To meet the requirement for multi container usecase which needs
a quicker reset and not causing VRAM lost, adding the Mode2
reset handler for sienna_cichlid.
v2: move skip mode2 flag part separately
v3: remove the use of asic_reset_res
Signed-off-by: Victor Zhao <Victor.Zhao@xxxxxxx>
---
drivers/gpu/drm/amd/amdgpu/Makefile | 2 +-
drivers/gpu/drm/amd/amdgpu/amdgpu_reset.c | 7 +
drivers/gpu/drm/amd/amdgpu/sienna_cichlid.c | 296 ++++++++++++++++++
drivers/gpu/drm/amd/amdgpu/sienna_cichlid.h | 32 ++
.../pm/swsmu/inc/pmfw_if/smu_v11_0_7_ppsmc.h | 4 +-
drivers/gpu/drm/amd/pm/swsmu/inc/smu_types.h | 3 +-
.../amd/pm/swsmu/smu11/sienna_cichlid_ppt.c | 54 ++++
7 files changed, 394 insertions(+), 4 deletions(-)
create mode 100644 drivers/gpu/drm/amd/amdgpu/sienna_cichlid.c
create mode 100644 drivers/gpu/drm/amd/amdgpu/sienna_cichlid.h
diff --git a/drivers/gpu/drm/amd/amdgpu/Makefile b/drivers/gpu/drm/amd/amdgpu/Makefile
index c7d0cd15b5ef..7030ac2d7d2c 100644
--- a/drivers/gpu/drm/amd/amdgpu/Makefile
+++ b/drivers/gpu/drm/amd/amdgpu/Makefile
@@ -75,7 +75,7 @@ amdgpu-y += \
vi.o mxgpu_vi.o nbio_v6_1.o soc15.o emu_soc.o mxgpu_ai.o nbio_v7_0.o vega10_reg_init.o \
vega20_reg_init.o nbio_v7_4.o nbio_v2_3.o nv.o arct_reg_init.o mxgpu_nv.o \
nbio_v7_2.o hdp_v4_0.o hdp_v5_0.o aldebaran_reg_init.o aldebaran.o soc21.o \
- nbio_v4_3.o hdp_v6_0.o nbio_v7_7.o hdp_v5_2.o lsdma_v6_0.o
+ sienna_cichlid.o nbio_v4_3.o hdp_v6_0.o nbio_v7_7.o hdp_v5_2.o lsdma_v6_0.o
# add DF block
amdgpu-y += \
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_reset.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_reset.c
index 32c86a0b145c..f778466bb9db 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_reset.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_reset.c
@@ -23,6 +23,7 @@
#include "amdgpu_reset.h"
#include "aldebaran.h"
+#include "sienna_cichlid.h"
int amdgpu_reset_add_handler(struct amdgpu_reset_control *reset_ctl,
struct amdgpu_reset_handler *handler)
@@ -40,6 +41,9 @@ int amdgpu_reset_init(struct amdgpu_device *adev)
case IP_VERSION(13, 0, 2):
ret = aldebaran_reset_init(adev);
break;
+ case IP_VERSION(11, 0, 7):
+ ret = sienna_cichlid_reset_init(adev);
+ break;
default:
break;
}
@@ -55,6 +59,9 @@ int amdgpu_reset_fini(struct amdgpu_device *adev)
case IP_VERSION(13, 0, 2):
ret = aldebaran_reset_fini(adev);
break;
+ case IP_VERSION(11, 0, 7):
+ ret = sienna_cichlid_reset_fini(adev);
+ break;
default:
break;
}
diff --git a/drivers/gpu/drm/amd/amdgpu/sienna_cichlid.c b/drivers/gpu/drm/amd/amdgpu/sienna_cichlid.c
new file mode 100644
index 000000000000..b61a8ddec7ef
--- /dev/null
+++ b/drivers/gpu/drm/amd/amdgpu/sienna_cichlid.c
@@ -0,0 +1,296 @@
+/*
