Merge airlied/drm-next into drm-misc-next

Bake in the conflict between the drm_print.h extraction and the
addition of DRM_DEBUG_LEASES since we lost it a few too many times.

Also fix a new use of drm_plane_helper_check_state in msm to follow
Ville's conversion in

commit a01cb8ba3f6282934cff65e89ab36b18b14cbe27
Author: Ville Syrjälä <ville.syrjala@linux.intel.com>
Date:   Wed Nov 1 22:16:19 2017 +0200

    drm: Move drm_plane_helper_check_state() into drm_atomic_helper.c

Acked-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

16 files changed, 57 insertions, 1044 deletions

diff --git a/drivers/gpu/drm/radeon/Makefile b/drivers/gpu/drm/radeon/Makefile
index be16c6390216..cf3e5985e3e7 100644
--- a/drivers/gpu/drm/radeon/Makefile
+++ b/drivers/gpu/drm/radeon/Makefile
@@ -102,8 +102,7 @@ radeon-y += \
 radeon-y += \
 	radeon_vce.o \
 	vce_v1_0.o \
-	vce_v2_0.o \
-	radeon_kfd.o
+	vce_v2_0.o
 
 radeon-$(CONFIG_VGA_SWITCHEROO) += radeon_atpx_handler.o
 radeon-$(CONFIG_ACPI) += radeon_acpi.o
diff --git a/drivers/gpu/drm/radeon/atombios_dp.c b/drivers/gpu/drm/radeon/atombios_dp.c
index 432cb46f6a34..3e798593e042 100644
--- a/drivers/gpu/drm/radeon/atombios_dp.c
+++ b/drivers/gpu/drm/radeon/atombios_dp.c
@@ -45,34 +45,32 @@ static char *pre_emph_names[] = {
 
 /***** radeon AUX functions *****/
 
-/* Atom needs data in little endian format
- * so swap as appropriate when copying data to
- * or from atom. Note that atom operates on
- * dw units.
+/* Atom needs data in little endian format so swap as appropriate when copying
+ * data to or from atom. Note that atom operates on dw units.
+ *
+ * Use to_le=true when sending data to atom and provide at least
+ * ALIGN(num_bytes,4) bytes in the dst buffer.
+ *
+ * Use to_le=false when receiving data from atom and provide ALIGN(num_bytes,4)
+ * byes in the src buffer.
  */
 void radeon_atom_copy_swap(u8 *dst, u8 *src, u8 num_bytes, bool to_le)
 {
 #ifdef __BIG_ENDIAN
-	u8 src_tmp[20], dst_tmp[20]; /* used for byteswapping */
-	u32 *dst32, *src32;
+	u32 src_tmp[5], dst_tmp[5];
 	int i;
+	u8 align_num_bytes = ALIGN(num_bytes, 4);
 
-	memcpy(src_tmp, src, num_bytes);
-	src32 = (u32 *)src_tmp;
-	dst32 = (u32 *)dst_tmp;
 	if (to_le) {
-		for (i = 0; i < ((num_bytes + 3) / 4); i++)
-			dst32[i] = cpu_to_le32(src32[i]);
-		memcpy(dst, dst_tmp, num_bytes);
+		memcpy(src_tmp, src, num_bytes);
+		for (i = 0; i < align_num_bytes / 4; i++)
+			dst_tmp[i] = cpu_to_le32(src_tmp[i]);
+		memcpy(dst, dst_tmp, align_num_bytes);
 	} else {
-		u8 dws = num_bytes & ~3;
-		for (i = 0; i < ((num_bytes + 3) / 4); i++)
-			dst32[i] = le32_to_cpu(src32[i]);
-		memcpy(dst, dst_tmp, dws);
-		if (num_bytes % 4) {
-			for (i = 0; i < (num_bytes % 4); i++)
-				dst[dws+i] = dst_tmp[dws+i];
-		}
+		memcpy(src_tmp, src, align_num_bytes);
+		for (i = 0; i < align_num_bytes / 4; i++)
+			dst_tmp[i] = le32_to_cpu(src_tmp[i]);
+		memcpy(dst, dst_tmp, num_bytes);
 	}
 #else
 	memcpy(dst, src, num_bytes);
@@ -304,10 +302,10 @@ static int convert_bpc_to_bpp(int bpc)
 
 /***** radeon specific DP functions *****/
 
-int radeon_dp_get_dp_link_config(struct drm_connector *connector,
-				 const u8 dpcd[DP_DPCD_SIZE],
-				 unsigned pix_clock,
-				 unsigned *dp_lanes, unsigned *dp_rate)
+static int radeon_dp_get_dp_link_config(struct drm_connector *connector,
+					const u8 dpcd[DP_DPCD_SIZE],
+					unsigned pix_clock,
+					unsigned *dp_lanes, unsigned *dp_rate)
 {
 	int bpp = convert_bpc_to_bpp(radeon_get_monitor_bpc(connector));
 	static const unsigned link_rates[3] = { 162000, 270000, 540000 };
diff --git a/drivers/gpu/drm/radeon/ci_dpm.c b/drivers/gpu/drm/radeon/ci_dpm.c
index c97fbb2ab48b..7e1b04dc5593 100644
--- a/drivers/gpu/drm/radeon/ci_dpm.c
+++ b/drivers/gpu/drm/radeon/ci_dpm.c
@@ -184,6 +184,7 @@ static int ci_set_overdrive_target_tdp(struct radeon_device *rdev,
 				       u32 target_tdp);
 static int ci_update_uvd_dpm(struct radeon_device *rdev, bool gate);
 
+static PPSMC_Result ci_send_msg_to_smc(struct radeon_device *rdev, PPSMC_Msg msg);
 static PPSMC_Result ci_send_msg_to_smc_with_parameter(struct radeon_device *rdev,
 						      PPSMC_Msg msg, u32 parameter);
 
@@ -1651,6 +1652,27 @@ static int ci_notify_hw_of_power_source(struct radeon_device *rdev,
 }
 #endif
 
+static PPSMC_Result ci_send_msg_to_smc(struct radeon_device *rdev, PPSMC_Msg msg)
+{
+	u32 tmp;
+	int i;
+
+	if (!ci_is_smc_running(rdev))
+		return PPSMC_Result_Failed;
+
+	WREG32(SMC_MESSAGE_0, msg);
+
+	for (i = 0; i < rdev->usec_timeout; i++) {
+		tmp = RREG32(SMC_RESP_0);
+		if (tmp != 0)
+			break;
+		udelay(1);
+	}
+	tmp = RREG32(SMC_RESP_0);
+
+	return (PPSMC_Result)tmp;
+}
+
 static PPSMC_Result ci_send_msg_to_smc_with_parameter(struct radeon_device *rdev,
 						      PPSMC_Msg msg, u32 parameter)
 {
diff --git a/drivers/gpu/drm/radeon/ci_dpm.h b/drivers/gpu/drm/radeon/ci_dpm.h
index 723220ffbea2..dff2a63df38f 100644
--- a/drivers/gpu/drm/radeon/ci_dpm.h
+++ b/drivers/gpu/drm/radeon/ci_dpm.h
@@ -330,7 +330,6 @@ int ci_program_jump_on_start(struct radeon_device *rdev);
 void ci_stop_smc_clock(struct radeon_device *rdev);
 void ci_start_smc_clock(struct radeon_device *rdev);
 bool ci_is_smc_running(struct radeon_device *rdev);
-PPSMC_Result ci_send_msg_to_smc(struct radeon_device *rdev, PPSMC_Msg msg);
 PPSMC_Result ci_wait_for_smc_inactive(struct radeon_device *rdev);
 int ci_load_smc_ucode(struct radeon_device *rdev, u32 limit);
 int ci_read_smc_sram_dword(struct radeon_device *rdev,
diff --git a/drivers/gpu/drm/radeon/ci_smc.c b/drivers/gpu/drm/radeon/ci_smc.c
index 3356a21d97ec..371121913756 100644
--- a/drivers/gpu/drm/radeon/ci_smc.c
+++ b/drivers/gpu/drm/radeon/ci_smc.c
@@ -163,27 +163,6 @@ bool ci_is_smc_running(struct radeon_device *rdev)
 	return false;
 }
 
