linux-stable/drivers/gpu/drm/amd/amdgpu/vce_v2_0.c
jimqu abc8c1ce48 drm/amd/amdgpu: S3 resumed failed after 4-5 times loop
Phenomenon: software hang when device resume back, read UVD fence is 0xffffffff
and read pcie pid is 0xffff.
The issue is caused by VCE reset when update cg setting. according to HW programming
guide, adjust update VCE cg sequence.
The patch apply to VCE2.0.

Signed-off-by: JimQu <Jim.Qu@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2016-08-19 12:32:55 -04:00

634 lines
15 KiB
C

/*
* Copyright 2013 Advanced Micro Devices, Inc.
* All Rights Reserved.
*
* 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, sub license, 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 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 NON-INFRINGEMENT. IN NO EVENT SHALL
* THE COPYRIGHT HOLDERS, AUTHORS AND/OR ITS SUPPLIERS 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.
*
* The above copyright notice and this permission notice (including the
* next paragraph) shall be included in all copies or substantial portions
* of the Software.
*
* Authors: Christian König <christian.koenig@amd.com>
*/
#include <linux/firmware.h>
#include <drm/drmP.h>
#include "amdgpu.h"
#include "amdgpu_vce.h"
#include "cikd.h"
#include "vce/vce_2_0_d.h"
#include "vce/vce_2_0_sh_mask.h"
#include "oss/oss_2_0_d.h"
#include "oss/oss_2_0_sh_mask.h"
#define VCE_V2_0_FW_SIZE (256 * 1024)
#define VCE_V2_0_STACK_SIZE (64 * 1024)
#define VCE_V2_0_DATA_SIZE (23552 * AMDGPU_MAX_VCE_HANDLES)
#define VCE_STATUS_VCPU_REPORT_FW_LOADED_MASK 0x02
static void vce_v2_0_mc_resume(struct amdgpu_device *adev);
static void vce_v2_0_set_ring_funcs(struct amdgpu_device *adev);
static void vce_v2_0_set_irq_funcs(struct amdgpu_device *adev);
static int vce_v2_0_wait_for_idle(void *handle);
/**
* vce_v2_0_ring_get_rptr - get read pointer
*
* @ring: amdgpu_ring pointer
*
* Returns the current hardware read pointer
*/
static uint32_t vce_v2_0_ring_get_rptr(struct amdgpu_ring *ring)
{
struct amdgpu_device *adev = ring->adev;
if (ring == &adev->vce.ring[0])
return RREG32(mmVCE_RB_RPTR);
else
return RREG32(mmVCE_RB_RPTR2);
}
/**
* vce_v2_0_ring_get_wptr - get write pointer
*
* @ring: amdgpu_ring pointer
*
* Returns the current hardware write pointer
*/
static uint32_t vce_v2_0_ring_get_wptr(struct amdgpu_ring *ring)
{
struct amdgpu_device *adev = ring->adev;
if (ring == &adev->vce.ring[0])
return RREG32(mmVCE_RB_WPTR);
else
return RREG32(mmVCE_RB_WPTR2);
}
/**
* vce_v2_0_ring_set_wptr - set write pointer
*
* @ring: amdgpu_ring pointer
*
* Commits the write pointer to the hardware
*/
static void vce_v2_0_ring_set_wptr(struct amdgpu_ring *ring)
{
struct amdgpu_device *adev = ring->adev;
if (ring == &adev->vce.ring[0])
WREG32(mmVCE_RB_WPTR, ring->wptr);
else
WREG32(mmVCE_RB_WPTR2, ring->wptr);
}
static int vce_v2_0_lmi_clean(struct amdgpu_device *adev)
{
int i, j;
for (i = 0; i < 10; ++i) {