+ * Copyright 2021 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 "sienna_cichlid.h"
+#include "amdgpu_reset.h"
+#include "amdgpu_amdkfd.h"
+#include "amdgpu_dpm.h"
+#include "amdgpu_job.h"
+#include "amdgpu_ring.h"
+#include "amdgpu_ras.h"
+#include "amdgpu_psp.h"
+#include "amdgpu_xgmi.h"
+
+static struct amdgpu_reset_handler *
+sienna_cichlid_get_reset_handler(struct amdgpu_reset_control *reset_ctl,
+ struct amdgpu_reset_context *reset_context)
+{
+ struct amdgpu_reset_handler *handler;
+ struct amdgpu_device *adev = (struct amdgpu_device *)reset_ctl->handle;
+
+ if (reset_context->method != AMD_RESET_METHOD_NONE) {
+ list_for_each_entry(handler, &reset_ctl->reset_handlers,
+ handler_list) {
+ if (handler->reset_method == reset_context->method)
+ return handler;
+ }
+ } else {
+ list_for_each_entry(handler, &reset_ctl->reset_handlers,
+ handler_list) {
+ if (handler->reset_method == AMD_RESET_METHOD_MODE2 &&
+ adev->pm.fw_version >= 0x3a5500 &&
+ !amdgpu_sriov_vf(adev)) {
+ reset_context->method = AMD_RESET_METHOD_MODE2;
+ return handler;
+ }
+ }
+ }
+
+ return NULL;
+}
+
+static int sienna_cichlid_mode2_suspend_ip(struct amdgpu_device *adev)
+{
+ int r, i;
+
+ amdgpu_device_set_pg_state(adev, AMD_PG_STATE_UNGATE);
+ amdgpu_device_set_cg_state(adev, AMD_CG_STATE_UNGATE);
+
+ for (i = adev->num_ip_blocks - 1; i >= 0; i--) {
+ if (!(adev->ip_blocks[i].version->type ==
+ AMD_IP_BLOCK_TYPE_GFX ||
+ adev->ip_blocks[i].version->type ==
+ AMD_IP_BLOCK_TYPE_SDMA))
+ continue;
+
+ r = adev->ip_blocks[i].version->funcs->suspend(adev);
+
+ if (r) {
+ dev_err(adev->dev,
+ "suspend of IP block <%s> failed %d\n",
+ adev->ip_blocks[i].version->funcs->name, r);
+ return r;
+ }
+ adev->ip_blocks[i].status.hw = false;
+ }
+
+ return r;
+}
+
+static int
+sienna_cichlid_mode2_prepare_hwcontext(struct amdgpu_reset_control *reset_ctl,
+ struct amdgpu_reset_context *reset_context)
+{
+ int r = 0;
+ struct amdgpu_device *adev = (struct amdgpu_device *)reset_ctl->handle;
+
+ if (!amdgpu_sriov_vf(adev))
+ r = sienna_cichlid_mode2_suspend_ip(adev);
+
+ return r;
+}
+
+static void sienna_cichlid_async_reset(struct work_struct *work)
+{
+ struct amdgpu_reset_handler *handler;
+ struct amdgpu_reset_control *reset_ctl =
+ container_of(work, struct amdgpu_reset_control, reset_work);
+ struct amdgpu_device *adev = (struct amdgpu_device *)reset_ctl->handle;
+
+ list_for_each_entry(handler, &reset_ctl->reset_handlers,
+ handler_list) {
+ if (handler->reset_method == reset_ctl->active_reset) {
+ dev_dbg(adev->dev, "Resetting device\n");
+ handler->do_reset(adev);
+ break;
+ }
+ }
+}
+
+static int sienna_cichlid_mode2_reset(struct amdgpu_device *adev)
+{
+ /* disable BM */
+ pci_clear_master(adev->pdev);
+ return amdgpu_dpm_mode2_reset(adev);