-PPSMC_Result ci_send_msg_to_smc(struct radeon_device *rdev, PPSMC_Msg msg)
-{
-	u32 tmp;
-	int i;
-
-	if (!ci_is_smc_running(rdev))
-		return PPSMC_Result_Failed;
-
-	WREG32(SMC_MESSAGE_0, msg);
-
-	for (i = 0; i < rdev->usec_timeout; i++) {
-		tmp = RREG32(SMC_RESP_0);
-		if (tmp != 0)
-			break;
-		udelay(1);
-	}
-	tmp = RREG32(SMC_RESP_0);
-
-	return (PPSMC_Result)tmp;
-}
-
 #if 0
 PPSMC_Result ci_wait_for_smc_inactive(struct radeon_device *rdev)
 {
diff --git a/drivers/gpu/drm/radeon/cik.c b/drivers/gpu/drm/radeon/cik.c
index 3cb6c55b268d..898f9a078830 100644
--- a/drivers/gpu/drm/radeon/cik.c
+++ b/drivers/gpu/drm/radeon/cik.c
@@ -33,7 +33,6 @@
 #include "cik_blit_shaders.h"
 #include "radeon_ucode.h"
 #include "clearstate_ci.h"
-#include "radeon_kfd.h"
 
 #define SH_MEM_CONFIG_GFX_DEFAULT \
 	ALIGNMENT_MODE(SH_MEM_ALIGNMENT_MODE_UNALIGNED)
@@ -5684,10 +5683,9 @@ int cik_vm_init(struct radeon_device *rdev)
 	/*
 	 * number of VMs
 	 * VMID 0 is reserved for System
-	 * radeon graphics/compute will use VMIDs 1-7
-	 * amdkfd will use VMIDs 8-15
+	 * radeon graphics/compute will use VMIDs 1-15
 	 */
-	rdev->vm_manager.nvm = RADEON_NUM_OF_VMIDS;
+	rdev->vm_manager.nvm = 16;
 	/* base offset of vram pages */
 	if (rdev->flags & RADEON_IS_IGP) {
 		u64 tmp = RREG32(MC_VM_FB_OFFSET);
@@ -7589,9 +7587,6 @@ restart_ih:
 		/* wptr/rptr are in bytes! */
 		ring_index = rptr / 4;
 
-		radeon_kfd_interrupt(rdev,
-				(const void *) &rdev->ih.ring[ring_index]);
-
 		src_id =  le32_to_cpu(rdev->ih.ring[ring_index]) & 0xff;
 		src_data = le32_to_cpu(rdev->ih.ring[ring_index + 1]) & 0xfffffff;
 		ring_id = le32_to_cpu(rdev->ih.ring[ring_index + 2]) & 0xff;
@@ -8486,10 +8481,6 @@ static int cik_startup(struct radeon_device *rdev)
 	if (r)
 		return r;
 
-	r = radeon_kfd_resume(rdev);
-	if (r)
-		return r;
-
 	return 0;
 }
 
@@ -8538,7 +8529,6 @@ int cik_resume(struct radeon_device *rdev)
  */
 int cik_suspend(struct radeon_device *rdev)
 {
-	radeon_kfd_suspend(rdev);
 	radeon_pm_suspend(rdev);
 	radeon_audio_fini(rdev);
 	radeon_vm_manager_fini(rdev);
diff --git a/drivers/gpu/drm/radeon/cikd.h b/drivers/gpu/drm/radeon/cikd.h
index e21015475ed5..cda16fcd43bb 100644
--- a/drivers/gpu/drm/radeon/cikd.h
+++ b/drivers/gpu/drm/radeon/cikd.h
@@ -30,8 +30,6 @@
 #define CIK_RB_BITMAP_WIDTH_PER_SH     2
 #define HAWAII_RB_BITMAP_WIDTH_PER_SH  4
 
-#define RADEON_NUM_OF_VMIDS	8
-
 /* DIDT IND registers */
 #define DIDT_SQ_CTRL0                                     0x0
 #       define DIDT_CTRL_EN                               (1 << 0)
diff --git a/drivers/gpu/drm/radeon/r100.c b/drivers/gpu/drm/radeon/r100.c
index c31e660e35db..7d39ed63e5be 100644
--- a/drivers/gpu/drm/radeon/r100.c
+++ b/drivers/gpu/drm/radeon/r100.c
@@ -1456,7 +1456,7 @@ int r100_cs_packet_parse_vline(struct radeon_cs_parser *p)
 	header = radeon_get_ib_value(p, h_idx);
 	crtc_id = radeon_get_ib_value(p, h_idx + 5);
 	reg = R100_CP_PACKET0_GET_REG(header);
-	crtc = drm_crtc_find(p->rdev->ddev, crtc_id);
+	crtc = drm_crtc_find(p->rdev->ddev, p->filp, crtc_id);
 	if (!crtc) {
 		DRM_ERROR("cannot find crtc %d\n", crtc_id);
 		return -ENOENT;
diff --git a/drivers/gpu/drm/radeon/r600_cs.c b/drivers/gpu/drm/radeon/r600_cs.c
index 97fd58e97043..c96b31950ca7 100644
--- a/drivers/gpu/drm/radeon/r600_cs.c
+++ b/drivers/gpu/drm/radeon/r600_cs.c
@@ -887,7 +887,7 @@ int r600_cs_common_vline_parse(struct radeon_cs_parser *p,
 	crtc_id = radeon_get_ib_value(p, h_idx + 2 + 7 + 1);
 	reg = R600_CP_PACKET0_GET_REG(header);
 
-	crtc = drm_crtc_find(p->rdev->ddev, crtc_id);
+	crtc = drm_crtc_find(p->rdev->ddev, p->filp, crtc_id);
 	if (!crtc) {
 		DRM_ERROR("cannot find crtc %d\n", crtc_id);
 		return -ENOENT;
diff --git a/drivers/gpu/drm/radeon/radeon.h b/drivers/gpu/drm/radeon/radeon.h
index 8cbaeec090c9..a8e546569858 100644
--- a/drivers/gpu/drm/radeon/radeon.h
+++ b/drivers/gpu/drm/radeon/radeon.h
@@ -2456,9 +2456,6 @@ struct radeon_device {
 	u64 vram_pin_size;
 	u64 gart_pin_size;
 
-	/* amdkfd interface */
-	struct kfd_dev		*kfd;
-
 	struct mutex	mn_lock;
 	DECLARE_HASHTABLE(mn_hash, 7);
 };
diff --git a/drivers/gpu/drm/radeon/radeon_connectors.c b/drivers/gpu/drm/radeon/radeon_connectors.c
index 9e45127329d7..5012f5e47a1e 100644
--- a/drivers/gpu/drm/radeon/radeon_connectors.c
+++ b/drivers/gpu/drm/radeon/radeon_connectors.c
@@ -263,7 +263,7 @@ radeon_connector_update_scratch_regs(struct drm_connector *connector, enum drm_c
 		if (connector->encoder_ids[i] == 0)
 			break;
 
