drm/amdgpu: Remove wrapper layer of cgs irq handling

v2: add Vega12 support

1. remove struct cgs_os_ops
2. delete cgs_linux.h
3. refine the irq code for vega10, can fix set pp table
failed issue.
4. add common smu irq process function

Acked-by: Christian König <[e-mail]>
Acked-by: Junwei Zhang <[e-mail]>
Signed-off-by: Rex Zhu <[e-mail]>
Signed-off-by: Alex Deucher <[e-mail]>

drivers/gpu/drm/amd/acp/include/acp_gfx_if.h

···
 #define _ACP_GFX_IF_H
 
 #include <linux/types.h>
-#include "cgs_linux.h"
 #include "cgs_common.h"
 
 int amd_acp_hw_init(struct cgs_device *cgs_device,

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

···
 #include <linux/firmware.h>
 #include <drm/amdgpu_drm.h>
 #include "amdgpu.h"
-#include "cgs_linux.h"
 #include "atom.h"
 #include "amdgpu_ucode.h"
 
···
 
 	return amdgpu_atom_execute_table(
 		adev->mode_info.atom_context, table, args);
-}
-
-struct cgs_irq_params {
-	unsigned src_id;
-	cgs_irq_source_set_func_t set;
-	cgs_irq_handler_func_t handler;
-	void *private_data;
-};
-
-static int cgs_set_irq_state(struct amdgpu_device *adev,
-			     struct amdgpu_irq_src *src,
-			     unsigned type,
-			     enum amdgpu_interrupt_state state)
-{
-	struct cgs_irq_params *irq_params =
-		(struct cgs_irq_params *)src->data;
-	if (!irq_params)
-		return -EINVAL;
-	if (!irq_params->set)
-		return -EINVAL;
-	return irq_params->set(irq_params->private_data,
-			       irq_params->src_id,
-			       type,
-			       (int)state);
-}
-
-static int cgs_process_irq(struct amdgpu_device *adev,
-			   struct amdgpu_irq_src *source,
-			   struct amdgpu_iv_entry *entry)
-{
-	struct cgs_irq_params *irq_params =
-		(struct cgs_irq_params *)source->data;
-	if (!irq_params)
-		return -EINVAL;
-	if (!irq_params->handler)
-		return -EINVAL;
-	return irq_params->handler(irq_params->private_data,
-				   irq_params->src_id,
-				   entry->iv_entry);
-}
-
-static const struct amdgpu_irq_src_funcs cgs_irq_funcs = {
-	.set = cgs_set_irq_state,
-	.process = cgs_process_irq,
-};
-
-static int amdgpu_cgs_add_irq_source(void *cgs_device,
-				     unsigned client_id,
-				     unsigned src_id,
-				     unsigned num_types,
-				     cgs_irq_source_set_func_t set,
-				     cgs_irq_handler_func_t handler,
-				     void *private_data)
-{
-	CGS_FUNC_ADEV;
-	int ret = 0;
-	struct cgs_irq_params *irq_params;
-	struct amdgpu_irq_src *source =
-		kzalloc(sizeof(struct amdgpu_irq_src), GFP_KERNEL);
-	if (!source)
-		return -ENOMEM;
-	irq_params =
-		kzalloc(sizeof(struct cgs_irq_params), GFP_KERNEL);
-	if (!irq_params) {
-		kfree(source);
-		return -ENOMEM;
-	}
-	source->num_types = num_types;
-	source->funcs = &cgs_irq_funcs;
-	irq_params->src_id = src_id;
-	irq_params->set = set;
-	irq_params->handler = handler;
-	irq_params->private_data = private_data;
-	source->data = (void *)irq_params;
-	ret = amdgpu_irq_add_id(adev, client_id, src_id, source);
-	if (ret) {
-		kfree(irq_params);
-		kfree(source);
-	}
-
-	return ret;
-}
-
-static int amdgpu_cgs_irq_get(void *cgs_device, unsigned client_id,
-			      unsigned src_id, unsigned type)
-{
-	CGS_FUNC_ADEV;
-
-	if (!adev->irq.client[client_id].sources)
-		return -EINVAL;
-
-	return amdgpu_irq_get(adev, adev->irq.client[client_id].sources[src_id], type);
-}
-
-static int amdgpu_cgs_irq_put(void *cgs_device, unsigned client_id,
-			      unsigned src_id, unsigned type)
-{
-	CGS_FUNC_ADEV;
-
-	if (!adev->irq.client[client_id].sources)
-		return -EINVAL;
-
-	return amdgpu_irq_put(adev, adev->irq.client[client_id].sources[src_id], type);
 }
 