+}
+
+static int
+sienna_cichlid_mode2_perform_reset(struct amdgpu_reset_control *reset_ctl,
+ struct amdgpu_reset_context *reset_context)
+{
+ struct amdgpu_device *adev = (struct amdgpu_device *)reset_ctl->handle;
+ int r;
+
+ r = sienna_cichlid_mode2_reset(adev);
+ if (r) {
+ dev_err(adev->dev,
+ "ASIC reset failed with error, %d ", r);
+ }
+ return r;
+}
+
+static int sienna_cichlid_mode2_restore_ip(struct amdgpu_device *adev)
+{
+ int i, r;
+ struct psp_context *psp = &adev->psp;
+
+ r = psp_rlc_autoload_start(psp);
+ if (r) {
+ dev_err(adev->dev, "Failed to start rlc autoload\n");
+ return r;
+ }
+
+ /* Reinit GFXHUB */
+ adev->gfxhub.funcs->init(adev);
+ r = adev->gfxhub.funcs->gart_enable(adev);
+ if (r) {
+ dev_err(adev->dev, "GFXHUB gart reenable failed after reset\n");
+ return r;
+ }
+
+ for (i = 0; i < adev->num_ip_blocks; i++) {
+ if (adev->ip_blocks[i].version->type == AMD_IP_BLOCK_TYPE_IH) {
+ r = adev->ip_blocks[i].version->funcs->resume(adev);
+ if (r) {
+ dev_err(adev->dev,
+ "resume of IP block <%s> failed %d\n",
+ adev->ip_blocks[i].version->funcs->name, r);
+ return r;
+ }
+
+ adev->ip_blocks[i].status.hw = true;
+ }
+ }
+
+ for (i = 0; i < adev->num_ip_blocks; i++) {
+ if (!(adev->ip_blocks[i].version->type ==
+ AMD_IP_BLOCK_TYPE_GFX ||
+ adev->ip_blocks[i].version->type ==
+ AMD_IP_BLOCK_TYPE_SDMA))
+ continue;
+ r = adev->ip_blocks[i].version->funcs->resume(adev);
+ if (r) {
+ dev_err(adev->dev,
+ "resume of IP block <%s> failed %d\n",
+ adev->ip_blocks[i].version->funcs->name, r);
+ return r;
+ }
+
+ adev->ip_blocks[i].status.hw = true;
+ }
+
+ for (i = 0; i < adev->num_ip_blocks; i++) {
+ if (!(adev->ip_blocks[i].version->type ==
+ AMD_IP_BLOCK_TYPE_GFX ||
+ adev->ip_blocks[i].version->type ==
+ AMD_IP_BLOCK_TYPE_SDMA))
+ continue;
+
+ if (adev->ip_blocks[i].version->funcs->late_init) {
+ r = adev->ip_blocks[i].version->funcs->late_init(
+ (void *)adev);
+ if (r) {
+ dev_err(adev->dev,
+ "late_init of IP block <%s> failed %d after reset\n",
+ adev->ip_blocks[i].version->funcs->name,
+ r);
+ return r;
+ }
+ }
+ adev->ip_blocks[i].status.late_initialized = true;
+ }
+
+ amdgpu_device_set_cg_state(adev, AMD_CG_STATE_GATE);
+ amdgpu_device_set_pg_state(adev, AMD_PG_STATE_GATE);
+
+ return r;
+}
+
+static int
+sienna_cichlid_mode2_restore_hwcontext(struct amdgpu_reset_control *reset_ctl,
+ struct amdgpu_reset_context *reset_context)
+{
+ int r;
+ struct amdgpu_device *tmp_adev = (struct amdgpu_device *)reset_ctl->handle;
+
+ dev_info(tmp_adev->dev,
+ "GPU reset succeeded, trying to resume\n");
+ r = sienna_cichlid_mode2_restore_ip(tmp_adev);
+ if (r)
+ goto end;
+
+ /*
+ * Add this ASIC as tracked as reset was already
+ * complete successfully.