-		encoder = drm_encoder_find(connector->dev,
+		encoder = drm_encoder_find(connector->dev, NULL,
 					   connector->encoder_ids[i]);
 		if (!encoder)
 			continue;
@@ -290,7 +290,7 @@ static struct drm_encoder *radeon_find_encoder(struct drm_connector *connector,
 		if (connector->encoder_ids[i] == 0)
 			break;
 
-		encoder = drm_encoder_find(connector->dev, connector->encoder_ids[i]);
+		encoder = drm_encoder_find(connector->dev, NULL, connector->encoder_ids[i]);
 		if (!encoder)
 			continue;
 
@@ -403,7 +403,7 @@ static struct drm_encoder *radeon_best_single_encoder(struct drm_connector *conn
 	int enc_id = connector->encoder_ids[0];
 	/* pick the encoder ids */
 	if (enc_id)
-		return drm_encoder_find(connector->dev, enc_id);
+		return drm_encoder_find(connector->dev, NULL, enc_id);
 	return NULL;
 }
 
@@ -1367,7 +1367,7 @@ radeon_dvi_detect(struct drm_connector *connector, bool force)
 			if (connector->encoder_ids[i] == 0)
 				break;
 
-			encoder = drm_encoder_find(connector->dev,
+			encoder = drm_encoder_find(connector->dev, NULL,
 						   connector->encoder_ids[i]);
 			if (!encoder)
 				continue;
@@ -1453,7 +1453,7 @@ static struct drm_encoder *radeon_dvi_encoder(struct drm_connector *connector)
 		if (connector->encoder_ids[i] == 0)
 			break;
 
-		encoder = drm_encoder_find(connector->dev, connector->encoder_ids[i]);
+		encoder = drm_encoder_find(connector->dev, NULL, connector->encoder_ids[i]);
 		if (!encoder)
 			continue;
 
@@ -1472,7 +1472,7 @@ static struct drm_encoder *radeon_dvi_encoder(struct drm_connector *connector)
 	/* then check use digitial */
 	/* pick the first one */
 	if (enc_id)
-		return drm_encoder_find(connector->dev, enc_id);
+		return drm_encoder_find(connector->dev, NULL, enc_id);
 	return NULL;
 }
 
@@ -1619,7 +1619,7 @@ u16 radeon_connector_encoder_get_dp_bridge_encoder_id(struct drm_connector *conn
 		if (connector->encoder_ids[i] == 0)
 			break;
 
-		encoder = drm_encoder_find(connector->dev, connector->encoder_ids[i]);
+		encoder = drm_encoder_find(connector->dev, NULL, connector->encoder_ids[i]);
 		if (!encoder)
 			continue;
 
@@ -1648,7 +1648,7 @@ static bool radeon_connector_encoder_is_hbr2(struct drm_connector *connector)
 		if (connector->encoder_ids[i] == 0)
 			break;
 
-		encoder = drm_encoder_find(connector->dev, connector->encoder_ids[i]);
+		encoder = drm_encoder_find(connector->dev, NULL, connector->encoder_ids[i]);
 		if (!encoder)
 			continue;
 
diff --git a/drivers/gpu/drm/radeon/radeon_drv.c b/drivers/gpu/drm/radeon/radeon_drv.c
index f4becad0a78c..31dd04f6baa1 100644
--- a/drivers/gpu/drm/radeon/radeon_drv.c
+++ b/drivers/gpu/drm/radeon/radeon_drv.c
@@ -43,7 +43,6 @@
 #include <drm/drm_fb_helper.h>
 
 #include <drm/drm_crtc_helper.h>
-#include "radeon_kfd.h"
 
 /*
  * KMS wrapper.
@@ -338,14 +337,6 @@ static int radeon_pci_probe(struct pci_dev *pdev,
 {
 	int ret;
 
-	/*
-	 * Initialize amdkfd before starting radeon. If it was not loaded yet,
-	 * defer radeon probing
-	 */
-	ret = radeon_kfd_init();
-	if (ret == -EPROBE_DEFER)
-		return ret;
-
 	if (vga_switcheroo_client_probe_defer(pdev))
 		return -EPROBE_DEFER;
 
@@ -645,7 +636,6 @@ static int __init radeon_init(void)
 
