drm/amdgpu: Add support for SRBM selection v3

Also remove code duplication in write and read regs functions.
This also fixes potential missing unlock in amdgpu_debugfs_regs_write
in case get_user would fail.

v2: Add SRBM mutex locking.
v3: Fix TO counter and fix comment location.
Signed-off-by: Andrey Grodzovsky <andrey.grodzovsky@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drivers/gpu/drm/amd/amdgpu/amdgpu.h

@@ -890,6 +890,7 @@ struct amdgpu_gfx_funcs {
 	void (*read_wave_data)(struct amdgpu_device *adev, uint32_t simd, uint32_t wave, uint32_t *dst, int *no_fields);
 	void (*read_wave_vgprs)(struct amdgpu_device *adev, uint32_t simd, uint32_t wave, uint32_t thread, uint32_t start, uint32_t size, uint32_t *dst);
 	void (*read_wave_sgprs)(struct amdgpu_device *adev, uint32_t simd, uint32_t wave, uint32_t start, uint32_t size, uint32_t *dst);
+	void (*select_me_pipe_q)(struct amdgpu_device *adev, u32 me, u32 pipe, u32 queue);
 };
 
 struct amdgpu_ngg_buf {
@@ -1812,6 +1813,7 @@ amdgpu_get_sdma_instance(struct amdgpu_ring *ring)
 #define amdgpu_gfx_select_se_sh(adev, se, sh, instance) (adev)->gfx.funcs->select_se_sh((adev), (se), (sh), (instance))
 #define amdgpu_gds_switch(adev, r, v, d, w, a) (adev)->gds.funcs->patch_gds_switch((r), (v), (d), (w), (a))
 #define amdgpu_psp_check_fw_loading_status(adev, i) (adev)->firmware.funcs->check_fw_loading_status((adev), (i))
+#define amdgpu_gfx_select_me_pipe_q(adev, me, pipe, q) (adev)->gfx.funcs->select_me_pipe_q((adev), (me), (pipe), (q))
 
 /* Common functions */
 int amdgpu_device_gpu_recover(struct amdgpu_device *adev,

drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c

@@ -64,16 +64,21 @@ int amdgpu_debugfs_add_files(struct amdgpu_device *adev,
 
 #if defined(CONFIG_DEBUG_FS)
 
-static ssize_t amdgpu_debugfs_regs_read(struct file *f, char __user *buf,
-					size_t size, loff_t *pos)
+
+static int  amdgpu_debugfs_process_reg_op(bool read, struct file *f,
+		char __user *buf, size_t size, loff_t *pos)
 {
 	struct amdgpu_device *adev = file_inode(f)->i_private;
 	ssize_t result = 0;
 	int r;
-	bool pm_pg_lock, use_bank;
-	unsigned instance_bank, sh_bank, se_bank;
+	bool pm_pg_lock, use_bank, use_ring;
+	unsigned instance_bank, sh_bank, se_bank, me, pipe, queue;
 
-	if (size & 0x3 || *pos & 0x3)
+	pm_pg_lock = use_bank = use_ring = false;
+	instance_bank = sh_bank = se_bank = me = pipe = queue = 0;
+
+	if (size & 0x3 || *pos & 0x3 ||
+			((*pos & (1ULL << 62)) && (*pos & (1ULL << 61))))
 		return -EINVAL;
 
 	/* are we reading registers for which a PG lock is necessary? */
@@ -91,8 +96,15 @@ static ssize_t amdgpu_debugfs_regs_read(struct file *f, char __user *buf,
 		if (instance_bank == 0x3FF)
 			instance_bank = 0xFFFFFFFF;
 		use_bank = 1;
+	} else if (*pos & (1ULL << 61)) {
+
+		me = (*pos & GENMASK_ULL(33, 24)) >> 24;
+		pipe = (*pos & GENMASK_ULL(43, 34)) >> 34;
+		queue = (*pos & GENMASK_ULL(53, 44)) >> 44;
+
+		use_ring = 1;
 	} else {
-		use_bank = 0;
+		use_bank = use_ring = 0;
 	}
 
 	*pos &= (1UL << 22) - 1;
@@ -104,6 +116,9 @@ static ssize_t amdgpu_debugfs_regs_read(struct file *f, char __user *buf,
 		mutex_lock(&adev->grbm_idx_mutex);
 		amdgpu_gfx_select_se_sh(adev, se_bank,
 					sh_bank, instance_bank);
+	} else if (use_ring) {
+		mutex_lock(&adev->srbm_mutex);
+		amdgpu_gfx_select_me_pipe_q(adev, me, pipe, queue);
 	}
 
 	if (pm_pg_lock)
@@ -115,8 +130,14 @@ static ssize_t amdgpu_debugfs_regs_read(struct file *f, char __user *buf,
 		if (*pos > adev->rmmio_size)
 			goto end;
 