 static int amdgpu_cgs_set_clockgating_state(struct cgs_device *cgs_device,
···
 	.lock_grbm_idx = amdgpu_cgs_lock_grbm_idx,
 };
 
-static const struct cgs_os_ops amdgpu_cgs_os_ops = {
-	.add_irq_source = amdgpu_cgs_add_irq_source,
-	.irq_get = amdgpu_cgs_irq_get,
-	.irq_put = amdgpu_cgs_irq_put
-};
-
 struct cgs_device *amdgpu_cgs_create_device(struct amdgpu_device *adev)
 {
 	struct amdgpu_cgs_device *cgs_device =
···
 	}
 
 	cgs_device->base.ops = &amdgpu_cgs_ops;
-	cgs_device->base.os_ops = &amdgpu_cgs_os_ops;
 	cgs_device->adev = adev;
 
 	return (struct cgs_device *)cgs_device;

drivers/gpu/drm/amd/display/dc/os_types.h

···
 
 #include <linux/kref.h>
 
-#include "cgs_linux.h"
+#include "cgs_common.h"
 
 #if defined(__BIG_ENDIAN) && !defined(BIGENDIAN_CPU)
 #define BIGENDIAN_CPU

drivers/gpu/drm/amd/include/cgs_common.h

···
 struct cgs_device
 {
 	const struct cgs_ops *ops;
-	const struct cgs_os_ops *os_ops;
 	/* to be embedded at the start of driver private structure */
 };
 

