#include "amdgpu_amdkfd.h"
#include "amd_shared.h"
- #include <drm/drmP.h>
+
#include "amdgpu.h"
#include "amdgpu_gfx.h"
+#include "amdgpu_dma_buf.h"
#include <linux/module.h>
#include <linux/dma-buf.h>
+#include "amdgpu_xgmi.h"
static const unsigned int compute_vmid_bitmap = 0xFF00;
--- /dev/null
- #include <drm/drmP.h>
-
+/*
+ * Copyright 2019 Advanced Micro Devices, Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * based on nouveau_prime.c
+ *
+ * Authors: Alex Deucher
+ */
+
+/**
+ * DOC: PRIME Buffer Sharing
+ *
+ * The following callback implementations are used for :ref:`sharing GEM buffer
+ * objects between different devices via PRIME <prime_buffer_sharing>`.
+ */
+
+#include "amdgpu.h"
+#include "amdgpu_display.h"
+#include "amdgpu_gem.h"
+#include <drm/amdgpu_drm.h>
+#include <linux/dma-buf.h>
+#include <linux/dma-fence-array.h>
+
+/**
+ * amdgpu_gem_prime_get_sg_table - &drm_driver.gem_prime_get_sg_table
+ * implementation
+ * @obj: GEM buffer object (BO)
+ *
+ * Returns:
+ * A scatter/gather table for the pinned pages of the BO's memory.
+ */
+struct sg_table *amdgpu_gem_prime_get_sg_table(struct drm_gem_object *obj)
+{
+ struct amdgpu_bo *bo = gem_to_amdgpu_bo(obj);
+ int npages = bo->tbo.num_pages;
+
+ return drm_prime_pages_to_sg(bo->tbo.ttm->pages, npages);
+}
+
+/**
+ * amdgpu_gem_prime_vmap - &dma_buf_ops.vmap implementation
+ * @obj: GEM BO
+ *
+ * Sets up an in-kernel virtual mapping of the BO's memory.
+ *
+ * Returns:
+ * The virtual address of the mapping or an error pointer.
+ */
+void *amdgpu_gem_prime_vmap(struct drm_gem_object *obj)
+{
+ struct amdgpu_bo *bo = gem_to_amdgpu_bo(obj);
+ int ret;
+
+ ret = ttm_bo_kmap(&bo->tbo, 0, bo->tbo.num_pages,
+ &bo->dma_buf_vmap);
+ if (ret)
+ return ERR_PTR(ret);
+
+ return bo->dma_buf_vmap.virtual;
+}
+
+/**
+ * amdgpu_gem_prime_vunmap - &dma_buf_ops.vunmap implementation
+ * @obj: GEM BO
+ * @vaddr: Virtual address (unused)
+ *
+ * Tears down the in-kernel virtual mapping of the BO's memory.
+ */
+void amdgpu_gem_prime_vunmap(struct drm_gem_object *obj, void *vaddr)
+{
+ struct amdgpu_bo *bo = gem_to_amdgpu_bo(obj);
+
+ ttm_bo_kunmap(&bo->dma_buf_vmap);
+}
+
+/**
+ * amdgpu_gem_prime_mmap - &drm_driver.gem_prime_mmap implementation
+ * @obj: GEM BO
+ * @vma: Virtual memory area
+ *
+ * Sets up a userspace mapping of the BO's memory in the given
+ * virtual memory area.
+ *
+ * Returns:
+ * 0 on success or a negative error code on failure.
