diff options
Diffstat (limited to 'drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c')
| -rw-r--r-- | drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c | 670 |
1 files changed, 466 insertions, 204 deletions
diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c b/drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c index 3747c3f1f0cc..1f9021f896a1 100644 --- a/drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c +++ b/drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c @@ -21,15 +21,17 @@ * */ +#include <linux/delay.h> #include <linux/firmware.h> -#include <drm/drmP.h> +#include <linux/module.h> +#include <linux/pci.h> + #include "amdgpu.h" #include "amdgpu_ucode.h" #include "amdgpu_trace.h" #include "gc/gc_10_1_0_offset.h" #include "gc/gc_10_1_0_sh_mask.h" -#include "hdp/hdp_5_0_0_offset.h" #include "ivsrcid/sdma0/irqsrcs_sdma0_5_0.h" #include "ivsrcid/sdma1/irqsrcs_sdma1_5_0.h" @@ -37,11 +39,21 @@ #include "soc15.h" #include "navi10_sdma_pkt_open.h" #include "nbio_v2_3.h" +#include "sdma_common.h" #include "sdma_v5_0.h" MODULE_FIRMWARE("amdgpu/navi10_sdma.bin"); MODULE_FIRMWARE("amdgpu/navi10_sdma1.bin"); +MODULE_FIRMWARE("amdgpu/navi14_sdma.bin"); +MODULE_FIRMWARE("amdgpu/navi14_sdma1.bin"); + +MODULE_FIRMWARE("amdgpu/navi12_sdma.bin"); +MODULE_FIRMWARE("amdgpu/navi12_sdma1.bin"); + +MODULE_FIRMWARE("amdgpu/cyan_skillfish2_sdma.bin"); +MODULE_FIRMWARE("amdgpu/cyan_skillfish2_sdma1.bin"); + #define SDMA1_REG_OFFSET 0x600 #define SDMA0_HYP_DEC_REG_START 0x5880 #define SDMA0_HYP_DEC_REG_END 0x5893 @@ -59,7 +71,7 @@ static const struct soc15_reg_golden golden_settings_sdma_5[] = { SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA0_RLC0_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000), SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA0_RLC1_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000), SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA0_RLC2_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000), - SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA0_RLC3_RB_WPTR_POLL_CNTL, 0x0000fff0, 0x00403000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA0_RLC3_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000), SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA0_RLC4_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000), SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA0_RLC5_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000), SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA0_RLC6_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000), @@ -71,7 +83,7 @@ static const struct soc15_reg_golden golden_settings_sdma_5[] = { SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA1_RLC0_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000), SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA1_RLC1_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000), SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA1_RLC2_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000), - SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA1_RLC3_RB_WPTR_POLL_CNTL, 0x0000fff0, 0x00403000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA1_RLC3_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000), SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA1_RLC4_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000), SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA1_RLC5_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000), SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA1_RLC6_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000), @@ -79,7 +91,77 @@ static const struct soc15_reg_golden golden_settings_sdma_5[] = { SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA1_UTCL1_PAGE, 0x00ffffff, 0x000c5c00) }; +static const struct soc15_reg_golden golden_settings_sdma_5_sriov[] = { + SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA0_GFX_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA0_PAGE_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA0_RLC0_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA0_RLC1_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA0_RLC2_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA0_RLC3_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA0_RLC4_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA0_RLC5_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA0_RLC6_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA0_RLC7_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA1_GFX_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA1_PAGE_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA1_RLC0_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA1_RLC1_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA1_RLC2_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA1_RLC3_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA1_RLC4_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA1_RLC5_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA1_RLC6_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA1_RLC7_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000), +}; + static const struct soc15_reg_golden golden_settings_sdma_nv10[] = { + SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA0_RLC3_RB_WPTR_POLL_CNTL, 0x0000fff0, 