drivers/gpu/drm/amd/include/cgs_linux.h

···
-/*
- * Copyright 2015 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.
- *
- *
- */
-#ifndef _CGS_LINUX_H
-#define _CGS_LINUX_H
-
-#include "cgs_common.h"
-
-/**
- * cgs_irq_source_set_func() - Callback for enabling/disabling interrupt sources
- * @private_data:  private data provided to cgs_add_irq_source
- * @src_id:        interrupt source ID
- * @type:          interrupt type
- * @enabled:       0 = disable source, non-0 = enable source
- *
- * Return:  0 on success, -errno otherwise
- */
-typedef int (*cgs_irq_source_set_func_t)(void *private_data,
-					 unsigned src_id, unsigned type,
-					 int enabled);
-
-/**
- * cgs_irq_handler_func() - Interrupt handler callback
- * @private_data:  private data provided to cgs_add_irq_source
- * @src_id:        interrupt source ID
- * @iv_entry:      pointer to raw ih ring entry
- *
- * This callback runs in interrupt context.
- *
- * Return:  0 on success, -errno otherwise
- */
-typedef int (*cgs_irq_handler_func_t)(void *private_data,
-				      unsigned src_id, const uint32_t *iv_entry);
-
-/**
- * cgs_add_irq_source() - Add an IRQ source
- * @cgs_device:    opaque device handle
- * @src_id:        interrupt source ID
- * @num_types:     number of interrupt types that can be independently enabled
- * @set:           callback function to enable/disable an interrupt type
- * @handler:       interrupt handler callback
- * @private_data:  private data to pass to callback functions
- *
- * The same IRQ source can be added only once. Adding an IRQ source
- * indicates ownership of that IRQ source and all its IRQ types.
- *
- * Return:  0 on success, -errno otherwise
- */
-typedef int (*cgs_add_irq_source_t)(void *cgs_device, unsigned client_id,
-				    unsigned src_id,
-				    unsigned num_types,
-				    cgs_irq_source_set_func_t set,
-				    cgs_irq_handler_func_t handler,
-				    void *private_data);
-
-/**
- * cgs_irq_get() - Request enabling an IRQ source and type
- * @cgs_device:  opaque device handle
- * @src_id:      interrupt source ID
- * @type:        interrupt type
- *
- * cgs_irq_get and cgs_irq_put calls must be balanced. They count
- * "references" to IRQ sources.
- *
- * Return:  0 on success, -errno otherwise
- */
-typedef int (*cgs_irq_get_t)(void *cgs_device, unsigned client_id, unsigned src_id, unsigned type);
-
-/**
- * cgs_irq_put() - Indicate IRQ source is no longer needed
- * @cgs_device:  opaque device handle
- * @src_id:      interrupt source ID
- * @type:        interrupt type
- *
- * cgs_irq_get and cgs_irq_put calls must be balanced. They count
- * "references" to IRQ sources. Even after cgs_irq_put is called, the
- * IRQ handler may still be called if there are more refecences to
- * the IRQ source.
- *
- * Return:  0 on success, -errno otherwise
- */
-typedef int (*cgs_irq_put_t)(void *cgs_device, unsigned client_id, unsigned src_id, unsigned type);
-
-struct cgs_os_ops {
-	/* IRQ handling */
-	cgs_add_irq_source_t add_irq_source;
-	cgs_irq_get_t irq_get;
-	cgs_irq_put_t irq_put;
-};
-
-#define cgs_add_irq_source(dev,client_id,src_id,num_types,set,handler,private_data) \
-	CGS_OS_CALL(add_irq_source,dev,client_id,src_id,num_types,set,handler, \
-		    private_data)
-#define cgs_irq_get(dev,client_id,src_id,type)	\
-	CGS_OS_CALL(irq_get,dev,client_id,src_id,type)
-#define cgs_irq_put(dev,client_id,src_id,type)	\
-	CGS_OS_CALL(irq_put,dev,client_id,src_id,type)
-
-#endif /* _CGS_LINUX_H */

drivers/gpu/drm/amd/powerplay/hwmgr/hwmgr.c

···
 static int topaz_set_asic_special_caps(struct pp_hwmgr *hwmgr);
 static int ci_set_asic_special_caps(struct pp_hwmgr *hwmgr);
 
-static int phm_thermal_l2h_irq(void *private_data,
-		 unsigned src_id, const uint32_t *iv_entry)
-{
-	struct pp_hwmgr *hwmgr = (struct pp_hwmgr *)private_data;
-	struct amdgpu_device *adev = hwmgr->adev;
-
-	pr_warn("GPU over temperature range detected on PCIe %d:%d.%d!\n",
-			PCI_BUS_NUM(adev->pdev->devfn),
-			PCI_SLOT(adev->pdev->devfn),
-			PCI_FUNC(adev->pdev->devfn));
-	return 0;
-}
-
-static int phm_thermal_h2l_irq(void *private_data,
-		 unsigned src_id, const uint32_t *iv_entry)
-{
-	struct pp_hwmgr *hwmgr = (struct pp_hwmgr *)private_data;
-	struct amdgpu_device *adev = hwmgr->adev;
-
-	pr_warn("GPU under temperature range detected on PCIe %d:%d.%d!\n",
-			PCI_BUS_NUM(adev->pdev->devfn),
-			PCI_SLOT(adev->pdev->devfn),
-			PCI_FUNC(adev->pdev->devfn));
-	return 0;
-}
-
-static int phm_ctf_irq(void *private_data,
-		 unsigned src_id, const uint32_t *iv_entry)
-{
-	struct pp_hwmgr *hwmgr = (struct pp_hwmgr *)private_data;
-	struct amdgpu_device *adev = hwmgr->adev;
-
-	pr_warn("GPU Critical Temperature Fault detected on PCIe %d:%d.%d!\n",
-			PCI_BUS_NUM(adev->pdev->devfn),
-			PCI_SLOT(adev->pdev->devfn),
-			PCI_FUNC(adev->pdev->devfn));
-	return 0;
-}
-
-static const struct cgs_irq_src_funcs thermal_irq_src[3] = {
-	{ .handler = phm_thermal_l2h_irq },
-	{ .handler = phm_thermal_h2l_irq },
-	{ .handler = phm_ctf_irq }
-};
 