+ */
+int amdgpu_gem_prime_mmap(struct drm_gem_object *obj,
+ struct vm_area_struct *vma)
+{
+ struct amdgpu_bo *bo = gem_to_amdgpu_bo(obj);
+ struct amdgpu_device *adev = amdgpu_ttm_adev(bo->tbo.bdev);
+ unsigned asize = amdgpu_bo_size(bo);
+ int ret;
+
+ if (!vma->vm_file)
+ return -ENODEV;
+
+ if (adev == NULL)
+ return -ENODEV;
+
+ /* Check for valid size. */
+ if (asize < vma->vm_end - vma->vm_start)
+ return -EINVAL;
+
+ if (amdgpu_ttm_tt_get_usermm(bo->tbo.ttm) ||
+ (bo->flags & AMDGPU_GEM_CREATE_NO_CPU_ACCESS)) {
+ return -EPERM;
+ }
+ vma->vm_pgoff += amdgpu_bo_mmap_offset(bo) >> PAGE_SHIFT;
+
+ /* prime mmap does not need to check access, so allow here */
+ ret = drm_vma_node_allow(&obj->vma_node, vma->vm_file->private_data);
+ if (ret)
+ return ret;
+
+ ret = ttm_bo_mmap(vma->vm_file, vma, &adev->mman.bdev);
+ drm_vma_node_revoke(&obj->vma_node, vma->vm_file->private_data);
+
+ return ret;
+}
+
+static int
+__reservation_object_make_exclusive(struct reservation_object *obj)
+{
+ struct dma_fence **fences;
+ unsigned int count;
+ int r;
+
+ if (!reservation_object_get_list(obj)) /* no shared fences to convert */
+ return 0;
+
+ r = reservation_object_get_fences_rcu(obj, NULL, &count, &fences);
+ if (r)
+ return r;
+
+ if (count == 0) {
+ /* Now that was unexpected. */
+ } else if (count == 1) {
+ reservation_object_add_excl_fence(obj, fences[0]);
+ dma_fence_put(fences[0]);
+ kfree(fences);
+ } else {
+ struct dma_fence_array *array;
+
+ array = dma_fence_array_create(count, fences,
+ dma_fence_context_alloc(1), 0,
+ false);
+ if (!array)
+ goto err_fences_put;
+
+ reservation_object_add_excl_fence(obj, &array->base);
+ dma_fence_put(&array->base);
+ }
+
+ return 0;
+
+err_fences_put:
+ while (count--)
+ dma_fence_put(fences[count]);
+ kfree(fences);
+ return -ENOMEM;
+}
+
+/**
+ * amdgpu_dma_buf_map_attach - &dma_buf_ops.attach implementation
+ * @dma_buf: Shared DMA buffer
+ * @attach: DMA-buf attachment
+ *
+ * Makes sure that the shared DMA buffer can be accessed by the target device.
+ * For now, simply pins it to the GTT domain, where it should be accessible by
+ * all DMA devices.
+ *
+ * Returns:
+ * 0 on success or a negative error code on failure.
+ */
+static int amdgpu_dma_buf_map_attach(struct dma_buf *dma_buf,
+ struct dma_buf_attachment *attach)
+{
+ struct drm_gem_object *obj = dma_buf->priv;
+ struct amdgpu_bo *bo = gem_to_amdgpu_bo(obj);
+ struct amdgpu_device *adev = amdgpu_ttm_adev(bo->tbo.bdev);
+ long r;
+
+ r = drm_gem_map_attach(dma_buf, attach);
+ if (r)
+ return r;
+
+ r = amdgpu_bo_reserve(bo, false);
+ if (unlikely(r != 0))
+ goto error_detach;
+
+
+ if (attach->dev->driver != adev->dev->driver) {
+ /*
+ * We only create shared fences for internal use, but importers
+ * of the dmabuf rely on exclusive fences for implicitly
+ * tracking write hazards. As any of the current fences may
+ * correspond to a write, we need to convert all existing
+ * fences on the reservation object into a single exclusive
+ * fence.
+ */
+ r = __reservation_object_make_exclusive(bo->tbo.resv);
+ if (r)
+ goto error_unreserve;
+ }
+
+ /* pin buffer into GTT */
+ r = amdgpu_bo_pin(bo, AMDGPU_GEM_DOMAIN_GTT);
+ if (r)
+ goto error_unreserve;
+
+ if (attach->dev->driver != adev->dev->driver)
+ bo->prime_shared_count++;
+
+error_unreserve:
+ amdgpu_bo_unreserve(bo);
+
+error_detach:
+ if (r)
+ drm_gem_map_detach(dma_buf, attach);
+ return r;
+}
+
+/**
+ * amdgpu_dma_buf_map_detach - &dma_buf_ops.detach implementation
+ * @dma_buf: Shared DMA buffer
+ * @attach: DMA-buf attachment
+ *
+ * This is called when a shared DMA buffer no longer needs to be accessible by
+ * another device. For now, simply unpins the buffer from GTT.