-		value = RREG32(*pos >> 2);
-		r = put_user(value, (uint32_t *)buf);
+		if (read) {
+			value = RREG32(*pos >> 2);
+			r = put_user(value, (uint32_t *)buf);
+		} else {
+			r = get_user(value, (uint32_t *)buf);
+			if (!r)
+				WREG32(*pos >> 2, value);
+		}
 		if (r) {
 			result = r;
 			goto end;
@@ -132,6 +153,9 @@ static ssize_t amdgpu_debugfs_regs_read(struct file *f, char __user *buf,
 	if (use_bank) {
 		amdgpu_gfx_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff);
 		mutex_unlock(&adev->grbm_idx_mutex);
+	} else if (use_ring) {
+		amdgpu_gfx_select_me_pipe_q(adev, 0, 0, 0);
+		mutex_unlock(&adev->srbm_mutex);
 	}
 
 	if (pm_pg_lock)
@@ -140,78 +164,17 @@ static ssize_t amdgpu_debugfs_regs_read(struct file *f, char __user *buf,
 	return result;
 }
 
+
+static ssize_t amdgpu_debugfs_regs_read(struct file *f, char __user *buf,
+					size_t size, loff_t *pos)
+{
+	return amdgpu_debugfs_process_reg_op(true, f, buf, size, pos);
+}
+
 static ssize_t amdgpu_debugfs_regs_write(struct file *f, const char __user *buf,
 					 size_t size, loff_t *pos)
 {
-	struct amdgpu_device *adev = file_inode(f)->i_private;
-	ssize_t result = 0;
-	int r;
-	bool pm_pg_lock, use_bank;
-	unsigned instance_bank, sh_bank, se_bank;
-
-	if (size & 0x3 || *pos & 0x3)
-		return -EINVAL;
-
-	/* are we reading registers for which a PG lock is necessary? */
-	pm_pg_lock = (*pos >> 23) & 1;
-
-	if (*pos & (1ULL << 62)) {
-		se_bank = (*pos & GENMASK_ULL(33, 24)) >> 24;
-		sh_bank = (*pos & GENMASK_ULL(43, 34)) >> 34;
-		instance_bank = (*pos & GENMASK_ULL(53, 44)) >> 44;
-
-		if (se_bank == 0x3FF)
-			se_bank = 0xFFFFFFFF;
-		if (sh_bank == 0x3FF)
-			sh_bank = 0xFFFFFFFF;
-		if (instance_bank == 0x3FF)
-			instance_bank = 0xFFFFFFFF;
-		use_bank = 1;
-	} else {
-		use_bank = 0;
-	}
-
-	*pos &= (1UL << 22) - 1;
-
-	if (use_bank) {
-		if ((sh_bank != 0xFFFFFFFF && sh_bank >= adev->gfx.config.max_sh_per_se) ||
-		    (se_bank != 0xFFFFFFFF && se_bank >= adev->gfx.config.max_shader_engines))
-			return -EINVAL;
-		mutex_lock(&adev->grbm_idx_mutex);
-		amdgpu_gfx_select_se_sh(adev, se_bank,
-					sh_bank, instance_bank);
-	}
-
-	if (pm_pg_lock)
-		mutex_lock(&adev->pm.mutex);
-
-	while (size) {
-		uint32_t value;
-
-		if (*pos > adev->rmmio_size)
-			return result;
-
-		r = get_user(value, (uint32_t *)buf);
-		if (r)
-			return r;
-
-		WREG32(*pos >> 2, value);
-
-		result += 4;
-		buf += 4;
-		*pos += 4;
-		size -= 4;
-	}
-
-	if (use_bank) {
-		amdgpu_gfx_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff);
-		mutex_unlock(&adev->grbm_idx_mutex);
-	}
-
-	if (pm_pg_lock)
-		mutex_unlock(&adev->pm.mutex);
-
-	return result;
+	return amdgpu_debugfs_process_reg_op(false, f, (char __user *)buf, size, pos);
 }
 
 static ssize_t amdgpu_debugfs_regs_pcie_read(struct file *f, char __user *buf,

drivers/gpu/drm/amd/amdgpu/gfx_v6_0.c

@@ -3061,11 +3061,18 @@ static void gfx_v6_0_read_wave_sgprs(struct amdgpu_device *adev, uint32_t simd,
 		start + SQIND_WAVE_SGPRS_OFFSET, size, dst);
 }
 