0x00403000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA1_RLC3_RB_WPTR_POLL_CNTL, 0x0000fff0, 0x00403000), +}; + +static const struct soc15_reg_golden golden_settings_sdma_nv14[] = { + SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA0_RLC3_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA1_RLC3_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000), +}; + +static const struct soc15_reg_golden golden_settings_sdma_nv12[] = { + SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA0_RLC3_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA0_GB_ADDR_CONFIG, 0x001877ff, 0x00000044), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA0_GB_ADDR_CONFIG_READ, 0x001877ff, 0x00000044), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA1_GB_ADDR_CONFIG, 0x001877ff, 0x00000044), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA1_GB_ADDR_CONFIG_READ, 0x001877ff, 0x00000044), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA1_RLC3_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000), +}; + +static const struct soc15_reg_golden golden_settings_sdma_cyan_skillfish[] = { + SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA0_CHICKEN_BITS, 0xffbf1f0f, 0x03ab0107), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA0_GB_ADDR_CONFIG, 0x001877ff, 0x00000044), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA0_GB_ADDR_CONFIG_READ, 0x001877ff, 0x00000044), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA0_GFX_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA0_PAGE_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA0_RLC0_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA0_RLC1_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA0_RLC2_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA0_RLC3_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA0_RLC4_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA0_RLC5_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA0_RLC6_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA0_RLC7_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA0_UTCL1_PAGE, 0x007fffff, 0x004c5c00), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA1_CHICKEN_BITS, 0xffbf1f0f, 0x03ab0107), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA1_GB_ADDR_CONFIG, 0x001877ff, 0x00000044), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA1_GB_ADDR_CONFIG_READ, 0x001877ff, 0x00000044), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA1_GFX_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA1_PAGE_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA1_RLC0_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA1_RLC1_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA1_RLC2_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA1_RLC3_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA1_RLC4_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA1_RLC5_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA1_RLC6_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA1_RLC7_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA1_UTCL1_PAGE, 0x007fffff, 0x004c5c00) }; static u32 sdma_v5_0_get_reg_offset(struct amdgpu_device *adev, u32 instance, u32 internal_offset) @@ -102,8 +184,8 @@ static u32 sdma_v5_0_get_reg_offset(struct amdgpu_device *adev, u32 instance, u3 static void sdma_v5_0_init_golden_registers(struct amdgpu_device *adev) { - switch (adev->asic_type) { - case CHIP_NAVI10: + switch (adev->ip_versions[SDMA0_HWIP][0]) { + case IP_VERSION(5, 0, 0): soc15_program_register_sequence(adev, golden_settings_sdma_5, (const u32)ARRAY_SIZE(golden_settings_sdma_5)); @@ -111,6 +193,32 @@ static void sdma_v5_0_init_golden_registers(struct amdgpu_device *adev) golden_settings_sdma_nv10, (const u32)ARRAY_SIZE(golden_settings_sdma_nv10)); break; + case IP_VERSION(5, 0, 2): + soc15_program_register_sequence(adev, + golden_settings_sdma_5, + (const u32)ARRAY_SIZE(golden_settings_sdma_5)); + soc15_program_register_sequence(adev, + golden_settings_sdma_nv14, + (const u32)ARRAY_SIZE(golden_settings_sdma_nv14)); + break; + case IP_VERSION(5, 0, 5): + if (amdgpu_sriov_vf(adev)) + soc15_program_register_sequence(adev, + golden_settings_sdma_5_sriov, + (const u32)ARRAY_SIZE(golden_settings_sdma_5_sriov)); + else + soc15_program_register_sequence(adev, + golden_settings_sdma_5, + (const u32)ARRAY_SIZE(golden_settings_sdma_5)); + soc15_program_register_sequence(adev, + golden_settings_sdma_nv12, + (const u32)ARRAY_SIZE(golden_settings_sdma_nv12)); + break; + case IP_VERSION(5, 0, 1): + soc15_program_register_sequence(adev, + golden_settings_sdma_cyan_skillfish, + (const u32)ARRAY_SIZE(golden_settings_sdma_cyan_skillfish)); + break; default: break; } @@ -131,18 +239,30 @@ static void sdma_v5_0_init_golden_registers(struct amdgpu_device *adev) static int sdma_v5_0_init_microcode(struct amdgpu_device *adev) { const char *chip_name; - char fw_name[30]; + char fw_name[40]; int err = 0, i; struct amdgpu_firmware_info *info = NULL; const struct common_firmware_header *header = NULL; const struct sdma_firmware_header_v1_0 *hdr; + if (amdgpu_sriov_vf(adev) && (adev->ip_versions[SDMA0_HWIP][0] == IP_VERSION(5, 0, 5))) + return 0; + DRM_DEBUG("\n"); - switch (adev->asic_type) { - case CHIP_NAVI10: + switch (adev->ip_versions[SDMA0_HWIP][0]) { + case IP_VERSION(5, 0, 0): chip_name = "navi10"; break; + case IP_VERSION(5, 0, 