 static void hwmgr_init_workload_prority(struct pp_hwmgr *hwmgr)
 {
···
 	if (ret)
 		goto err2;
 
-	ret = phm_register_thermal_interrupt(hwmgr, &thermal_irq_src);
+	ret = phm_register_thermal_interrupt(hwmgr, NULL);
 	if (ret)
 		goto err2;
 

drivers/gpu/drm/amd/powerplay/hwmgr/smu_helper.c

···
 }
 
 
+int phm_irq_process(struct amdgpu_device *adev,
+			   struct amdgpu_irq_src *source,
+			   struct amdgpu_iv_entry *entry)
+{
+	uint32_t client_id = entry->client_id;
+	uint32_t src_id = entry->src_id;
+
+	if (client_id == AMDGPU_IH_CLIENTID_LEGACY) {
+		if (src_id == 230)
+			pr_warn("GPU over temperature range detected on PCIe %d:%d.%d!\n",
+						PCI_BUS_NUM(adev->pdev->devfn),
+						PCI_SLOT(adev->pdev->devfn),
+						PCI_FUNC(adev->pdev->devfn));
+		else if (src_id == 231)
+			pr_warn("GPU under temperature range detected on PCIe %d:%d.%d!\n",
+					PCI_BUS_NUM(adev->pdev->devfn),
+					PCI_SLOT(adev->pdev->devfn),
+					PCI_FUNC(adev->pdev->devfn));
+		else if (src_id == 83)
+			pr_warn("GPU Critical Temperature Fault detected on PCIe %d:%d.%d!\n",
+					PCI_BUS_NUM(adev->pdev->devfn),
+					PCI_SLOT(adev->pdev->devfn),
+					PCI_FUNC(adev->pdev->devfn));
+	} else if (client_id == SOC15_IH_CLIENTID_THM) {
+		if (src_id == 0)
+			pr_warn("GPU over temperature range detected on PCIe %d:%d.%d!\n",
+						PCI_BUS_NUM(adev->pdev->devfn),
+						PCI_SLOT(adev->pdev->devfn),
+						PCI_FUNC(adev->pdev->devfn));
+		else
+			pr_warn("GPU under temperature range detected on PCIe %d:%d.%d!\n",
+					PCI_BUS_NUM(adev->pdev->devfn),
+					PCI_SLOT(adev->pdev->devfn),
+					PCI_FUNC(adev->pdev->devfn));
+	} else if (client_id == SOC15_IH_CLIENTID_ROM_SMUIO)
+		pr_warn("GPU Critical Temperature Fault detected on PCIe %d:%d.%d!\n",
+				PCI_BUS_NUM(adev->pdev->devfn),
+				PCI_SLOT(adev->pdev->devfn),
+				PCI_FUNC(adev->pdev->devfn));
+
+	return 0;
+}
+
+static const struct amdgpu_irq_src_funcs smu9_irq_funcs = {
+	.process = phm_irq_process,
+};
+
+int smu9_register_thermal_interrupt(struct pp_hwmgr *hwmgr,
+		const void *info)
+{
+	struct amdgpu_irq_src *source =
+		kzalloc(sizeof(struct amdgpu_irq_src), GFP_KERNEL);
+
+	if (!source)
+		return -ENOMEM;
+
+	source->funcs = &smu9_irq_funcs;
+
+	amdgpu_irq_add_id((struct amdgpu_device *)(hwmgr->adev),
+			SOC15_IH_CLIENTID_THM,
+			0,
+			source);
+	amdgpu_irq_add_id((struct amdgpu_device *)(hwmgr->adev),
+			SOC15_IH_CLIENTID_THM,
+			1,
+			source);
+
+	/* Register CTF(GPIO_19) interrupt */
+	amdgpu_irq_add_id((struct amdgpu_device *)(hwmgr->adev),
+			SOC15_IH_CLIENTID_ROM_SMUIO,
+			83,
+			source);
+
+	return 0;
+}