+ */
+static void amdgpu_dma_buf_map_detach(struct dma_buf *dma_buf,
+ struct dma_buf_attachment *attach)
+{
+ struct drm_gem_object *obj = dma_buf->priv;
+ struct amdgpu_bo *bo = gem_to_amdgpu_bo(obj);
+ struct amdgpu_device *adev = amdgpu_ttm_adev(bo->tbo.bdev);
+ int ret = 0;
+
+ ret = amdgpu_bo_reserve(bo, true);
+ if (unlikely(ret != 0))
+ goto error;
+
+ amdgpu_bo_unpin(bo);
+ if (attach->dev->driver != adev->dev->driver && bo->prime_shared_count)
+ bo->prime_shared_count--;
+ amdgpu_bo_unreserve(bo);
+
+error:
+ drm_gem_map_detach(dma_buf, attach);
+}
+
+/**
+ * amdgpu_gem_prime_res_obj - &drm_driver.gem_prime_res_obj implementation
+ * @obj: GEM BO
+ *
+ * Returns:
+ * The BO's reservation object.
+ */
+struct reservation_object *amdgpu_gem_prime_res_obj(struct drm_gem_object *obj)
+{
+ struct amdgpu_bo *bo = gem_to_amdgpu_bo(obj);
+
+ return bo->tbo.resv;
+}
+
+/**
+ * amdgpu_dma_buf_begin_cpu_access - &dma_buf_ops.begin_cpu_access implementation
+ * @dma_buf: Shared DMA buffer
+ * @direction: Direction of DMA transfer
+ *
+ * This is called before CPU access to the shared DMA buffer's memory. If it's
+ * a read access, the buffer is moved to the GTT domain if possible, for optimal
+ * CPU read performance.
+ *
+ * Returns:
+ * 0 on success or a negative error code on failure.
+ */
+static int amdgpu_dma_buf_begin_cpu_access(struct dma_buf *dma_buf,
+ enum dma_data_direction direction)
+{
+ struct amdgpu_bo *bo = gem_to_amdgpu_bo(dma_buf->priv);
+ struct amdgpu_device *adev = amdgpu_ttm_adev(bo->tbo.bdev);
+ struct ttm_operation_ctx ctx = { true, false };
+ u32 domain = amdgpu_display_supported_domains(adev);
+ int ret;
+ bool reads = (direction == DMA_BIDIRECTIONAL ||
+ direction == DMA_FROM_DEVICE);
+
+ if (!reads || !(domain & AMDGPU_GEM_DOMAIN_GTT))
+ return 0;
+
+ /* move to gtt */
+ ret = amdgpu_bo_reserve(bo, false);
+ if (unlikely(ret != 0))
+ return ret;
+
+ if (!bo->pin_count && (bo->allowed_domains & AMDGPU_GEM_DOMAIN_GTT)) {
+ amdgpu_bo_placement_from_domain(bo, AMDGPU_GEM_DOMAIN_GTT);
+ ret = ttm_bo_validate(&bo->tbo, &bo->placement, &ctx);
+ }
+
+ amdgpu_bo_unreserve(bo);
+ return ret;
+}
+
+const struct dma_buf_ops amdgpu_dmabuf_ops = {
+ .attach = amdgpu_dma_buf_map_attach,
+ .detach = amdgpu_dma_buf_map_detach,
+ .map_dma_buf = drm_gem_map_dma_buf,
+ .unmap_dma_buf = drm_gem_unmap_dma_buf,
+ .release = drm_gem_dmabuf_release,
+ .begin_cpu_access = amdgpu_dma_buf_begin_cpu_access,
+ .mmap = drm_gem_dmabuf_mmap,
+ .vmap = drm_gem_dmabuf_vmap,
+ .vunmap = drm_gem_dmabuf_vunmap,
+};
+
+/**
+ * amdgpu_gem_prime_export - &drm_driver.gem_prime_export implementation
+ * @dev: DRM device
+ * @gobj: GEM BO
+ * @flags: Flags such as DRM_CLOEXEC and DRM_RDWR.
+ *
+ * The main work is done by the &drm_gem_prime_export helper, which in turn
+ * uses &amdgpu_gem_prime_res_obj.
+ *
+ * Returns:
+ * Shared DMA buffer representing the GEM BO from the given device.