 static void __exit radeon_exit(void)
 {
-	radeon_kfd_fini();
 	pci_unregister_driver(pdriver);
 	radeon_unregister_atpx_handler();
 }
diff --git a/drivers/gpu/drm/radeon/radeon_kfd.c b/drivers/gpu/drm/radeon/radeon_kfd.c
deleted file mode 100644
index a2ac8ac0930d..000000000000
--- a/drivers/gpu/drm/radeon/radeon_kfd.c
+++ /dev/null
@@ -1,901 +0,0 @@
-/*
- * Copyright 2014 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.
- */
-
-#include <linux/module.h>
-#include <linux/fdtable.h>
-#include <linux/uaccess.h>
-#include <drm/drmP.h>
-#include "radeon.h"
-#include "cikd.h"
-#include "cik_reg.h"
-#include "radeon_kfd.h"
-#include "radeon_ucode.h"
-#include <linux/firmware.h>
-#include "cik_structs.h"
-
-#define CIK_PIPE_PER_MEC	(4)
-
-static const uint32_t watchRegs[MAX_WATCH_ADDRESSES * ADDRESS_WATCH_REG_MAX] = {
-	TCP_WATCH0_ADDR_H, TCP_WATCH0_ADDR_L, TCP_WATCH0_CNTL,
-	TCP_WATCH1_ADDR_H, TCP_WATCH1_ADDR_L, TCP_WATCH1_CNTL,
-	TCP_WATCH2_ADDR_H, TCP_WATCH2_ADDR_L, TCP_WATCH2_CNTL,
-	TCP_WATCH3_ADDR_H, TCP_WATCH3_ADDR_L, TCP_WATCH3_CNTL
-};
-
-struct kgd_mem {
-	struct radeon_bo *bo;
-	uint64_t gpu_addr;
-	void *cpu_ptr;
-};
-
-
-static int alloc_gtt_mem(struct kgd_dev *kgd, size_t size,
-			void **mem_obj, uint64_t *gpu_addr,
-			void **cpu_ptr);
-
-static void free_gtt_mem(struct kgd_dev *kgd, void *mem_obj);
-
-static uint64_t get_vmem_size(struct kgd_dev *kgd);
-static uint64_t get_gpu_clock_counter(struct kgd_dev *kgd);
-
-static uint32_t get_max_engine_clock_in_mhz(struct kgd_dev *kgd);
-
-static int alloc_pasid(unsigned int bits);
-static void free_pasid(unsigned int pasid);
-
-static uint16_t get_fw_version(struct kgd_dev *kgd, enum kgd_engine_type type);
-
-/*
- * Register access functions
- */
-
-static void kgd_program_sh_mem_settings(struct kgd_dev *kgd, uint32_t vmid,
-		uint32_t sh_mem_config,	uint32_t sh_mem_ape1_base,
-		uint32_t sh_mem_ape1_limit, uint32_t sh_mem_bases);
-
-static int kgd_set_pasid_vmid_mapping(struct kgd_dev *kgd, unsigned int pasid,
-					unsigned int vmid);
-
-static int kgd_init_pipeline(struct kgd_dev *kgd, uint32_t pipe_id,
-				uint32_t hpd_size, uint64_t hpd_gpu_addr);
-static int kgd_init_interrupts(struct kgd_dev *kgd, uint32_t pipe_id);
-static int kgd_hqd_load(struct kgd_dev *kgd, void *mqd, uint32_t pipe_id,
-			uint32_t queue_id, uint32_t __user *wptr,
-			uint32_t wptr_shift, uint32_t wptr_mask,
-			struct mm_struct *mm);
-static int kgd_hqd_sdma_load(struct kgd_dev *kgd, void *mqd);
-static bool kgd_hqd_is_occupied(struct kgd_dev *kgd, uint64_t queue_address,
-				uint32_t pipe_id, uint32_t queue_id);
-
-static int kgd_hqd_destroy(struct kgd_dev *kgd, void *mqd, uint32_t reset_type,
-				unsigned int timeout, uint32_t pipe_id,
-				uint32_t queue_id);
-static bool kgd_hqd_sdma_is_occupied(struct kgd_dev *kgd, void *mqd);
-static int kgd_hqd_sdma_destroy(struct kgd_dev *kgd, void *mqd,
-				unsigned int timeout);
-static int kgd_address_watch_disable(struct kgd_dev *kgd);
-static int kgd_address_watch_execute(struct kgd_dev *kgd,
-					unsigned int watch_point_id,
-					uint32_t cntl_val,
-					uint32_t addr_hi,
-					uint32_t addr_lo);
-static int kgd_wave_control_execute(struct kgd_dev *kgd,
-					uint32_t gfx_index_val,
-					uint32_t sq_cmd);
-static uint32_t kgd_address_watch_get_offset(struct kgd_dev *kgd,
-					unsigned int watch_point_id,
-					unsigned int reg_offset);
-
-static bool get_atc_vmid_pasid_mapping_valid(struct kgd_dev *kgd, uint8_t vmid);
-static uint16_t get_atc_vmid_pasid_mapping_pasid(struct kgd_dev *kgd,
-							uint8_t vmid);
-static void write_vmid_invalidate_request(struct kgd_dev *kgd, uint8_t vmid);
-
-static const struct kfd2kgd_calls kfd2kgd = {
-	.init_gtt_mem_allocation = alloc_gtt_mem,
-	.free_gtt_mem = free_gtt_mem,
-	.get_vmem_size = get_vmem_size,
-	.get_gpu_clock_counter = get_gpu_clock_counter,
-	.get_max_engine_clock_in_mhz = get_max_engine_clock_in_mhz,
-	.alloc_pasid = alloc_pasid,
-	.free_pasid = free_pasid,
-	.program_sh_mem_settings = kgd_program_sh_mem_settings,
-	.set_pasid_vmid_mapping = kgd_set_pasid_vmid_mapping,
-	.init_pipeline = kgd_init_pipeline,
-	.init_interrupts = kgd_init_interrupts,
-	.hqd_load = kgd_hqd_load,
-	.hqd_sdma_load = kgd_hqd_sdma_load,
-	.hqd_is_occupied = kgd_hqd_is_occupied,
-	.hqd_sdma_is_occupied = kgd_hqd_sdma_is_occupied,
-	.hqd_destroy = kgd_hqd_destroy,
-	.hqd_sdma_destroy = kgd_hqd_sdma_destroy,
-	.address_watch_disable = kgd_address_watch_disable,
-	.address_watch_execute = kgd_address_watch_execute,
-	.wave_control_execute = kgd_wave_control_execute,
-	.address_watch_get_offset = kgd_address_watch_get_offset,
-	.get_atc_vmid_pasid_mapping_pasid = get_atc_vmid_pasid_mapping_pasid,
-	.get_atc_vmid_pasid_mapping_valid = get_atc_vmid_pasid_mapping_valid,
-	.write_vmid_invalidate_request = write_vmid_invalidate_request,
-	.get_fw_version = get_fw_version
-};
-
-static const struct kgd2kfd_calls *kgd2kfd;
-
-int radeon_kfd_init(void)
-{
-	int ret;
-
-#if defined(CONFIG_HSA_AMD_MODULE)
-	int (*kgd2kfd_init_p)(unsigned, const struct kgd2kfd_calls**);
-
-	kgd2kfd_init_p = symbol_request(kgd2kfd_init);
-
-	if (kgd2kfd_init_p == NULL)
-		return -ENOENT;
-
-	ret = kgd2kfd_init_p(KFD_INTERFACE_VERSION, &kgd2kfd);
-	if (ret) {
-		symbol_put(kgd2kfd_init);
-		kgd2kfd = NULL;
-	}
-
-#elif defined(CONFIG_HSA_AMD)
-	ret = kgd2kfd_init(KFD_INTERFACE_VERSION, &kgd2kfd);
-	if (ret)
-		kgd2kfd = NULL;
-
-#else
-	ret = -ENOENT;
-#endif
-
-	return ret;
-}
-
-void radeon_kfd_fini(void)
-{
-	if (kgd2kfd) {
-		kgd2kfd->exit();
-		symbol_put(kgd2kfd_init);
-	}
-}
-
-void radeon_kfd_device_probe(struct radeon_device *rdev)
-{
-	if (kgd2kfd)
-		rdev->kfd = kgd2kfd->probe((struct kgd_dev *)rdev,
-			rdev->pdev, &kfd2kgd);
-}
-
-void radeon_kfd_device_init(struct radeon_device *rdev)
-{
-	int i, queue, pipe, mec;
-
-	if (rdev->kfd) {
-		struct kgd2kfd_shared_resources gpu_resources = {
-			.compute_vmid_bitmap = 0xFF00,
-			.num_pipe_per_mec = 4,
-			.num_queue_per_pipe = 8
-		};
-
-		bitmap_zero(gpu_resources.queue_bitmap, KGD_MAX_QUEUES);
-
-		for (i = 0; i < KGD_MAX_QUEUES; ++i) {
-			queue = i % gpu_resources.num_queue_per_pipe;
-			pipe = (i / gpu_resources.num_queue_per_pipe)
-				% gpu_resources.num_pipe_per_mec;
-			mec = (i / gpu_resources.num_queue_per_pipe)
-				/ gpu_resources.num_pipe_per_mec;