2): + chip_name = "navi14"; + break; + case IP_VERSION(5, 0, 5): + chip_name = "navi12"; + break; + case IP_VERSION(5, 0, 1): + chip_name = "cyan_skillfish2"; + break; default: BUG(); } @@ -227,7 +347,7 @@ static uint64_t sdma_v5_0_ring_get_rptr(struct amdgpu_ring *ring) u64 *rptr; /* XXX check if swapping is necessary on BE */ - rptr = ((u64 *)&ring->adev->wb.wb[ring->rptr_offs]); + rptr = (u64 *)ring->rptr_cpu_addr; DRM_DEBUG("rptr before shift == 0x%016llx\n", *rptr); return ((*rptr) >> 2); @@ -243,30 +363,20 @@ static uint64_t sdma_v5_0_ring_get_rptr(struct amdgpu_ring *ring) static uint64_t sdma_v5_0_ring_get_wptr(struct amdgpu_ring *ring) { struct amdgpu_device *adev = ring->adev; - u64 *wptr = NULL; - uint64_t local_wptr = 0; + u64 wptr; if (ring->use_doorbell) { /* XXX check if swapping is necessary on BE */ - wptr = ((u64 *)&adev->wb.wb[ring->wptr_offs]); - DRM_DEBUG("wptr/doorbell before shift == 0x%016llx\n", *wptr); - *wptr = (*wptr) >> 2; - DRM_DEBUG("wptr/doorbell after shift == 0x%016llx\n", *wptr); + wptr = READ_ONCE(*((u64 *)ring->wptr_cpu_addr)); + DRM_DEBUG("wptr/doorbell before shift == 0x%016llx\n", wptr); } else { - u32 lowbit, highbit; - - wptr = &local_wptr; - lowbit = RREG32(sdma_v5_0_get_reg_offset(adev, ring->me, mmSDMA0_GFX_RB_WPTR)) >> 2; - highbit = RREG32(sdma_v5_0_get_reg_offset(adev, ring->me, mmSDMA0_GFX_RB_WPTR_HI)) >> 2; - - DRM_DEBUG("wptr [%i]high== 0x%08x low==0x%08x\n", - ring->me, highbit, lowbit); - *wptr = highbit; - *wptr = (*wptr) << 32; - *wptr |= lowbit; + wptr = RREG32_SOC15_IP(GC, sdma_v5_0_get_reg_offset(adev, ring->me, mmSDMA0_GFX_RB_WPTR_HI)); + wptr = wptr << 32; + wptr |= RREG32_SOC15_IP(GC, sdma_v5_0_get_reg_offset(adev, ring->me, mmSDMA0_GFX_RB_WPTR)); + DRM_DEBUG("wptr before shift [%i] wptr == 0x%016llx\n", ring->me, wptr); } - return *wptr; + return wptr >> 2; } /** @@ -284,14 +394,14 @@ static void sdma_v5_0_ring_set_wptr(struct amdgpu_ring *ring) if (ring->use_doorbell) { DRM_DEBUG("Using doorbell -- " "wptr_offs == 0x%08x " - "lower_32_bits(ring->wptr) << 2 == 0x%08x " - "upper_32_bits(ring->wptr) << 2 == 0x%08x\n", + "lower_32_bits(ring->wptr << 2) == 0x%08x " + "upper_32_bits(ring->wptr << 2) == 0x%08x\n", ring->wptr_offs, lower_32_bits(ring->wptr << 2), upper_32_bits(ring->wptr << 2)); /* XXX check if swapping is necessary on BE */ - adev->wb.wb[ring->wptr_offs] = lower_32_bits(ring->wptr << 2); - adev->wb.wb[ring->wptr_offs + 1] = upper_32_bits(ring->wptr << 2); + atomic64_set((atomic64_t *)ring->wptr_cpu_addr, + ring->wptr << 2); DRM_DEBUG("calling WDOORBELL64(0x%08x, 0x%016llx)\n", ring->doorbell_index, ring->wptr << 2); WDOORBELL64(ring->doorbell_index, ring->wptr << 2); @@ -303,9 +413,9 @@ static void sdma_v5_0_ring_set_wptr(struct amdgpu_ring *ring) lower_32_bits(ring->wptr << 2), ring->me, upper_32_bits(ring->wptr << 2)); - WREG32(sdma_v5_0_get_reg_offset(adev, ring->me, mmSDMA0_GFX_RB_WPTR), + WREG32_SOC15_IP(GC, sdma_v5_0_get_reg_offset(adev, ring->me, mmSDMA0_GFX_RB_WPTR), lower_32_bits(ring->wptr << 2)); - WREG32(sdma_v5_0_get_reg_offset(adev, ring->me, mmSDMA0_GFX_RB_WPTR_HI), + WREG32_SOC15_IP(GC, sdma_v5_0_get_reg_offset(adev, ring->me, mmSDMA0_GFX_RB_WPTR_HI), upper_32_bits(ring->wptr << 2)); } } @@ -327,7 +437,9 @@ static void sdma_v5_0_ring_insert_nop(struct amdgpu_ring *ring, uint32_t count) * sdma_v5_0_ring_emit_ib - Schedule an IB on the DMA engine * * @ring: amdgpu ring pointer + * @job: job to retrieve vmid from * @ib: IB object to schedule + * @flags: unused * * Schedule an IB in the DMA ring (NAVI10). */ @@ -339,8 +451,15 @@ static void sdma_v5_0_ring_emit_ib(struct amdgpu_ring *ring, unsigned vmid = AMDGPU_JOB_GET_VMID(job); uint64_t csa_mc_addr = amdgpu_sdma_get_csa_mc_addr(ring, vmid); - /* IB packet must end on a 8 DW boundary */ - sdma_v5_0_ring_insert_nop(ring, (10 - (lower_32_bits(ring->wptr) & 7)) % 8); + /* An IB packet must end on a 8 DW boundary--the next dword + * must be on a 8-dword boundary. Our IB packet below is 6 + * dwords long, thus add x number of NOPs, such that, in + * modular arithmetic, + * wptr + 6 + x = 8k, k >= 0, which in C is, + * (wptr + 6 + x) % 8 = 0. + * The expression below, is a solution of x. + */ + sdma_v5_0_ring_insert_nop(ring, (2 - lower_32_bits(ring->wptr)) & 7); amdgpu_ring_write(ring, SDMA_PKT_HEADER_OP(SDMA_OP_INDIRECT) | SDMA_PKT_INDIRECT_HEADER_VMID(vmid & 0xf)); @@ -353,6 +472,30 @@ static void sdma_v5_0_ring_emit_ib(struct amdgpu_ring *ring, } /** + * sdma_v5_0_ring_emit_mem_sync - flush the IB by graphics cache rinse + * + * @ring: amdgpu ring pointer + * + * flush the IB by graphics cache rinse. + */ +static void sdma_v5_0_ring_emit_mem_sync(struct amdgpu_ring *ring) +{ + uint32_t gcr_cntl = SDMA_GCR_GL2_INV | SDMA_GCR_GL2_WB | SDMA_GCR_GLM_INV | + SDMA_GCR_GL1_INV | SDMA_GCR_GLV_INV | SDMA_GCR_GLK_INV | + SDMA_GCR_GLI_INV(1); + + /* flush entire cache L0/L1/L2, this can be optimized by performance requirement */ + amdgpu_ring_write(ring, SDMA_PKT_HEADER_OP(SDMA_OP_GCR_REQ)); + amdgpu_ring_write(ring, SDMA_PKT_GCR_REQ_PAYLOAD1_BASE_VA_31_7(0)); + amdgpu_ring_write(ring, SDMA_PKT_GCR_REQ_PAYLOAD2_GCR_CONTROL_15_0(gcr_cntl) | + SDMA_PKT_GCR_REQ_PAYLOAD2_BASE_VA_47_32(0)); + amdgpu_ring_write(ring, SDMA_PKT_GCR_REQ_PAYLOAD3_LIMIT_VA_31_7(0) | + SDMA_PKT_GCR_REQ_PAYLOAD3_GCR_CONTROL_18_16(gcr_cntl >> 