + block_offset, reg); r = RREG32(AZ_F0_CODEC_ENDPOINT_DATA + block_offset); + spin_unlock_irqrestore(&rdev->end_idx_lock, flags); + return r; } static void dce6_endpoint_wreg(struct radeon_device *rdev, u32 block_offset, u32 reg, u32 v) { + unsigned long flags; + + spin_lock_irqsave(&rdev->end_idx_lock, flags); if (ASIC_IS_DCE8(rdev)) WREG32(AZ_F0_CODEC_ENDPOINT_INDEX + block_offset, reg); else WREG32(AZ_F0_CODEC_ENDPOINT_INDEX + block_offset, AZ_ENDPOINT_REG_WRITE_EN | AZ_ENDPOINT_REG_INDEX(reg)); WREG32(AZ_F0_CODEC_ENDPOINT_DATA + block_offset, v); + spin_unlock_irqrestore(&rdev->end_idx_lock, flags); } #define RREG32_ENDPOINT(block, reg) dce6_endpoint_rreg(rdev, (block), (reg)) @@ -86,12 +94,12 @@ void dce6_afmt_select_pin(struct drm_encoder *encoder) struct radeon_encoder *radeon_encoder = to_radeon_encoder(encoder); struct radeon_encoder_atom_dig *dig = radeon_encoder->enc_priv; u32 offset = dig->afmt->offset; - u32 id = dig->afmt->pin->id; if (!dig->afmt->pin) return; - WREG32(AFMT_AUDIO_SRC_CONTROL + offset, AFMT_AUDIO_SRC_SELECT(id)); + WREG32(AFMT_AUDIO_SRC_CONTROL + offset, + AFMT_AUDIO_SRC_SELECT(dig->afmt->pin->id)); } void dce6_afmt_write_speaker_allocation(struct drm_encoder *encoder) diff --git a/drivers/gpu/drm/radeon/kv_dpm.c b/drivers/gpu/drm/radeon/kv_dpm.c index ecd60809db4..71399065db0 100644 --- a/drivers/gpu/drm/radeon/kv_dpm.c +++ b/drivers/gpu/drm/radeon/kv_dpm.c @@ -40,6 +40,7 @@ static int kv_calculate_dpm_settings(struct radeon_device *rdev); static void kv_enable_new_levels(struct radeon_device *rdev); static void kv_program_nbps_index_settings(struct radeon_device *rdev, struct radeon_ps *new_rps); +static int kv_set_enabled_level(struct radeon_device *rdev, u32 level); static int kv_set_enabled_levels(struct radeon_device *rdev); static int kv_force_dpm_highest(struct radeon_device *rdev); static int kv_force_dpm_lowest(struct radeon_device *rdev); @@ -519,7 +520,7 @@ static int kv_set_dpm_boot_state(struct radeon_device *rdev) static void kv_program_vc(struct radeon_device *rdev) { - WREG32_SMC(CG_FTV_0, 0x3FFFC000); + WREG32_SMC(CG_FTV_0, 0x3FFFC100); } static void kv_clear_vc(struct radeon_device *rdev) @@ -638,7 +639,10 @@ static int kv_force_lowest_valid(struct radeon_device *rdev) static int kv_unforce_levels(struct radeon_device *rdev) { - return kv_notify_message_to_smu(rdev, PPSMC_MSG_NoForcedLevel); + if (rdev->family == CHIP_KABINI) + return kv_notify_message_to_smu(rdev, PPSMC_MSG_NoForcedLevel); + else + return kv_set_enabled_levels(rdev); } static int kv_update_sclk_t(struct radeon_device *rdev) @@ -667,9 +671,8 @@ static int kv_program_bootup_state(struct radeon_device *rdev) &rdev->pm.dpm.dyn_state.vddc_dependency_on_sclk; if (table && table->count) { - for (i = pi->graphics_dpm_level_count - 1; i >= 0; i--) { - if ((table->entries[i].clk == pi->boot_pl.sclk) || - (i == 0)) + for (i = pi->graphics_dpm_level_count - 1; i > 0; i--) { + if (table->entries[i].clk == pi->boot_pl.sclk) break; } @@ -682,9 +685,8 @@ static int kv_program_bootup_state(struct radeon_device *rdev) if (table->num_max_dpm_entries == 0) return -EINVAL; - for (i = pi->graphics_dpm_level_count - 1; i >= 0; i--) { - if ((table->entries[i].sclk_frequency == pi->boot_pl.sclk) || - (i == 0)) + for (i = pi->graphics_dpm_level_count - 1; i > 0; i--) { + if (table->entries[i].sclk_frequency == pi->boot_pl.sclk) break; } @@ -1078,6 +1080,13 @@ static int kv_enable_ulv(struct radeon_device *rdev, bool enable) PPSMC_MSG_EnableULV : PPSMC_MSG_DisableULV); } +static void kv_reset_acp_boot_level(struct radeon_device *rdev) +{ + struct kv_power_info *pi = kv_get_pi(rdev); + + pi->acp_boot_level = 0xff; +} + static void kv_update_current_ps(struct radeon_device *rdev, struct radeon_ps *rps) { @@ -1100,6 +1109,18 @@ static void kv_update_requested_ps(struct radeon_device *rdev, pi->requested_rps.ps_priv = &pi->requested_ps; } +void kv_dpm_enable_bapm(struct radeon_device *rdev, bool enable) +{ + struct kv_power_info *pi = kv_get_pi(rdev); + int ret; + + if (pi->bapm_enable) { + ret = kv_smc_bapm_enable(rdev, enable); + if (ret) + DRM_ERROR("kv_smc_bapm_enable failed\n"); + } +} + int kv_dpm_enable(struct radeon_device *rdev) { struct kv_power_info *pi = kv_get_pi(rdev); @@ -1192,6 +1213,8 @@ int kv_dpm_enable(struct radeon_device *rdev) return ret; } + kv_reset_acp_boot_level(rdev); + if (rdev->irq.installed && r600_is_internal_thermal_sensor(rdev->pm.int_thermal_type)) { ret = kv_set_thermal_temperature_range(rdev, R600_TEMP_RANGE_MIN, R600_TEMP_RANGE_MAX); @@ -1203,6 +1226,12 @@ int kv_dpm_enable(struct radeon_device *rdev) radeon_irq_set(rdev); } + ret = kv_smc_bapm_enable(rdev, false); + if (ret) { + DRM_ERROR("kv_smc_bapm_enable failed\n"); + return ret; + } + /* powerdown unused blocks for now */ kv_dpm_powergate_acp(rdev, true); kv_dpm_powergate_samu(rdev, true); @@ -1226,6 +1255,8 @@ void kv_dpm_disable(struct radeon_device *rdev) RADEON_CG_BLOCK_BIF | RADEON_CG_BLOCK_HDP), false); + kv_smc_bapm_enable(rdev, false); + /* powerup blocks */ kv_dpm_powergate_acp(rdev, false); kv_dpm_powergate_samu(rdev, false); @@ -1450,6 +1481,39 @@ static int kv_update_samu_dpm(struct radeon_device *rdev, bool gate) return kv_enable_samu_dpm(rdev, !gate); } +static u8 kv_get_acp_boot_level(struct radeon_device *rdev) +{ + u8 i; + struct radeon_clock_voltage_dependency_table *table = + &rdev->pm.dpm.dyn_state.acp_clock_voltage_dependency_table; + + for (i = 0; i < table->count; i++) { + if (table->entries[i].clk >= 0) /* XXX */ + break; + } + + if (i >= table->count) + i = table->count - 1; + + return i; +} + +static void kv_update_acp_boot_level(struct radeon_device *rdev) +{ + struct kv_power_info *pi = kv_get_pi(rdev); + u8 acp_boot_level; + + if (!pi->caps_stable_p_state) { + acp_boot_level = kv_get_acp_boot_level(rdev); + if (acp_boot_level != pi->acp_boot_level) { + pi->acp_boot_level = acp_boot_level; + kv_send_msg_to_smc_with_parameter(rdev, + PPSMC_MSG_ACPDPM_SetEnabledMask, + (1 << pi->acp_boot_level)); + } + } +} + static int kv_update_acp_dpm(struct radeon_device *rdev, bool gate) { struct kv_power_info *pi = kv_get_pi(rdev); @@ -1461,7 +1525,7 @@ static int kv_update_acp_dpm(struct radeon_device *rdev, bool gate) if (pi->caps_stable_p_state) pi->acp_boot_level = table->count - 1; else - pi->acp_boot_level = 0; + pi->acp_boot_level = kv_get_acp_boot_level(rdev); ret = kv_copy_bytes_to_smc(rdev, pi->dpm_table_start + @@ -1588,13 +1652,11 @@ static void kv_set_valid_clock_range(struct radeon_device *rdev, } } - for (i = pi->graphics_dpm_level_count - 1; i >= 0; i--) { - if ((table->entries[i].clk <= new_ps->levels[new_ps->num_levels -1].sclk) || - (i == 0)) { - pi->highest_valid = i; + for (i = pi->graphics_dpm_level_count - 1; i > 0; i--) { + if (table->entries[i].clk <= new_ps->levels[new_ps->num_levels - 1].sclk) break; - } } + pi->highest_valid = i; if (pi->lowest_valid > pi->highest_valid) { if ((new_ps->levels[0].sclk - table->entries[pi->highest_valid].clk) > @@ -1615,14 +1677,12 @@ static void kv_set_valid_clock_range(struct radeon_device *rdev, } } - for (i = pi->graphics_dpm_level_count - 1; i >= 0; i--) { + for (i = pi->graphics_dpm_level_count - 1; i > 0; i--) { if (table->entries[i].sclk_frequency <= - new_ps->levels[new_ps->num_levels - 1].sclk || - i == 0) { - pi->highest_valid = i; + new_ps->levels[new_ps->num_levels - 1].sclk) break; - } } + pi->highest_valid = i; if (pi->lowest_valid > pi->highest_valid) { if ((new_ps->levels[0].sclk - @@ -1724,6 +1784,14 @@ int kv_dpm_set_power_state(struct radeon_device *rdev) RADEON_CG_BLOCK_BIF | RADEON_CG_BLOCK_HDP), false); + if (pi->bapm_enable) { + ret = kv_smc_bapm_enable(rdev, rdev->pm.dpm.ac_power); + if (ret) { + DRM_ERROR("kv_smc_bapm_enable failed\n"); + return ret; + } + } + if (rdev->family == CHIP_KABINI) { if (pi->enable_dpm) { kv_set_valid_clock_range(rdev, new_ps); @@ -1775,6 +1843,7 @@ int kv_dpm_set_power_state(struct radeon_device *rdev) return ret; } #endif + kv_update_acp_boot_level(rdev); kv_update_sclk_t(rdev); kv_enable_nb_dpm(rdev); } @@ -1785,7 +1854,6 @@ int kv_dpm_set_power_state(struct radeon_device *rdev) RADEON_CG_BLOCK_BIF | RADEON_CG_BLOCK_HDP), true); - rdev->pm.dpm.forced_level = RADEON_DPM_FORCED_LEVEL_AUTO; return 0; } @@ -1806,12 +1874,23 @@ void kv_dpm_setup_asic(struct radeon_device *rdev) void kv_dpm_reset_asic(struct radeon_device *rdev) { - kv_force_lowest_valid(rdev); - kv_init_graphics_levels(rdev); - kv_program_bootup_state(rdev); - kv_upload_dpm_settings(rdev); - kv_force_lowest_valid(rdev); - kv_unforce_levels(rdev); + struct kv_power_info *pi = kv_get_pi(rdev); + + if (rdev->family == CHIP_KABINI) { + kv_force_lowest_valid(rdev); + kv_init_graphics_levels(rdev); + kv_program_bootup_state(rdev); + kv_upload_dpm_settings(rdev); + kv_force_lowest_valid(rdev); + kv_unforce_levels(rdev); + } else { + kv_init_graphics_levels(rdev); + kv_program_bootup_state(rdev); + kv_freeze_sclk_dpm(rdev, true); + kv_upload_dpm_settings(rdev); + kv_freeze_sclk_dpm(rdev, false); + kv_set_enabled_level(rdev, pi->graphics_boot_level); + } } //XXX use sumo_dpm_display_configuration_changed @@ -1871,12 +1950,15 @@ static int kv_force_dpm_highest(struct radeon_device *rdev) if (ret) return ret; - for (i = SMU7_MAX_LEVELS_GRAPHICS - 1; i >= 0; i--) { + for (i = SMU7_MAX_LEVELS_GRAPHICS - 1; i > 0; i--) { if (enable_mask & (1 << i)) break; } - return kv_send_msg_to_smc_with_parameter(rdev, PPSMC_MSG_DPM_ForceState, i); + if (rdev->family == CHIP_KABINI) + return kv_send_msg_to_smc_with_parameter(rdev, PPSMC_MSG_DPM_ForceState, i); + else + return kv_set_enabled_level(rdev, i); } static int kv_force_dpm_lowest(struct radeon_device *rdev) @@ -1893,7 +1975,10 @@ static int kv_force_dpm_lowest(struct radeon_device *rdev) break; } - return kv_send_msg_to_smc_with_parameter(rdev, PPSMC_MSG_DPM_ForceState, i); + if (rdev->family == CHIP_KABINI) + return kv_send_msg_to_smc_with_parameter(rdev, PPSMC_MSG_DPM_ForceState, i); + else + return kv_set_enabled_level(rdev, i); } static u8 kv_get_sleep_divider_id_from_clock(struct radeon_device *rdev, @@ -1911,9 +1996,9 @@ static u8 kv_get_sleep_divider_id_from_clock(struct radeon_device *rdev, if (!pi->caps_sclk_ds) return 0; - for (i = KV_MAX_DEEPSLEEP_DIVIDER_ID; i <= 0; i--) { + for (i = KV_MAX_DEEPSLEEP_DIVIDER_ID; i > 0; i--) { temp = sclk / sumo_get_sleep_divider_from_id(i); - if ((temp >= min) || (i == 0)) + if (temp >= min) break; } @@ -2039,12 +2124,12 @@ static void kv_apply_state_adjust_rules(struct radeon_device *rdev, ps->dpmx_nb_ps_lo = 0x1; ps->dpmx_nb_ps_hi = 0x0; } else { - ps->dpm0_pg_nb_ps_lo = 0x1; + ps->dpm0_pg_nb_ps_lo = 0x3; ps->dpm0_pg_nb_ps_hi = 0x0; - ps->dpmx_nb_ps_lo = 0x2; - ps->dpmx_nb_ps_hi = 0x1; + ps->dpmx_nb_ps_lo = 0x3; + ps->dpmx_nb_ps_hi = 0x0; - if (pi->sys_info.nb_dpm_enable && pi->battery_state) { + if (pi->sys_info.nb_dpm_enable) { force_high = (mclk >= pi->sys_info.nbp_memory_clock[3]) || pi->video_start || (rdev->pm.dpm.new_active_crtc_count >= 3) || pi->disable_nb_ps3_in_battery; @@ -2210,6 +2295,15 @@ static void kv_enable_new_levels(struct radeon_device *rdev) } } +static int kv_set_enabled_level(struct radeon_device *rdev, u32 level) +{ + u32 new_mask = (1 << level); + + return