+static void gfx_v6_0_select_me_pipe_q(struct amdgpu_device *adev,
+				  u32 me, u32 pipe, u32 q)
+{
+	DRM_INFO("Not implemented\n");
+}
+
 static const struct amdgpu_gfx_funcs gfx_v6_0_gfx_funcs = {
 	.get_gpu_clock_counter = &gfx_v6_0_get_gpu_clock_counter,
 	.select_se_sh = &gfx_v6_0_select_se_sh,
 	.read_wave_data = &gfx_v6_0_read_wave_data,
 	.read_wave_sgprs = &gfx_v6_0_read_wave_sgprs,
+	.select_me_pipe_q = &gfx_v6_0_select_me_pipe_q
 };
 
 static int gfx_v6_0_early_init(void *handle)

drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c

@@ -4270,11 +4270,18 @@ static void gfx_v7_0_read_wave_sgprs(struct amdgpu_device *adev, uint32_t simd,
 		start + SQIND_WAVE_SGPRS_OFFSET, size, dst);
 }
 
+static void gfx_v7_0_select_me_pipe_q(struct amdgpu_device *adev,
+				  u32 me, u32 pipe, u32 q)
+{
+	cik_srbm_select(adev, me, pipe, q, 0);
+}
+
 static const struct amdgpu_gfx_funcs gfx_v7_0_gfx_funcs = {
 	.get_gpu_clock_counter = &gfx_v7_0_get_gpu_clock_counter,
 	.select_se_sh = &gfx_v7_0_select_se_sh,
 	.read_wave_data = &gfx_v7_0_read_wave_data,
 	.read_wave_sgprs = &gfx_v7_0_read_wave_sgprs,
+	.select_me_pipe_q = &gfx_v7_0_select_me_pipe_q
 };
 
 static const struct amdgpu_rlc_funcs gfx_v7_0_rlc_funcs = {

drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c

@@ -3475,6 +3475,12 @@ static void gfx_v8_0_select_se_sh(struct amdgpu_device *adev,
 	WREG32(mmGRBM_GFX_INDEX, data);
 }
 
+static void gfx_v8_0_select_me_pipe_q(struct amdgpu_device *adev,
+				  u32 me, u32 pipe, u32 q)
+{
+	vi_srbm_select(adev, me, pipe, q, 0);
+}
+
 static u32 gfx_v8_0_get_rb_active_bitmap(struct amdgpu_device *adev)
 {
 	u32 data, mask;
@@ -5442,6 +5448,7 @@ static const struct amdgpu_gfx_funcs gfx_v8_0_gfx_funcs = {
 	.select_se_sh = &gfx_v8_0_select_se_sh,
 	.read_wave_data = &gfx_v8_0_read_wave_data,
 	.read_wave_sgprs = &gfx_v8_0_read_wave_sgprs,
+	.select_me_pipe_q = &gfx_v8_0_select_me_pipe_q
 };
 
 static int gfx_v8_0_early_init(void *handle)

drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c

@@ -998,12 +998,19 @@ static void gfx_v9_0_read_wave_vgprs(struct amdgpu_device *adev, uint32_t simd,
 		start + SQIND_WAVE_VGPRS_OFFSET, size, dst);
 }
 
+static void gfx_v9_0_select_me_pipe_q(struct amdgpu_device *adev,
+				  u32 me, u32 pipe, u32 q)
+{
+	soc15_grbm_select(adev, me, pipe, q, 0);
+}
+
 static const struct amdgpu_gfx_funcs gfx_v9_0_gfx_funcs = {
 	.get_gpu_clock_counter = &gfx_v9_0_get_gpu_clock_counter,
 	.select_se_sh = &gfx_v9_0_select_se_sh,
 	.read_wave_data = &gfx_v9_0_read_wave_data,
 	.read_wave_sgprs = &gfx_v9_0_read_wave_sgprs,
 	.read_wave_vgprs = &gfx_v9_0_read_wave_vgprs,
+	.select_me_pipe_q = &gfx_v9_0_select_me_pipe_q
 };
 
 static void gfx_v9_0_gpu_early_init(struct amdgpu_device *adev)
@@ -2773,13 +2780,13 @@ static int gfx_v9_0_kiq_fini_register(struct amdgpu_ring *ring)
 			udelay(1);
 		}
 
-		if (adev->usec_timeout == AMDGPU_MAX_USEC_TIMEOUT) {
+		if (j == AMDGPU_MAX_USEC_TIMEOUT) {
 			DRM_DEBUG("KIQ dequeue request failed.\n");
 
+			/* Manual disable if dequeue request times out */
 			WREG32_SOC15(GC, 0, mmCP_HQD_ACTIVE, 0);
 		}
 
-		/* Manual disable if dequeue request times out */
 		WREG32_SOC15(GC, 0, mmCP_HQD_DEQUEUE_REQUEST,
 		      0);
 	}