for (j = 0; j < 100; ++j) {
uint32_t status = RREG32(mmVCE_LMI_STATUS);
if (status & 0x337f)
return 0;
mdelay(10);
}
}
return -ETIMEDOUT;
}
static int vce_v2_0_firmware_loaded(struct amdgpu_device *adev)
{
int i, j;
for (i = 0; i < 10; ++i) {
for (j = 0; j < 100; ++j) {
uint32_t status = RREG32(mmVCE_STATUS);
if (status & VCE_STATUS_VCPU_REPORT_FW_LOADED_MASK)
return 0;
mdelay(10);
}
DRM_ERROR("VCE not responding, trying to reset the ECPU!!!\n");
WREG32_P(mmVCE_SOFT_RESET,
VCE_SOFT_RESET__ECPU_SOFT_RESET_MASK,
~VCE_SOFT_RESET__ECPU_SOFT_RESET_MASK);
mdelay(10);
WREG32_P(mmVCE_SOFT_RESET, 0,
~VCE_SOFT_RESET__ECPU_SOFT_RESET_MASK);
mdelay(10);
}
return -ETIMEDOUT;
}
/**
* vce_v2_0_start - start VCE block
*
* @adev: amdgpu_device pointer
*
* Setup and start the VCE block
*/
static int vce_v2_0_start(struct amdgpu_device *adev)
{
struct amdgpu_ring *ring;
int r;
vce_v2_0_mc_resume(adev);
/* set BUSY flag */
WREG32_P(mmVCE_STATUS, 1, ~1);
ring = &adev->vce.ring[0];
WREG32(mmVCE_RB_RPTR, ring->wptr);
WREG32(mmVCE_RB_WPTR, ring->wptr);
WREG32(mmVCE_RB_BASE_LO, ring->gpu_addr);
WREG32(mmVCE_RB_BASE_HI, upper_32_bits(ring->gpu_addr));
WREG32(mmVCE_RB_SIZE, ring->ring_size / 4);
ring = &adev->vce.ring[1];
WREG32(mmVCE_RB_RPTR2, ring->wptr);
WREG32(mmVCE_RB_WPTR2, ring->wptr);
WREG32(mmVCE_RB_BASE_LO2, ring->gpu_addr);
WREG32(mmVCE_RB_BASE_HI2, upper_32_bits(ring->gpu_addr));
WREG32(mmVCE_RB_SIZE2, ring->ring_size / 4);
WREG32_FIELD(VCE_VCPU_CNTL, CLK_EN, 1);
WREG32_FIELD(VCE_SOFT_RESET, ECPU_SOFT_RESET, 1);
mdelay(100);
WREG32_FIELD(VCE_SOFT_RESET, ECPU_SOFT_RESET, 0);
r = vce_v2_0_firmware_loaded(adev);
/* clear BUSY flag */
WREG32_P(mmVCE_STATUS, 0, ~1);
if (r) {
DRM_ERROR("VCE not responding, giving up!!!\n");
return r;
}
return 0;
}
static int vce_v2_0_early_init(void *handle)
{
struct amdgpu_device *adev = (struct amdgpu_device *)handle;
vce_v2_0_set_ring_funcs(adev);
vce_v2_0_set_irq_funcs(adev);
return 0;
}
static int vce_v2_0_sw_init(void *handle)
{
struct amdgpu_ring *ring;
int r;
struct amdgpu_device *adev = (struct amdgpu_device *)handle;
/* VCE */
r = amdgpu_irq_add_id(adev, 167, &adev->vce.irq);
if (r)
return r;
r = amdgpu_vce_sw_init(adev, VCE_V2_0_FW_SIZE +
VCE_V2_0_STACK_SIZE + VCE_V2_0_DATA_SIZE);
if (r)
return r;
r = amdgpu_vce_resume(adev);
if (r)
return r;
ring = &adev->vce.ring[0];
sprintf(ring->name, "vce0");
r = amdgpu_ring_init(adev, ring, 512, VCE_CMD_NO_OP, 0xf,
&adev->vce.irq, 0, AMDGPU_RING_TYPE_VCE);
if (r)
return r;
ring = &adev->vce.ring[1];
sprintf(ring->name, "vce1");
r = amdgpu_ring_init(adev, ring, 512, VCE_CMD_NO_OP, 0xf,
&adev->vce.irq, 0, AMDGPU_RING_TYPE_VCE);
if (r)
return r;
return r;
}
static int vce_v2_0_sw_fini(void *handle)
{
int r;
struct amdgpu_device *adev = (struct amdgpu_device *)handle;
r = amdgpu_vce_suspend(adev);
if (r)
return r;
r = amdgpu_vce_sw_fini(adev);
if (r)
return r;
return r;
}
static int vce_v2_0_hw_init(void *handle)
{