-
-			if (mec == 0 && pipe > 0)
-				set_bit(i, gpu_resources.queue_bitmap);
-		}
-
-		radeon_doorbell_get_kfd_info(rdev,
-				&gpu_resources.doorbell_physical_address,
-				&gpu_resources.doorbell_aperture_size,
-				&gpu_resources.doorbell_start_offset);
-
-		kgd2kfd->device_init(rdev->kfd, &gpu_resources);
-	}
-}
-
-void radeon_kfd_device_fini(struct radeon_device *rdev)
-{
-	if (rdev->kfd) {
-		kgd2kfd->device_exit(rdev->kfd);
-		rdev->kfd = NULL;
-	}
-}
-
-void radeon_kfd_interrupt(struct radeon_device *rdev, const void *ih_ring_entry)
-{
-	if (rdev->kfd)
-		kgd2kfd->interrupt(rdev->kfd, ih_ring_entry);
-}
-
-void radeon_kfd_suspend(struct radeon_device *rdev)
-{
-	if (rdev->kfd)
-		kgd2kfd->suspend(rdev->kfd);
-}
-
-int radeon_kfd_resume(struct radeon_device *rdev)
-{
-	int r = 0;
-
-	if (rdev->kfd)
-		r = kgd2kfd->resume(rdev->kfd);
-
-	return r;
-}
-
-static int alloc_gtt_mem(struct kgd_dev *kgd, size_t size,
-			void **mem_obj, uint64_t *gpu_addr,
-			void **cpu_ptr)
-{
-	struct radeon_device *rdev = (struct radeon_device *)kgd;
-	struct kgd_mem **mem = (struct kgd_mem **) mem_obj;
-	int r;
-
-	BUG_ON(kgd == NULL);
-	BUG_ON(gpu_addr == NULL);
-	BUG_ON(cpu_ptr == NULL);
-
-	*mem = kmalloc(sizeof(struct kgd_mem), GFP_KERNEL);
-	if ((*mem) == NULL)
-		return -ENOMEM;
-
-	r = radeon_bo_create(rdev, size, PAGE_SIZE, true, RADEON_GEM_DOMAIN_GTT,
-				RADEON_GEM_GTT_WC, NULL, NULL, &(*mem)->bo);
-	if (r) {
-		dev_err(rdev->dev,
-			"failed to allocate BO for amdkfd (%d)\n", r);
-		return r;
-	}
-
-	/* map the buffer */
-	r = radeon_bo_reserve((*mem)->bo, true);
-	if (r) {
-		dev_err(rdev->dev, "(%d) failed to reserve bo for amdkfd\n", r);
-		goto allocate_mem_reserve_bo_failed;
-	}
-
-	r = radeon_bo_pin((*mem)->bo, RADEON_GEM_DOMAIN_GTT,
-				&(*mem)->gpu_addr);
-	if (r) {
-		dev_err(rdev->dev, "(%d) failed to pin bo for amdkfd\n", r);
-		goto allocate_mem_pin_bo_failed;
-	}
-	*gpu_addr = (*mem)->gpu_addr;
-
-	r = radeon_bo_kmap((*mem)->bo, &(*mem)->cpu_ptr);
-	if (r) {
-		dev_err(rdev->dev,
-			"(%d) failed to map bo to kernel for amdkfd\n", r);
-		goto allocate_mem_kmap_bo_failed;
-	}
-	*cpu_ptr = (*mem)->cpu_ptr;
-
-	radeon_bo_unreserve((*mem)->bo);
-
-	return 0;
-
-allocate_mem_kmap_bo_failed:
-	radeon_bo_unpin((*mem)->bo);
-allocate_mem_pin_bo_failed:
-	radeon_bo_unreserve((*mem)->bo);
-allocate_mem_reserve_bo_failed:
-	radeon_bo_unref(&(*mem)->bo);
-
-	return r;
-}
-
-static void free_gtt_mem(struct kgd_dev *kgd, void *mem_obj)
-{
-	struct kgd_mem *mem = (struct kgd_mem *) mem_obj;
-
-	BUG_ON(mem == NULL);
-
-	radeon_bo_reserve(mem->bo, true);
-	radeon_bo_kunmap(mem->bo);
-	radeon_bo_unpin(mem->bo);
-	radeon_bo_unreserve(mem->bo);
-	radeon_bo_unref(&(mem->bo));
-	kfree(mem);
-}
-
-static uint64_t get_vmem_size(struct kgd_dev *kgd)
-{
-	struct radeon_device *rdev = (struct radeon_device *)kgd;
-
-	BUG_ON(kgd == NULL);
-
-	return rdev->mc.real_vram_size;
-}
-
-static uint64_t get_gpu_clock_counter(struct kgd_dev *kgd)
-{
-	struct radeon_device *rdev = (struct radeon_device *)kgd;
-
-	return rdev->asic->get_gpu_clock_counter(rdev);
-}
-
-static uint32_t get_max_engine_clock_in_mhz(struct kgd_dev *kgd)
-{
-	struct radeon_device *rdev = (struct radeon_device *)kgd;
-
-	/* The sclk is in quantas of 10kHz */
-	return rdev->pm.dpm.dyn_state.max_clock_voltage_on_ac.sclk / 100;
-}
-
-/*
- * PASID manager
- */
-static DEFINE_IDA(pasid_ida);
-
-int alloc_pasid(unsigned int bits)
-{
-	int pasid = -EINVAL;
-
-	for (bits = min(bits, 31U); bits > 0; bits--) {
-		pasid = ida_simple_get(&pasid_ida,
-				       1U << (bits - 1), 1U << bits,
-				       GFP_KERNEL);
-		if (pasid != -ENOSPC)
-			break;
-	}
-
-	return pasid;
-}
-
-void free_pasid(unsigned int pasid)
-{
-	ida_simple_remove(&pasid_ida, pasid);
-}
-
-static inline struct radeon_device *get_radeon_device(struct kgd_dev *kgd)
-{
-	return (struct radeon_device *)kgd;
-}
-
-static void write_register(struct kgd_dev *kgd, uint32_t offset, uint32_t value)
-{
-	struct radeon_device *rdev = get_radeon_device(kgd);
-
-	writel(value, (void __iomem *)(rdev->rmmio + offset));
-}
-
-static uint32_t read_register(struct kgd_dev *kgd, uint32_t offset)
-{
-	struct radeon_device *rdev = get_radeon_device(kgd);
-
-	return readl((void __iomem *)(rdev->rmmio + offset));
-}
-
-static void lock_srbm(struct kgd_dev *kgd, uint32_t mec, uint32_t pipe,
-			uint32_t queue, uint32_t vmid)
-{
-	struct radeon_device *rdev = get_radeon_device(kgd);
-	uint32_t value = PIPEID(pipe) | MEID(mec) | VMID(vmid) | QUEUEID(queue);
-
-	mutex_lock(&rdev->srbm_mutex);
-	write_register(kgd, SRBM_GFX_CNTL, value);
-}
-
-static void unlock_srbm(struct kgd_dev *kgd)
-{
-	struct radeon_device *rdev = get_radeon_device(kgd);
-
-	write_register(kgd, SRBM_GFX_CNTL, 0);
-	mutex_unlock(&rdev->srbm_mutex);
-}
-
-static void acquire_queue(struct kgd_dev *kgd, uint32_t pipe_id,
-				uint32_t queue_id)
-{
-	uint32_t mec = (++pipe_id / CIK_PIPE_PER_MEC) + 1;
-	uint32_t pipe = (pipe_id % CIK_PIPE_PER_MEC);
-
-	lock_srbm(kgd, mec, pipe, queue_id, 0);
-}
-
-static void release_queue(struct kgd_dev *kgd)
-{
-	unlock_srbm(kgd);
-}
-
-static void kgd_program_sh_mem_settings(struct kgd_dev *kgd, uint32_t vmid,
-					uint32_t sh_mem_config,
-					uint32_t sh_mem_ape1_base,
-					uint32_t sh_mem_ape1_limit,
-					uint32_t sh_mem_bases)
-{
-	lock_srbm(kgd, 0, 0, 0, vmid);
-
-	write_register(kgd, SH_MEM_CONFIG, sh_mem_config);
-	write_register(kgd, SH_MEM_APE1_BASE, sh_mem_ape1_base);
-	write_register(kgd, SH_MEM_APE1_LIMIT, sh_mem_ape1_limit);
-	write_register(kgd, SH_MEM_BASES, sh_mem_bases);
-
-	unlock_srbm(kgd);
-}
-
-static int kgd_set_pasid_vmid_mapping(struct kgd_dev *kgd, unsigned int pasid,
-					unsigned int vmid)
-{
-	/*
-	 * We have to assume that there is no outstanding mapping.
-	 * The ATC_VMID_PASID_MAPPING_UPDATE_STATUS bit could be 0
-	 * because a mapping is in progress or because a mapping finished and
-	 * the SW cleared it.
-	 * So the protocol is to always wait & clear.
-	 */
-	uint32_t pasid_mapping = (pasid == 0) ? 0 : (uint32_t)pasid |
-					ATC_VMID_PASID_MAPPING_VALID_MASK;
-
-	write_register(kgd, ATC_VMID0_PASID_MAPPING + vmid*sizeof(uint32_t),
-			pasid_mapping);
-
-	while (!(read_register(kgd, ATC_VMID_PASID_MAPPING_UPDATE_STATUS) &
-								(1U << vmid)))
-		cpu_relax();
