From 017f3a894c19840fdf883b656985edd534bd1a44 Mon Sep 17 00:00:00 2001 From: Amber Lin Date: Thu, 23 Aug 2018 10:52:34 -0400 Subject: [PATCH 2814/2940] drm/amdgpu: Move KFD parameters to amdgpu (v3) After merging KFD into amdgpu, move module parameters defined in KFD to amdgpu_drv.c, where other module parameters are declared. v2: add kernel-doc comments v3: rebase and fix parameter variable name (Alex) Signed-off-by: Amber Lin Reviewed-by: Felix Kuehling Signed-off-by: Alex Deucher Signed-off-by: Chaudhary Amit Kumar --- drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c | 122 ++++++++++++++++++ drivers/gpu/drm/amd/amdkfd/kfd_module.c | 45 ------- drivers/gpu/drm/amd/amdkfd/kfd_priv.h | 29 ----- .../gpu/drm/amd/include/kgd_kfd_interface.h | 28 ++++ 4 files changed, 150 insertions(+), 74 deletions(-) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c index 00a1f426a68a..2a2014fca297 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c @@ -646,6 +646,128 @@ uint amdgpu_dm_abm_level = 0; MODULE_PARM_DESC(abmlevel, "ABM level (0 = off (default), 1-4 = backlight reduction level) "); module_param_named(abmlevel, amdgpu_dm_abm_level, uint, 0444); +#ifdef CONFIG_HSA_AMD +/** + * DOC: sched_policy (int) + * Set scheduling policy. Default is HWS(hardware scheduling) with over-subscription. + * Setting 1 disables over-subscription. Setting 2 disables HWS and statically + * assigns queues to HQDs. + */ +int sched_policy = KFD_SCHED_POLICY_HWS; +module_param(sched_policy, int, 0444); +MODULE_PARM_DESC(sched_policy, + "Scheduling policy (0 = HWS (Default), 1 = HWS without over-subscription, 2 = Non-HWS (Used for debugging only)"); + +/** + * DOC: hws_max_conc_proc (int) + * Maximum number of processes that HWS can schedule concurrently. The maximum is the + * number of VMIDs assigned to the HWS, which is also the default. + */ +int hws_max_conc_proc = 8; +module_param(hws_max_conc_proc, int, 0444); +MODULE_PARM_DESC(hws_max_conc_proc, + "Max # processes HWS can execute concurrently when sched_policy=0 (0 = no concurrency, #VMIDs for KFD = Maximum(default))"); + +/** + * DOC: cwsr_enable (int) + * CWSR(compute wave store and resume) allows the GPU to preempt shader execution in + * the middle of a compute wave. Default is 1 to enable this feature. Setting 0 + * disables it. + */ +int cwsr_enable = 1; +module_param(cwsr_enable, int, 0444); +MODULE_PARM_DESC(cwsr_enable, "CWSR enable (0 = Off, 1 = On (Default))"); + +/** + * DOC: max_num_of_queues_per_device (int) + * Maximum number of queues per device. Valid setting is between 1 and 4096. Default + * is 4096. + */ +int max_num_of_queues_per_device = KFD_MAX_NUM_OF_QUEUES_PER_DEVICE_DEFAULT; +module_param(max_num_of_queues_per_device, int, 0444); +MODULE_PARM_DESC(max_num_of_queues_per_device, + "Maximum number of supported queues per device (1 = Minimum, 4096 = default)"); + +/** + * DOC: send_sigterm (int) + * Send sigterm to HSA process on unhandled exceptions. Default is not to send sigterm + * but just print errors on dmesg. Setting 1 enables sending sigterm. + */ +int send_sigterm; +module_param(send_sigterm, int, 0444); +MODULE_PARM_DESC(send_sigterm, + "Send sigterm to HSA process on unhandled exception (0 = disable, 1 = enable)"); + +/** + * DOC: debug_largebar (int) + * Set debug_largebar as 1 to enable simulating large-bar capability on non-large bar + * system. This limits the VRAM size reported to ROCm applications to the visible + * size, usually 256MB. + * Default value is 0, diabled. + */ +int debug_largebar; +module_param(debug_largebar, int, 0444); +MODULE_PARM_DESC(debug_largebar, + "Debug large-bar flag used to simulate large-bar capability on non-large bar machine (0 = disable, 1 = enable)"); + +/** + * DOC: ignore_crat (int) + * Ignore CRAT table during KFD initialization. By default, KFD uses the ACPI CRAT + * table to get information about AMD APUs. This option can serve as a workaround on + * systems with a broken CRAT table. + */ +int ignore_crat; +module_param(ignore_crat, int, 0444); +MODULE_PARM_DESC(ignore_crat, + "Ignore CRAT table during KFD initialization (0 = use CRAT (default), 1 = ignore CRAT)"); + +/** + * DOC: noretry (int) + * This parameter sets sh_mem_config.retry_disable. Default value, 0, enables retry. + * Setting 1 disables retry. + * Retry is needed for recoverable page faults. + */ +int noretry; +module_param(noretry, int, 0644); +MODULE_PARM_DESC(noretry, + "Set sh_mem_config.retry_disable on Vega10 (0 = retry enabled (default), 1 = retry disabled)"); + +/** + * DOC: halt_if_hws_hang (int) + * Halt if HWS hang is detected. Default value, 0, disables the halt on hang. + * Setting 1 enables halt on hang. + */ +int halt_if_hws_hang; +module_param(halt_if_hws_hang, int, 0644); +MODULE_PARM_DESC(halt_if_hws_hang, "Halt if HWS hang is detected (0 = off (default), 1 = on)"); + +/* + * DOC: hws_gws_support(bool) + * Whether HWS support gws barriers. Default value: false (not supported) + * This will be replaced with a MEC firmware version check once firmware + * is ready + */ +bool hws_gws_support; +module_param(hws_gws_support, bool, 0444); +MODULE_PARM_DESC(hws_gws_support, "MEC FW support gws barriers (false = not supported (Default), true = supported)"); + +/** + * DOC: priv_cp_queues (int) + * Enable privileged mode for CP queues. Default value: 0 (off) + */ +int priv_cp_queues; +module_param(priv_cp_queues, int, 0644); +MODULE_PARM_DESC(priv_cp_queues, "Enable privileged mode for CP queues (0 = off (default), 1 = on)"); + +/** + * DOC: keep_udle_process_evicted (bool) + * Keep an evicted process evicted if it is idle. Default value: false (off) + */ +bool keep_idle_process_evicted; +module_param(keep_idle_process_evicted, bool, 0444); +MODULE_PARM_DESC(keep_idle_process_evicted, "Restore evicted process only if queues are active (N = off(default), Y = on)"); +#endif + static const struct pci_device_id pciidlist[] = { #ifdef CONFIG_DRM_AMDGPU_SI {0x1002, 0x6780, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_TAHITI}, diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_module.c b/drivers/gpu/drm/amd/amdkfd/kfd_module.c index e0ce8f31efb8..9c6dfb856797 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_module.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_module.c @@ -21,55 +21,10 @@ */ #include -#include #include #include "kfd_priv.h" #include "amdgpu_amdkfd.h" -int sched_policy = KFD_SCHED_POLICY_HWS; -module_param(sched_policy, int, 0444); -MODULE_PARM_DESC(sched_policy, - "Scheduling policy (0 = HWS (Default), 1 = HWS without over-subscription, 2 = Non-HWS (Used for debugging only)"); - -int hws_max_conc_proc = 8; -module_param(hws_max_conc_proc, int, 0444); -MODULE_PARM_DESC(hws_max_conc_proc, - "Max # processes HWS can execute concurrently when sched_policy=0 (0 = no concurrency, #VMIDs for KFD = Maximum(default))"); - -int cwsr_enable = 1; -module_param(cwsr_enable, int, 0444); -MODULE_PARM_DESC(cwsr_enable, "CWSR enable (0 = off, 1 = on (default))"); - -int max_num_of_queues_per_device = KFD_MAX_NUM_OF_QUEUES_PER_DEVICE_DEFAULT; -module_param(max_num_of_queues_per_device, int, 0444); -MODULE_PARM_DESC(max_num_of_queues_per_device, - "Maximum number of supported queues per device (1 = Minimum, 4096 = default)"); - -int send_sigterm; -module_param(send_sigterm, int, 0444); -MODULE_PARM_DESC(send_sigterm, - "Send sigterm to HSA process on unhandled exception (0 = disable, 1 = enable)"); - -int debug_largebar; -module_param(debug_largebar, int, 0444); -MODULE_PARM_DESC(debug_largebar, - "Debug large-bar flag used to simulate large-bar capability on non-large bar machine (0 = disable, 1 = enable)"); - -int ignore_crat; -module_param(ignore_crat, int, 