16)); + amdgpu_ring_write(ring, SDMA_PKT_GCR_REQ_PAYLOAD4_LIMIT_VA_47_32(0) | + SDMA_PKT_GCR_REQ_PAYLOAD4_VMID(0)); +} + +/** * sdma_v5_0_ring_emit_hdp_flush - emit an hdp flush on the DMA ring * * @ring: amdgpu ring pointer @@ -363,7 +506,7 @@ static void sdma_v5_0_ring_emit_hdp_flush(struct amdgpu_ring *ring) { struct amdgpu_device *adev = ring->adev; u32 ref_and_mask = 0; - const struct nbio_hdp_flush_reg *nbio_hf_reg = adev->nbio_funcs->hdp_flush_reg; + const struct nbio_hdp_flush_reg *nbio_hf_reg = adev->nbio.hdp_flush_reg; if (ring->me == 0) ref_and_mask = nbio_hf_reg->ref_and_mask_sdma0; @@ -373,8 +516,8 @@ static void sdma_v5_0_ring_emit_hdp_flush(struct amdgpu_ring *ring) amdgpu_ring_write(ring, SDMA_PKT_HEADER_OP(SDMA_OP_POLL_REGMEM) | SDMA_PKT_POLL_REGMEM_HEADER_HDP_FLUSH(1) | SDMA_PKT_POLL_REGMEM_HEADER_FUNC(3)); /* == */ - amdgpu_ring_write(ring, (adev->nbio_funcs->get_hdp_flush_done_offset(adev)) << 2); - amdgpu_ring_write(ring, (adev->nbio_funcs->get_hdp_flush_req_offset(adev)) << 2); + amdgpu_ring_write(ring, (adev->nbio.funcs->get_hdp_flush_done_offset(adev)) << 2); + amdgpu_ring_write(ring, (adev->nbio.funcs->get_hdp_flush_req_offset(adev)) << 2); amdgpu_ring_write(ring, ref_and_mask); /* reference */ amdgpu_ring_write(ring, ref_and_mask); /* mask */ amdgpu_ring_write(ring, SDMA_PKT_POLL_REGMEM_DW5_RETRY_COUNT(0xfff) | @@ -385,7 +528,9 @@ static void sdma_v5_0_ring_emit_hdp_flush(struct amdgpu_ring *ring) * sdma_v5_0_ring_emit_fence - emit a fence on the DMA ring * * @ring: amdgpu ring pointer - * @fence: amdgpu fence object + * @addr: address + * @seq: sequence number + * @flags: fence related flags * * Add a DMA fence packet to the ring to write * the fence seq number and DMA trap packet to generate @@ -394,7 +539,6 @@ static void sdma_v5_0_ring_emit_hdp_flush(struct amdgpu_ring *ring) static void sdma_v5_0_ring_emit_fence(struct amdgpu_ring *ring, u64 addr, u64 seq, unsigned flags) { - struct amdgpu_device *adev = ring->adev; bool write64bit = flags & AMDGPU_FENCE_FLAG_64BIT; /* write the fence */ amdgpu_ring_write(ring, SDMA_PKT_HEADER_OP(SDMA_OP_FENCE) | @@ -417,11 +561,12 @@ static void sdma_v5_0_ring_emit_fence(struct amdgpu_ring *ring, u64 addr, u64 se amdgpu_ring_write(ring, upper_32_bits(seq)); } - /* Interrupt not work fine on GFX10.1 model yet. Use fallback instead */ - if ((flags & AMDGPU_FENCE_FLAG_INT) && adev->pdev->device != 0x50) { + if (flags & AMDGPU_FENCE_FLAG_INT) { + uint32_t ctx = ring->is_mes_queue ? + (ring->hw_queue_id | AMDGPU_FENCE_MES_QUEUE_FLAG) : 0; /* generate an interrupt */ amdgpu_ring_write(ring, SDMA_PKT_HEADER_OP(SDMA_OP_TRAP)); - amdgpu_ring_write(ring, SDMA_PKT_TRAP_INT_CONTEXT_INT_CONTEXT(0)); + amdgpu_ring_write(ring, SDMA_PKT_TRAP_INT_CONTEXT_INT_CONTEXT(ctx)); } } @@ -445,16 +590,13 @@ static void sdma_v5_0_gfx_stop(struct amdgpu_device *adev) amdgpu_ttm_set_buffer_funcs_status(adev, false); for (i = 0; i < adev->sdma.num_instances; i++) { - rb_cntl = RREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_CNTL)); + rb_cntl = RREG32_SOC15_IP(GC, sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_CNTL)); rb_cntl = REG_SET_FIELD(rb_cntl, SDMA0_GFX_RB_CNTL, RB_ENABLE, 0); - WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_CNTL), rb_cntl); - ib_cntl = RREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_IB_CNTL)); + WREG32_SOC15_IP(GC, sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_CNTL), rb_cntl); + ib_cntl = RREG32_SOC15_IP(GC, sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_IB_CNTL)); ib_cntl = REG_SET_FIELD(ib_cntl, SDMA0_GFX_IB_CNTL, IB_ENABLE, 0); - WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_IB_CNTL), ib_cntl); + WREG32_SOC15_IP(GC, sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_IB_CNTL), ib_cntl); } - - sdma0->sched.ready = false; - sdma1->sched.ready = false; } /** @@ -470,7 +612,7 @@ static void sdma_v5_0_rlc_stop(struct amdgpu_device *adev) } /** - * sdma_v_0_ctx_switch_enable - stop the async dma engines context switch + * sdma_v5_0_ctx_switch_enable - stop the async dma engines context switch * * @adev: amdgpu_device pointer * @enable: enable/disable the DMA MEs context switch. @@ -479,7 +621,7 @@ static void sdma_v5_0_rlc_stop(struct amdgpu_device *adev) */ static void sdma_v5_0_ctx_switch_enable(struct amdgpu_device *adev, bool enable) { - u32 f32_cntl, phase_quantum = 0; + u32 f32_cntl = 0, phase_quantum = 0; int i; if (amdgpu_sdma_phase_quantum) { @@ -507,18 +649,22 @@ static void sdma_v5_0_ctx_switch_enable(struct amdgpu_device *adev, bool enable) } for (i = 0; i < adev->sdma.num_instances; i++) { - f32_cntl = RREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_CNTL)); - f32_cntl = REG_SET_FIELD(f32_cntl, SDMA0_CNTL, - AUTO_CTXSW_ENABLE, enable ? 1 : 0); + if (!amdgpu_sriov_vf(adev)) { + f32_cntl = RREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_CNTL)); + f32_cntl = REG_SET_FIELD(f32_cntl, SDMA0_CNTL, + AUTO_CTXSW_ENABLE, enable ? 