+ */
+ amdgpu_register_gpu_instance(tmp_adev);
+
+ /* Resume RAS */
+ amdgpu_ras_resume(tmp_adev);
+
+ amdgpu_irq_gpu_reset_resume_helper(tmp_adev);
+
+ r = amdgpu_ib_ring_tests(tmp_adev);
+ if (r) {
+ dev_err(tmp_adev->dev,
+ "ib ring test failed (%d).\n", r);
+ r = -EAGAIN;
+ goto end;
+ }
+
+end:
+ if (r)
+ return -EAGAIN;
+ else
+ return r;
+}
+
+static struct amdgpu_reset_handler sienna_cichlid_mode2_handler = {
+ .reset_method = AMD_RESET_METHOD_MODE2,
+ .prepare_env = NULL,
+ .prepare_hwcontext = sienna_cichlid_mode2_prepare_hwcontext,
+ .perform_reset = sienna_cichlid_mode2_perform_reset,
+ .restore_hwcontext = sienna_cichlid_mode2_restore_hwcontext,
+ .restore_env = NULL,
+ .do_reset = sienna_cichlid_mode2_reset,
+};
+
+int sienna_cichlid_reset_init(struct amdgpu_device *adev)
+{
+ struct amdgpu_reset_control *reset_ctl;
+
+ reset_ctl = kzalloc(sizeof(*reset_ctl), GFP_KERNEL);
+ if (!reset_ctl)
+ return -ENOMEM;
+
+ reset_ctl->handle = adev;
+ reset_ctl->async_reset = sienna_cichlid_async_reset;
+ reset_ctl->active_reset = AMD_RESET_METHOD_NONE;
+ reset_ctl->get_reset_handler = sienna_cichlid_get_reset_handler;
+
+ INIT_LIST_HEAD(&reset_ctl->reset_handlers);
+ INIT_WORK(&reset_ctl->reset_work, reset_ctl->async_reset);
+ /* Only mode2 is handled through reset control now */
+ amdgpu_reset_add_handler(reset_ctl, &sienna_cichlid_mode2_handler);
+
+ adev->reset_cntl = reset_ctl;
+
+ return 0;
+}
+
+int sienna_cichlid_reset_fini(struct amdgpu_device *adev)
+{
+ kfree(adev->reset_cntl);
+ adev->reset_cntl = NULL;
+ return 0;
+}
diff --git a/drivers/gpu/drm/amd/amdgpu/sienna_cichlid.h b/drivers/gpu/drm/amd/amdgpu/sienna_cichlid.h
new file mode 100644
index 000000000000..5213b162dacd
--- /dev/null
+++ b/drivers/gpu/drm/amd/amdgpu/sienna_cichlid.h
@@ -0,0 +1,32 @@
+/*
+ * Copyright 2021 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 __SIENNA_CICHLID_H__
+#define __SIENNA_CICHLID_H__
+
+#include "amdgpu.h"
+
+int sienna_cichlid_reset_init(struct amdgpu_device *adev);
+int sienna_cichlid_reset_fini(struct amdgpu_device *adev);
+
+#endif
diff --git a/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v11_0_7_ppsmc.h b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v11_0_7_ppsmc.h
index d2e10a724560..82cf9e563065 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v11_0_7_ppsmc.h
+++ b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v11_0_7_ppsmc.h
@@ -137,7 +137,7 @@
#define PPSMC_MSG_DisallowGpo 0x56
#define PPSMC_MSG_Enable2ndUSB20Port 0x57
-
-#define PPSMC_Message_Count 0x58
+#define PPSMC_MSG_DriverMode2Reset 0x5D
+#define PPSMC_Message_Count 0x5E
#endif
diff --git a/drivers/gpu/drm/amd/pm/swsmu/inc/smu_types.h b/drivers/gpu/drm/amd/pm/swsmu/inc/smu_types.h
index 19084a4fcb2b..28f6a1eb6945 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/inc/smu_types.h
+++ b/drivers/gpu/drm/amd/pm/swsmu/inc/smu_types.h
@@ -235,7 +235,8 @@