kv_send_msg_to_smc_with_parameter(rdev, + PPSMC_MSG_SCLKDPM_SetEnabledMask, + new_mask); +} + static int kv_set_enabled_levels(struct radeon_device *rdev) { struct kv_power_info *pi = kv_get_pi(rdev); diff --git a/drivers/gpu/drm/radeon/kv_dpm.h b/drivers/gpu/drm/radeon/kv_dpm.h index 32bb079572d..8cef7525d7a 100644 --- a/drivers/gpu/drm/radeon/kv_dpm.h +++ b/drivers/gpu/drm/radeon/kv_dpm.h @@ -192,6 +192,7 @@ int kv_send_msg_to_smc_with_parameter(struct radeon_device *rdev, int kv_read_smc_sram_dword(struct radeon_device *rdev, u32 smc_address, u32 *value, u32 limit); int kv_smc_dpm_enable(struct radeon_device *rdev, bool enable); +int kv_smc_bapm_enable(struct radeon_device *rdev, bool enable); int kv_copy_bytes_to_smc(struct radeon_device *rdev, u32 smc_start_address, const u8 *src, u32 byte_count, u32 limit); diff --git a/drivers/gpu/drm/radeon/kv_smc.c b/drivers/gpu/drm/radeon/kv_smc.c index 34a226d7e34..0000b59a6d0 100644 --- a/drivers/gpu/drm/radeon/kv_smc.c +++ b/drivers/gpu/drm/radeon/kv_smc.c @@ -107,6 +107,14 @@ int kv_smc_dpm_enable(struct radeon_device *rdev, bool enable) return kv_notify_message_to_smu(rdev, PPSMC_MSG_DPM_Disable); } +int kv_smc_bapm_enable(struct radeon_device *rdev, bool enable) +{ + if (enable) + return kv_notify_message_to_smu(rdev, PPSMC_MSG_EnableBAPM); + else + return kv_notify_message_to_smu(rdev, PPSMC_MSG_DisableBAPM); +} + int kv_copy_bytes_to_smc(struct radeon_device *rdev, u32 smc_start_address, const u8 *src, u32 byte_count, u32 limit) diff --git a/drivers/gpu/drm/radeon/ni_dpm.c b/drivers/gpu/drm/radeon/ni_dpm.c index f7b625c9e0e..6c398a456d7 100644 --- a/drivers/gpu/drm/radeon/ni_dpm.c +++ b/drivers/gpu/drm/radeon/ni_dpm.c @@ -3865,12 +3865,6 @@ int ni_dpm_set_power_state(struct radeon_device *rdev) return ret; } - ret = ni_dpm_force_performance_level(rdev, RADEON_DPM_FORCED_LEVEL_AUTO); - if (ret) { - DRM_ERROR("ni_dpm_force_performance_level failed\n"); - return ret; - } - return 0; } diff --git a/drivers/gpu/drm/radeon/ppsmc.h b/drivers/gpu/drm/radeon/ppsmc.h index 682842804bc..5670b829128 100644 --- a/drivers/gpu/drm/radeon/ppsmc.h +++ b/drivers/gpu/drm/radeon/ppsmc.h @@ -163,6 +163,8 @@ typedef uint8_t PPSMC_Result; #define PPSMC_MSG_VCEPowerON ((uint32_t) 0x10f) #define PPSMC_MSG_DCE_RemoveVoltageAdjustment ((uint32_t) 0x11d) #define PPSMC_MSG_DCE_AllowVoltageAdjustment ((uint32_t) 0x11e) +#define PPSMC_MSG_EnableBAPM ((uint32_t) 0x120) +#define PPSMC_MSG_DisableBAPM ((uint32_t) 0x121) #define PPSMC_MSG_UVD_DPM_Config ((uint32_t) 0x124) diff --git a/drivers/gpu/drm/radeon/r100.c b/drivers/gpu/drm/radeon/r100.c index 9fc61dd68bc..24175717307 100644 --- a/drivers/gpu/drm/radeon/r100.c +++ b/drivers/gpu/drm/radeon/r100.c @@ -2853,21 +2853,28 @@ static void r100_pll_errata_after_data(struct radeon_device *rdev) uint32_t r100_pll_rreg(struct radeon_device *rdev, uint32_t reg) { + unsigned long flags; uint32_t data; + spin_lock_irqsave(&rdev->pll_idx_lock, flags); WREG8(RADEON_CLOCK_CNTL_INDEX, reg & 0x3f); r100_pll_errata_after_index(rdev); data = RREG32(RADEON_CLOCK_CNTL_DATA); r100_pll_errata_after_data(rdev); + spin_unlock_irqrestore(&rdev->pll_idx_lock, flags); return data; } void r100_pll_wreg(struct radeon_device *rdev, uint32_t reg, uint32_t v) { + unsigned long flags; + + spin_lock_irqsave(&rdev->pll_idx_lock, flags); WREG8(RADEON_CLOCK_CNTL_INDEX, ((reg & 0x3f) | RADEON_PLL_WR_EN)); r100_pll_errata_after_index(rdev); WREG32(RADEON_CLOCK_CNTL_DATA, v); r100_pll_errata_after_data(rdev); + spin_unlock_irqrestore(&rdev->pll_idx_lock, flags); } static void r100_set_safe_registers(struct radeon_device *rdev) diff --git a/drivers/gpu/drm/radeon/r300.c b/drivers/gpu/drm/radeon/r300.c index b9b776f1e58..d8dd269b915 100644 --- a/drivers/gpu/drm/radeon/r300.c +++ b/drivers/gpu/drm/radeon/r300.c @@ -1541,7 +1541,7 @@ int r300_init(struct radeon_device *rdev) rdev->accel_working = true; r = r300_startup(rdev); if (r) { - /* Somethings want wront with the accel init stop accel */ + /* Something went wrong with the accel init, so stop accel */ dev_err(rdev->dev, "Disabling GPU acceleration\n"); r100_cp_fini(rdev); radeon_wb_fini(rdev); diff --git a/drivers/gpu/drm/radeon/r420.c b/drivers/gpu/drm/radeon/r420.c index 4e796ecf9ea..6edf2b3a52b 100644 --- a/drivers/gpu/drm/radeon/r420.c +++ b/drivers/gpu/drm/radeon/r420.c @@ -160,18 +160,25 @@ void r420_pipes_init(struct radeon_device *rdev) u32 r420_mc_rreg(struct radeon_device *rdev, u32 reg) { + unsigned long flags; u32 r; + spin_lock_irqsave(&rdev->mc_idx_lock, flags); WREG32(R_0001F8_MC_IND_INDEX, S_0001F8_MC_IND_ADDR(reg)); r = RREG32(R_0001FC_MC_IND_DATA); + spin_unlock_irqrestore(&rdev->mc_idx_lock, flags); return r; } void r420_mc_wreg(struct radeon_device *rdev, u32 reg, u32 v) { + unsigned long flags; + + spin_lock_irqsave(&rdev->mc_idx_lock, flags); WREG32(R_0001F8_MC_IND_INDEX, S_0001F8_MC_IND_ADDR(reg) | S_0001F8_MC_IND_WR_EN(1)); WREG32(R_0001FC_MC_IND_DATA, v); + spin_unlock_irqrestore(&rdev->mc_idx_lock, flags); } static void r420_debugfs(struct radeon_device *rdev) diff --git a/drivers/gpu/drm/radeon/r600.c b/drivers/gpu/drm/radeon/r600.c index ea4d3734e6d..2a1b1876b43 100644 --- a/drivers/gpu/drm/radeon/r600.c +++ b/drivers/gpu/drm/radeon/r600.c @@ -119,6 +119,11 @@ u32 r600_get_xclk(struct radeon_device *rdev) return rdev->clock.spll.reference_freq; } +int r600_set_uvd_clocks(struct radeon_device *rdev, u32 vclk, u32 dclk) +{ + return 0; +} + /* get temperature in millidegrees */ int rv6xx_get_temp(struct radeon_device *rdev) { @@ -1045,20 +1050,27 @@ int r600_mc_wait_for_idle(struct radeon_device *rdev) uint32_t rs780_mc_rreg(struct radeon_device *rdev, uint32_t reg) { + unsigned long flags; uint32_t r; + spin_lock_irqsave(&rdev->mc_idx_lock, flags); WREG32(R_0028F8_MC_INDEX, S_0028F8_MC_IND_ADDR(reg)); r = RREG32(R_0028FC_MC_DATA); WREG32(R_0028F8_MC_INDEX, ~C_0028F8_MC_IND_ADDR); + spin_unlock_irqrestore(&rdev->mc_idx_lock, flags); return r; } void rs780_mc_wreg(struct radeon_device *rdev, uint32_t reg, uint32_t v) { + unsigned long flags; + + spin_lock_irqsave(&rdev->mc_idx_lock, flags); WREG32(R_0028F8_MC_INDEX, S_0028F8_MC_IND_ADDR(reg) | S_0028F8_MC_IND_WR_EN(1)); WREG32(R_0028FC_MC_DATA, v); WREG32(R_0028F8_MC_INDEX, 0x7F); + spin_unlock_irqrestore(&rdev->mc_idx_lock, flags); } static void r600_mc_program(struct radeon_device *rdev) @@ -2092,20 +2104,27 @@ static void r600_gpu_init(struct radeon_device *rdev) */ u32 r600_pciep_rreg(struct radeon_device *rdev, u32 reg) { + unsigned long flags; u32 r; + spin_lock_irqsave(&rdev->pciep_idx_lock, flags); WREG32(PCIE_PORT_INDEX, ((reg) & 0xff)); (void)RREG32(PCIE_PORT_INDEX); r = RREG32(PCIE_PORT_DATA); + spin_unlock_irqrestore(&rdev->pciep_idx_lock, flags); return r; } void r600_pciep_wreg(struct radeon_device *rdev, u32 reg, u32 v) { + unsigned long flags; + + spin_lock_irqsave(&rdev->pciep_idx_lock, flags); WREG32(PCIE_PORT_INDEX, ((reg) & 0xff)); (void)RREG32(PCIE_PORT_INDEX); WREG32(PCIE_PORT_DATA, (v)); (void)RREG32(PCIE_PORT_DATA); + spin_unlock_irqrestore(&rdev->pciep_idx_lock, flags); } /* diff --git a/drivers/gpu/drm/radeon/r600_dpm.c b/drivers/gpu/drm/radeon/r600_dpm.c index fa0de46fcc0..e65f211a7be 100644 --- a/drivers/gpu/drm/radeon/r600_dpm.c +++ b/drivers/gpu/drm/radeon/r600_dpm.c @@ -1219,30 +1219,20 @@ int r600_parse_extended_power_table(struct radeon_device *rdev) void r600_free_extended_power_table(struct radeon_device *rdev) { - if (rdev->pm.dpm.dyn_state.vddc_dependency_on_sclk.entries) - kfree(rdev->pm.dpm.dyn_state.vddc_dependency_on_sclk.entries); - if (rdev->pm.dpm.dyn_state.vddci_dependency_on_mclk.entries) - kfree(rdev->pm.dpm.dyn_state.vddci_dependency_on_mclk.entries); - if (rdev->pm.dpm.dyn_state.vddc_dependency_on_mclk.entries) - kfree(rdev->pm.dpm.dyn_state.vddc_dependency_on_mclk.entries); - if (rdev->pm.dpm.dyn_state.mvdd_dependency_on_mclk.entries) - kfree(rdev->pm.dpm.dyn_state.mvdd_dependency_on_mclk.entries); - if (rdev->pm.dpm.dyn_state.cac_leakage_table.entries) - kfree(rdev->pm.dpm.dyn_state.cac_leakage_table.entries); - if (rdev->pm.dpm.dyn_state.phase_shedding_limits_table.entries) - kfree(rdev->pm.dpm.dyn_state.phase_shedding_limits_table.entries); - if (rdev->pm.dpm.dyn_state.ppm_table) - kfree(rdev->pm.dpm.dyn_state.ppm_table); - if (rdev->pm.dpm.dyn_state.cac_tdp_table) - kfree(rdev->pm.dpm.dyn_state.cac_tdp_table); - if (rdev->pm.dpm.dyn_state.vce_clock_voltage_dependency_table.entries) - kfree(rdev->pm.dpm.dyn_state.vce_clock_voltage_dependency_table.entries); - if (rdev->pm.dpm.dyn_state.uvd_clock_voltage_dependency_table.entries) - kfree(rdev->pm.dpm.dyn_state.uvd_clock_voltage_dependency_table.entries); - if (rdev->pm.dpm.dyn_state.samu_clock_voltage_dependency_table.entries) - kfree(rdev->pm.dpm.dyn_state.samu_clock_voltage_dependency_table.entries); - if (rdev->pm.dpm.dyn_state.acp_clock_voltage_dependency_table.entries) - kfree(rdev->pm.dpm.dyn_state.acp_clock_voltage_dependency_table.entries); + struct radeon_dpm_dynamic_state *dyn_state = &rdev->pm.dpm.dyn_state; + + kfree(dyn_state->vddc_dependency_on_sclk.entries); + kfree(dyn_state->vddci_dependency_on_mclk.entries); + kfree(dyn_state->vddc_dependency_on_mclk.entries); + kfree(dyn_state->mvdd_dependency_on_mclk.entries); + kfree(dyn_state->cac_leakage_table.entries); + kfree(dyn_state->phase_shedding_limits_table.entries); + kfree(dyn_state->ppm_table); + kfree(dyn_state->cac_tdp_table); + kfree(dyn_state->vce_clock_voltage_dependency_table.entries); + kfree(dyn_state->uvd_clock_voltage_dependency_table.entries); + kfree(dyn_state->samu_clock_voltage_dependency_table.entries); + kfree(dyn_state->acp_clock_voltage_dependency_table.entries); } enum radeon_pcie_gen r600_get_pcie_gen_support(struct radeon_device *rdev, diff --git a/drivers/gpu/drm/radeon/r600d.h b/drivers/gpu/drm/radeon/r600d.h index 454f90a849e..e673fe26ea8 100644 --- a/drivers/gpu/drm/radeon/r600d.h +++ b/drivers/gpu/drm/radeon/r600d.h @@ -1040,7 +1040,7 @@ # define HDMI0_AVI_INFO_CONT (1 << 1) # define HDMI0_AUDIO_INFO_SEND (1 << 4) # define HDMI0_AUDIO_INFO_CONT (1 << 5) -# define HDMI0_AUDIO_INFO_SOURCE (1 << 6) /* 0 - sound block; 1 - hmdi regs */ +# define HDMI0_AUDIO_INFO_SOURCE (1 << 6) /* 0 - sound block; 1 - hdmi regs */ # define HDMI0_AUDIO_INFO_UPDATE (1 << 7) # define HDMI0_MPEG_INFO_SEND (1 << 8) # define HDMI0_MPEG_INFO_CONT (1 << 9) diff --git a/drivers/gpu/drm/radeon/radeon.h b/drivers/gpu/drm/radeon/radeon.h index ff8b564ce2b..a400ac1c414 100644 --- a/drivers/gpu/drm/radeon/radeon.h +++ b/drivers/gpu/drm/radeon/radeon.h @@ -181,7 +181,7 @@ extern int radeon_aspm; #define RADEON_CG_SUPPORT_HDP_MGCG (1 << 16) /* PG flags */ -#define RADEON_PG_SUPPORT_GFX_CG (1 << 0) +#define RADEON_PG_SUPPORT_GFX_PG (1 << 0) #define RADEON_PG_SUPPORT_GFX_SMG (1 << 1) #define RADEON_PG_SUPPORT_GFX_DMG (1 << 2) #define RADEON_PG_SUPPORT_UVD (1 << 3) @@ -1778,6 +1778,7 @@ struct radeon_asic { int (*force_performance_level)(struct radeon_device *rdev, enum radeon_dpm_forced_level level); bool (*vblank_too_short)(struct radeon_device *rdev); void (*powergate_uvd)(struct radeon_device *rdev, bool gate); + void (*enable_bapm)(struct radeon_device *rdev, bool enable); } dpm; /* pageflipping */ struct { @@ -2110,6 +2111,28 @@ struct radeon_device { resource_size_t rmmio_size; /* protects