1 : 0); + } + if (enable && amdgpu_sdma_phase_quantum) { - WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_PHASE0_QUANTUM), + WREG32_SOC15_IP(GC, sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_PHASE0_QUANTUM), phase_quantum); - WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_PHASE1_QUANTUM), + WREG32_SOC15_IP(GC, sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_PHASE1_QUANTUM), phase_quantum); - WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_PHASE2_QUANTUM), + WREG32_SOC15_IP(GC, sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_PHASE2_QUANTUM), phase_quantum); } - WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_CNTL), f32_cntl); + if (!amdgpu_sriov_vf(adev)) + WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_CNTL), f32_cntl); } } @@ -536,11 +682,14 @@ static void sdma_v5_0_enable(struct amdgpu_device *adev, bool enable) u32 f32_cntl; int i; - if (enable == false) { + if (!enable) { sdma_v5_0_gfx_stop(adev); sdma_v5_0_rlc_stop(adev); } + if (amdgpu_sriov_vf(adev)) + return; + for (i = 0; i < adev->sdma.num_instances; i++) { f32_cntl = RREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_F32_CNTL)); f32_cntl = REG_SET_FIELD(f32_cntl, SDMA0_F32_CNTL, HALT, enable ? 0 : 1); @@ -561,7 +710,6 @@ static int sdma_v5_0_gfx_resume(struct amdgpu_device *adev) struct amdgpu_ring *ring; u32 rb_cntl, ib_cntl; u32 rb_bufsz; - u32 wb_offset; u32 doorbell; u32 doorbell_offset; u32 temp; @@ -571,64 +719,69 @@ static int sdma_v5_0_gfx_resume(struct amdgpu_device *adev) for (i = 0; i < adev->sdma.num_instances; i++) { ring = &adev->sdma.instance[i].ring; - wb_offset = (ring->rptr_offs * 4); - WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_SEM_WAIT_FAIL_TIMER_CNTL), 0); + if (!amdgpu_sriov_vf(adev)) + WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_SEM_WAIT_FAIL_TIMER_CNTL), 0); /* Set ring buffer size in dwords */ rb_bufsz = order_base_2(ring->ring_size / 4); - rb_cntl = RREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_CNTL)); + rb_cntl = RREG32_SOC15_IP(GC, sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_CNTL)); rb_cntl = REG_SET_FIELD(rb_cntl, SDMA0_GFX_RB_CNTL, RB_SIZE, rb_bufsz); #ifdef __BIG_ENDIAN rb_cntl = REG_SET_FIELD(rb_cntl, SDMA0_GFX_RB_CNTL, RB_SWAP_ENABLE, 1); rb_cntl = REG_SET_FIELD(rb_cntl, SDMA0_GFX_RB_CNTL, RPTR_WRITEBACK_SWAP_ENABLE, 1); #endif - WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_CNTL), rb_cntl); + WREG32_SOC15_IP(GC, sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_CNTL), rb_cntl); /* Initialize the ring buffer's read and write pointers */ - WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_RPTR), 0); - WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_RPTR_HI), 0); - WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_WPTR), 0); - WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_WPTR_HI), 0); + WREG32_SOC15_IP(GC, sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_RPTR), 0); + WREG32_SOC15_IP(GC, sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_RPTR_HI), 0); + WREG32_SOC15_IP(GC, sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_WPTR), 0); + WREG32_SOC15_IP(GC, sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_WPTR_HI), 0); /* setup the wptr shadow polling */ - wptr_gpu_addr = adev->wb.gpu_addr + (ring->wptr_offs * 4); - WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_WPTR_POLL_ADDR_LO), + wptr_gpu_addr = ring->wptr_gpu_addr; + WREG32_SOC15_IP(GC, sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_WPTR_POLL_ADDR_LO), lower_32_bits(wptr_gpu_addr)); - WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_WPTR_POLL_ADDR_HI), + WREG32_SOC15_IP(GC, sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_WPTR_POLL_ADDR_HI), upper_32_bits(wptr_gpu_addr)); - wptr_poll_cntl = RREG32(sdma_v5_0_get_reg_offset(adev, i, + wptr_poll_cntl = RREG32_SOC15_IP(GC, sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_WPTR_POLL_CNTL)); wptr_poll_cntl = REG_SET_FIELD(wptr_poll_cntl, SDMA0_GFX_RB_WPTR_POLL_CNTL, F32_POLL_ENABLE, 1); - WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_WPTR_POLL_CNTL), + WREG32_SOC15_IP(GC, sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_WPTR_POLL_CNTL), wptr_poll_cntl); /* set the wb address whether it's enabled or not */ - WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_RPTR_ADDR_HI), - upper_32_bits(adev->wb.gpu_addr + wb_offset) & 0xFFFFFFFF); - WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_RPTR_ADDR_LO), - lower_32_bits(adev->wb.gpu_addr + wb_offset) & 0xFFFFFFFC); + WREG32_SOC15_IP(GC, sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_RPTR_ADDR_HI), + upper_32_bits(ring->rptr_gpu_addr) & 0xFFFFFFFF); + WREG32_SOC15_IP(GC, sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_RPTR_ADDR_LO), + lower_32_bits(ring->rptr_gpu_addr) & 0xFFFFFFFC); rb_cntl = REG_SET_FIELD(rb_cntl, SDMA0_GFX_RB_CNTL, RPTR_WRITEBACK_ENABLE, 1); - WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_BASE), ring->gpu_addr >> 8); - WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_BASE_HI), ring->gpu_addr >> 40); + WREG32_SOC15_IP(GC, sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_BASE), + ring->gpu_addr >> 8); + WREG32_SOC15_IP(GC, sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_BASE_HI), + ring->gpu_addr >> 40); ring->wptr = 0; /* before programing wptr to a less value, need set minor_ptr_update first */ - WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_MINOR_PTR_UPDATE), 1); + WREG32_SOC15_IP(GC, sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_MINOR_PTR_UPDATE), 1); if (!amdgpu_sriov_vf(adev)) { /* only bare-metal use register write for wptr */ - WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_WPTR), lower_32_bits(ring->wptr) << 2); - WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_WPTR_HI), upper_32_bits(ring->wptr) << 2); + WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_WPTR), + lower_32_bits(ring->wptr << 2)); + WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_WPTR_HI), + upper_32_bits(ring->wptr << 2)); } - doorbell = RREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_DOORBELL)); - doorbell_offset = RREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_DOORBELL_OFFSET)); + doorbell = RREG32_SOC15_IP(GC, sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_DOORBELL)); + doorbell_offset = RREG32_SOC15_IP(GC, sdma_v5_0_get_reg_offset(adev, i, + mmSDMA0_GFX_DOORBELL_OFFSET)); if (ring->use_doorbell) { doorbell = REG_SET_FIELD(doorbell, SDMA0_GFX_DOORBELL, ENABLE, 1); @@ -637,38 +790,41 @@ static int sdma_v5_0_gfx_resume(struct amdgpu_device *adev) } else { doorbell = REG_SET_FIELD(doorbell, SDMA0_GFX_DOORBELL, ENABLE, 0); } - WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_DOORBELL), doorbell); - WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_DOORBELL_OFFSET), doorbell_offset); + WREG32_SOC15_IP(GC, sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_DOORBELL), doorbell); + WREG32_SOC15_IP(GC, sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_DOORBELL_OFFSET), + doorbell_offset); - adev->nbio_funcs->sdma_doorbell_range(adev, i, ring->use_doorbell, + adev->nbio.funcs->sdma_doorbell_range(adev, i, ring->use_doorbell, ring->doorbell_index, 20); if (amdgpu_sriov_vf(adev)) sdma_v5_0_ring_set_wptr(ring); /* set minor_ptr_update to 0 after wptr programed */ - WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_MINOR_PTR_UPDATE), 0); - - /* set utc l1 enable flag always to 1 */ - temp = RREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_CNTL)); - temp = REG_SET_FIELD(temp, SDMA0_CNTL, UTC_L1_ENABLE, 1); - - /* enable MCBP */ - temp = REG_SET_FIELD(temp, SDMA0_CNTL, MIDCMD_PREEMPT_ENABLE, 1); - WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_CNTL), temp); - - /* Set up RESP_MODE to non-copy addresses */ - temp = RREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_UTCL1_CNTL)); - temp = REG_SET_FIELD(temp, SDMA0_UTCL1_CNTL, RESP_MODE, 3); - temp = REG_SET_FIELD(temp, SDMA0_UTCL1_CNTL, REDO_DELAY, 9); - WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_UTCL1_CNTL), temp); + WREG32_SOC15_IP(GC, sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_MINOR_PTR_UPDATE), 0); - /* program default cache read and write policy */ - temp = RREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_UTCL1_PAGE)); - /* clean read policy and write policy bits */ - temp &= 0xFF0FFF; - temp |= ((CACHE_READ_POLICY_L2__DEFAULT << 12) | (CACHE_WRITE_POLICY_L2__DEFAULT << 14)); - WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_UTCL1_PAGE), temp); + if (!amdgpu_sriov_vf(adev)) { + /* set utc l1 enable flag always to 1 */ + temp = RREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_CNTL)); + temp = REG_SET_FIELD(temp, SDMA0_CNTL, UTC_L1_ENABLE, 1); + + /* enable MCBP */ + temp = REG_SET_FIELD(temp, SDMA0_CNTL, MIDCMD_PREEMPT_ENABLE, 1); + WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_CNTL), temp); + + /* Set up RESP_MODE to non-copy addresses */ + temp = RREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_UTCL1_CNTL)); + temp = REG_SET_FIELD(temp, SDMA0_UTCL1_CNTL, RESP_MODE, 3); + temp = REG_SET_FIELD(temp, SDMA0_UTCL1_CNTL, REDO_DELAY, 9); + WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_UTCL1_CNTL), temp); + + /* program default cache read and write policy */ + temp = RREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_UTCL1_PAGE)); + /* clean read policy and write policy bits */ + temp &= 0xFF0FFF; + temp |= ((CACHE_READ_POLICY_L2__DEFAULT << 12) | (CACHE_WRITE_POLICY_L2__DEFAULT << 14)); + WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_UTCL1_PAGE), temp); + } if (!amdgpu_sriov_vf(adev)) { /* unhalt engine */ @@ -679,15 +835,15 @@ static int sdma_v5_0_gfx_resume(struct amdgpu_device *adev) /* enable DMA RB */ rb_cntl = REG_SET_FIELD(rb_cntl, SDMA0_GFX_RB_CNTL, RB_ENABLE, 1); - WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_CNTL), rb_cntl); + WREG32_SOC15_IP(GC, sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_CNTL), rb_cntl); - ib_cntl = RREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_IB_CNTL)); + ib_cntl = RREG32_SOC15_IP(GC, sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_IB_CNTL)); ib_cntl = REG_SET_FIELD(ib_cntl, SDMA0_GFX_IB_CNTL, IB_ENABLE, 1); #ifdef __BIG_ENDIAN ib_cntl = REG_SET_FIELD(ib_cntl, SDMA0_GFX_IB_CNTL, IB_SWAP_ENABLE, 1); #endif /* enable DMA IBs */ - WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_IB_CNTL), ib_cntl); + WREG32_SOC15_IP(GC, sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_IB_CNTL), ib_cntl); ring->sched.ready = true; @@ -696,11 +852,9 @@ static int sdma_v5_0_gfx_resume(struct amdgpu_device *adev) sdma_v5_0_enable(adev, true); } - r = amdgpu_ring_test_ring(ring); - if (r) { - ring->sched.ready = false; + r = amdgpu_ring_test_helper(ring); + if (r) return r; - } if (adev->mman.buffer_funcs_ring == ring) amdgpu_ttm_set_buffer_funcs_status(adev, true); @@ -791,10 +945,6 @@ static int sdma_v5_0_start(struct amdgpu_device *adev) r = sdma_v5_0_load_microcode(adev); if (r) return r; - - /* The value of mmSDMA_F32_CNTL