0444); -MODULE_PARM_DESC(ignore_crat, - "Ignore CRAT table during KFD initialization (0 = use CRAT (default), 1 = ignore CRAT)"); - -int noretry; -module_param(noretry, int, 0644); -MODULE_PARM_DESC(noretry, - "Set sh_mem_config.retry_disable on GFXv9+ dGPUs (0 = retry enabled (default), 1 = retry disabled)"); - -int halt_if_hws_hang; -module_param(halt_if_hws_hang, int, 0644); -MODULE_PARM_DESC(halt_if_hws_hang, "Halt if HWS hang is detected (0 = off (default), 1 = on)"); - - static int kfd_init(void) { int err; diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h index 62675c0e9a91..67967a893a55 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h +++ b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h @@ -128,8 +128,6 @@ struct drm_device; */ extern int max_num_of_queues_per_device; -#define KFD_MAX_NUM_OF_QUEUES_PER_DEVICE_DEFAULT 4096 - /* Kernel module parameter to specify the scheduling policy */ extern int sched_policy; @@ -185,33 +183,6 @@ extern bool hws_gws_support; extern bool keep_idle_process_evicted; -/** - * enum kfd_sched_policy - * - * @KFD_SCHED_POLICY_HWS: H/W scheduling policy known as command processor (cp) - * scheduling. In this scheduling mode we're using the firmware code to - * schedule the user mode queues and kernel queues such as HIQ and DIQ. - * the HIQ queue is used as a special queue that dispatches the configuration - * to the cp and the user mode queues list that are currently running. - * the DIQ queue is a debugging queue that dispatches debugging commands to the - * firmware. - * in this scheduling mode user mode queues over subscription feature is - * enabled. - * - * @KFD_SCHED_POLICY_HWS_NO_OVERSUBSCRIPTION: The same as above but the over - * subscription feature disabled. - * - * @KFD_SCHED_POLICY_NO_HWS: no H/W scheduling policy is a mode which directly - * set the command processor registers and sets the queues "manually". This - * mode is used *ONLY* for debugging proposes. - * - */ -enum kfd_sched_policy { - KFD_SCHED_POLICY_HWS = 0, - KFD_SCHED_POLICY_HWS_NO_OVERSUBSCRIPTION, - KFD_SCHED_POLICY_NO_HWS -}; - enum cache_policy { cache_policy_coherent, cache_policy_noncoherent diff --git a/drivers/gpu/drm/amd/include/kgd_kfd_interface.h b/drivers/gpu/drm/amd/include/kgd_kfd_interface.h index fca84236d847..ffc93d12f6a5 100644 --- a/drivers/gpu/drm/amd/include/kgd_kfd_interface.h +++ b/drivers/gpu/drm/amd/include/kgd_kfd_interface.h @@ -100,6 +100,33 @@ enum kgd_engine_type { KGD_ENGINE_MAX }; +/** + * enum kfd_sched_policy + * + * @KFD_SCHED_POLICY_HWS: H/W scheduling policy known as command processor (cp) + * scheduling. In this scheduling mode we're using the firmware code to + * schedule the user mode queues and kernel queues such as HIQ and DIQ. + * the HIQ queue is used as a special queue that dispatches the configuration + * to the cp and the user mode queues list that are currently running. + * the DIQ queue is a debugging queue that dispatches debugging commands to the + * firmware. + * in this scheduling mode user mode queues over subscription feature is + * enabled. + * + * @KFD_SCHED_POLICY_HWS_NO_OVERSUBSCRIPTION: The same as above but the over + * subscription feature disabled. + * + * @KFD_SCHED_POLICY_NO_HWS: no H/W scheduling policy is a mode which directly + * set the command processor registers and sets the queues "manually". This + * mode is used *ONLY* for debugging proposes. + * + */ +enum kfd_sched_policy { + KFD_SCHED_POLICY_HWS = 0, + KFD_SCHED_POLICY_HWS_NO_OVERSUBSCRIPTION, + KFD_SCHED_POLICY_NO_HWS +}; + struct kgd2kfd_shared_resources { /* Bit n == 1 means VMID n is available for KFD. */ unsigned int compute_vmid_bitmap; @@ -152,6 +179,7 @@ struct tile_config { uint32_t num_ranks; }; +#define KFD_MAX_NUM_OF_QUEUES_PER_DEVICE_DEFAULT 4096 /* * Allocation flag domains -- 2.17.1