struct amdgpu_ring *ring;
int r;
struct amdgpu_device *adev = (struct amdgpu_device *)handle;
r = vce_v2_0_start(adev);
if (r)
/* this error mean vcpu not in running state, so just skip ring test, not stop driver initialize */
return 0;
ring = &adev->vce.ring[0];
ring->ready = true;
r = amdgpu_ring_test_ring(ring);
if (r) {
ring->ready = false;
return r;
}
ring = &adev->vce.ring[1];
ring->ready = true;
r = amdgpu_ring_test_ring(ring);
if (r) {
ring->ready = false;
return r;
}
DRM_INFO("VCE initialized successfully.\n");
return 0;
}
static int vce_v2_0_hw_fini(void *handle)
{
return 0;
}
static int vce_v2_0_suspend(void *handle)
{
int r;
struct amdgpu_device *adev = (struct amdgpu_device *)handle;
r = vce_v2_0_hw_fini(adev);
if (r)
return r;
r = amdgpu_vce_suspend(adev);
if (r)
return r;
return r;
}
static int vce_v2_0_resume(void *handle)
{
int r;
struct amdgpu_device *adev = (struct amdgpu_device *)handle;
r = amdgpu_vce_resume(adev);
if (r)
return r;
r = vce_v2_0_hw_init(adev);
if (r)
return r;
return r;
}
static void vce_v2_0_set_sw_cg(struct amdgpu_device *adev, bool gated)
{
u32 tmp;
if (gated) {
tmp = RREG32(mmVCE_CLOCK_GATING_B);
tmp |= 0xe70000;
WREG32(mmVCE_CLOCK_GATING_B, tmp);
tmp = RREG32(mmVCE_UENC_CLOCK_GATING);
tmp |= 0xff000000;
WREG32(mmVCE_UENC_CLOCK_GATING, tmp);
tmp = RREG32(mmVCE_UENC_REG_CLOCK_GATING);
tmp &= ~0x3fc;
WREG32(mmVCE_UENC_REG_CLOCK_GATING, tmp);
WREG32(mmVCE_CGTT_CLK_OVERRIDE, 0);
} else {
tmp = RREG32(mmVCE_CLOCK_GATING_B);
tmp |= 0xe7;
tmp &= ~0xe70000;
WREG32(mmVCE_CLOCK_GATING_B, tmp);
tmp = RREG32(mmVCE_UENC_CLOCK_GATING);
tmp |= 0x1fe000;
tmp &= ~0xff000000;
WREG32(mmVCE_UENC_CLOCK_GATING, tmp);
tmp = RREG32(mmVCE_UENC_REG_CLOCK_GATING);
tmp |= 0x3fc;
WREG32(mmVCE_UENC_REG_CLOCK_GATING, tmp);
}
}
static void vce_v2_0_set_dyn_cg(struct amdgpu_device *adev, bool gated)
{
if (vce_v2_0_wait_for_idle(adev)) {
DRM_INFO("VCE is busy, Can't set clock gateing");
return;
}
WREG32_P(mmVCE_LMI_CTRL2, 0x100, ~0x100);
if (vce_v2_0_lmi_clean(adev)) {
DRM_INFO("LMI is busy, Can't set clock gateing");
return;
}
WREG32_P(mmVCE_VCPU_CNTL, 0, ~VCE_VCPU_CNTL__CLK_EN_MASK);
WREG32_P(mmVCE_SOFT_RESET,
VCE_SOFT_RESET__ECPU_SOFT_RESET_MASK,
~VCE_SOFT_RESET__ECPU_SOFT_RESET_MASK);
WREG32(mmVCE_STATUS, 0);
if (gated)
WREG32(mmVCE_CGTT_CLK_OVERRIDE, 0);
/* LMI_MC/LMI_UMC always set in dynamic, set {CGC_*_GATE_MODE, CGC_*_SW_GATE} = {0, 0} */
if (gated) {
/* Force CLOCK OFF , set {CGC_*_GATE_MODE, CGC_*_SW_GATE} = {*, 1} */
WREG32(mmVCE_CLOCK_GATING_B, 0xe90010);
} else {
/* Force CLOCK ON, set {CGC_*_GATE_MODE, CGC_*_SW_GATE} = {1, 0} */
WREG32(mmVCE_CLOCK_GATING_B, 0x800f1);
}
/* Set VCE_UENC_CLOCK_GATING always in dynamic mode {*_FORCE_ON, *_FORCE_OFF} = {0, 0}*/;
WREG32(mmVCE_UENC_CLOCK_GATING, 0x40);
/* set VCE_UENC_REG_CLOCK_GATING always in dynamic mode */
WREG32(mmVCE_UENC_REG_CLOCK_GATING, 0x00);
WREG32_P(mmVCE_LMI_CTRL2, 0, ~0x100);
if(!gated) {