+ */
+struct dma_buf *amdgpu_gem_prime_export(struct drm_device *dev,
+ struct drm_gem_object *gobj,
+ int flags)
+{
+ struct amdgpu_bo *bo = gem_to_amdgpu_bo(gobj);
+ struct dma_buf *buf;
+
+ if (amdgpu_ttm_tt_get_usermm(bo->tbo.ttm) ||
+ bo->flags & AMDGPU_GEM_CREATE_VM_ALWAYS_VALID)
+ return ERR_PTR(-EPERM);
+
+ buf = drm_gem_prime_export(dev, gobj, flags);
+ if (!IS_ERR(buf)) {
+ buf->file->f_mapping = dev->anon_inode->i_mapping;
+ buf->ops = &amdgpu_dmabuf_ops;
+ }
+
+ return buf;
+}
+
+/**
+ * amdgpu_gem_prime_import_sg_table - &drm_driver.gem_prime_import_sg_table
+ * implementation
+ * @dev: DRM device
+ * @attach: DMA-buf attachment
+ * @sg: Scatter/gather table
+ *
+ * Imports shared DMA buffer memory exported by another device.
+ *
+ * Returns:
+ * A new GEM BO of the given DRM device, representing the memory
+ * described by the given DMA-buf attachment and scatter/gather table.
+ */
+struct drm_gem_object *
+amdgpu_gem_prime_import_sg_table(struct drm_device *dev,
+ struct dma_buf_attachment *attach,
+ struct sg_table *sg)
+{
+ struct reservation_object *resv = attach->dmabuf->resv;
+ struct amdgpu_device *adev = dev->dev_private;
+ struct amdgpu_bo *bo;
+ struct amdgpu_bo_param bp;
+ int ret;
+
+ memset(&bp, 0, sizeof(bp));
+ bp.size = attach->dmabuf->size;
+ bp.byte_align = PAGE_SIZE;
+ bp.domain = AMDGPU_GEM_DOMAIN_CPU;
+ bp.flags = 0;
+ bp.type = ttm_bo_type_sg;
+ bp.resv = resv;
+ ww_mutex_lock(&resv->lock, NULL);
+ ret = amdgpu_bo_create(adev, &bp, &bo);
+ if (ret)
+ goto error;
+
+ bo->tbo.sg = sg;
+ bo->tbo.ttm->sg = sg;
+ bo->allowed_domains = AMDGPU_GEM_DOMAIN_GTT;
+ bo->preferred_domains = AMDGPU_GEM_DOMAIN_GTT;
+ if (attach->dmabuf->ops != &amdgpu_dmabuf_ops)
+ bo->prime_shared_count = 1;
+
+ ww_mutex_unlock(&resv->lock);
+ return &bo->gem_base;
+
+error:
+ ww_mutex_unlock(&resv->lock);
+ return ERR_PTR(ret);
+}
+
+/**
+ * amdgpu_gem_prime_import - &drm_driver.gem_prime_import implementation
+ * @dev: DRM device
+ * @dma_buf: Shared DMA buffer
+ *
+ * The main work is done by the &drm_gem_prime_import helper, which in turn
+ * uses &amdgpu_gem_prime_import_sg_table.
+ *
+ * Returns:
+ * GEM BO representing the shared DMA buffer for the given device.
+ */
+struct drm_gem_object *amdgpu_gem_prime_import(struct drm_device *dev,
+ struct dma_buf *dma_buf)
+{
+ struct drm_gem_object *obj;
+
+ if (dma_buf->ops == &amdgpu_dmabuf_ops) {
+ obj = dma_buf->priv;
+ if (obj->dev == dev) {
+ /*
+ * Importing dmabuf exported from out own gem increases
+ * refcount on gem itself instead of f_count of dmabuf.