concurrent MM_INDEX/DATA based register access */ spinlock_t mmio_idx_lock; + /* protects concurrent SMC based register access */ + spinlock_t smc_idx_lock; + /* protects concurrent PLL register access */ + spinlock_t pll_idx_lock; + /* protects concurrent MC register access */ + spinlock_t mc_idx_lock; + /* protects concurrent PCIE register access */ + spinlock_t pcie_idx_lock; + /* protects concurrent PCIE_PORT register access */ + spinlock_t pciep_idx_lock; + /* protects concurrent PIF register access */ + spinlock_t pif_idx_lock; + /* protects concurrent CG register access */ + spinlock_t cg_idx_lock; + /* protects concurrent UVD register access */ + spinlock_t uvd_idx_lock; + /* protects concurrent RCU register access */ + spinlock_t rcu_idx_lock; + /* protects concurrent DIDT register access */ + spinlock_t didt_idx_lock; + /* protects concurrent ENDPOINT (audio) register access */ + spinlock_t end_idx_lock; void __iomem *rmmio; radeon_rreg_t mc_rreg; radeon_wreg_t mc_wreg; @@ -2277,123 +2300,179 @@ void cik_mm_wdoorbell(struct radeon_device *rdev, u32 offset, u32 v); */ static inline uint32_t rv370_pcie_rreg(struct radeon_device *rdev, uint32_t reg) { + unsigned long flags; uint32_t r; + spin_lock_irqsave(&rdev->pcie_idx_lock, flags); WREG32(RADEON_PCIE_INDEX, ((reg) & rdev->pcie_reg_mask)); r = RREG32(RADEON_PCIE_DATA); + spin_unlock_irqrestore(&rdev->pcie_idx_lock, flags); return r; } static inline void rv370_pcie_wreg(struct radeon_device *rdev, uint32_t reg, uint32_t v) { + unsigned long flags; + + spin_lock_irqsave(&rdev->pcie_idx_lock, flags); WREG32(RADEON_PCIE_INDEX, ((reg) & rdev->pcie_reg_mask)); WREG32(RADEON_PCIE_DATA, (v)); + spin_unlock_irqrestore(&rdev->pcie_idx_lock, flags); } static inline u32 tn_smc_rreg(struct radeon_device *rdev, u32 reg) { + unsigned long flags; u32 r; + spin_lock_irqsave(&rdev->smc_idx_lock, flags); WREG32(TN_SMC_IND_INDEX_0, (reg)); r = RREG32(TN_SMC_IND_DATA_0); + spin_unlock_irqrestore(&rdev->smc_idx_lock, flags); return r; } static inline void tn_smc_wreg(struct radeon_device *rdev, u32 reg, u32 v) { + unsigned long flags; + + spin_lock_irqsave(&rdev->smc_idx_lock, flags); WREG32(TN_SMC_IND_INDEX_0, (reg)); WREG32(TN_SMC_IND_DATA_0, (v)); + spin_unlock_irqrestore(&rdev->smc_idx_lock, flags); } static inline u32 r600_rcu_rreg(struct radeon_device *rdev, u32 reg) { + unsigned long flags; u32 r; + spin_lock_irqsave(&rdev->rcu_idx_lock, flags); WREG32(R600_RCU_INDEX, ((reg) & 0x1fff)); r = RREG32(R600_RCU_DATA); + spin_unlock_irqrestore(&rdev->rcu_idx_lock, flags); return r; } static inline void r600_rcu_wreg(struct radeon_device *rdev, u32 reg, u32 v) { + unsigned long flags; + + spin_lock_irqsave(&rdev->rcu_idx_lock, flags); WREG32(R600_RCU_INDEX, ((reg) & 0x1fff)); WREG32(R600_RCU_DATA, (v)); + spin_unlock_irqrestore(&rdev->rcu_idx_lock, flags); } static inline u32 eg_cg_rreg(struct radeon_device *rdev, u32 reg) { + unsigned long flags; u32 r; + spin_lock_irqsave(&rdev->cg_idx_lock, flags); WREG32(EVERGREEN_CG_IND_ADDR, ((reg) & 0xffff)); r = RREG32(EVERGREEN_CG_IND_DATA); + spin_unlock_irqrestore(&rdev->cg_idx_lock, flags); return r; } static inline void eg_cg_wreg(struct radeon_device *rdev, u32 reg, u32 v) { + unsigned long flags; + + spin_lock_irqsave(&rdev->cg_idx_lock, flags); WREG32(EVERGREEN_CG_IND_ADDR, ((reg) & 0xffff)); WREG32(EVERGREEN_CG_IND_DATA, (v)); + spin_unlock_irqrestore(&rdev->cg_idx_lock, flags); } static inline u32 eg_pif_phy0_rreg(struct radeon_device *rdev, u32 reg) { + unsigned long flags; u32 r; + spin_lock_irqsave(&rdev->pif_idx_lock, flags); WREG32(EVERGREEN_PIF_PHY0_INDEX, ((reg) & 0xffff)); r = RREG32(EVERGREEN_PIF_PHY0_DATA); + spin_unlock_irqrestore(&rdev->pif_idx_lock, flags); return r; } static inline void eg_pif_phy0_wreg(struct radeon_device *rdev, u32 reg, u32 v) { + unsigned long flags; + + spin_lock_irqsave(&rdev->pif_idx_lock, flags); WREG32(EVERGREEN_PIF_PHY0_INDEX, ((reg) & 0xffff)); WREG32(EVERGREEN_PIF_PHY0_DATA, (v)); + spin_unlock_irqrestore(&rdev->pif_idx_lock, flags); } static inline u32 eg_pif_phy1_rreg(struct radeon_device *rdev, u32 reg) { + unsigned long flags; u32 r; + spin_lock_irqsave(&rdev->pif_idx_lock, flags); WREG32(EVERGREEN_PIF_PHY1_INDEX, ((reg) & 0xffff)); r = RREG32(EVERGREEN_PIF_PHY1_DATA); + spin_unlock_irqrestore(&rdev->pif_idx_lock, flags); return r; } static inline void eg_pif_phy1_wreg(struct radeon_device *rdev, u32 reg, u32 v) { + unsigned long flags; + + spin_lock_irqsave(&rdev->pif_idx_lock, flags); WREG32(EVERGREEN_PIF_PHY1_INDEX, ((reg) & 0xffff)); WREG32(EVERGREEN_PIF_PHY1_DATA, (v)); + spin_unlock_irqrestore(&rdev->pif_idx_lock, flags); } static inline u32 r600_uvd_ctx_rreg(struct radeon_device *rdev, u32 reg) { + unsigned long flags; u32 r; + spin_lock_irqsave(&rdev->uvd_idx_lock, flags); WREG32(R600_UVD_CTX_INDEX, ((reg) & 0x1ff)); r = RREG32(R600_UVD_CTX_DATA); + spin_unlock_irqrestore(&rdev->uvd_idx_lock, flags); return r; } static inline void r600_uvd_ctx_wreg(struct radeon_device *rdev, u32 reg, u32 v) { + unsigned long flags; + + spin_lock_irqsave(&rdev->uvd_idx_lock, flags); WREG32(R600_UVD_CTX_INDEX, ((reg) & 0x1ff)); WREG32(R600_UVD_CTX_DATA, (v)); + spin_unlock_irqrestore(&rdev->uvd_idx_lock, flags); } static inline u32 cik_didt_rreg(struct radeon_device *rdev, u32 reg) { + unsigned long flags; u32 r; + spin_lock_irqsave(&rdev->didt_idx_lock, flags); WREG32(CIK_DIDT_IND_INDEX, (reg)); r = RREG32(CIK_DIDT_IND_DATA); + spin_unlock_irqrestore(&rdev->didt_idx_lock, flags); return r; } static inline void cik_didt_wreg(struct radeon_device *rdev, u32 reg, u32 v) { + unsigned long flags; + + spin_lock_irqsave(&rdev->didt_idx_lock, flags); WREG32(CIK_DIDT_IND_INDEX, (reg)); WREG32(CIK_DIDT_IND_DATA, (v)); + spin_unlock_irqrestore(&rdev->didt_idx_lock, flags); } void r100_pll_errata_after_index(struct radeon_device *rdev); @@ -2569,6 +2648,7 @@ void radeon_ring_write(struct radeon_ring *ring, uint32_t v); #define radeon_dpm_force_performance_level(rdev, l) rdev->asic->dpm.force_performance_level((rdev), (l)) #define radeon_dpm_vblank_too_short(rdev) rdev->asic->dpm.vblank_too_short((rdev)) #define radeon_dpm_powergate_uvd(rdev, g) rdev->asic->dpm.powergate_uvd((rdev), (g)) +#define radeon_dpm_enable_bapm(rdev, e) rdev->asic->dpm.enable_bapm((rdev), (e)) /* Common functions */ /* AGP */ diff --git a/drivers/gpu/drm/radeon/radeon_asic.c b/drivers/gpu/drm/radeon/radeon_asic.c index 630853b9684..5003385a751 100644 --- a/drivers/gpu/drm/radeon/radeon_asic.c +++ b/drivers/gpu/drm/radeon/radeon_asic.c @@ -1037,6 +1037,7 @@ static struct radeon_asic rv6xx_asic = { .set_pcie_lanes = &r600_set_pcie_lanes, .set_clock_gating = NULL, .get_temperature = &rv6xx_get_temp, + .set_uvd_clocks = &r600_set_uvd_clocks, }, .dpm = { .init = &rv6xx_dpm_init, @@ -1126,6 +1127,7 @@ static struct radeon_asic rs780_asic = { .set_pcie_lanes = NULL, .set_clock_gating = NULL, .get_temperature = &rv6xx_get_temp, + .set_uvd_clocks = &r600_set_uvd_clocks, }, .dpm = { .init = &rs780_dpm_init, @@ -1141,6 +1143,7 @@ static struct radeon_asic rs780_asic = { .get_mclk = &rs780_dpm_get_mclk, .print_power_state = &rs780_dpm_print_power_state, .debugfs_print_current_performance_level = &rs780_dpm_debugfs_print_current_performance_level, + .force_performance_level = &rs780_dpm_force_performance_level, }, .pflip = { .pre_page_flip = &rs600_pre_page_flip, @@ -1791,6 +1794,7 @@ static struct radeon_asic trinity_asic = { .print_power_state = &trinity_dpm_print_power_state, .debugfs_print_current_performance_level = &trinity_dpm_debugfs_print_current_performance_level, .force_performance_level = &trinity_dpm_force_performance_level, + .enable_bapm = &trinity_dpm_enable_bapm, }, .pflip = { .pre_page_flip = &evergreen_pre_page_flip, @@ -2166,6 +2170,7 @@ static struct radeon_asic kv_asic = { .debugfs_print_current_performance_level = &kv_dpm_debugfs_print_current_performance_level, .force_performance_level = &kv_dpm_force_performance_level, .powergate_uvd = &kv_dpm_powergate_uvd, + .enable_bapm = &kv_dpm_enable_bapm, }, .pflip = { .pre_page_flip = &evergreen_pre_page_flip, @@ -2390,7 +2395,7 @@ int radeon_asic_init(struct radeon_device *rdev) RADEON_CG_SUPPORT_HDP_LS | RADEON_CG_SUPPORT_HDP_MGCG; rdev->pg_flags = 0 | - /*RADEON_PG_SUPPORT_GFX_CG | */ + /*RADEON_PG_SUPPORT_GFX_PG | */ RADEON_PG_SUPPORT_SDMA; break; case CHIP_OLAND: @@ -2479,7 +2484,7 @@ int radeon_asic_init(struct radeon_device *rdev) RADEON_CG_SUPPORT_HDP_LS | RADEON_CG_SUPPORT_HDP_MGCG; rdev->pg_flags = 0; - /*RADEON_PG_SUPPORT_GFX_CG | + /*RADEON_PG_SUPPORT_GFX_PG | RADEON_PG_SUPPORT_GFX_SMG | RADEON_PG_SUPPORT_GFX_DMG | RADEON_PG_SUPPORT_UVD | @@ -2507,7 +2512,7 @@ int radeon_asic_init(struct radeon_device *rdev) RADEON_CG_SUPPORT_HDP_LS | RADEON_CG_SUPPORT_HDP_MGCG; rdev->pg_flags = 0; - /*RADEON_PG_SUPPORT_GFX_CG | + /*RADEON_PG_SUPPORT_GFX_PG | RADEON_PG_SUPPORT_GFX_SMG | RADEON_PG_SUPPORT_UVD | RADEON_PG_SUPPORT_VCE | diff --git a/drivers/gpu/drm/radeon/radeon_asic.h b/drivers/gpu/drm/radeon/radeon_asic.h index 818bbe6b884..70c29d5e080 100644 --- a/drivers/gpu/drm/radeon/radeon_asic.h +++ b/drivers/gpu/drm/radeon/radeon_asic.h @@ -389,6 +389,7 @@ int r600_mc_wait_for_idle(struct radeon_device *rdev); u32 r600_get_xclk(struct radeon_device *rdev); uint64_t r600_get_gpu_clock_counter(struct radeon_device *rdev); int rv6xx_get_temp(struct radeon_device *rdev); +int r600_set_uvd_clocks(struct radeon_device *rdev, u32 vclk, u32 dclk); int r600_dpm_pre_set_power_state(struct radeon_device *rdev); void r600_dpm_post_set_power_state(struct radeon_device *rdev); /* r600 dma */ @@ -428,6 +429,8 @@ void rs780_dpm_print_power_state(struct radeon_device *rdev, struct radeon_ps *ps); void rs780_dpm_debugfs_print_current_performance_level(struct radeon_device *rdev, struct seq_file *m); +int rs780_dpm_force_performance_level(struct radeon_device *rdev, + enum radeon_dpm_forced_level level); /* * rv770,rv730,rv710,rv740 @@ -625,6 +628,7 @@ void trinity_dpm_debugfs_print_current_performance_level(struct radeon_device *r struct seq_file *m); int trinity_dpm_force_performance_level(struct radeon_device *rdev, enum radeon_dpm_forced_level level); +void trinity_dpm_enable_bapm(struct radeon_device *rdev, bool enable); /* DCE6 - SI */ void dce6_bandwidth_update(struct radeon_device *rdev); @@ -781,6 +785,7 @@ void kv_dpm_debugfs_print_current_performance_level(struct radeon_device *rdev, int kv_dpm_force_performance_level(struct radeon_device *rdev, enum radeon_dpm_forced_level level); void kv_dpm_powergate_uvd(struct radeon_device *rdev, bool gate); +void kv_dpm_enable_bapm(struct radeon_device *rdev, bool enable); /* uvd v1.0 */ uint32_t uvd_v1_0_get_rptr(struct radeon_device *rdev, diff --git a/drivers/gpu/drm/radeon/radeon_connectors.c b/drivers/gpu/drm/radeon/radeon_connectors.c index 2399f25ec03..79159b5da05 100644 --- a/drivers/gpu/drm/radeon/radeon_connectors.c +++ b/drivers/gpu/drm/radeon/radeon_connectors.c @@ -396,6 +396,21 @@ static int radeon_connector_set_property(struct drm_connector *connector, struct } } + if (property == rdev->mode_info.audio_property) { + struct radeon_connector *radeon_connector = to_radeon_connector(connector); + /* need to find digital encoder on connector */ + encoder = radeon_find_encoder(connector, DRM_MODE_ENCODER_TMDS); + if (!encoder) + return 0; + + radeon_encoder = to_radeon_encoder(encoder); + + if (radeon_connector->audio != val) { + radeon_connector->audio = val; + radeon_property_change_mode(&radeon_encoder->base); + } + } + if (property == rdev->mode_info.underscan_property) { /* need to find digital encoder on connector */ encoder = radeon_find_encoder(connector, DRM_MODE_ENCODER_TMDS); @@ -1420,7 +1435,7 @@ radeon_dp_detect(struct drm_connector *connector, bool force) if (radeon_dp_getdpcd(radeon_connector)) ret = connector_status_connected; } else { - /* try non-aux ddc (DP to DVI/HMDI/etc. adapter) */ + /* try non-aux ddc (DP to DVI/HDMI/etc. adapter) */ if (radeon_ddc_probe(radeon_connector, false)) ret = connector_status_connected; } @@ -1489,6 +1504,24 @@ static const struct drm_connector_funcs radeon_dp_connector_funcs = { .force = radeon_dvi_force, }; +static const struct drm_connector_funcs radeon_edp_connector_funcs = { + .dpms = drm_helper_connector_dpms, + .detect = radeon_dp_detect, + .fill_modes = drm_helper_probe_single_connector_modes, + .set_property = radeon_lvds_set_property, + .destroy = radeon_dp_connector_destroy, + .force = radeon_dvi_force, +}; + +static const struct drm_connector_funcs radeon_lvds_bridge_connector_funcs = { + .dpms = drm_helper_connector_dpms, + .detect = radeon_dp_detect, + .fill_modes = drm_helper_probe_single_connector_modes, + .set_property = radeon_lvds_set_property, + .destroy = radeon_dp_connector_destroy, + .force = radeon_dvi_force, +}; + void radeon_add_atom_connector(struct drm_device *dev, uint32_t connector_id, @@ -1580,8 +1613,6 @@ radeon_add_atom_connector(struct drm_device *dev, goto failed; radeon_dig_connector->igp_lane_info = igp_lane_info; radeon_connector->con_priv = radeon_dig_connector; - drm_connector_init(dev, &radeon_connector->base, &radeon_dp_connector_funcs, connector_type); - drm_connector_helper_add(&radeon_connector->base, &radeon_dp_connector_helper_funcs); if (i2c_bus->valid) { /* add DP i2c bus */ if (connector_type == DRM_MODE_CONNECTOR_eDP) @@ -1598,6 +1629,10 @@ radeon_add_atom_connector(struct drm_device *dev, case DRM_MODE_CONNECTOR_VGA: case DRM_MODE_CONNECTOR_DVIA: default: + drm_connector_init(dev, &radeon_connector->base, + &radeon_dp_connector_funcs, connector_type); + drm_connector_helper_add(&radeon_connector->base, + &radeon_dp_connector_helper_funcs); connector->interlace_allowed = true; connector->doublescan_allowed = true; radeon_connector->dac_load_detect = true; @@ -1610,6 +1645,10 @@ radeon_add_atom_connector(struct drm_device *dev, case DRM_MODE_CONNECTOR_HDMIA: case DRM_MODE_CONNECTOR_HDMIB: case DRM_MODE_CONNECTOR_DisplayPort: + drm_connector_init(dev, &radeon_connector->base, + &radeon_dp_connector_funcs, connector_type); + drm_connector_helper_add(&radeon_connector->base, + &radeon_dp_connector_helper_funcs); drm_object_attach_property(&radeon_connector->base.base, rdev->mode_info.underscan_property, UNDERSCAN_OFF); @@ -1619,6 +1658,9 @@ radeon_add_atom_connector(struct drm_device *dev, drm_object_attach_property(&radeon_connector->base.base, rdev->mode_info.underscan_vborder_property, 0); + drm_object_attach_property(&radeon_connector->base.base, + rdev->mode_info.audio_property, + RADEON_AUDIO_DISABLE); subpixel_order = SubPixelHorizontalRGB; connector->interlace_allowed = true; if (connector_type == DRM_MODE_CONNECTOR_HDMIB) @@ -1634,6 +1676,10 @@ radeon_add_atom_connector(struct drm_device *dev, break; case DRM_MODE_CONNECTOR_LVDS: case DRM_MODE_CONNECTOR_eDP: + drm_connector_init(dev, &radeon_connector->base, + &radeon_lvds_bridge_connector_funcs, connector_type); + drm_connector_helper_add(&radeon_connector->base, + &radeon_dp_connector_helper_funcs); drm_object_attach_property(&radeon_connector->base.base, dev->mode_config.scaling_mode_property, DRM_MODE_SCALE_FULLSCREEN); @@ -1708,6 +1754,11 @@ radeon_add_atom_connector(struct drm_device *dev, rdev->mode_info.underscan_vborder_property, 0); } + if (ASIC_IS_DCE2(rdev)) { + drm_object_attach_property(&radeon_connector->base.base, + rdev->mode_info.audio_property, + RADEON_AUDIO_DISABLE); + } if (connector_type == DRM_MODE_CONNECTOR_DVII) { radeon_connector->dac_load_detect = true; drm_object_attach_property(&radeon_connector->base.base, @@ -1748,6 +1799,11 @@ radeon_add_atom_connector(struct drm_device *dev, rdev->mode_info.underscan_vborder_property, 0); } + if (ASIC_IS_DCE2(rdev)) { + drm_object_attach_property(&radeon_connector->base.base, + rdev->mode_info.audio_property, + RADEON_AUDIO_DISABLE); + } subpixel_order = SubPixelHorizontalRGB; connector->interlace_allowed = true; if (connector_type == DRM_MODE_CONNECTOR_HDMIB) @@ -1787,6 +1843,11 @@ radeon_add_atom_connector(struct drm_device *dev, rdev->mode_info.underscan_vborder_property, 0); } + if (ASIC_IS_DCE2(rdev)) { + drm_object_attach_property(&radeon_connector->base.base, + rdev->mode_info.audio_property, + RADEON_AUDIO_DISABLE); + } connector->interlace_allowed = true; /* in theory with a DP to VGA converter... */ connector->doublescan_allowed = false; @@ -1797,7 +1858,7 @@ radeon_add_atom_connector(struct drm_device *dev, goto failed; radeon_dig_connector->igp_lane_info = igp_lane_info; radeon_connector->con_priv = radeon_dig_connector; - drm_connector_init(dev, &radeon_connector->base, &radeon_dp_connector_funcs, connector_type); + drm_connector_init(dev, &radeon_connector->base, &radeon_edp_connector_funcs, connector_type); drm_connector_helper_add(&radeon_connector->base, &radeon_dp_connector_helper_funcs); if (i2c_bus->valid) { /* add DP i2c bus */ diff --git a/drivers/gpu/drm/radeon/radeon_cs.c b/drivers/gpu/drm/radeon/radeon_cs.c index a5608441037..ac6ece61a47 100644 --- a/drivers/gpu/drm/radeon/radeon_cs.c +++ b/drivers/gpu/drm/radeon/radeon_cs.c @@ -28,6 +28,7 @@ #include <drm/radeon_drm.h> #include "radeon_reg.h" #include "radeon.h" +#include "radeon_trace.h" static int radeon_cs_parser_relocs(struct radeon_cs_parser *p) { @@ -80,9 +81,11 @@ static int radeon_cs_parser_relocs(struct radeon_cs_parser *p) p->relocs[i].lobj.bo = p->relocs[i].robj; p->relocs[i].lobj.written = !!r->write_domain; - /* the first reloc of an UVD job is the - msg and that must be in VRAM */ - if (p->ring == R600_RING_TYPE_UVD_INDEX && i == 0) { + /* the first reloc of an UVD job is the msg and that must be in + VRAM, also but everything into VRAM on AGP cards to avoid + image corruptions */ + if (p->ring == R600_RING_TYPE_UVD_INDEX && + (i == 0 || p->rdev->flags & RADEON_IS_AGP)) { /* TODO: is this still needed for NI+ ? */ p->relocs[i].lobj.domain = RADEON_GEM_DOMAIN_VRAM; @@ -559,6 +562,8 @@ int radeon_cs_ioctl(struct drm_device *dev, void *data, struct drm_file *filp) return r; } + trace_radeon_cs(&parser); + r = radeon_cs_ib_chunk(rdev, &parser); if (r) { goto out; diff --git a/drivers/gpu/drm/radeon/radeon_device.c b/drivers/gpu/drm/radeon/radeon_device.c index 16cb8792b1e..e29faa73b57 100644 --- a/drivers/gpu/drm/radeon/radeon_device.c +++ b/drivers/gpu/drm/radeon/radeon_device.c @@ -1249,6 +1249,17 @@ int radeon_device_init(struct radeon_device *rdev, /* Registers mapping */ /* TODO: block userspace mapping of io register */ spin_lock_init(&rdev->mmio_idx_lock); + spin_lock_init(&rdev->smc_idx_lock); + spin_lock_init(&rdev->pll_idx_lock); + spin_lock_init(&rdev->mc_idx_lock); + spin_lock_init(&rdev->pcie_idx_lock); + spin_lock_init(&rdev->pciep_idx_lock); + spin_lock_init(&rdev->pif_idx_lock); + spin_lock_init(&rdev->cg_idx_lock); + spin_lock_init(&rdev->uvd_idx_lock); + spin_lock_init(&rdev->rcu_idx_lock); + spin_lock_init(&rdev->didt_idx_lock); + spin_lock_init(&rdev->end_idx_lock); if (rdev->family >= CHIP_BONAIRE) { rdev->rmmio_base = pci_resource_start(rdev->pdev, 5); rdev->rmmio_size = pci_resource_len(rdev->pdev, 5); diff --git a/drivers/gpu/drm/radeon/radeon_display.c b/drivers/gpu/drm/radeon/radeon_display.c index b055bddaa94..0d1aa050d41 100644 --- a/drivers/gpu/drm/radeon/radeon_display.c +++ b/drivers/gpu/drm/radeon/radeon_display.c @@ -1172,6 +1172,12 @@ static struct drm_prop_enum_list radeon_underscan_enum_list[] = { UNDERSCAN_AUTO, "auto" }, }; +static struct drm_prop_enum_list radeon_audio_enum_list[] = +{ { RADEON_AUDIO_DISABLE, "off" }, + { RADEON_AUDIO_ENABLE, "on" }, + { RADEON_AUDIO_AUTO, "auto" }, +}; + static int radeon_modeset_create_props(struct radeon_device *rdev) { int sz; @@ -1222,6 +1228,12 @@ static int radeon_modeset_create_props(struct radeon_device *rdev) if (!rdev->mode_info.underscan_vborder_property) return -ENOMEM; + sz = ARRAY_SIZE(radeon_audio_enum_list); + rdev->mode_info.audio_property = + drm_property_create_enum(rdev->ddev, 0, + "audio", + radeon_audio_enum_list, sz); + return 0; } diff --git a/drivers/gpu/drm/radeon/radeon_drv.c b/drivers/gpu/drm/radeon/radeon_drv.c index cb4445f55a9..cdd12dcd988 100644 --- a/drivers/gpu/drm/radeon/radeon_drv.c +++ b/drivers/gpu/drm/radeon/radeon_drv.c @@ -153,7 +153,7 @@ int radeon_benchmarking = 0; int radeon_testing = 0; int radeon_connector_table = 0; int radeon_tv = 1; -int radeon_audio = 0; +int radeon_audio = 1; int radeon_disp_priority = 0; int radeon_hw_i2c = 0; int radeon_pcie_gen2 = -1; diff --git a/drivers/gpu/drm/radeon/radeon_mode.h b/drivers/gpu/drm/radeon/radeon_mode.h index d908d8d68f6..ef63d3f00b2 100644 --- a/drivers/gpu/drm/radeon/radeon_mode.h +++ b/drivers/gpu/drm/radeon/radeon_mode.h @@ -247,6 +247,8 @@ struct radeon_mode_info { struct drm_property *underscan_property; struct drm_property *underscan_hborder_property; struct drm_property *underscan_vborder_property; + /* audio */ + struct drm_property *audio_property; /* hardcoded DFP edid from BIOS */ struct edid *bios_hardcoded_edid; int bios_hardcoded_edid_size; @@ -471,6 +473,12 @@ struct radeon_router { u8 cd_mux_state; }; +enum radeon_connector_audio { + RADEON_AUDIO_DISABLE = 0, + RADEON_AUDIO_ENABLE = 1, + RADEON_AUDIO_AUTO = 2 +}; + struct radeon_connector { struct drm_connector base; uint32_t connector_id; @@ -489,6 +497,7 @@ struct radeon_connector { struct radeon_hpd hpd; struct radeon_router router; struct radeon_i2c_chan *router_bus; + enum radeon_connector_audio audio; }; struct radeon_framebuffer { diff --git a/drivers/gpu/drm/radeon/radeon_pm.c b/drivers/gpu/drm/radeon/radeon_pm.c index d7555369a3e..87e1d69e8fd 100644 --- a/drivers/gpu/drm/radeon/radeon_pm.c +++ b/drivers/gpu/drm/radeon/radeon_pm.c @@ -67,7 +67,16 @@ int radeon_pm_get_type_index(struct radeon_device *rdev, void radeon_pm_acpi_event_handler(struct radeon_device *rdev) { - if (rdev->pm.pm_method == PM_METHOD_PROFILE) { + if ((rdev->pm.pm_method == PM_METHOD_DPM) && rdev->pm.dpm_enabled) { + mutex_lock(&rdev->pm.mutex); + if (power_supply_is_system_supplied() > 0) + rdev->pm.dpm.ac_power = true; + else + rdev->pm.dpm.ac_power = false; + if (rdev->asic->dpm.enable_bapm) + radeon_dpm_enable_bapm(rdev, rdev->pm.dpm.ac_power); + mutex_unlock(&rdev->pm.mutex); + } else if (rdev->pm.pm_method == PM_METHOD_PROFILE) { if (rdev->pm.profile == PM_PROFILE_AUTO) { mutex_lock(&rdev->pm.mutex); radeon_pm_update_profile(rdev); @@ -333,7 +342,7 @@ static ssize_t radeon_get_pm_profile(struct device *dev, struct device_attribute *attr, char *buf) { - struct drm_device *ddev = pci_get_drvdata(to_pci_dev(dev)); + struct drm_device *ddev = dev_get_drvdata(dev); struct radeon_device *rdev = ddev->dev_private; int cp = rdev->pm.profile; @@ -349,7 +358,7 @@ static ssize_t radeon_set_pm_profile(struct device *dev, const char *buf, size_t count) { - struct drm_device *ddev = pci_get_drvdata(to_pci_dev(dev)); + struct drm_device *ddev = dev_get_drvdata(dev); struct radeon_device *rdev = ddev->dev_private; mutex_lock(&rdev->pm.mutex); @@ -383,7 +392,7 @@ static ssize_t radeon_get_pm_method(struct device *dev, struct device_attribute *attr, char *buf) { - struct drm_device *ddev = pci_get_drvdata(to_pci_dev(dev)); + struct drm_device *ddev = dev_get_drvdata(dev); struct radeon_device *rdev = ddev->dev_private; int pm = rdev->pm.pm_method; @@ -397,7 +406,7 @@ static ssize_t radeon_set_pm_method(struct device *dev, const char *buf, size_t count) { - struct drm_device *ddev = pci_get_drvdata(to_pci_dev(dev)); + struct drm_device *ddev = dev_get_drvdata(dev); struct radeon_device *rdev = ddev->dev_private; /* we don't support the legacy modes with dpm */ @@ -433,7 +442,7 @@ static ssize_t radeon_get_dpm_state(struct device *dev, struct device_attribute *attr, char *buf) { - struct drm_device *ddev = pci_get_drvdata(to_pci_dev(dev)); + struct drm_device *ddev = dev_get_drvdata(dev); struct radeon_device *rdev = ddev->dev_private; enum radeon_pm_state_type pm = rdev->pm.dpm.user_state; @@ -447,7 +456,7 @@ static ssize_t radeon_set_dpm_state(struct device *dev, const char *buf, size_t count) { - struct drm_device *ddev = pci_get_drvdata(to_pci_dev(dev)); + struct drm_device *ddev = dev_get_drvdata(dev); struct radeon_device *rdev = ddev->dev_private; mutex_lock(&rdev->pm.mutex); @@ -472,7 +481,7 @@ static ssize_t radeon_get_dpm_forced_performance_level(struct device *dev, struct device_attribute *attr, char *buf) { - struct drm_device *ddev = pci_get_drvdata(to_pci_dev(dev)); + struct drm_device *ddev = dev_get_drvdata(dev); struct radeon_device *rdev = ddev->dev_private; enum radeon_dpm_forced_level level = rdev->pm.dpm.forced_level; @@ -486,7 +495,7 @@ static ssize_t radeon_set_dpm_forced_performance_level(struct device *dev, const char *buf, size_t count) { - struct drm_device *ddev = pci_get_drvdata(to_pci_dev(dev)); + struct drm_device *ddev = dev_get_drvdata(dev); struct radeon_device *rdev = ddev->dev_private; enum radeon_dpm_forced_level level; int ret = 0; @@ -524,7 +533,7 @@ static ssize_t radeon_hwmon_show_temp(struct device *dev, struct device_attribute *attr, char *buf) { - struct drm_device *ddev = pci_get_drvdata(to_pci_dev(dev)); + struct drm_device *ddev = dev_get_drvdata(dev); struct radeon_device *rdev = ddev->dev_private; int temp; @@ -536,6 +545,23 @@ static ssize_t radeon_hwmon_show_temp(struct device *dev, return snprintf(buf, PAGE_SIZE, "%d\n", temp); } +static ssize_t radeon_hwmon_show_temp_thresh(struct device *dev, + struct device_attribute *attr, + char *buf) +{ + struct drm_device *ddev = dev_get_drvdata(dev); + struct radeon_device *rdev = ddev->dev_private; + int hyst = to_sensor_dev_attr(attr)->index; + int temp; + + if (hyst) + temp = rdev->pm.dpm.thermal.min_temp; + else + temp = rdev->pm.dpm.thermal.max_temp; + + return snprintf(buf, PAGE_SIZE, "%d\n", temp); +} + static ssize_t radeon_hwmon_show_name(struct device *dev, struct device_attribute *attr, char *buf) @@ -544,16 +570,37 @@ static ssize_t radeon_hwmon_show_name(struct device *dev, } static SENSOR_DEVICE_ATTR(temp1_input, S_IRUGO, radeon_hwmon_show_temp, NULL, 0); +static SENSOR_DEVICE_ATTR(temp1_crit, S_IRUGO, radeon_hwmon_show_temp_thresh, NULL, 0); +static SENSOR_DEVICE_ATTR(temp1_crit_hyst, S_IRUGO, radeon_hwmon_show_temp_thresh, NULL, 1); static SENSOR_DEVICE_ATTR(name, S_IRUGO, radeon_hwmon_show_name, NULL, 0); static struct attribute *hwmon_attributes[] = { &sensor_dev_attr_temp1_input.dev_attr.attr, + &sensor_dev_attr_temp1_crit.dev_attr.attr, + &sensor_dev_attr_temp1_crit_hyst.dev_attr.attr, &sensor_dev_attr_name.dev_attr.attr, NULL }; +static umode_t hwmon_attributes_visible(struct kobject *kobj, + struct attribute *attr, int index) +{ + struct device *dev = container_of(kobj, struct device, kobj); + struct drm_device *ddev = dev_get_drvdata(dev); + struct radeon_device *rdev = ddev->dev_private; + + /* Skip limit attributes if DPM is not enabled */ + if (rdev->pm.pm_method != PM_METHOD_DPM && + (attr == &sensor_dev_attr_temp1_crit.dev_attr.attr || + attr == &sensor_dev_attr_temp1_crit_hyst.dev_attr.attr)) + return 0; + + return attr->mode; +} + static const struct attribute_group hwmon_attrgroup = { .attrs = hwmon_attributes, + .is_visible = hwmon_attributes_visible, }; static int radeon_hwmon_init(struct radeon_device *rdev) @@ -870,10 +917,13 @@ static void radeon_dpm_change_power_state_locked(struct radeon_device *rdev) radeon_dpm_post_set_power_state(rdev); - /* force low perf level for thermal */ - if (rdev->pm.dpm.thermal_active && - rdev->asic->dpm.force_performance_level) { - radeon_dpm_force_performance_level(rdev, RADEON_DPM_FORCED_LEVEL_LOW); + if (rdev->asic->dpm.force_performance_level) { + if (rdev->pm.dpm.thermal_active) + /* force low perf level for thermal */ + radeon_dpm_force_performance_level(rdev, RADEON_DPM_FORCED_LEVEL_LOW); + else + /* otherwise, enable auto */ + radeon_dpm_force_performance_level(rdev, RADEON_DPM_FORCED_LEVEL_AUTO); } done: @@ -1102,9 +1152,10 @@ static int radeon_pm_init_dpm(struct radeon_device *rdev) { int ret; - /* default to performance state */ + /* default to balanced state */ rdev->pm.dpm.state = POWER_STATE_TYPE_BALANCED; rdev->pm.dpm.user_state = POWER_STATE_TYPE_BALANCED; + rdev->pm.dpm.forced_level = RADEON_DPM_FORCED_LEVEL_AUTO; rdev->pm.default_sclk = rdev->clock.default_sclk; rdev->pm.default_mclk = rdev->clock.default_mclk; rdev->pm.current_sclk = rdev->clock.default_sclk; diff --git a/drivers/gpu/drm/radeon/radeon_trace.h b/drivers/gpu/drm/radeon/radeon_trace.h index eafd8160a15..f7e36781596 100644 --- a/drivers/gpu/drm/radeon/radeon_trace.h +++ b/drivers/gpu/drm/radeon/radeon_trace.h @@ -27,6 +27,26 @@ TRACE_EVENT(radeon_bo_create, TP_printk("bo=%p, pages=%u", __entry->bo, __entry->pages) ); +TRACE_EVENT(radeon_cs, + TP_PROTO(struct radeon_cs_parser *p), + TP_ARGS(p), + TP_STRUCT__entry( + __field(u32, ring) + __field(u32, dw) + __field(u32, fences) + ), + + TP_fast_assign( + __entry->ring = p->ring; + __entry->dw = p->chunks[p->chunk_ib_idx].length_dw; + __entry->fences = radeon_fence_count_emitted( + p->rdev, p->ring); + ), + TP_printk("ring=%u, dw=%u, fences=%u", + __entry->ring, __entry->dw, + __entry->fences) +); + DECLARE_EVENT_CLASS(radeon_fence_request, TP_PROTO(struct drm_device *dev, u32 seqno), @@ -53,13 +73,6 @@ DEFINE_EVENT(radeon_fence_request, radeon_fence_emit, TP_ARGS(dev, seqno) ); -DEFINE_EVENT(radeon_fence_request, radeon_fence_retire, - - TP_PROTO(struct drm_device *dev, u32 seqno), - - TP_ARGS(dev, seqno) -); - DEFINE_EVENT(radeon_fence_request, radeon_fence_wait_begin, TP_PROTO(struct drm_device *dev, u32 seqno), diff --git a/drivers/gpu/drm/radeon/rs400.c b/drivers/gpu/drm/radeon/rs400.c index b8074a8ec75..9566b5940a5 100644 --- a/drivers/gpu/drm/radeon/rs400.c +++ b/drivers/gpu/drm/radeon/rs400.c @@ -274,19 +274,26 @@ static void rs400_mc_init(struct radeon_device *rdev) uint32_t rs400_mc_rreg(struct radeon_device *rdev, uint32_t reg) { + unsigned long flags; uint32_t r; + spin_lock_irqsave(&rdev->mc_idx_lock, flags); WREG32(RS480_NB_MC_INDEX, reg & 0xff); r = RREG32(RS480_NB_MC_DATA); WREG32(RS480_NB_MC_INDEX, 0xff); + spin_unlock_irqrestore(&rdev->mc_idx_lock, flags); return r; } void rs400_mc_wreg(struct radeon_device *rdev, uint32_t reg, uint32_t v) { + unsigned long flags; + + spin_lock_irqsave(&rdev->mc_idx_lock, flags); WREG32(RS480_NB_MC_INDEX, ((reg) & 0xff) | RS480_NB_MC_IND_WR_EN); WREG32(RS480_NB_MC_DATA, (v)); WREG32(RS480_NB_MC_INDEX, 0xff); + spin_unlock_irqrestore(&rdev->mc_idx_lock, flags); } #if defined(CONFIG_DEBUG_FS) diff --git a/drivers/gpu/drm/radeon/rs600.c b/drivers/gpu/drm/radeon/rs600.c index 670b555d2ca..6acba8017b9 100644 --- a/drivers/gpu/drm/radeon/rs600.c +++ b/drivers/gpu/drm/radeon/rs600.c @@ -847,16 +847,26 @@ void rs600_bandwidth_update(struct radeon_device *rdev) uint32_t rs600_mc_rreg(struct radeon_device *rdev, uint32_t reg) { + unsigned long flags; + u32 r; + + spin_lock_irqsave(&rdev->mc_idx_lock, flags); WREG32(R_000070_MC_IND_INDEX, S_000070_MC_IND_ADDR(reg) | S_000070_MC_IND_CITF_ARB0(1)); - return RREG32(R_000074_MC_IND_DATA); + r = RREG32(R_000074_MC_IND_DATA); + spin_unlock_irqrestore(&rdev->mc_idx_lock, flags); + return r; } void rs600_mc_wreg(struct radeon_device *rdev, uint32_t reg, uint32_t v) { + unsigned long flags; + + spin_lock_irqsave(&rdev->mc_idx_lock, flags); WREG32(R_000070_MC_IND_INDEX, S_000070_MC_IND_ADDR(reg) | S_000070_MC_IND_CITF_ARB0(1) | S_000070_MC_IND_WR_EN(1)); WREG32(R_000074_MC_IND_DATA, v); + spin_unlock_irqrestore(&rdev->mc_idx_lock, flags); } static void rs600_debugfs(struct radeon_device *rdev) diff --git a/drivers/gpu/drm/radeon/rs690.c b/drivers/gpu/drm/radeon/rs690.c index d8ddfb34545..1447d794c22 100644 --- a/drivers/gpu/drm/radeon/rs690.c +++ b/drivers/gpu/drm/radeon/rs690.c @@ -631,20 +631,27 @@ void rs690_bandwidth_update(struct radeon_device *rdev) uint32_t rs690_mc_rreg(struct radeon_device *rdev, uint32_t reg) { + unsigned long flags; uint32_t r; + spin_lock_irqsave(&rdev->mc_idx_lock, flags); WREG32(R_000078_MC_INDEX, S_000078_MC_IND_ADDR(reg)); r = RREG32(R_00007C_MC_DATA); WREG32(R_000078_MC_INDEX, ~C_000078_MC_IND_ADDR); + spin_unlock_irqrestore(&rdev->mc_idx_lock, flags); return r; } void rs690_mc_wreg(struct radeon_device *rdev, uint32_t reg, uint32_t v) { + unsigned long flags; + + spin_lock_irqsave(&rdev->mc_idx_lock, flags); WREG32(R_000078_MC_INDEX, S_000078_MC_IND_ADDR(reg) | S_000078_MC_IND_WR_EN(1)); WREG32(R_00007C_MC_DATA, v); WREG32(R_000078_MC_INDEX, 0x7F); + spin_unlock_irqrestore(&rdev->mc_idx_lock, flags); } static void rs690_mc_program(struct radeon_device *rdev) diff --git a/drivers/gpu/drm/radeon/rs780_dpm.c b/drivers/gpu/drm/radeon/rs780_dpm.c index d1a1ce73bd4..6af8505cf4d 100644 --- a/drivers/gpu/drm/radeon/rs780_dpm.c +++ b/drivers/gpu/drm/radeon/rs780_dpm.c @@ -62,9 +62,7 @@ static void rs780_get_pm_mode_parameters(struct radeon_device *rdev) radeon_crtc = to_radeon_crtc(crtc); pi->crtc_id = radeon_crtc->crtc_id; if (crtc->mode.htotal && crtc->mode.vtotal) - pi->refresh_rate = - (crtc->mode.clock * 1000) / - (crtc->mode.htotal * crtc->mode.vtotal); + pi->refresh_rate = drm_mode_vrefresh(&crtc->mode); break; } } @@ -376,9 +374,8 @@ static void rs780_disable_vbios_powersaving(struct radeon_device *rdev) WREG32_P(CG_INTGFX_MISC, 0, ~0xFFF00000); } -static void rs780_force_voltage_to_high(struct radeon_device *rdev) +static void rs780_force_voltage(struct radeon_device *rdev, u16 voltage) { - struct igp_power_info *pi = rs780_get_pi(rdev); struct igp_ps *current_state = rs780_get_ps(rdev->pm.dpm.current_ps); if ((current_state->max_voltage == RS780_VDDC_LEVEL_HIGH) && @@ -390,7 +387,7 @@ static void rs780_force_voltage_to_high(struct radeon_device *rdev) udelay(1); WREG32_P(FVTHROT_PWM_CTRL_REG0, - STARTING_PWM_HIGHTIME(pi->max_voltage), + STARTING_PWM_HIGHTIME(voltage), ~STARTING_PWM_HIGHTIME_MASK); WREG32_P(FVTHROT_PWM_CTRL_REG0, @@ -404,6 +401,26 @@ static void rs780_force_voltage_to_high(struct radeon_device *rdev) WREG32_P(GFX_MACRO_BYPASS_CNTL, 0, ~SPLL_BYPASS_CNTL); } +static void rs780_force_fbdiv(struct radeon_device *rdev, u32 fb_div) +{ + struct igp_ps *current_state = rs780_get_ps(rdev->pm.dpm.current_ps); + + if (current_state->sclk_low == current_state->sclk_high) + return; + + WREG32_P(GFX_MACRO_BYPASS_CNTL, SPLL_BYPASS_CNTL, ~SPLL_BYPASS_CNTL); + + WREG32_P(FVTHROT_FBDIV_REG2, FORCED_FEEDBACK_DIV(fb_div), + ~FORCED_FEEDBACK_DIV_MASK); + WREG32_P(FVTHROT_FBDIV_REG1, STARTING_FEEDBACK_DIV(fb_div), + ~STARTING_FEEDBACK_DIV_MASK); + WREG32_P(FVTHROT_FBDIV_REG1, FORCE_FEEDBACK_DIV, ~FORCE_FEEDBACK_DIV); + + udelay(100); + + WREG32_P(GFX_MACRO_BYPASS_CNTL, 0, ~SPLL_BYPASS_CNTL); +} + static int rs780_set_engine_clock_scaling(struct radeon_device *rdev, struct radeon_ps *new_ps, struct radeon_ps *old_ps) @@ -432,17 +449,13 @@ static int rs780_set_engine_clock_scaling(struct radeon_device *rdev, if (ret) return ret; - WREG32_P(GFX_MACRO_BYPASS_CNTL, SPLL_BYPASS_CNTL, ~SPLL_BYPASS_CNTL); - - WREG32_P(FVTHROT_FBDIV_REG2, FORCED_FEEDBACK_DIV(max_dividers.fb_div), - ~FORCED_FEEDBACK_DIV_MASK); - WREG32_P(FVTHROT_FBDIV_REG1, STARTING_FEEDBACK_DIV(max_dividers.fb_div), - ~STARTING_FEEDBACK_DIV_MASK); - WREG32_P(FVTHROT_FBDIV_REG1, FORCE_FEEDBACK_DIV, ~FORCE_FEEDBACK_DIV); - - udelay(100); + if ((min_dividers.ref_div != max_dividers.ref_div) || + (min_dividers.post_div != max_dividers.post_div) || + (max_dividers.ref_div != current_max_dividers.ref_div) || + (max_dividers.post_div != current_max_dividers.post_div)) + return -EINVAL; - WREG32_P(GFX_MACRO_BYPASS_CNTL, 0, ~SPLL_BYPASS_CNTL); + rs780_force_fbdiv(rdev, max_dividers.fb_div); if (max_dividers.fb_div > min_dividers.fb_div) { WREG32_P(FVTHROT_FBDIV_REG0, @@ -486,6 +499,9 @@ static void rs780_activate_engine_clk_scaling(struct radeon_device *rdev, (new_state->sclk_low == old_state->sclk_low)) return; + if (new_state->sclk_high == new_state->sclk_low) + return; + rs780_clk_scaling_enable(rdev, true); } @@ -649,7 +665,7 @@ int rs780_dpm_set_power_state(struct radeon_device *rdev) rs780_set_uvd_clock_before_set_eng_clock(rdev, new_ps, old_ps); if (pi->voltage_control) { - rs780_force_voltage_to_high(rdev); + rs780_force_voltage(rdev, pi->max_voltage); mdelay(5); } @@ -717,14 +733,18 @@ static void rs780_parse_pplib_non_clock_info(struct radeon_device *rdev, if (ATOM_PPLIB_NONCLOCKINFO_VER1 < table_rev) { rps->vclk = le32_to_cpu(non_clock_info->ulVCLK); rps->dclk = le32_to_cpu(non_clock_info->ulDCLK); - } else if (r600_is_uvd_state(rps->class, rps->class2)) { - rps->vclk = RS780_DEFAULT_VCLK_FREQ; - rps->dclk = RS780_DEFAULT_DCLK_FREQ; } else { rps->vclk = 0; rps->dclk = 0; } + if (r600_is_uvd_state(rps->class, rps->class2)) { + if ((rps->vclk == 0) || (rps->dclk == 0)) { + rps->vclk = RS780_DEFAULT_VCLK_FREQ; + rps->dclk = RS780_DEFAULT_DCLK_FREQ; + } + } + if (rps->class & ATOM_PPLIB_CLASSIFICATION_BOOT) rdev->pm.dpm.boot_ps = rps; if (rps->class & ATOM_PPLIB_CLASSIFICATION_UVDSTATE) @@ -986,3 +1006,55 @@ void rs780_dpm_debugfs_print_current_performance_level(struct radeon_device *rde seq_printf(m, "power level 1 sclk: %u vddc_index: %d\n", ps->sclk_high, ps->max_voltage); } + +int rs780_dpm_force_performance_level(struct radeon_device *rdev, + enum radeon_dpm_forced_level level) +{ + struct igp_power_info *pi = rs780_get_pi(rdev); + struct radeon_ps *rps = rdev->pm.dpm.current_ps; + struct igp_ps *ps = rs780_get_ps(rps); + struct atom_clock_dividers dividers; + int ret; + + rs780_clk_scaling_enable(rdev, false); + rs780_voltage_scaling_enable(rdev, false); + + if (level == RADEON_DPM_FORCED_LEVEL_HIGH) { + if (pi->voltage_control) + rs780_force_voltage(rdev, pi->max_voltage); + + ret = radeon_atom_get_clock_dividers(rdev, COMPUTE_ENGINE_PLL_PARAM, + ps->sclk_high, false, ÷rs); + if (ret) + return ret; + + rs780_force_fbdiv(rdev, dividers.fb_div); + } else if (level == RADEON_DPM_FORCED_LEVEL_LOW) { + ret = radeon_atom_get_clock_dividers(rdev, COMPUTE_ENGINE_PLL_PARAM, + ps->sclk_low, false, ÷rs); + if (ret) + return ret; + + rs780_force_fbdiv(rdev, dividers.fb_div); + + if (pi->voltage_control) + rs780_force_voltage(rdev, pi->min_voltage); + } else { + if (pi->voltage_control) + rs780_force_voltage(rdev, pi->max_voltage); + + if (ps->sclk_high != ps->sclk_low) { + WREG32_P(FVTHROT_FBDIV_REG1, 0, ~FORCE_FEEDBACK_DIV); + rs780_clk_scaling_enable(rdev, true); + } + + if (pi->voltage_control) { + rs780_voltage_scaling_enable(rdev, true); + rs780_enable_voltage_scaling(rdev, rps); + } + } + + rdev->pm.dpm.forced_level = level; + + return 0; +} diff --git a/drivers/gpu/drm/radeon/rv515.c b/drivers/gpu/drm/radeon/rv515.c index 8ea1573ae82..873eb4b193b 100644 --- a/drivers/gpu/drm/radeon/rv515.c +++ b/drivers/gpu/drm/radeon/rv515.c @@ -209,19 +209,27 @@ static void rv515_mc_init(struct radeon_device *rdev) uint32_t rv515_mc_rreg(struct radeon_device *rdev, uint32_t reg) { + unsigned long flags; uint32_t r; + spin_lock_irqsave(&rdev->mc_idx_lock, flags); WREG32(MC_IND_INDEX, 0x7f0000 | (reg & 0xffff)); r = RREG32(MC_IND_DATA); WREG32(MC_IND_INDEX, 0); + spin_unlock_irqrestore(&rdev->mc_idx_lock, flags); + return r; } void rv515_mc_wreg(struct radeon_device *rdev, uint32_t reg, uint32_t v) { + unsigned long flags; + + spin_lock_irqsave(&rdev->mc_idx_lock, flags); WREG32(MC_IND_INDEX, 0xff0000 | ((reg) & 0xffff)); WREG32(MC_IND_DATA, (v)); WREG32(MC_IND_INDEX, 0); + spin_unlock_irqrestore(&rdev->mc_idx_lock, flags); } #if defined(CONFIG_DEBUG_FS) diff --git a/drivers/gpu/drm/radeon/rv6xx_dpm.c b/drivers/gpu/drm/radeon/rv6xx_dpm.c index ab1f2016f21..5811d277a36 100644 --- a/drivers/gpu/drm/radeon/rv6xx_dpm.c +++ b/drivers/gpu/drm/radeon/rv6xx_dpm.c @@ -1758,8 +1758,6 @@ int rv6xx_dpm_set_power_state(struct radeon_device *rdev) rv6xx_set_uvd_clock_after_set_eng_clock(rdev, new_ps, old_ps); - rdev->pm.dpm.forced_level = RADEON_DPM_FORCED_LEVEL_AUTO; - return 0; } diff --git a/drivers/gpu/drm/radeon/rv770_dpm.c b/drivers/gpu/drm/radeon/rv770_dpm.c index 8cbb85dae5a..913b025ae9b 100644 --- a/drivers/gpu/drm/radeon/rv770_dpm.c +++ b/drivers/gpu/drm/radeon/rv770_dpm.c @@ -2064,12 +2064,6 @@ int rv770_dpm_set_power_state(struct radeon_device *rdev) rv770_program_dcodt_after_state_switch(rdev, new_ps, old_ps); rv770_set_uvd_clock_after_set_eng_clock(rdev, new_ps, old_ps); - ret = rv770_dpm_force_performance_level(rdev, RADEON_DPM_FORCED_LEVEL_AUTO); - if (ret) { - DRM_ERROR("rv770_dpm_force_performance_level failed\n"); - return ret; - } - return 0; } @@ -2147,14 +2141,18 @@ static void rv7xx_parse_pplib_non_clock_info(struct radeon_device *rdev, if (ATOM_PPLIB_NONCLOCKINFO_VER1 < table_rev) { rps->vclk = le32_to_cpu(non_clock_info->ulVCLK); rps->dclk = le32_to_cpu(non_clock_info->ulDCLK); - } else if (r600_is_uvd_state(rps->class, rps->class2)) { - rps->vclk = RV770_DEFAULT_VCLK_FREQ; - rps->dclk = RV770_DEFAULT_DCLK_FREQ; } else { rps->vclk = 0; rps->dclk = 0; } + if (r600_is_uvd_state(rps->class, rps->class2)) { + if ((rps->vclk == 0) || (rps->dclk == 0)) { + rps->vclk = RV770_DEFAULT_VCLK_FREQ; + rps->dclk = RV770_DEFAULT_DCLK_FREQ; + } + } + if (rps->class & ATOM_PPLIB_CLASSIFICATION_BOOT) rdev->pm.dpm.boot_ps = rps; if (rps->class & ATOM_PPLIB_CLASSIFICATION_UVDSTATE) diff --git a/drivers/gpu/drm/radeon/rv770_smc.c b/drivers/gpu/drm/radeon/rv770_smc.c index ab95da57021..b2a22440736 100644 --- a/drivers/gpu/drm/radeon/rv770_smc.c +++ b/drivers/gpu/drm/radeon/rv770_smc.c @@ -274,8 +274,8 @@ static const u8 cayman_smc_int_vectors[] = 0x08, 0x72, 0x08, 0x72 }; -int rv770_set_smc_sram_address(struct radeon_device *rdev, - u16 smc_address, u16 limit) +static int rv770_set_smc_sram_address(struct radeon_device *rdev, + u16 smc_address, u16 limit) { u32 addr; @@ -296,9 +296,10 @@ int rv770_copy_bytes_to_smc(struct radeon_device *rdev, u16 smc_start_address, const u8 *src, u16 byte_count, u16 limit) { + unsigned long flags; u32 data, original_data, extra_shift; u16 addr; - int ret; + int ret = 0; if (smc_start_address & 3) return -EINVAL; @@ -307,13 +308,14 @@ int rv770_copy_bytes_to_smc(struct radeon_device *rdev, addr = smc_start_address; + spin_lock_irqsave(&rdev->smc_idx_lock, flags); while (byte_count >= 4) { /* SMC address space is BE */ data = (src[0] << 24) | (src[1] << 16) | (src[2] << 8) | src[3]; ret = rv770_set_smc_sram_address(rdev, addr, limit); if (ret) - return ret; + goto done; WREG32(SMC_SRAM_DATA, data); @@ -328,7 +330,7 @@ int rv770_copy_bytes_to_smc(struct radeon_device *rdev, ret = rv770_set_smc_sram_address(rdev, addr, limit); if (ret) - return ret; + goto done; original_data = RREG32(SMC_SRAM_DATA); @@ -346,12 +348,15 @@ int rv770_copy_bytes_to_smc(struct radeon_device *rdev, ret = rv770_set_smc_sram_address(rdev, addr, limit); if (ret) - return ret; + goto done; WREG32(SMC_SRAM_DATA, data); } - return 0; +done: + spin_unlock_irqrestore(&rdev->smc_idx_lock, flags); + + return ret; } static int rv770_program_interrupt_vectors(struct radeon_device *rdev, @@ -461,12 +466,15 @@ PPSMC_Result rv770_wait_for_smc_inactive(struct radeon_device *rdev) static void rv770_clear_smc_sram(struct radeon_device *rdev, u16 limit) { + unsigned long flags; u16 i; + spin_lock_irqsave(&rdev->smc_idx_lock, flags); for (i = 0; i < limit; i += 4) { rv770_set_smc_sram_address(rdev, i, limit); WREG32(SMC_SRAM_DATA, 0); } + spin_unlock_irqrestore(&rdev->smc_idx_lock, flags); } int rv770_load_smc_ucode(struct radeon_device *rdev, @@ -595,27 +603,29 @@ int rv770_load_smc_ucode(struct radeon_device *rdev, int rv770_read_smc_sram_dword(struct radeon_device *rdev, u16 smc_address, u32 *value, u16 limit) { + unsigned long flags; int ret; + spin_lock_irqsave(&rdev->smc_idx_lock, flags); ret = rv770_set_smc_sram_address(rdev, smc_address, limit); - if (ret) - return ret; - - *value = RREG32(SMC_SRAM_DATA); + if (ret == 0) + *value = RREG32(SMC_SRAM_DATA); + spin_unlock_irqrestore(&rdev->smc_idx_lock, flags); - return 0; + return ret; } int rv770_write_smc_sram_dword(struct radeon_device *rdev, u16 smc_address, u32 value, u16 limit) { + unsigned long flags; int ret; + spin_lock_irqsave(&rdev->smc_idx_lock, flags); ret = rv770_set_smc_sram_address(rdev, smc_address, limit); - if (ret) - return ret; + if (ret == 0) + WREG32(SMC_SRAM_DATA, value); + spin_unlock_irqrestore(&rdev->smc_idx_lock, flags); - WREG32(SMC_SRAM_DATA, value); - - return 0; + return ret; } diff --git a/drivers/gpu/drm/radeon/rv770_smc.h b/drivers/gpu/drm/radeon/rv770_smc.h index f78d92a4b32..3b2c963c488 100644 --- a/drivers/gpu/drm/radeon/rv770_smc.h +++ b/drivers/gpu/drm/radeon/rv770_smc.h @@ -187,8 +187,6 @@ typedef struct RV770_SMC_STATETABLE RV770_SMC_STATETABLE; #define RV770_SMC_SOFT_REGISTER_uvd_enabled 0x9C #define RV770_SMC_SOFT_REGISTER_is_asic_lombok 0xA0 -int rv770_set_smc_sram_address(struct radeon_device *rdev, - u16 smc_address, u16 limit); int rv770_copy_bytes_to_smc(struct radeon_device *rdev, u16 smc_start_address, const u8 *src, u16 byte_count, u16 limit); diff --git a/drivers/gpu/drm/radeon/rv770d.h b/drivers/gpu/drm/radeon/rv770d.h index 9fe60e54292..1ae277152cc 100644 --- a/drivers/gpu/drm/radeon/rv770d.h +++ b/drivers/gpu/drm/radeon/rv770d.h @@ -852,7 +852,7 @@ #define