WREG32_P(mmVCE_VCPU_CNTL, VCE_VCPU_CNTL__CLK_EN_MASK, ~VCE_VCPU_CNTL__CLK_EN_MASK);
mdelay(100);
WREG32_P(mmVCE_SOFT_RESET, 0, ~VCE_SOFT_RESET__ECPU_SOFT_RESET_MASK);
vce_v2_0_firmware_loaded(adev);
WREG32_P(mmVCE_STATUS, 0, ~VCE_STATUS__JOB_BUSY_MASK);
}
}
static void vce_v2_0_disable_cg(struct amdgpu_device *adev)
{
WREG32(mmVCE_CGTT_CLK_OVERRIDE, 7);
}
static void vce_v2_0_enable_mgcg(struct amdgpu_device *adev, bool enable)
{
bool sw_cg = false;
if (enable && (adev->cg_flags & AMD_CG_SUPPORT_VCE_MGCG)) {
if (sw_cg)
vce_v2_0_set_sw_cg(adev, true);
else
vce_v2_0_set_dyn_cg(adev, true);
} else {
vce_v2_0_disable_cg(adev);
if (sw_cg)
vce_v2_0_set_sw_cg(adev, false);
else
vce_v2_0_set_dyn_cg(adev, false);
}
}
static void vce_v2_0_init_cg(struct amdgpu_device *adev)
{
u32 tmp;
tmp = RREG32(mmVCE_CLOCK_GATING_A);
tmp &= ~0xfff;
tmp |= ((0 << 0) | (4 << 4));
tmp |= 0x40000;
WREG32(mmVCE_CLOCK_GATING_A, tmp);
tmp = RREG32(mmVCE_UENC_CLOCK_GATING);
tmp &= ~0xfff;
tmp |= ((0 << 0) | (4 << 4));
WREG32(mmVCE_UENC_CLOCK_GATING, tmp);
tmp = RREG32(mmVCE_CLOCK_GATING_B);
tmp |= 0x10;
tmp &= ~0x100000;
WREG32(mmVCE_CLOCK_GATING_B, tmp);
}
static void vce_v2_0_mc_resume(struct amdgpu_device *adev)
{
uint64_t addr = adev->vce.gpu_addr;
uint32_t size;
WREG32_P(mmVCE_CLOCK_GATING_A, 0, ~(1 << 16));
WREG32_P(mmVCE_UENC_CLOCK_GATING, 0x1FF000, ~0xFF9FF000);
WREG32_P(mmVCE_UENC_REG_CLOCK_GATING, 0x3F, ~0x3F);
WREG32(mmVCE_CLOCK_GATING_B, 0xf7);
WREG32(mmVCE_LMI_CTRL, 0x00398000);
WREG32_P(mmVCE_LMI_CACHE_CTRL, 0x0, ~0x1);
WREG32(mmVCE_LMI_SWAP_CNTL, 0);
WREG32(mmVCE_LMI_SWAP_CNTL1, 0);
WREG32(mmVCE_LMI_VM_CTRL, 0);
addr += AMDGPU_VCE_FIRMWARE_OFFSET;
size = VCE_V2_0_FW_SIZE;
WREG32(mmVCE_VCPU_CACHE_OFFSET0, addr & 0x7fffffff);
WREG32(mmVCE_VCPU_CACHE_SIZE0, size);
addr += size;
size = VCE_V2_0_STACK_SIZE;
WREG32(mmVCE_VCPU_CACHE_OFFSET1, addr & 0x7fffffff);
WREG32(mmVCE_VCPU_CACHE_SIZE1, size);
addr += size;
size = VCE_V2_0_DATA_SIZE;
WREG32(mmVCE_VCPU_CACHE_OFFSET2, addr & 0x7fffffff);
WREG32(mmVCE_VCPU_CACHE_SIZE2, size);
WREG32_P(mmVCE_LMI_CTRL2, 0x0, ~0x100);
WREG32_FIELD(VCE_SYS_INT_EN, VCE_SYS_INT_TRAP_INTERRUPT_EN, 1);
vce_v2_0_init_cg(adev);
}
static bool vce_v2_0_is_idle(void *handle)
{
struct amdgpu_device *adev = (struct amdgpu_device *)handle;
return !(RREG32(mmSRBM_STATUS2) & SRBM_STATUS2__VCE_BUSY_MASK);
}
static int vce_v2_0_wait_for_idle(void *handle)
{
struct amdgpu_device *adev = (struct amdgpu_device *)handle;
unsigned i;
for (i = 0; i < adev->usec_timeout; i++) {
if (vce_v2_0_is_idle(handle))
return 0;
}
return -ETIMEDOUT;
}
static int vce_v2_0_soft_reset(void *handle)
{
struct amdgpu_device *adev = (struct amdgpu_device *)handle;
WREG32_FIELD(SRBM_SOFT_RESET, SOFT_RESET_VCE, 1);
mdelay(5);
return vce_v2_0_start(adev);
}
static int vce_v2_0_set_interrupt_state(struct amdgpu_device *adev,
struct amdgpu_irq_src *source,
unsigned type,
enum amdgpu_interrupt_state state)
{