+ */
+ drm_gem_object_get(obj);
+ return obj;
+ }
+ }
+
+ return drm_gem_prime_import(dev, dma_buf);
+}
#include <linux/firmware.h>
#include <linux/module.h>
-#include <linux/mmu_notifier.h>
+#include <linux/hmm.h>
#include <linux/interval_tree.h>
- #include <drm/drmP.h>
+
#include <drm/drm.h>
#include "amdgpu.h"
* Thomas Hellstrom <thomas-at-tungstengraphics-dot-com>
* Dave Airlie
*/
+
+ #include <linux/dma-mapping.h>
+ #include <linux/iommu.h>
++#include <linux/hmm.h>
+ #include <linux/pagemap.h>
+ #include <linux/sched/task.h>
+ #include <linux/seq_file.h>
+ #include <linux/slab.h>
+ #include <linux/swap.h>
+ #include <linux/swiotlb.h>
+
#include <drm/ttm/ttm_bo_api.h>
#include <drm/ttm/ttm_bo_driver.h>
#include <drm/ttm/ttm_placement.h>
#include <drm/ttm/ttm_module.h>
#include <drm/ttm/ttm_page_alloc.h>
- #include <drm/drmP.h>
+
+ #include <drm/drm_debugfs.h>
#include <drm/amdgpu_drm.h>
- #include <linux/seq_file.h>
- #include <linux/slab.h>
- #include <linux/swiotlb.h>
- #include <linux/swap.h>
- #include <linux/pagemap.h>
- #include <linux/debugfs.h>
- #include <linux/iommu.h>
- #include <linux/hmm.h>
++
#include "amdgpu.h"
#include "amdgpu_object.h"
#include "amdgpu_trace.h"
--- /dev/null
+/*
+ * Copyright 2012-16 Advanced Micro Devices, Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: AMD
+ *
+ */
+
++#include <linux/slab.h>
++
+#include "dal_asic_id.h"
+#include "dc_types.h"
+#include "dccg.h"
+#include "clk_mgr_internal.h"
+
+#include "dce100/dce_clk_mgr.h"
+#include "dce110/dce110_clk_mgr.h"
+#include "dce112/dce112_clk_mgr.h"
+#include "dce120/dce120_clk_mgr.h"
+#include "dcn10/rv1_clk_mgr.h"
+#include "dcn10/rv2_clk_mgr.h"
+
+
+int clk_mgr_helper_get_active_display_cnt(
+ struct dc *dc,
+ struct dc_state *context)
+{
+ int i, display_count;
+
+ display_count = 0;
+ for (i = 0; i < context->stream_count; i++) {
+ const struct dc_stream_state *stream = context->streams[i];
+
+ /*
+ * Only notify active stream or virtual stream.
+ * Need to notify virtual stream to work around
+ * headless case. HPD does not fire when system is in
+ * S0i2.
+ */
+ if (!stream->dpms_off || stream->signal == SIGNAL_TYPE_VIRTUAL)
+ display_count++;
+ }
+
+ return display_count;
+}
+
+
+struct clk_mgr *dc_clk_mgr_create(struct dc_context *ctx, struct pp_smu_funcs *pp_smu, struct dccg *dccg)
+{
+ struct hw_asic_id asic_id = ctx->asic_id;
+
+ struct clk_mgr_internal *clk_mgr = kzalloc(sizeof(*clk_mgr), GFP_KERNEL);
+
+ if (clk_mgr == NULL) {
+ BREAK_TO_DEBUGGER();
+ return NULL;
+ }
+
+ switch (asic_id.chip_family) {
+ case FAMILY_CI:
+ case FAMILY_KV:
+ dce_clk_mgr_construct(ctx, clk_mgr);
+ break;
+ case FAMILY_CZ:
+ dce110_clk_mgr_construct(ctx, clk_mgr);
+ break;
+ case FAMILY_VI:
+ if (ASIC_REV_IS_TONGA_P(asic_id.hw_internal_rev) ||
+ ASIC_REV_IS_FIJI_P(asic_id.hw_internal_rev)) {
+ dce_clk_mgr_construct(ctx, clk_mgr);
+ break;
+ }
+ if (ASIC_REV_IS_POLARIS10_P(asic_id.hw_internal_rev) ||
+ ASIC_REV_IS_POLARIS11_M(asic_id.hw_internal_rev) ||
+ ASIC_REV_IS_POLARIS12_V(asic_id.hw_internal_rev)) {
+ dce112_clk_mgr_construct(ctx, clk_mgr);
+ break;