AFMT_VBI_PACKET_CONTROL 0x7608 # define AFMT_GENERIC0_UPDATE (1 << 2) #define AFMT_INFOFRAME_CONTROL0 0x760c -# define AFMT_AUDIO_INFO_SOURCE (1 << 6) /* 0 - sound block; 1 - hmdi regs */ +# define AFMT_AUDIO_INFO_SOURCE (1 << 6) /* 0 - sound block; 1 - hdmi regs */ # define AFMT_AUDIO_INFO_UPDATE (1 << 7) # define AFMT_MPEG_INFO_UPDATE (1 << 10) #define AFMT_GENERIC0_7 0x7610 diff --git a/drivers/gpu/drm/radeon/si.c b/drivers/gpu/drm/radeon/si.c index 3e23b757dcf..c354c109496 100644 --- a/drivers/gpu/drm/radeon/si.c +++ b/drivers/gpu/drm/radeon/si.c @@ -83,6 +83,8 @@ extern void si_dma_vm_set_page(struct radeon_device *rdev, uint64_t pe, uint64_t addr, unsigned count, uint32_t incr, uint32_t flags); +static void si_enable_gui_idle_interrupt(struct radeon_device *rdev, + bool enable); static const u32 verde_rlc_save_restore_register_list[] = { @@ -3386,6 +3388,8 @@ static int si_cp_resume(struct radeon_device *rdev) u32 rb_bufsz; int r; + si_enable_gui_idle_interrupt(rdev, false); + WREG32(CP_SEM_WAIT_TIMER, 0x0); WREG32(CP_SEM_INCOMPLETE_TIMER_CNTL, 0x0); @@ -3501,6 +3505,8 @@ static int si_cp_resume(struct radeon_device *rdev) rdev->ring[CAYMAN_RING_TYPE_CP2_INDEX].ready = false; } + si_enable_gui_idle_interrupt(rdev, true); + return 0; } @@ -4888,7 +4894,7 @@ static void si_enable_gfx_cgpg(struct radeon_device *rdev, { u32 tmp; - if (enable && (rdev->pg_flags & RADEON_PG_SUPPORT_GFX_CG)) { + if (enable && (rdev->pg_flags & RADEON_PG_SUPPORT_GFX_PG)) { tmp = RLC_PUD(0x10) | RLC_PDD(0x10) | RLC_TTPD(0x10) | RLC_MSD(0x10); WREG32(RLC_TTOP_D, tmp); @@ -5250,6 +5256,7 @@ void si_update_cg(struct radeon_device *rdev, u32 block, bool enable) { if (block & RADEON_CG_BLOCK_GFX) { + si_enable_gui_idle_interrupt(rdev, false); /* order matters! */ if (enable) { si_enable_mgcg(rdev, true); @@ -5258,6 +5265,7 @@ void si_update_cg(struct radeon_device *rdev, si_enable_cgcg(rdev, false); si_enable_mgcg(rdev, false); } + si_enable_gui_idle_interrupt(rdev, true); } if (block & RADEON_CG_BLOCK_MC) { @@ -5408,7 +5416,7 @@ static void si_init_pg(struct radeon_device *rdev) si_init_dma_pg(rdev); } si_init_ao_cu_mask(rdev); - if (rdev->pg_flags & RADEON_PG_SUPPORT_GFX_CG) { + if (rdev->pg_flags & RADEON_PG_SUPPORT_GFX_PG) { si_init_gfx_cgpg(rdev); } si_enable_dma_pg(rdev, true); @@ -5560,7 +5568,9 @@ static void si_disable_interrupt_state(struct radeon_device *rdev) { u32 tmp; - WREG32(CP_INT_CNTL_RING0, CNTX_BUSY_INT_ENABLE | CNTX_EMPTY_INT_ENABLE); + tmp = RREG32(CP_INT_CNTL_RING0) & + (CNTX_BUSY_INT_ENABLE | CNTX_EMPTY_INT_ENABLE); + WREG32(CP_INT_CNTL_RING0, tmp); WREG32(CP_INT_CNTL_RING1, 0); WREG32(CP_INT_CNTL_RING2, 0); tmp = RREG32(DMA_CNTL + DMA0_REGISTER_OFFSET) & ~TRAP_ENABLE; @@ -5685,7 +5695,7 @@ static int si_irq_init(struct radeon_device *rdev) int si_irq_set(struct radeon_device *rdev) { - u32 cp_int_cntl = CNTX_BUSY_INT_ENABLE | CNTX_EMPTY_INT_ENABLE; + u32 cp_int_cntl; u32 cp_int_cntl1 = 0, cp_int_cntl2 = 0; u32 crtc1 = 0, crtc2 = 0, crtc3 = 0, crtc4 = 0, crtc5 = 0, crtc6 = 0; u32 hpd1 = 0, hpd2 = 0, hpd3 = 0, hpd4 = 0, hpd5 = 0, hpd6 = 0; @@ -5706,6 +5716,9 @@ int si_irq_set(struct radeon_device *rdev) return 0; } + cp_int_cntl = RREG32(CP_INT_CNTL_RING0) & + (CNTX_BUSY_INT_ENABLE | CNTX_EMPTY_INT_ENABLE); + if (!ASIC_IS_NODCE(rdev)) { hpd1 = RREG32(DC_HPD1_INT_CONTROL) & ~DC_HPDx_INT_EN; hpd2 = RREG32(DC_HPD2_INT_CONTROL) & ~DC_HPDx_INT_EN; diff --git a/drivers/gpu/drm/radeon/si_dpm.c b/drivers/gpu/drm/radeon/si_dpm.c index 5be9b4e7235..cfe5d4d2891 100644 --- a/drivers/gpu/drm/radeon/si_dpm.c +++ b/drivers/gpu/drm/radeon/si_dpm.c @@ -6075,12 +6075,6 @@ int si_dpm_set_power_state(struct radeon_device *rdev) return ret; } - ret = si_dpm_force_performance_level(rdev, RADEON_DPM_FORCED_LEVEL_AUTO); - if (ret) { - DRM_ERROR("si_dpm_force_performance_level failed\n"); - return ret; - } - si_update_cg(rdev, (RADEON_CG_BLOCK_GFX | RADEON_CG_BLOCK_MC | RADEON_CG_BLOCK_SDMA | diff --git a/drivers/gpu/drm/radeon/si_smc.c b/drivers/gpu/drm/radeon/si_smc.c index 5f524c0a541..d422a1cbf72 100644 --- a/drivers/gpu/drm/radeon/si_smc.c +++ b/drivers/gpu/drm/radeon/si_smc.c @@ -29,8 +29,8 @@ #include "ppsmc.h" #include "radeon_ucode.h" -int si_set_smc_sram_address(struct radeon_device *rdev, - u32 smc_address, u32 limit) +static int si_set_smc_sram_address(struct radeon_device *rdev, + u32 smc_address, u32 limit) { if (smc_address & 3) return -EINVAL; @@ -47,7 +47,8 @@ int si_copy_bytes_to_smc(struct radeon_device *rdev, u32 smc_start_address, const u8 *src, u32 byte_count, u32 limit) { - int ret; + unsigned long flags; + int ret = 0; u32 data, original_data, addr, extra_shift; if (smc_start_address & 3) @@ -57,13 +58,14 @@ int si_copy_bytes_to_smc(struct radeon_device *rdev, addr = smc_start_address; + spin_lock_irqsave(&rdev->smc_idx_lock, flags); while (byte_count >= 4) { /* SMC address space is BE */ data = (src[0] << 24) | (src[1] << 16) | (src[2] << 8) | src[3]; ret = si_set_smc_sram_address(rdev, addr, limit); if (ret) - return ret; + goto done; WREG32(SMC_IND_DATA_0, data); @@ -78,7 +80,7 @@ int si_copy_bytes_to_smc(struct radeon_device *rdev, ret = si_set_smc_sram_address(rdev, addr, limit); if (ret) - return ret; + goto done; original_data = RREG32(SMC_IND_DATA_0); @@ -96,11 +98,15 @@ int si_copy_bytes_to_smc(struct radeon_device *rdev, ret = si_set_smc_sram_address(rdev, addr, limit); if (ret) - return ret; + goto done; WREG32(SMC_IND_DATA_0, data); } - return 0; + +done: + spin_unlock_irqrestore(&rdev->smc_idx_lock, flags); + + return ret; } void si_start_smc(struct radeon_device *rdev) @@ -203,6 +209,7 @@ PPSMC_Result si_wait_for_smc_inactive(struct radeon_device *rdev) int si_load_smc_ucode(struct radeon_device *rdev, u32 limit) { + unsigned long flags; u32 ucode_start_address; u32 ucode_size; const u8 *src; @@ -241,6 +248,7 @@ int si_load_smc_ucode(struct radeon_device *rdev, u32 limit) return -EINVAL; src = (const u8 *)rdev->smc_fw->data; + spin_lock_irqsave(&rdev->smc_idx_lock, flags); WREG32(SMC_IND_INDEX_0, ucode_start_address); WREG32_P(SMC_IND_ACCESS_CNTL, AUTO_INCREMENT_IND_0, ~AUTO_INCREMENT_IND_0); while (ucode_size >= 4) { @@ -253,6 +261,7 @@ int si_load_smc_ucode(struct radeon_device *rdev, u32 limit) ucode_size -= 4; } WREG32_P(SMC_IND_ACCESS_CNTL, 0, ~AUTO_INCREMENT_IND_0); + spin_unlock_irqrestore(&rdev->smc_idx_lock, flags); return 0; } @@ -260,25 +269,29 @@ int si_load_smc_ucode(struct radeon_device *rdev, u32 limit) int si_read_smc_sram_dword(struct radeon_device *rdev, u32 smc_address, u32 *value, u32 limit) { + unsigned long flags; int ret; + spin_lock_irqsave(&rdev->smc_idx_lock, flags); ret = si_set_smc_sram_address(rdev, smc_address, limit); - if (ret) - return ret; + if (ret == 0) + *value = RREG32(SMC_IND_DATA_0); + spin_unlock_irqrestore(&rdev->smc_idx_lock, flags); - *value = RREG32(SMC_IND_DATA_0); - return 0; + return ret; } int si_write_smc_sram_dword(struct radeon_device *rdev, u32 smc_address, u32 value, u32 limit) { + unsigned long flags; int ret; + spin_lock_irqsave(&rdev->smc_idx_lock, flags); ret = si_set_smc_sram_address(rdev, smc_address, limit); - if (ret) - return ret; + if (ret == 0) + WREG32(SMC_IND_DATA_0, value); + spin_unlock_irqrestore(&rdev->smc_idx_lock, flags); - WREG32(SMC_IND_DATA_0, value); - return 0; + return ret; } diff --git a/drivers/gpu/drm/radeon/sumo_dpm.c b/drivers/gpu/drm/radeon/sumo_dpm.c index 864761c0120..96ea6db8bf5 100644 --- a/drivers/gpu/drm/radeon/sumo_dpm.c +++ b/drivers/gpu/drm/radeon/sumo_dpm.c @@ -1319,8 +1319,6 @@ int sumo_dpm_set_power_state(struct radeon_device *rdev) if (pi->enable_dpm) sumo_set_uvd_clock_after_set_eng_clock(rdev, new_ps, old_ps); - rdev->pm.dpm.forced_level = RADEON_DPM_FORCED_LEVEL_AUTO; - return 0; } diff --git a/drivers/gpu/drm/radeon/trinity_dpm.c b/drivers/gpu/drm/radeon/trinity_dpm.c index b07b7b8f1af..7f998bf1cc9 100644 --- a/drivers/gpu/drm/radeon/trinity_dpm.c +++ b/drivers/gpu/drm/radeon/trinity_dpm.c @@ -1068,6 +1068,17 @@ static void trinity_update_requested_ps(struct radeon_device *rdev, pi->requested_rps.ps_priv = &pi->requested_ps; } +void trinity_dpm_enable_bapm(struct radeon_device *rdev, bool enable) +{ + struct trinity_power_info *pi = trinity_get_pi(rdev); + + if (pi->enable_bapm) { + trinity_acquire_mutex(rdev); + trinity_dpm_bapm_enable(rdev, enable); + trinity_release_mutex(rdev); + } +} + int trinity_dpm_enable(struct radeon_device *rdev) { struct trinity_power_info *pi = trinity_get_pi(rdev); @@ -1091,6 +1102,7 @@ int trinity_dpm_enable(struct radeon_device *rdev) trinity_program_sclk_dpm(rdev); trinity_start_dpm(rdev); trinity_wait_for_dpm_enabled(rdev); + trinity_dpm_bapm_enable(rdev, false); trinity_release_mutex(rdev); if (rdev->irq.installed && @@ -1116,6 +1128,7 @@ void trinity_dpm_disable(struct radeon_device *rdev) trinity_release_mutex(rdev); return; } + trinity_dpm_bapm_enable(rdev, false); trinity_disable_clock_power_gating(rdev); sumo_clear_vc(rdev); trinity_wait_for_level_0(rdev); @@ -1212,6 +1225,8 @@ int trinity_dpm_set_power_state(struct radeon_device *rdev) trinity_acquire_mutex(rdev); if (pi->enable_dpm) { + if (pi->enable_bapm) + trinity_dpm_bapm_enable(rdev, rdev->pm.dpm.ac_power); trinity_set_uvd_clock_before_set_eng_clock(rdev, new_ps, old_ps); trinity_enable_power_level_0(rdev); trinity_force_level_0(rdev); @@ -1221,7 +1236,6 @@ int trinity_dpm_set_power_state(struct radeon_device *rdev) trinity_force_level_0(rdev); trinity_unforce_levels(rdev); trinity_set_uvd_clock_after_set_eng_clock(rdev, new_ps, old_ps); - rdev->pm.dpm.forced_level = RADEON_DPM_FORCED_LEVEL_AUTO; } trinity_release_mutex(rdev); @@ -1854,6 +1868,7 @@ int trinity_dpm_init(struct radeon_device *rdev) for (i = 0; i < SUMO_MAX_HARDWARE_POWERLEVELS; i++) pi->at[i] = TRINITY_AT_DFLT; + pi->enable_bapm = true; pi->enable_nbps_policy = true; pi->enable_sclk_ds = true; pi->enable_gfx_power_gating = true; diff --git a/drivers/gpu/drm/radeon/trinity_dpm.h b/drivers/gpu/drm/radeon/trinity_dpm.h index e82df071f8b..c261657750c 100644 --- a/drivers/gpu/drm/radeon/trinity_dpm.h +++ b/drivers/gpu/drm/radeon/trinity_dpm.h @@ -108,6 +108,7 @@ struct trinity_power_info { bool enable_auto_thermal_throttling; bool enable_dpm; bool enable_sclk_ds; + bool enable_bapm; bool uvd_dpm; struct radeon_ps current_rps; struct trinity_ps current_ps; @@ -118,6 +119,7 @@ struct trinity_power_info { #define TRINITY_AT_DFLT 30 /* trinity_smc.c */ +int trinity_dpm_bapm_enable(struct radeon_device *rdev, bool enable); int trinity_dpm_config(struct radeon_device *rdev, bool enable); int trinity_uvd_dpm_config(struct radeon_device *rdev); int trinity_dpm_force_state(struct radeon_device *rdev, u32 n); diff --git a/drivers/gpu/drm/radeon/trinity_smc.c b/drivers/gpu/drm/radeon/trinity_smc.c index a42d89f1830..9672bcbc731 100644 --- a/drivers/gpu/drm/radeon/trinity_smc.c +++ b/drivers/gpu/drm/radeon/trinity_smc.c @@ -56,6 +56,14 @@ static int trinity_notify_message_to_smu(struct radeon_device *rdev, u32 id) return 0; } +int trinity_dpm_bapm_enable(struct radeon_device *rdev, bool enable) +{ + if (enable) + return trinity_notify_message_to_smu(rdev, PPSMC_MSG_EnableBAPM); + else + return trinity_notify_message_to_smu(rdev, PPSMC_MSG_DisableBAPM); +} + int trinity_dpm_config(struct radeon_device *rdev, bool enable) { if (enable) diff --git a/drivers/gpu/drm/tilcdc/tilcdc_slave.c b/drivers/gpu/drm/tilcdc/tilcdc_slave.c index 23b3203d824..595068ba2d5 100644 --- a/drivers/gpu/drm/tilcdc/tilcdc_slave.c +++ b/drivers/gpu/drm/tilcdc/tilcdc_slave.c @@ -16,7 +16,6 @@ */ #include <linux/i2c.h> -#include <linux/of_i2c.h> #include <linux/pinctrl/pinmux.h> #include <linux/pinctrl/consumer.h> #include <drm/drm_encoder_slave.h> diff --git a/drivers/gpu/drm/tilcdc/tilcdc_tfp410.c b/drivers/gpu/drm/tilcdc/tilcdc_tfp410.c index 925c7cddeff..c38b56b268a 100644 --- a/drivers/gpu/drm/tilcdc/tilcdc_tfp410.c +++ b/drivers/gpu/drm/tilcdc/tilcdc_tfp410.c @@ -16,7 +16,6 @@ */ #include <linux/i2c.h> -#include <linux/of_i2c.h> #include <linux/gpio.h> #include <linux/of_gpio.h> #include <linux/pinctrl/pinmux.h> diff --git a/drivers/gpu/drm/ttm/ttm_object.c b/drivers/gpu/drm/ttm/ttm_object.c index 58a5f3261c0..a868176c258 100644 --- a/drivers/gpu/drm/ttm/ttm_object.c +++ b/drivers/gpu/drm/ttm/ttm_object.c @@ -218,7 +218,7 @@ struct ttm_base_object *ttm_base_object_lookup(struct ttm_object_file *tfile, uint32_t key) { struct ttm_object_device *tdev = tfile->tdev; - struct ttm_base_object *base; + struct ttm_base_object *uninitialized_var(base); struct drm_hash_item *hash; int ret; diff --git a/drivers/gpu/drm/ttm/ttm_page_alloc.c b/drivers/gpu/drm/ttm/ttm_page_alloc.c index bd2a3b40cd1..863bef9f923 100644 --- a/drivers/gpu/drm/ttm/ttm_page_alloc.c +++ b/drivers/gpu/drm/ttm/ttm_page_alloc.c @@ -377,28 +377,26 @@ out: return nr_free; } -/* Get good estimation how many pages are free in pools */ -static int ttm_pool_get_num_unused_pages(void) -{ - unsigned i; - int total = 0; - for (i = 0; i < NUM_POOLS; ++i) - total += _manager->pools[i].npages; - - return total; -} - /** * Callback for mm to request pool to reduce number of page held. + * + * XXX: (dchinner) Deadlock warning! + * + * ttm_page_pool_free() does memory allocation using GFP_KERNEL. that means + * this can deadlock when called a sc->gfp_mask that is not equal to + * GFP_KERNEL. + * + * This code is crying out for a shrinker per pool.... */ -static int ttm_pool_mm_shrink(struct shrinker *shrink, - struct shrink_control *sc) +static unsigned long +ttm_pool_shrink_scan(struct shrinker *shrink, struct shrink_control *sc) { static atomic_t start_pool = ATOMIC_INIT(0); unsigned i; unsigned pool_offset = atomic_add_return(1, &start_pool); struct ttm_page_pool *pool; int shrink_pages = sc->nr_to_scan; + unsigned long freed = 0; pool_offset = pool_offset % NUM_POOLS; /* select start pool in round robin fashion */ @@ -408,14 +406,28 @@ static int ttm_pool_mm_shrink(struct shrinker *shrink, break; pool = &_manager->pools[(i + pool_offset)%NUM_POOLS]; shrink_pages = ttm_page_pool_free(pool, nr_free); + freed += nr_free - shrink_pages; } - /* return estimated number of unused pages in pool */ - return ttm_pool_get_num_unused_pages(); + return freed; +} + + +static unsigned long +ttm_pool_shrink_count(struct shrinker *shrink, struct shrink_control *sc) +{ + unsigned i; + unsigned long count = 0; + + for (i = 0; i < NUM_POOLS; ++i) + count += _manager->pools[i].npages; + + return count; } static void ttm_pool_mm_shrink_init(struct ttm_pool_manager *manager) { - manager->mm_shrink.shrink = &ttm_pool_mm_shrink; + manager->mm_shrink.count_objects = ttm_pool_shrink_count; + manager->mm_shrink.scan_objects = ttm_pool_shrink_scan; manager->mm_shrink.seeks = 1; register_shrinker(&manager->mm_shrink); } diff --git a/drivers/gpu/drm/ttm/ttm_page_alloc_dma.c b/drivers/gpu/drm/ttm/ttm_page_alloc_dma.c index b8b394319b4..7957beeeaf7 100644 --- a/drivers/gpu/drm/ttm/ttm_page_alloc_dma.c +++ b/drivers/gpu/drm/ttm/ttm_page_alloc_dma.c @@ -918,19 +918,6 @@ int ttm_dma_populate(struct ttm_dma_tt *ttm_dma, struct device *dev) } EXPORT_SYMBOL_GPL(ttm_dma_populate); -/* Get good estimation how many pages are free in pools */ -static int ttm_dma_pool_get_num_unused_pages(void) -{ - struct device_pools *p; - unsigned total = 0; - - mutex_lock(&_manager->lock); - list_for_each_entry(p, &_manager->pools, pools) - total += p->pool->npages_free; - mutex_unlock(&_manager->lock); - return total; -} - /* Put all pages in pages list to correct pool to wait for reuse */ void ttm_dma_unpopulate(struct ttm_dma_tt *ttm_dma, struct device *dev) { @@ -1002,18 +989,29 @@ EXPORT_SYMBOL_GPL(ttm_dma_unpopulate); /** * Callback for mm to request pool to reduce number of page held. + * + * XXX: (dchinner) Deadlock warning! + * + * ttm_dma_page_pool_free() does GFP_KERNEL memory allocation, and so attention + * needs to be paid to sc->gfp_mask to determine if this can be done or not. + * GFP_KERNEL memory allocation in a GFP_ATOMIC reclaim context woul dbe really + * bad. + * + * I'm getting sadder as I hear more pathetical whimpers about needing per-pool + * shrinkers */ -static int ttm_dma_pool_mm_shrink(struct shrinker *shrink, - struct shrink_control *sc) +static unsigned long +ttm_dma_pool_shrink_scan(struct shrinker *shrink, struct shrink_control *sc) { static atomic_t start_pool = ATOMIC_INIT(0); unsigned idx = 0; unsigned pool_offset = atomic_add_return(1, &start_pool); unsigned shrink_pages = sc->nr_to_scan; struct device_pools *p; + unsigned long freed = 0; if (list_empty(&_manager->pools)) - return 0; + return SHRINK_STOP; mutex_lock(&_manager->lock); pool_offset = pool_offset % _manager->npools; @@ -1029,18 +1027,33 @@ static int ttm_dma_pool_mm_shrink(struct shrinker *shrink, continue; nr_free = shrink_pages; shrink_pages = ttm_dma_page_pool_free(p->pool, nr_free); + freed += nr_free - shrink_pages; + pr_debug("%s: (%s:%d) Asked to shrink %d, have %d more to go\n", p->pool->dev_name, p->pool->name, current->pid, nr_free, shrink_pages); } mutex_unlock(&_manager->lock); - /* return estimated number of unused pages in pool */ - return ttm_dma_pool_get_num_unused_pages(); + return freed; +} + +static unsigned long +ttm_dma_pool_shrink_count(struct shrinker *shrink, struct shrink_control *sc) +{ + struct device_pools *p; + unsigned long count = 0; + + mutex_lock(&_manager->lock); + list_for_each_entry(p, &_manager->pools, pools) + count += p->pool->npages_free; + mutex_unlock(&_manager->lock); + return count; } static void ttm_dma_pool_mm_shrink_init(struct ttm_pool_manager *manager) { - manager->mm_shrink.shrink = &ttm_dma_pool_mm_shrink; + manager->mm_shrink.count_objects = ttm_dma_pool_shrink_count; + manager->mm_shrink.scan_objects = &ttm_dma_pool_shrink_scan; manager->mm_shrink.seeks = 1; register_shrinker(&manager->mm_shrink); } diff --git a/drivers/gpu/drm/ttm/ttm_tt.c b/drivers/gpu/drm/ttm/ttm_tt.c index 5e93a52d4f2..210d5036516 100644 --- a/drivers/gpu/drm/ttm/ttm_tt.c +++ b/drivers/gpu/drm/ttm/ttm_tt.c @@ -170,7 +170,7 @@ void ttm_tt_destroy(struct ttm_tt *ttm) ttm_tt_unbind(ttm); } - if (likely(ttm->pages != NULL)) { + if (ttm->state == tt_unbound) { ttm->bdev->driver->ttm_tt_unpopulate(ttm); } diff --git a/drivers/gpu/drm/udl/udl_gem.c b/drivers/gpu/drm/udl/udl_gem.c index 8dbe9d0ae9a..8bf646183ba 100644 --- a/drivers/gpu/drm/udl/udl_gem.c +++ b/drivers/gpu/drm/udl/udl_gem.c @@ -97,7 +97,6 @@ int udl_gem_fault(struct vm_area_struct *vma, struct vm_fault *vmf) ret = vm_insert_page(vma, (unsigned long)vmf->virtual_address, page); switch (ret) { case -EAGAIN: - set_need_resched(); case 0: case -ERESTARTSYS: return VM_FAULT_NOPAGE; diff --git a/drivers/gpu/host1x/dev.c b/drivers/gpu/host1x/dev.c index 28e28a23d44..47163029987 100644 --- a/drivers/gpu/host1x/dev.c +++ b/drivers/gpu/host1x/dev.c @@ -43,7 +43,7 @@ void host1x_set_drm_data(struct device *dev, void *data) void *host1x_get_drm_data(struct device *dev) { struct host1x *host1x = dev_get_drvdata(dev); - return host1x->drm_data; + return host1x ? host1x->drm_data : NULL; } void host1x_sync_writel(struct host1x *host1x, u32 v, u32 r) diff --git a/drivers/gpu/host1x/dev.h b/drivers/gpu/host1x/dev.h index 790ddf114e5..bed90a8131b 100644 --- a/drivers/gpu/host1x/dev.h +++ b/drivers/gpu/host1x/dev.h @@ -301,8 +301,8 @@ static inline void host1x_hw_show_mlocks(struct host1x *host, struct output *o) host->debug_op->show_mlocks(host, o); } -extern struct platform_driver tegra_hdmi_driver; extern struct platform_driver tegra_dc_driver; +extern struct platform_driver tegra_hdmi_driver; extern struct platform_driver tegra_gr2d_driver; #endif diff --git a/drivers/gpu/host1x/drm/hdmi.c b/drivers/gpu/host1x/drm/hdmi.c index 52e3c9641a0..644d95c7d48 100644 --- a/drivers/gpu/host1x/drm/hdmi.c +++ b/drivers/gpu/host1x/drm/hdmi.c @@ -888,6 +888,11 @@ static int tegra_hdmi_show_regs(struct seq_file *s, void *data) { struct drm_info_node *node = s->private; struct tegra_hdmi *hdmi = node->info_ent->data; + int err; + + err = clk_enable(hdmi->clk); + if (err) + return err; #define DUMP_REG(name) \ seq_printf(s, "%-56s %#05x %08lx\n", #name, name, \ @@ -1053,6 +1058,8 @@ static int tegra_hdmi_show_regs(struct seq_file *s, void *data) #undef DUMP_REG + clk_disable(hdmi->clk); + return 0; } diff --git a/drivers/gpu/host1x/drm/output.c b/drivers/gpu/host1x/drm/output.c index 8140fc6c34d..137ae81ab80 100644 --- a/drivers/gpu/host1x/drm/output.c +++ b/drivers/gpu/host1x/drm/output.c @@ -9,7 +9,7 @@ #include <linux/module.h> #include <linux/of_gpio.h> -#include <linux/of_i2c.h> +#include <linux/i2c.h> #include "drm.h" diff --git a/drivers/gpu/host1x/drm/rgb.c b/drivers/gpu/host1x/drm/rgb.c index ed4416f2026..5aa66ef7a94 100644 --- a/drivers/gpu/host1x/drm/rgb.c +++ b/drivers/gpu/host1x/drm/rgb.c @@ -147,6 +147,13 @@ int tegra_dc_rgb_probe(struct tegra_dc *dc) if (!rgb) return -ENOMEM; + rgb->output.dev = dc->dev; + rgb->output.of_node = np; + + err = tegra_output_parse_dt(&rgb->output); + if (err < 0) + return err; + rgb->clk = devm_clk_get(dc->dev, NULL); if (IS_ERR(rgb->clk)) { dev_err(dc->dev, "failed to get clock\n"); @@ -165,13 +172,6 @@ int tegra_dc_rgb_probe(struct tegra_dc *dc) return err; } - rgb->output.dev = dc->dev; - rgb->output.of_node = np; - - err = tegra_output_parse_dt(&rgb->output); - if (err < 0) - return err; - dc->rgb = &rgb->output; return 0; diff --git a/drivers/gpu/host1x/job.c b/drivers/gpu/host1x/job.c index cc807667d8f..c4e1050f225 100644 --- a/drivers/gpu/host1x/job.c +++ b/drivers/gpu/host1x/job.c @@ -42,12 +42,12 @@ struct host1x_job *host1x_job_alloc(struct host1x_channel *ch, /* Check that we're not going to overflow */ total = sizeof(struct host1x_job) + - num_relocs * sizeof(struct host1x_reloc) + - num_unpins * sizeof(struct host1x_job_unpin_data) + - num_waitchks * sizeof(struct host1x_waitchk) + - num_cmdbufs * sizeof(struct host1x_job_gather) + - num_unpins * sizeof(dma_addr_t) + - num_unpins * sizeof(u32 *); + (u64)num_relocs * sizeof(struct host1x_reloc) + + (u64)num_unpins * sizeof(struct host1x_job_unpin_data) + + (u64)num_waitchks * sizeof(struct host1x_waitchk) + + (u64)num_cmdbufs * sizeof(struct host1x_job_gather) + + (u64)num_unpins * sizeof(dma_addr_t) + + (u64)num_unpins * sizeof(u32 *); if (total > ULONG_MAX) return NULL; @@ -466,9 +466,8 @@ static inline int copy_gathers(struct host1x_job *job, struct device *dev) &job->gather_copy, GFP_KERNEL); if (!job->gather_copy_mapped) { - int err = PTR_ERR(job->gather_copy_mapped); job->gather_copy_mapped = NULL; - return err; + return -ENOMEM; } job->gather_copy_size = size; |