uint32_t val = 0;
if (state == AMDGPU_IRQ_STATE_ENABLE)
val |= VCE_SYS_INT_EN__VCE_SYS_INT_TRAP_INTERRUPT_EN_MASK;
WREG32_P(mmVCE_SYS_INT_EN, val, ~VCE_SYS_INT_EN__VCE_SYS_INT_TRAP_INTERRUPT_EN_MASK);
return 0;
}
static int vce_v2_0_process_interrupt(struct amdgpu_device *adev,
struct amdgpu_irq_src *source,
struct amdgpu_iv_entry *entry)
{
DRM_DEBUG("IH: VCE\n");
switch (entry->src_data) {
case 0:
case 1:
amdgpu_fence_process(&adev->vce.ring[entry->src_data]);
break;
default:
DRM_ERROR("Unhandled interrupt: %d %d\n",
entry->src_id, entry->src_data);
break;
}
return 0;
}
static int vce_v2_0_set_clockgating_state(void *handle,
enum amd_clockgating_state state)
{
bool gate = false;
struct amdgpu_device *adev = (struct amdgpu_device *)handle;
if (state == AMD_CG_STATE_GATE)
gate = true;
vce_v2_0_enable_mgcg(adev, gate);
return 0;
}
static int vce_v2_0_set_powergating_state(void *handle,
enum amd_powergating_state state)
{
/* This doesn't actually powergate the VCE block.
* That's done in the dpm code via the SMC. This
* just re-inits the block as necessary. The actual
* gating still happens in the dpm code. We should
* revisit this when there is a cleaner line between
* the smc and the hw blocks
*/
struct amdgpu_device *adev = (struct amdgpu_device *)handle;
if (!(adev->pg_flags & AMD_PG_SUPPORT_VCE))
return 0;
if (state == AMD_PG_STATE_GATE)
/* XXX do we need a vce_v2_0_stop()? */
return 0;
else
return vce_v2_0_start(adev);
}
const struct amd_ip_funcs vce_v2_0_ip_funcs = {
.name = "vce_v2_0",
.early_init = vce_v2_0_early_init,
.late_init = NULL,
.sw_init = vce_v2_0_sw_init,
.sw_fini = vce_v2_0_sw_fini,
.hw_init = vce_v2_0_hw_init,
.hw_fini = vce_v2_0_hw_fini,
.suspend = vce_v2_0_suspend,
.resume = vce_v2_0_resume,
.is_idle = vce_v2_0_is_idle,
.wait_for_idle = vce_v2_0_wait_for_idle,
.soft_reset = vce_v2_0_soft_reset,
.set_clockgating_state = vce_v2_0_set_clockgating_state,
.set_powergating_state = vce_v2_0_set_powergating_state,
};
static const struct amdgpu_ring_funcs vce_v2_0_ring_funcs = {
.get_rptr = vce_v2_0_ring_get_rptr,
.get_wptr = vce_v2_0_ring_get_wptr,
.set_wptr = vce_v2_0_ring_set_wptr,
.parse_cs = amdgpu_vce_ring_parse_cs,
.emit_ib = amdgpu_vce_ring_emit_ib,
.emit_fence = amdgpu_vce_ring_emit_fence,
.test_ring = amdgpu_vce_ring_test_ring,
.test_ib = amdgpu_vce_ring_test_ib,
.insert_nop = amdgpu_ring_insert_nop,
.pad_ib = amdgpu_ring_generic_pad_ib,
.begin_use = amdgpu_vce_ring_begin_use,
.end_use = amdgpu_vce_ring_end_use,
};
static void vce_v2_0_set_ring_funcs(struct amdgpu_device *adev)
{
adev->vce.ring[0].funcs = &vce_v2_0_ring_funcs;
adev->vce.ring[1].funcs = &vce_v2_0_ring_funcs;
}
static const struct amdgpu_irq_src_funcs vce_v2_0_irq_funcs = {
.set = vce_v2_0_set_interrupt_state,
.process = vce_v2_0_process_interrupt,
};
static void vce_v2_0_set_irq_funcs(struct amdgpu_device *adev)
{
adev->vce.irq.num_types = 1;
adev->vce.irq.funcs = &vce_v2_0_irq_funcs;
};