+ }
+ if (ASIC_REV_IS_VEGAM(asic_id.hw_internal_rev)) {
+ dce112_clk_mgr_construct(ctx, clk_mgr);
+ break;
+ }
+ break;
+ case FAMILY_AI:
+ if (ASICREV_IS_VEGA20_P(asic_id.hw_internal_rev))
+ dce121_clk_mgr_construct(ctx, clk_mgr);
+ else
+ dce120_clk_mgr_construct(ctx, clk_mgr);
+ break;
+
+#if defined(CONFIG_DRM_AMD_DC_DCN1_0)
+ case FAMILY_RV:
+ if (ASICREV_IS_RAVEN2(asic_id.hw_internal_rev)) {
+ rv2_clk_mgr_construct(ctx, clk_mgr, pp_smu);
+ break;
+ }
+ if (ASICREV_IS_RAVEN(asic_id.hw_internal_rev) ||
+ ASICREV_IS_PICASSO(asic_id.hw_internal_rev)) {
+ rv1_clk_mgr_construct(ctx, clk_mgr, pp_smu);
+ break;
+ }
+ break;
+#endif /* Family RV */
+
+ default:
+ ASSERT(0); /* Unknown Asic */
+ break;
+ }
+
+ return &clk_mgr->base;
+}
+
+void dc_destroy_clk_mgr(struct clk_mgr *clk_mgr_base)
+{
+ struct clk_mgr_internal *clk_mgr = TO_CLK_MGR_INTERNAL(clk_mgr_base);
+
+ kfree(clk_mgr);
+}
+
--- /dev/null
+/*
+ * Copyright 2018 Advanced Micro Devices, Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: AMD
+ *
+ */
+
++#include <linux/slab.h>
++
++#include "reg_helper.h"
+#include "core_types.h"
+#include "clk_mgr_internal.h"
+#include "rv1_clk_mgr.h"
+#include "dce100/dce_clk_mgr.h"
+#include "dce112/dce112_clk_mgr.h"
+#include "rv1_clk_mgr_vbios_smu.h"
+#include "rv1_clk_mgr_clk.h"
+
+static int rv1_determine_dppclk_threshold(struct clk_mgr_internal *clk_mgr, struct dc_clocks *new_clocks)
+{
+ bool request_dpp_div = new_clocks->dispclk_khz > new_clocks->dppclk_khz;
+ bool dispclk_increase = new_clocks->dispclk_khz > clk_mgr->base.clks.dispclk_khz;
+ int disp_clk_threshold = new_clocks->max_supported_dppclk_khz;
+ bool cur_dpp_div = clk_mgr->base.clks.dispclk_khz > clk_mgr->base.clks.dppclk_khz;
+
+ /* increase clock, looking for div is 0 for current, request div is 1*/
+ if (dispclk_increase) {
+ /* already divided by 2, no need to reach target clk with 2 steps*/
+ if (cur_dpp_div)
+ return new_clocks->dispclk_khz;
+
+ /* request disp clk is lower than maximum supported dpp clk,
+ * no need to reach target clk with two steps.
+ */
+ if (new_clocks->dispclk_khz <= disp_clk_threshold)
+ return new_clocks->dispclk_khz;
+
+ /* target dpp clk not request divided by 2, still within threshold */
+ if (!request_dpp_div)
+ return new_clocks->dispclk_khz;
+
+ } else {
+ /* decrease clock, looking for current dppclk divided by 2,
+ * request dppclk not divided by 2.
+ */
+
+ /* current dpp clk not divided by 2, no need to ramp*/
+ if (!cur_dpp_div)
+ return new_clocks->dispclk_khz;
+
+ /* current disp clk is lower than current maximum dpp clk,
+ * no need to ramp
+ */
+ if (clk_mgr->base.clks.dispclk_khz <= disp_clk_threshold)
+ return new_clocks->dispclk_khz;
+
+ /* request dpp clk need to be divided by 2 */
+ if (request_dpp_div)
+ return new_clocks->dispclk_khz;
+ }
+
+ return disp_clk_threshold;
+}
+
+static void ramp_up_dispclk_with_dpp(struct clk_mgr_internal *clk_mgr, struct dc *dc, struct dc_clocks *new_clocks)
+{
+ int i;
+ int dispclk_to_dpp_threshold = rv1_determine_dppclk_threshold(clk_mgr, new_clocks);