__SMU_DUMMY_MAP(UnforceGfxVid), \
__SMU_DUMMY_MAP(HeavySBR), \
__SMU_DUMMY_MAP(SetBadHBMPagesRetiredFlagsPerChannel), \
- __SMU_DUMMY_MAP(EnableGfxImu),
+ __SMU_DUMMY_MAP(EnableGfxImu), \
+ __SMU_DUMMY_MAP(DriverMode2Reset),
#undef __SMU_DUMMY_MAP
#define __SMU_DUMMY_MAP(type) SMU_MSG_##type
diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu11/sienna_cichlid_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu11/sienna_cichlid_ppt.c
index fa520d79ef67..a73d241bb64f 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu11/sienna_cichlid_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu11/sienna_cichlid_ppt.c
@@ -154,6 +154,7 @@ static struct cmn2asic_msg_mapping sienna_cichlid_message_map[SMU_MSG_MAX_COUNT]
MSG_MAP(SetGpoFeaturePMask, PPSMC_MSG_SetGpoFeaturePMask, 0),
MSG_MAP(DisallowGpo, PPSMC_MSG_DisallowGpo, 0),
MSG_MAP(Enable2ndUSB20Port, PPSMC_MSG_Enable2ndUSB20Port, 0),
+ MSG_MAP(DriverMode2Reset, PPSMC_MSG_DriverMode2Reset, 0),
};
static struct cmn2asic_mapping sienna_cichlid_clk_map[SMU_CLK_COUNT] = {
@@ -4254,6 +4255,57 @@ static int sienna_cichlid_stb_get_data_direct(struct smu_context *smu,
return 0;
}
+static bool sienna_cichlid_is_mode2_reset_supported(struct smu_context *smu)
+{
+ return true;
+}
+
+static int sienna_cichlid_mode2_reset(struct smu_context *smu)
+{
+ u32 smu_version;
+ int ret = 0, index;
+ struct amdgpu_device *adev = smu->adev;
+ int timeout = 100;
+
+ smu_cmn_get_smc_version(smu, NULL, &smu_version);
+
+ index = smu_cmn_to_asic_specific_index(smu, CMN2ASIC_MAPPING_MSG,
+ SMU_MSG_DriverMode2Reset);
+
+ mutex_lock(&smu->message_lock);
+
+ ret = smu_cmn_send_msg_without_waiting(smu, (uint16_t)index,
+ SMU_RESET_MODE_2);
+
+ ret = smu_cmn_wait_for_response(smu);
+ while (ret != 0 && timeout) {
+ ret = smu_cmn_wait_for_response(smu);
+ /* Wait a bit more time for getting ACK */
+ if (ret != 0) {
+ --timeout;
+ usleep_range(500, 1000);
+ continue;
+ } else {
+ break;
+ }
+ }
+
+ if (!timeout) {
+ dev_err(adev->dev,
+ "failed to send mode2 message \tparam: 0x%08x response %#x\n",
+ SMU_RESET_MODE_2, ret);
+ goto out;
+ }
+
+ dev_info(smu->adev->dev, "restore config space...\n");
+ /* Restore the config space saved during init */
+ amdgpu_device_load_pci_state(adev->pdev);
+out:
+ mutex_unlock(&smu->message_lock);
+
+ return ret;
+}
+
static const struct pptable_funcs sienna_cichlid_ppt_funcs = {
.get_allowed_feature_mask = sienna_cichlid_get_allowed_feature_mask,
.set_default_dpm_table = sienna_cichlid_set_default_dpm_table,
@@ -4348,6 +4400,8 @@ static const struct pptable_funcs sienna_cichlid_ppt_funcs = {
.get_default_config_table_settings = sienna_cichlid_get_default_config_table_settings,
.set_config_table = sienna_cichlid_set_config_table,
.get_unique_id = sienna_cichlid_get_unique_id,
+ .mode2_reset_is_support = sienna_cichlid_is_mode2_reset_supported,
+ .mode2_reset = sienna_cichlid_mode2_reset,
};
void sienna_cichlid_set_ppt_funcs(struct smu_context *smu)