is invalid the moment after loading fw */ - if (amdgpu_emu_mode == 1 && adev->pdev->device == 0x4d) - msleep(1000); } /* unhalt the MEs */ @@ -811,6 +961,49 @@ static int sdma_v5_0_start(struct amdgpu_device *adev) return r; } +static int sdma_v5_0_mqd_init(struct amdgpu_device *adev, void *mqd, + struct amdgpu_mqd_prop *prop) +{ + struct v10_sdma_mqd *m = mqd; + uint64_t wb_gpu_addr; + + m->sdmax_rlcx_rb_cntl = + order_base_2(prop->queue_size / 4) << SDMA0_RLC0_RB_CNTL__RB_SIZE__SHIFT | + 1 << SDMA0_RLC0_RB_CNTL__RPTR_WRITEBACK_ENABLE__SHIFT | + 6 << SDMA0_RLC0_RB_CNTL__RPTR_WRITEBACK_TIMER__SHIFT | + 1 << SDMA0_RLC0_RB_CNTL__RB_PRIV__SHIFT; + + m->sdmax_rlcx_rb_base = lower_32_bits(prop->hqd_base_gpu_addr >> 8); + m->sdmax_rlcx_rb_base_hi = upper_32_bits(prop->hqd_base_gpu_addr >> 8); + + m->sdmax_rlcx_rb_wptr_poll_cntl = RREG32(sdma_v5_0_get_reg_offset(adev, 0, + mmSDMA0_GFX_RB_WPTR_POLL_CNTL)); + + wb_gpu_addr = prop->wptr_gpu_addr; + m->sdmax_rlcx_rb_wptr_poll_addr_lo = lower_32_bits(wb_gpu_addr); + m->sdmax_rlcx_rb_wptr_poll_addr_hi = upper_32_bits(wb_gpu_addr); + + wb_gpu_addr = prop->rptr_gpu_addr; + m->sdmax_rlcx_rb_rptr_addr_lo = lower_32_bits(wb_gpu_addr); + m->sdmax_rlcx_rb_rptr_addr_hi = upper_32_bits(wb_gpu_addr); + + m->sdmax_rlcx_ib_cntl = RREG32(sdma_v5_0_get_reg_offset(adev, 0, + mmSDMA0_GFX_IB_CNTL)); + + m->sdmax_rlcx_doorbell_offset = + prop->doorbell_index << SDMA0_RLC0_DOORBELL_OFFSET__OFFSET__SHIFT; + + m->sdmax_rlcx_doorbell = REG_SET_FIELD(0, SDMA0_RLC0_DOORBELL, ENABLE, 1); + + return 0; +} + +static void sdma_v5_0_set_mqd_funcs(struct amdgpu_device *adev) +{ + adev->mqds[AMDGPU_HW_IP_DMA].mqd_size = sizeof(struct v10_sdma_mqd); + adev->mqds[AMDGPU_HW_IP_DMA].init_mqd = sdma_v5_0_mqd_init; +} + /** * sdma_v5_0_ring_test_ring - simple async dma engine test * @@ -828,18 +1021,29 @@ static int sdma_v5_0_ring_test_ring(struct amdgpu_ring *ring) int r; u32 tmp; u64 gpu_addr; + volatile uint32_t *cpu_ptr = NULL; - r = amdgpu_device_wb_get(adev, &index); - if (r) { - dev_err(adev->dev, "(%d) failed to allocate wb slot\n", r); - return r; - } - - gpu_addr = adev->wb.gpu_addr + (index * 4); tmp = 0xCAFEDEAD; - adev->wb.wb[index] = cpu_to_le32(tmp); - r = amdgpu_ring_alloc(ring, 5); + if (ring->is_mes_queue) { + uint32_t offset = 0; + offset = amdgpu_mes_ctx_get_offs(ring, + AMDGPU_MES_CTX_PADDING_OFFS); + gpu_addr = amdgpu_mes_ctx_get_offs_gpu_addr(ring, offset); + cpu_ptr = amdgpu_mes_ctx_get_offs_cpu_addr(ring, offset); + *cpu_ptr = tmp; + } else { + r = amdgpu_device_wb_get(adev, &index); + if (r) { + dev_err(adev->dev, "(%d) failed to allocate wb slot\n", r); + return r; + } + + gpu_addr = adev->wb.gpu_addr + (index * 4); + adev->wb.wb[index] = cpu_to_le32(tmp); + } + + r = amdgpu_ring_alloc(ring, 20); if (r) { DRM_ERROR("amdgpu: dma failed to lock ring %d (%d).\n", ring->idx, r); amdgpu_device_wb_free(adev, index); @@ -855,26 +1059,23 @@ static int sdma_v5_0_ring_test_ring(struct amdgpu_ring *ring) amdgpu_ring_commit(ring); for (i = 0; i < adev->usec_timeout; i++) { - tmp = le32_to_cpu(adev->wb.wb[index]); + if (ring->is_mes_queue) + tmp = le32_to_cpu(*cpu_ptr); + else + tmp = le32_to_cpu(adev->wb.wb[index]); if (tmp == 0xDEADBEEF) break; if (amdgpu_emu_mode == 1) msleep(1); else - DRM_UDELAY(1); + udelay(1); } - if (i < adev->usec_timeout) { - if (amdgpu_emu_mode == 1) - DRM_INFO("ring test on %d succeeded in %d msecs\n", ring->idx, i); - else - DRM_INFO("ring test on %d succeeded in %d usecs\n", ring->idx, i); - } else { - DRM_ERROR("amdgpu: ring %d test failed (0x%08X)\n", - ring->idx, tmp); - r = -EINVAL; - } - amdgpu_device_wb_free(adev, index); + if (i >= adev->usec_timeout) + r = -ETIMEDOUT; + + if (!ring->is_mes_queue) + amdgpu_device_wb_free(adev, index); return r; } @@ -883,6 +1084,7 @@ static int sdma_v5_0_ring_test_ring(struct amdgpu_ring *ring) * sdma_v5_0_ring_test_ib - test an IB on the DMA engine * * @ring: amdgpu_ring structure holding ring information + * @timeout: timeout value in jiffies, or MAX_SCHEDULE_TIMEOUT * * Test a simple IB in the DMA ring (NAVI10). * Returns 0 on success, error on failure. @@ -896,21 +1098,38 @@ static int sdma_v5_0_ring_test_ib(struct amdgpu_ring *ring, long timeout) long r; u32 tmp = 0; u64 gpu_addr; + volatile uint32_t *cpu_ptr = NULL; - r = amdgpu_device_wb_get(adev, &index); - if (r) { - dev_err(adev->dev, "(%ld) failed to allocate wb slot\n", r); - return r; - } - - gpu_addr = adev->wb.gpu_addr + (index * 4); tmp = 0xCAFEDEAD; - adev->wb.wb[index] = cpu_to_le32(tmp); memset(&ib, 0, sizeof(ib)); - r = amdgpu_ib_get(adev, NULL, 256, &ib); - if (r) { - DRM_ERROR("amdgpu: failed to get ib (%ld).\n", r); - goto err0; + + if (ring->is_mes_queue) { + uint32_t offset = 0; + offset = amdgpu_mes_ctx_get_offs(ring, AMDGPU_MES_CTX_IB_OFFS); + ib.gpu_addr = amdgpu_mes_ctx_get_offs_gpu_addr(ring, offset); + ib.ptr = (void *)amdgpu_mes_ctx_get_offs_cpu_addr(ring, offset); + + offset = amdgpu_mes_ctx_get_offs(ring, + AMDGPU_MES_CTX_PADDING_OFFS); + gpu_addr = amdgpu_mes_ctx_get_offs_gpu_addr(ring, offset); + cpu_ptr = amdgpu_mes_ctx_get_offs_cpu_addr(ring, offset); + *cpu_ptr = tmp; + } else { + r = amdgpu_device_wb_get(adev, &index); + if (r) { + dev_err(adev->dev, "(%ld) failed to allocate wb slot\n", r); + return r; + } + + gpu_addr = adev->wb.gpu_addr + (index * 4); + adev->wb.wb[index] = cpu_to_le32(tmp); + + r = amdgpu_ib_get(adev, NULL, 256, + AMDGPU_IB_POOL_DIRECT, &ib); + if (r) { + DRM_ERROR("amdgpu: failed to get ib (%ld).