+ bool request_dpp_div = new_clocks->dispclk_khz > new_clocks->dppclk_khz;
+
+ /* set disp clk to dpp clk threshold */
+
+ clk_mgr->funcs->set_dispclk(clk_mgr, dispclk_to_dpp_threshold);
+ clk_mgr->funcs->set_dprefclk(clk_mgr);
+
+
+ /* update request dpp clk division option */
+ for (i = 0; i < dc->res_pool->pipe_count; i++) {
+ struct pipe_ctx *pipe_ctx = &dc->current_state->res_ctx.pipe_ctx[i];
+
+ if (!pipe_ctx->plane_state)
+ continue;
+
+ pipe_ctx->plane_res.dpp->funcs->dpp_dppclk_control(
+ pipe_ctx->plane_res.dpp,
+ request_dpp_div,
+ true);
+ }
+
+ /* If target clk not same as dppclk threshold, set to target clock */
+ if (dispclk_to_dpp_threshold != new_clocks->dispclk_khz) {
+ clk_mgr->funcs->set_dispclk(clk_mgr, new_clocks->dispclk_khz);
+ clk_mgr->funcs->set_dprefclk(clk_mgr);
+ }
+
+
+ clk_mgr->base.clks.dispclk_khz = new_clocks->dispclk_khz;
+ clk_mgr->base.clks.dppclk_khz = new_clocks->dppclk_khz;
+ clk_mgr->base.clks.max_supported_dppclk_khz = new_clocks->max_supported_dppclk_khz;
+}
+
+static void rv1_update_clocks(struct clk_mgr *clk_mgr_base,
+ struct dc_state *context,
+ bool safe_to_lower)
+{
+ struct clk_mgr_internal *clk_mgr = TO_CLK_MGR_INTERNAL(clk_mgr_base);
+ struct dc *dc = clk_mgr_base->ctx->dc;
+ struct dc_debug_options *debug = &dc->debug;
+ struct dc_clocks *new_clocks = &context->bw_ctx.bw.dcn.clk;
+ struct pp_smu_funcs_rv *pp_smu = NULL;
+ bool send_request_to_increase = false;
+ bool send_request_to_lower = false;
+ int display_count;
+
+ bool enter_display_off = false;
+
+ ASSERT(clk_mgr->pp_smu);
+
+ pp_smu = &clk_mgr->pp_smu->rv_funcs;
+
+ display_count = clk_mgr_helper_get_active_display_cnt(dc, context);
+
+ if (display_count == 0)
+ enter_display_off = true;
+
+ if (enter_display_off == safe_to_lower) {
+ /*
+ * Notify SMU active displays
+ * if function pointer not set up, this message is
+ * sent as part of pplib_apply_display_requirements.
+ */
+ if (pp_smu->set_display_count)
+ pp_smu->set_display_count(&pp_smu->pp_smu, display_count);
+ }
+
+ if (new_clocks->dispclk_khz > clk_mgr_base->clks.dispclk_khz
+ || new_clocks->phyclk_khz > clk_mgr_base->clks.phyclk_khz
+ || new_clocks->fclk_khz > clk_mgr_base->clks.fclk_khz
+ || new_clocks->dcfclk_khz > clk_mgr_base->clks.dcfclk_khz)
+ send_request_to_increase = true;
+
+ if (should_set_clock(safe_to_lower, new_clocks->phyclk_khz, clk_mgr_base->clks.phyclk_khz)) {
+ clk_mgr_base->clks.phyclk_khz = new_clocks->phyclk_khz;
+ send_request_to_lower = true;
+ }
+
+ // F Clock
+ if (debug->force_fclk_khz != 0)
+ new_clocks->fclk_khz = debug->force_fclk_khz;
+
+ if (should_set_clock(safe_to_lower, new_clocks->fclk_khz, clk_mgr_base->clks.fclk_khz)) {
+ clk_mgr_base->clks.fclk_khz = new_clocks->fclk_khz;
+ send_request_to_lower = true;
+ }
+
+ //DCF Clock
+ if (should_set_clock(safe_to_lower, new_clocks->dcfclk_khz, clk_mgr_base->clks.dcfclk_khz)) {
+ clk_mgr_base->clks.dcfclk_khz = new_clocks->dcfclk_khz;
+ send_request_to_lower = true;
+ }
+
+ if (should_set_clock(safe_to_lower,
+ new_clocks->dcfclk_deep_sleep_khz, clk_mgr_base->clks.dcfclk_deep_sleep_khz)) {