-	write_register(kgd, ATC_VMID_PASID_MAPPING_UPDATE_STATUS, 1U << vmid);
-
-	/* Mapping vmid to pasid also for IH block */
-	write_register(kgd, IH_VMID_0_LUT + vmid * sizeof(uint32_t),
-			pasid_mapping);
-
-	return 0;
-}
-
-static int kgd_init_pipeline(struct kgd_dev *kgd, uint32_t pipe_id,
-				uint32_t hpd_size, uint64_t hpd_gpu_addr)
-{
-	/* nothing to do here */
-	return 0;
-}
-
-static int kgd_init_interrupts(struct kgd_dev *kgd, uint32_t pipe_id)
-{
-	uint32_t mec;
-	uint32_t pipe;
-
-	mec = (pipe_id / CIK_PIPE_PER_MEC) + 1;
-	pipe = (pipe_id % CIK_PIPE_PER_MEC);
-
-	lock_srbm(kgd, mec, pipe, 0, 0);
-
-	write_register(kgd, CPC_INT_CNTL,
-			TIME_STAMP_INT_ENABLE | OPCODE_ERROR_INT_ENABLE);
-
-	unlock_srbm(kgd);
-
-	return 0;
-}
-
-static inline uint32_t get_sdma_base_addr(struct cik_sdma_rlc_registers *m)
-{
-	uint32_t retval;
-
-	retval = m->sdma_engine_id * SDMA1_REGISTER_OFFSET +
-			m->sdma_queue_id * KFD_CIK_SDMA_QUEUE_OFFSET;
-
-	pr_debug("kfd: sdma base address: 0x%x\n", retval);
-
-	return retval;
-}
-
-static inline struct cik_mqd *get_mqd(void *mqd)
-{
-	return (struct cik_mqd *)mqd;
-}
-
-static inline struct cik_sdma_rlc_registers *get_sdma_mqd(void *mqd)
-{
-	return (struct cik_sdma_rlc_registers *)mqd;
-}
-
-static int kgd_hqd_load(struct kgd_dev *kgd, void *mqd, uint32_t pipe_id,
-			uint32_t queue_id, uint32_t __user *wptr,
-			uint32_t wptr_shift, uint32_t wptr_mask,
-			struct mm_struct *mm)
-{
-	uint32_t wptr_shadow, is_wptr_shadow_valid;
-	struct cik_mqd *m;
-
-	m = get_mqd(mqd);
-
-	is_wptr_shadow_valid = !get_user(wptr_shadow, wptr);
-
-	acquire_queue(kgd, pipe_id, queue_id);
-	write_register(kgd, CP_MQD_BASE_ADDR, m->cp_mqd_base_addr_lo);
-	write_register(kgd, CP_MQD_BASE_ADDR_HI, m->cp_mqd_base_addr_hi);
-	write_register(kgd, CP_MQD_CONTROL, m->cp_mqd_control);
-
-	write_register(kgd, CP_HQD_PQ_BASE, m->cp_hqd_pq_base_lo);
-	write_register(kgd, CP_HQD_PQ_BASE_HI, m->cp_hqd_pq_base_hi);
-	write_register(kgd, CP_HQD_PQ_CONTROL, m->cp_hqd_pq_control);
-
-	write_register(kgd, CP_HQD_IB_CONTROL, m->cp_hqd_ib_control);
-	write_register(kgd, CP_HQD_IB_BASE_ADDR, m->cp_hqd_ib_base_addr_lo);
-	write_register(kgd, CP_HQD_IB_BASE_ADDR_HI, m->cp_hqd_ib_base_addr_hi);
-
-	write_register(kgd, CP_HQD_IB_RPTR, m->cp_hqd_ib_rptr);
-
-	write_register(kgd, CP_HQD_PERSISTENT_STATE,
-			m->cp_hqd_persistent_state);
-	write_register(kgd, CP_HQD_SEMA_CMD, m->cp_hqd_sema_cmd);
-	write_register(kgd, CP_HQD_MSG_TYPE, m->cp_hqd_msg_type);
-
-	write_register(kgd, CP_HQD_ATOMIC0_PREOP_LO,
-			m->cp_hqd_atomic0_preop_lo);
-
-	write_register(kgd, CP_HQD_ATOMIC0_PREOP_HI,
-			m->cp_hqd_atomic0_preop_hi);
-
-	write_register(kgd, CP_HQD_ATOMIC1_PREOP_LO,
-			m->cp_hqd_atomic1_preop_lo);
-
-	write_register(kgd, CP_HQD_ATOMIC1_PREOP_HI,
-			m->cp_hqd_atomic1_preop_hi);
-
-	write_register(kgd, CP_HQD_PQ_RPTR_REPORT_ADDR,
-			m->cp_hqd_pq_rptr_report_addr_lo);
-
-	write_register(kgd, CP_HQD_PQ_RPTR_REPORT_ADDR_HI,
-			m->cp_hqd_pq_rptr_report_addr_hi);
-
-	write_register(kgd, CP_HQD_PQ_RPTR, m->cp_hqd_pq_rptr);
-
-	write_register(kgd, CP_HQD_PQ_WPTR_POLL_ADDR,
-			m->cp_hqd_pq_wptr_poll_addr_lo);
-
-	write_register(kgd, CP_HQD_PQ_WPTR_POLL_ADDR_HI,
-			m->cp_hqd_pq_wptr_poll_addr_hi);
-
-	write_register(kgd, CP_HQD_PQ_DOORBELL_CONTROL,
-			m->cp_hqd_pq_doorbell_control);
-
-	write_register(kgd, CP_HQD_VMID, m->cp_hqd_vmid);
-
-	write_register(kgd, CP_HQD_QUANTUM, m->cp_hqd_quantum);
-
-	write_register(kgd, CP_HQD_PIPE_PRIORITY, m->cp_hqd_pipe_priority);
-	write_register(kgd, CP_HQD_QUEUE_PRIORITY, m->cp_hqd_queue_priority);
-
-	write_register(kgd, CP_HQD_IQ_RPTR, m->cp_hqd_iq_rptr);
-
-	if (is_wptr_shadow_valid)
-		write_register(kgd, CP_HQD_PQ_WPTR, wptr_shadow);
-
-	write_register(kgd, CP_HQD_ACTIVE, m->cp_hqd_active);
-	release_queue(kgd);
-
-	return 0;
-}
-
-static int kgd_hqd_sdma_load(struct kgd_dev *kgd, void *mqd)
-{
-	struct cik_sdma_rlc_registers *m;
-	uint32_t sdma_base_addr;
-
-	m = get_sdma_mqd(mqd);
-	sdma_base_addr = get_sdma_base_addr(m);
-
-	write_register(kgd,
-			sdma_base_addr + SDMA0_RLC0_VIRTUAL_ADDR,
-			m->sdma_rlc_virtual_addr);
-
-	write_register(kgd,
-			sdma_base_addr + SDMA0_RLC0_RB_BASE,
-			m->sdma_rlc_rb_base);
-
-	write_register(kgd,
-			sdma_base_addr + SDMA0_RLC0_RB_BASE_HI,
-			m->sdma_rlc_rb_base_hi);
-
-	write_register(kgd,
-			sdma_base_addr + SDMA0_RLC0_RB_RPTR_ADDR_LO,
-			m->sdma_rlc_rb_rptr_addr_lo);
-
-	write_register(kgd,
-			sdma_base_addr + SDMA0_RLC0_RB_RPTR_ADDR_HI,
-			m->sdma_rlc_rb_rptr_addr_hi);
-
-	write_register(kgd,
-			sdma_base_addr + SDMA0_RLC0_DOORBELL,
-			m->sdma_rlc_doorbell);
-
-	write_register(kgd,
-			sdma_base_addr + SDMA0_RLC0_RB_CNTL,
-			m->sdma_rlc_rb_cntl);
-
-	return 0;
-}
-
-static bool kgd_hqd_is_occupied(struct kgd_dev *kgd, uint64_t queue_address,
-				uint32_t pipe_id, uint32_t queue_id)
-{
-	uint32_t act;
-	bool retval = false;
-	uint32_t low, high;
-
-	acquire_queue(kgd, pipe_id, queue_id);
-	act = read_register(kgd, CP_HQD_ACTIVE);
-	if (act) {
-		low = lower_32_bits(queue_address >> 8);
-		high = upper_32_bits(queue_address >> 8);
-
-		if (low == read_register(kgd, CP_HQD_PQ_BASE) &&
-				high == read_register(kgd, CP_HQD_PQ_BASE_HI))
-			retval = true;
-	}
-	release_queue(kgd);
-	return retval;
-}
-
-static bool kgd_hqd_sdma_is_occupied(struct kgd_dev *kgd, void *mqd)
-{
-	struct cik_sdma_rlc_registers *m;
-	uint32_t sdma_base_addr;
-	uint32_t sdma_rlc_rb_cntl;
-
-	m = get_sdma_mqd(mqd);
-	sdma_base_addr = get_sdma_base_addr(m);
-
-	sdma_rlc_rb_cntl = read_register(kgd,
-					sdma_base_addr + SDMA0_RLC0_RB_CNTL);
-
-	if (sdma_rlc_rb_cntl & SDMA_RB_ENABLE)
-		return true;
-
-	return false;
-}
-
-static int kgd_hqd_destroy(struct kgd_dev *kgd, void *mqd, uint32_t reset_type,
-				unsigned int timeout, uint32_t pipe_id,
-				uint32_t queue_id)
-{
-	uint32_t temp;
-
-	acquire_queue(kgd, pipe_id, queue_id);
-	write_register(kgd, CP_HQD_PQ_DOORBELL_CONTROL, 0);
-
-	write_register(kgd, CP_HQD_DEQUEUE_REQUEST, reset_type);
-
-	while (true) {
-		temp = read_register(kgd, CP_HQD_ACTIVE);
-		if (temp & 0x1)
-			break;
-		if (timeout == 0) {
-			pr_err("kfd: cp queue preemption time out (%dms)\n",
-				temp);
-			release_queue(kgd);
-			return -ETIME;
-		}
-		msleep(20);
-		timeout -= 20;
-	}
-
-	release_queue(kgd);
-	return 0;
-}