drivers/gpu/drm/amd/powerplay/hwmgr/smu_helper.h

···
 				uint32_t value,
 				uint32_t mask);
 
+int phm_irq_process(struct amdgpu_device *adev,
+			   struct amdgpu_irq_src *source,
+			   struct amdgpu_iv_entry *entry);
+
+int smu9_register_thermal_interrupt(struct pp_hwmgr *hwmgr,
+		const void *info);
+
 #define PHM_FIELD_SHIFT(reg, field) reg##__##field##__SHIFT
 #define PHM_FIELD_MASK(reg, field) reg##__##field##_MASK
 

drivers/gpu/drm/amd/powerplay/hwmgr/vega10_hwmgr.c

···
 #include "vega10_thermal.h"
 #include "pp_debug.h"
 #include "amd_pcie_helpers.h"
-#include "cgs_linux.h"
 #include "ppinterrupt.h"
 #include "pp_overdriver.h"
 #include "pp_thermal.h"
···
 	return 0;
 }
 
-static int vega10_register_thermal_interrupt(struct pp_hwmgr *hwmgr,
-		const void *info)
-{
-	struct cgs_irq_src_funcs *irq_src =
-			(struct cgs_irq_src_funcs *)info;
-
-	if (hwmgr->thermal_controller.ucType ==
-			ATOM_VEGA10_PP_THERMALCONTROLLER_VEGA10 ||
-		hwmgr->thermal_controller.ucType ==
-			ATOM_VEGA10_PP_THERMALCONTROLLER_EMC2103_WITH_INTERNAL) {
-		PP_ASSERT_WITH_CODE(!cgs_add_irq_source(hwmgr->device,
-				SOC15_IH_CLIENTID_THM,
-				0, 0, irq_src[0].set, irq_src[0].handler, hwmgr),
-				"Failed to register high thermal interrupt!",
-				return -EINVAL);
-		PP_ASSERT_WITH_CODE(!cgs_add_irq_source(hwmgr->device,
-				SOC15_IH_CLIENTID_THM,
-				1, 0, irq_src[1].set, irq_src[1].handler, hwmgr),
-				"Failed to register low thermal interrupt!",
-				return -EINVAL);
-	}
-
-	/* Register CTF(GPIO_19) interrupt */
-	PP_ASSERT_WITH_CODE(!cgs_add_irq_source(hwmgr->device,
-			SOC15_IH_CLIENTID_ROM_SMUIO,
-			83, 0, irq_src[2].set, irq_src[2].handler, hwmgr),
-			"Failed to register CTF thermal interrupt!",
-			return -EINVAL);
-
-	return 0;
-}
-
 static int vega10_get_power_profile_mode(struct pp_hwmgr *hwmgr, char *buf)
 {
 	struct vega10_hwmgr *data = hwmgr->backend;
···
 	.avfs_control = vega10_avfs_enable,
 	.notify_cac_buffer_info = vega10_notify_cac_buffer_info,
 	.get_thermal_temperature_range = vega10_get_thermal_temperature_range,
-	.register_internal_thermal_interrupt = vega10_register_thermal_interrupt,
+	.register_internal_thermal_interrupt = smu9_register_thermal_interrupt,
 	.start_thermal_controller = vega10_start_thermal_controller,
 	.get_power_profile_mode = vega10_get_power_profile_mode,
 	.set_power_profile_mode = vega10_set_power_profile_mode,

drivers/gpu/drm/amd/powerplay/hwmgr/vega12_hwmgr.c

···
 #include "vega12_ppsmc.h"
 #include "pp_debug.h"
 #include "amd_pcie_helpers.h"
-#include "cgs_linux.h"
 #include "ppinterrupt.h"
 #include "pp_overdriver.h"
 #include "pp_thermal.h"
···
 	return 0;
 }
 