+ clk_mgr_base->clks.dcfclk_deep_sleep_khz = new_clocks->dcfclk_deep_sleep_khz;
+ send_request_to_lower = true;
+ }
+
+ /* make sure dcf clk is before dpp clk to
+ * make sure we have enough voltage to run dpp clk
+ */
+ if (send_request_to_increase) {
+ /*use dcfclk to request voltage*/
+ if (pp_smu->set_hard_min_fclk_by_freq &&
+ pp_smu->set_hard_min_dcfclk_by_freq &&
+ pp_smu->set_min_deep_sleep_dcfclk) {
+ pp_smu->set_hard_min_fclk_by_freq(&pp_smu->pp_smu, new_clocks->fclk_khz / 1000);
+ pp_smu->set_hard_min_dcfclk_by_freq(&pp_smu->pp_smu, new_clocks->dcfclk_khz / 1000);
+ pp_smu->set_min_deep_sleep_dcfclk(&pp_smu->pp_smu, (new_clocks->dcfclk_deep_sleep_khz + 999) / 1000);
+ }
+ }
+
+ /* dcn1 dppclk is tied to dispclk */
+ /* program dispclk on = as a w/a for sleep resume clock ramping issues */
+ if (should_set_clock(safe_to_lower, new_clocks->dispclk_khz, clk_mgr_base->clks.dispclk_khz)
+ || new_clocks->dispclk_khz == clk_mgr_base->clks.dispclk_khz) {
+ ramp_up_dispclk_with_dpp(clk_mgr, dc, new_clocks);
+ clk_mgr_base->clks.dispclk_khz = new_clocks->dispclk_khz;
+ send_request_to_lower = true;
+ }
+
+ if (!send_request_to_increase && send_request_to_lower) {
+ /*use dcfclk to request voltage*/
+ if (pp_smu->set_hard_min_fclk_by_freq &&
+ pp_smu->set_hard_min_dcfclk_by_freq &&
+ pp_smu->set_min_deep_sleep_dcfclk) {
+ pp_smu->set_hard_min_fclk_by_freq(&pp_smu->pp_smu, new_clocks->fclk_khz / 1000);
+ pp_smu->set_hard_min_dcfclk_by_freq(&pp_smu->pp_smu, new_clocks->dcfclk_khz / 1000);
+ pp_smu->set_min_deep_sleep_dcfclk(&pp_smu->pp_smu, (new_clocks->dcfclk_deep_sleep_khz + 999) / 1000);
+ }
+ }
+}
+
+static struct clk_mgr_funcs rv1_clk_funcs = {
+ .get_dp_ref_clk_frequency = dce12_get_dp_ref_freq_khz,
+ .update_clocks = rv1_update_clocks,
+};
+
+static struct clk_mgr_internal_funcs rv1_clk_internal_funcs = {
+ .set_dispclk = rv1_vbios_smu_set_dispclk,
+ .set_dprefclk = dce112_set_dprefclk
+};
+
+void rv1_clk_mgr_construct(struct dc_context *ctx, struct clk_mgr_internal *clk_mgr, struct pp_smu_funcs *pp_smu)
+{
+ struct dc_debug_options *debug = &ctx->dc->debug;
+ struct dc_bios *bp = ctx->dc_bios;
+ struct dc_firmware_info fw_info = { { 0 } };
+
+ clk_mgr->base.ctx = ctx;
+ clk_mgr->pp_smu = pp_smu;
+ clk_mgr->base.funcs = &rv1_clk_funcs;
+ clk_mgr->funcs = &rv1_clk_internal_funcs;
+
+ clk_mgr->dfs_bypass_disp_clk = 0;
+
+ clk_mgr->dprefclk_ss_percentage = 0;
+ clk_mgr->dprefclk_ss_divider = 1000;
+ clk_mgr->ss_on_dprefclk = false;
+ clk_mgr->base.dprefclk_khz = 600000;
+
+ if (bp->integrated_info)
+ clk_mgr->dentist_vco_freq_khz = bp->integrated_info->dentist_vco_freq;
+ if (clk_mgr->dentist_vco_freq_khz == 0) {
+ bp->funcs->get_firmware_info(bp, &fw_info);
+ clk_mgr->dentist_vco_freq_khz = fw_info.smu_gpu_pll_output_freq;
+ if (clk_mgr->dentist_vco_freq_khz == 0)
+ clk_mgr->dentist_vco_freq_khz = 3600000;
+ }
+
+ if (!debug->disable_dfs_bypass && bp->integrated_info)
+ if (bp->integrated_info->gpu_cap_info & DFS_BYPASS_ENABLE)
+ clk_mgr->dfs_bypass_enabled = true;
+
+ dce_clock_read_ss_info(clk_mgr);
+}
+
+
projects / openwrt / staging / blogic.git / commitdiff