-
-static int kgd_hqd_sdma_destroy(struct kgd_dev *kgd, void *mqd,
-				unsigned int timeout)
-{
-	struct cik_sdma_rlc_registers *m;
-	uint32_t sdma_base_addr;
-	uint32_t temp;
-
-	m = get_sdma_mqd(mqd);
-	sdma_base_addr = get_sdma_base_addr(m);
-
-	temp = read_register(kgd, sdma_base_addr + SDMA0_RLC0_RB_CNTL);
-	temp = temp & ~SDMA_RB_ENABLE;
-	write_register(kgd, sdma_base_addr + SDMA0_RLC0_RB_CNTL, temp);
-
-	while (true) {
-		temp = read_register(kgd, sdma_base_addr +
-						SDMA0_RLC0_CONTEXT_STATUS);
-		if (temp & SDMA_RLC_IDLE)
-			break;
-		if (timeout == 0)
-			return -ETIME;
-		msleep(20);
-		timeout -= 20;
-	}
-
-	write_register(kgd, sdma_base_addr + SDMA0_RLC0_DOORBELL, 0);
-	write_register(kgd, sdma_base_addr + SDMA0_RLC0_RB_RPTR, 0);
-	write_register(kgd, sdma_base_addr + SDMA0_RLC0_RB_WPTR, 0);
-	write_register(kgd, sdma_base_addr + SDMA0_RLC0_RB_BASE, 0);
-
-	return 0;
-}
-
-static int kgd_address_watch_disable(struct kgd_dev *kgd)
-{
-	union TCP_WATCH_CNTL_BITS cntl;
-	unsigned int i;
-
-	cntl.u32All = 0;
-
-	cntl.bitfields.valid = 0;
-	cntl.bitfields.mask = ADDRESS_WATCH_REG_CNTL_DEFAULT_MASK;
-	cntl.bitfields.atc = 1;
-
-	/* Turning off this address until we set all the registers */
-	for (i = 0; i < MAX_WATCH_ADDRESSES; i++)
-		write_register(kgd,
-				watchRegs[i * ADDRESS_WATCH_REG_MAX +
-					ADDRESS_WATCH_REG_CNTL],
-				cntl.u32All);
-
-	return 0;
-}
-
-static int kgd_address_watch_execute(struct kgd_dev *kgd,
-					unsigned int watch_point_id,
-					uint32_t cntl_val,
-					uint32_t addr_hi,
-					uint32_t addr_lo)
-{
-	union TCP_WATCH_CNTL_BITS cntl;
-
-	cntl.u32All = cntl_val;
-
-	/* Turning off this watch point until we set all the registers */
-	cntl.bitfields.valid = 0;
-	write_register(kgd,
-			watchRegs[watch_point_id * ADDRESS_WATCH_REG_MAX +
-				ADDRESS_WATCH_REG_CNTL],
-			cntl.u32All);
-
-	write_register(kgd,
-			watchRegs[watch_point_id * ADDRESS_WATCH_REG_MAX +
-				ADDRESS_WATCH_REG_ADDR_HI],
-			addr_hi);
-
-	write_register(kgd,
-			watchRegs[watch_point_id * ADDRESS_WATCH_REG_MAX +
-				ADDRESS_WATCH_REG_ADDR_LO],
-			addr_lo);
-
-	/* Enable the watch point */
-	cntl.bitfields.valid = 1;
-
-	write_register(kgd,
-			watchRegs[watch_point_id * ADDRESS_WATCH_REG_MAX +
-				ADDRESS_WATCH_REG_CNTL],
-			cntl.u32All);
-
-	return 0;
-}
-
-static int kgd_wave_control_execute(struct kgd_dev *kgd,
-					uint32_t gfx_index_val,
-					uint32_t sq_cmd)
-{
-	struct radeon_device *rdev = get_radeon_device(kgd);
-	uint32_t data;
-
-	mutex_lock(&rdev->grbm_idx_mutex);
-
-	write_register(kgd, GRBM_GFX_INDEX, gfx_index_val);
-	write_register(kgd, SQ_CMD, sq_cmd);
-
-	/*  Restore the GRBM_GFX_INDEX register  */
-
-	data = INSTANCE_BROADCAST_WRITES | SH_BROADCAST_WRITES |
-		SE_BROADCAST_WRITES;
-
-	write_register(kgd, GRBM_GFX_INDEX, data);
-
-	mutex_unlock(&rdev->grbm_idx_mutex);
-
-	return 0;
-}
-
-static uint32_t kgd_address_watch_get_offset(struct kgd_dev *kgd,
-					unsigned int watch_point_id,
-					unsigned int reg_offset)
-{
-	return watchRegs[watch_point_id * ADDRESS_WATCH_REG_MAX + reg_offset]
-		/ 4;
-}
-
-static bool get_atc_vmid_pasid_mapping_valid(struct kgd_dev *kgd, uint8_t vmid)
-{
-	uint32_t reg;
-	struct radeon_device *rdev = (struct radeon_device *) kgd;
-
-	reg = RREG32(ATC_VMID0_PASID_MAPPING + vmid*4);
-	return reg & ATC_VMID_PASID_MAPPING_VALID_MASK;
-}
-
-static uint16_t get_atc_vmid_pasid_mapping_pasid(struct kgd_dev *kgd,
-							uint8_t vmid)
-{
-	uint32_t reg;
-	struct radeon_device *rdev = (struct radeon_device *) kgd;
-
-	reg = RREG32(ATC_VMID0_PASID_MAPPING + vmid*4);
-	return reg & ATC_VMID_PASID_MAPPING_PASID_MASK;
-}
-
-static void write_vmid_invalidate_request(struct kgd_dev *kgd, uint8_t vmid)
-{
-	struct radeon_device *rdev = (struct radeon_device *) kgd;
-
-	return WREG32(VM_INVALIDATE_REQUEST, 1 << vmid);
-}
-
-static uint16_t get_fw_version(struct kgd_dev *kgd, enum kgd_engine_type type)
-{
-	struct radeon_device *rdev = (struct radeon_device *) kgd;
-	const union radeon_firmware_header *hdr;
-
-	BUG_ON(kgd == NULL || rdev->mec_fw == NULL);
-
-	switch (type) {
-	case KGD_ENGINE_PFP:
-		hdr = (const union radeon_firmware_header *) rdev->pfp_fw->data;
-		break;
-
-	case KGD_ENGINE_ME:
-		hdr = (const union radeon_firmware_header *) rdev->me_fw->data;
-		break;
-
-	case KGD_ENGINE_CE:
-		hdr = (const union radeon_firmware_header *) rdev->ce_fw->data;
-		break;
-
-	case KGD_ENGINE_MEC1:
-		hdr = (const union radeon_firmware_header *) rdev->mec_fw->data;
-		break;
-
-	case KGD_ENGINE_MEC2:
-		hdr = (const union radeon_firmware_header *)
-							rdev->mec2_fw->data;
-		break;
-
-	case KGD_ENGINE_RLC:
-		hdr = (const union radeon_firmware_header *) rdev->rlc_fw->data;
-		break;
-
-	case KGD_ENGINE_SDMA1:
-	case KGD_ENGINE_SDMA2:
-		hdr = (const union radeon_firmware_header *)
-							rdev->sdma_fw->data;
-		break;
-
-	default:
-		return 0;
-	}
-
-	if (hdr == NULL)
-		return 0;
-
-	/* Only 12 bit in use*/
-	return hdr->common.ucode_version;
-}
diff --git a/drivers/gpu/drm/radeon/radeon_kfd.h b/drivers/gpu/drm/radeon/radeon_kfd.h
deleted file mode 100644
index 9df1fea8e971..000000000000
--- a/drivers/gpu/drm/radeon/radeon_kfd.h
+++ /dev/null
@@ -1,47 +0,0 @@
-/*
- * Copyright 2014 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.
- */
-
-/*
- * radeon_kfd.h defines the private interface between the
- * AMD kernel graphics drivers and the AMD KFD.
- */
-
-#ifndef RADEON_KFD_H_INCLUDED
-#define RADEON_KFD_H_INCLUDED
-
-#include <linux/types.h>
-#include "kgd_kfd_interface.h"
-
-struct radeon_device;
-
-int radeon_kfd_init(void);
-void radeon_kfd_fini(void);
-
-void radeon_kfd_suspend(struct radeon_device *rdev);
-int radeon_kfd_resume(struct radeon_device *rdev);
-void radeon_kfd_interrupt(struct radeon_device *rdev,
-			const void *ih_ring_entry);
-void radeon_kfd_device_probe(struct radeon_device *rdev);
-void radeon_kfd_device_init(struct radeon_device *rdev);
-void radeon_kfd_device_fini(struct radeon_device *rdev);
-
-#endif /* RADEON_KFD_H_INCLUDED */
diff --git a/drivers/gpu/drm/radeon/radeon_kms.c b/drivers/gpu/drm/radeon/radeon_kms.c
index dfee8f7d94ae..cde037f213d7 100644
--- a/drivers/gpu/drm/radeon/radeon_kms.c
+++ b/drivers/gpu/drm/radeon/radeon_kms.c
@@ -34,8 +34,6 @@
 #include <linux/slab.h>
 #include <linux/pm_runtime.h>
 