-static int vega12_is_hardware_ctf_enabled(struct pp_hwmgr *hwmgr)
-{
-	uint32_t reg;
-
-	reg = soc15_get_register_offset(THM_HWID, 0,
-			mmTHM_TCON_THERM_TRIP_BASE_IDX,
-			mmTHM_TCON_THERM_TRIP);
-
-	return (((cgs_read_register(hwmgr->device, reg) &
-		THM_TCON_THERM_TRIP__THERM_TP_EN_MASK) >>
-		THM_TCON_THERM_TRIP__THERM_TP_EN__SHIFT) == 1);
-}
-
-static int vega12_register_thermal_interrupt(struct pp_hwmgr *hwmgr,
-		const void *info)
-{
-	struct cgs_irq_src_funcs *irq_src =
-			(struct cgs_irq_src_funcs *)info;
-
-	if (hwmgr->thermal_controller.ucType ==
-			ATOM_VEGA12_PP_THERMALCONTROLLER_VEGA12) {
-		PP_ASSERT_WITH_CODE(!cgs_add_irq_source(hwmgr->device,
-				0xf, /* AMDGPU_IH_CLIENTID_THM */
-				0, 0, irq_src[0].set, irq_src[0].handler, hwmgr),
-				"Failed to register high thermal interrupt!",
-				return -EINVAL);
-		PP_ASSERT_WITH_CODE(!cgs_add_irq_source(hwmgr->device,
-				0xf, /* AMDGPU_IH_CLIENTID_THM */
-				1, 0, irq_src[1].set, irq_src[1].handler, hwmgr),
-				"Failed to register low thermal interrupt!",
-				return -EINVAL);
-	}
-
-	if (vega12_is_hardware_ctf_enabled(hwmgr))
-		/* Register CTF(GPIO_19) interrupt */
-		PP_ASSERT_WITH_CODE(!cgs_add_irq_source(hwmgr->device,
-				0x16, /* AMDGPU_IH_CLIENTID_ROM_SMUIO, */
-				83, 0, irq_src[2].set, irq_src[2].handler, hwmgr),
-				"Failed to register CTF thermal interrupt!",
-				return -EINVAL);
-
-	return 0;
-}
-
-
 static const struct pp_hwmgr_func vega12_hwmgr_funcs = {
 	.backend_init = vega12_hwmgr_backend_init,
 	.backend_fini = vega12_hwmgr_backend_fini,
···
 #endif
 	.notify_cac_buffer_info = vega12_notify_cac_buffer_info,
 	.get_thermal_temperature_range = vega12_get_thermal_temperature_range,
-	.register_internal_thermal_interrupt = vega12_register_thermal_interrupt,
+	.register_internal_thermal_interrupt = smu9_register_thermal_interrupt,
 	.start_thermal_controller = vega12_start_thermal_controller,
 };
 

drivers/gpu/drm/amd/powerplay/inc/hwmgr.h

···
 	uint32_t workload_setting[Workload_Policy_Max];
 };
 
-struct cgs_irq_src_funcs {
-	cgs_irq_source_set_func_t set;
-	cgs_irq_handler_func_t handler;
-};
-
 extern int hwmgr_early_init(struct pp_hwmgr *hwmgr);
 extern int hwmgr_hw_init(struct pp_hwmgr *hwmgr);
 extern int hwmgr_hw_fini(struct pp_hwmgr *hwmgr);