-#include "radeon_kfd.h"
-
 #if defined(CONFIG_VGA_SWITCHEROO)
 bool radeon_has_atpx(void);
 #else
@@ -68,8 +66,6 @@ void radeon_driver_unload_kms(struct drm_device *dev)
 		pm_runtime_forbid(dev->dev);
 	}
 
-	radeon_kfd_device_fini(rdev);
-
 	radeon_acpi_fini(rdev);
 	
 	radeon_modeset_fini(rdev);
@@ -174,9 +170,6 @@ int radeon_driver_load_kms(struct drm_device *dev, unsigned long flags)
 				"Error during ACPI methods call\n");
 	}
 
-	radeon_kfd_device_probe(rdev);
-	radeon_kfd_device_init(rdev);
-
 	if (radeon_is_px(dev)) {
 		pm_runtime_use_autosuspend(dev->dev);
 		pm_runtime_set_autosuspend_delay(dev->dev, 5000);
diff --git a/drivers/gpu/drm/radeon/radeon_mode.h b/drivers/gpu/drm/radeon/radeon_mode.h
index da44ac234f64..ca0a7ed28c9b 100644
--- a/drivers/gpu/drm/radeon/radeon_mode.h
+++ b/drivers/gpu/drm/radeon/radeon_mode.h
@@ -762,10 +762,6 @@ extern u8 radeon_dp_getsinktype(struct radeon_connector *radeon_connector);
 extern bool radeon_dp_getdpcd(struct radeon_connector *radeon_connector);
 extern int radeon_dp_get_panel_mode(struct drm_encoder *encoder,
 				    struct drm_connector *connector);
-extern int radeon_dp_get_dp_link_config(struct drm_connector *connector,
-					const u8 *dpcd,
-					unsigned pix_clock,
-					unsigned *dp_lanes, unsigned *dp_rate);
 extern void radeon_dp_set_rx_power_state(struct drm_connector *connector,
 					 u8 power_state);
 extern void radeon_dp_aux_init(struct radeon_connector *radeon_connector);