\n", r); + goto err0; + } } ib.ptr[0] = SDMA_PKT_HEADER_OP(SDMA_OP_WRITE) | @@ -937,20 +1156,23 @@ static int sdma_v5_0_ring_test_ib(struct amdgpu_ring *ring, long timeout) DRM_ERROR("amdgpu: fence wait failed (%ld).\n", r); goto err1; } - tmp = le32_to_cpu(adev->wb.wb[index]); - if (tmp == 0xDEADBEEF) { - DRM_INFO("ib test on ring %d succeeded\n", ring->idx); + + if (ring->is_mes_queue) + tmp = le32_to_cpu(*cpu_ptr); + else + tmp = le32_to_cpu(adev->wb.wb[index]); + + if (tmp == 0xDEADBEEF) r = 0; - } else { - DRM_ERROR("amdgpu: ib test failed (0x%08X)\n", tmp); + else r = -EINVAL; - } err1: amdgpu_ib_free(adev, &ib, NULL); dma_fence_put(f); err0: - amdgpu_device_wb_free(adev, index); + if (!ring->is_mes_queue) + amdgpu_device_wb_free(adev, index); return r; } @@ -987,10 +1209,9 @@ static void sdma_v5_0_vm_copy_pte(struct amdgpu_ib *ib, * * @ib: indirect buffer to fill with commands * @pe: addr of the page entry - * @addr: dst addr to write into pe + * @value: dst addr to write into pe * @count: number of page entries to update * @incr: increase next addr by incr bytes - * @flags: access flags * * Update PTEs by writing them manually using sDMA (NAVI10). */ @@ -1043,10 +1264,11 @@ static void sdma_v5_0_vm_set_pte_pde(struct amdgpu_ib *ib, } /** - * sdma_v5_0_ring_pad_ib - pad the IB to the required number of dw - * + * sdma_v5_0_ring_pad_ib - pad the IB + * @ring: amdgpu_ring structure holding ring information * @ib: indirect buffer to fill with padding * + * Pad the IB with NOPs to a boundary multiple of 8. */ static void sdma_v5_0_ring_pad_ib(struct amdgpu_ring *ring, struct amdgpu_ib *ib) { @@ -1054,7 +1276,7 @@ static void sdma_v5_0_ring_pad_ib(struct amdgpu_ring *ring, struct amdgpu_ib *ib u32 pad_count; int i; - pad_count = (8 - (ib->length_dw & 0x7)) % 8; + pad_count = (-ib->length_dw) & 0x7; for (i = 0; i < pad_count; i++) if (sdma && sdma->burst_nop && (i == 0)) ib->ptr[ib->length_dw++] = @@ -1086,7 +1308,7 @@ static void sdma_v5_0_ring_emit_pipeline_sync(struct amdgpu_ring *ring) amdgpu_ring_write(ring, addr & 0xfffffffc); amdgpu_ring_write(ring, upper_32_bits(addr) & 0xffffffff); amdgpu_ring_write(ring, seq); /* reference */ - amdgpu_ring_write(ring, 0xfffffff); /* mask */ + amdgpu_ring_write(ring, 0xffffffff); /* mask */ amdgpu_ring_write(ring, SDMA_PKT_POLL_REGMEM_DW5_RETRY_COUNT(0xfff) | SDMA_PKT_POLL_REGMEM_DW5_INTERVAL(4)); /* retry count, poll interval */ } @@ -1096,7 +1318,8 @@ static void sdma_v5_0_ring_emit_pipeline_sync(struct amdgpu_ring *ring) * sdma_v5_0_ring_emit_vm_flush - vm flush using sDMA * * @ring: amdgpu_ring pointer - * @vm: amdgpu_vm pointer + * @vmid: vmid number to use + * @pd_addr: address * * Update the page table base and flush the VM TLB * using sDMA (NAVI10). @@ -1130,16 +1353,25 @@ static void sdma_v5_0_ring_emit_reg_wait(struct amdgpu_ring *ring, uint32_t reg, SDMA_PKT_POLL_REGMEM_DW5_INTERVAL(10)); } +static void sdma_v5_0_ring_emit_reg_write_reg_wait(struct amdgpu_ring *ring, + uint32_t reg0, uint32_t reg1, + uint32_t ref, uint32_t mask) +{ + amdgpu_ring_emit_wreg(ring, reg0, ref); + /* wait for a cycle to reset vm_inv_eng*_ack */ + amdgpu_ring_emit_reg_wait(ring, reg0, 0, 0); + amdgpu_ring_emit_reg_wait(ring, reg1, mask, mask); +} + static int sdma_v5_0_early_init(void *handle) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; - adev->sdma.num_instances = 2; - sdma_v5_0_set_ring_funcs(adev); sdma_v5_0_set_buffer_funcs(adev); sdma_v5_0_set_vm_pte_funcs(adev); sdma_v5_0_set_irq_funcs(adev); + sdma_v5_0_set_mqd_funcs(adev); return 0; } @@ -1176,7 +1408,7 @@ static int sdma_v5_0_sw_init(void *handle) ring->ring_obj = NULL; ring->use_doorbell = true; - DRM_INFO("use_doorbell being set to: [%s]\n", + DRM_DEBUG("SDMA %d use_doorbell being set to: [%s]\n", i, ring->use_doorbell?"true":"false"); ring->doorbell_index = (i == 0) ? @@ -1184,11 +1416,10 @@ static int sdma_v5_0_sw_init(void *handle) : (adev->doorbell_index.sdma_engine[1] << 1); // get DWORD offset sprintf(ring->name, "sdma%d", i); - r = amdgpu_ring_init(adev, ring, 1024, - &adev->sdma.trap_irq, - (i == 0) ? - AMDGPU_SDMA_IRQ_INSTANCE0 : - AMDGPU_SDMA_IRQ_INSTANCE1); + r = amdgpu_ring_init(adev, ring, 1024, &adev->sdma.trap_irq, + (i == 0) ? AMDGPU_SDMA_IRQ_INSTANCE0 : + AMDGPU_SDMA_IRQ_INSTANCE1, + AMDGPU_RING_PRIO_DEFAULT, NULL); if (r) return r; } @@ -1201,8 +1432,12 @@ static int sdma_v5_0_sw_fini(void *handle) struct amdgpu_device *adev = (struct amdgpu_device *)handle; int i; - for (i = 0; i < adev->sdma.num_instances; i++) + for (i = 0; i < adev->sdma.num_instances; i++) { + release_firmware(adev->sdma.instance[i].fw); + adev->sdma.instance[i].fw = NULL; + amdgpu_ring_fini(&adev->sdma.instance[i].ring); + } return 0; } @@ -1316,7 +1551,7 @@ static int sdma_v5_0_ring_preempt_ib(struct amdgpu_ring *ring) if (ring->trail_seq == le32_to_cpu(*(ring->trail_fence_cpu_addr))) break; - DRM_UDELAY(1); + udelay(1); } if (i >= adev->usec_timeout) { @@ -1339,14 +1574,16 @@ static int sdma_v5_0_set_trap_irq_state(struct amdgpu_device *adev, { u32 sdma_cntl; - u32 reg_offset = (type == AMDGPU_SDMA_IRQ_INSTANCE0) ? - sdma_v5_0_get_reg_offset(adev, 0, mmSDMA0_CNTL) : - sdma_v5_0_get_reg_offset(adev, 1, mmSDMA0_CNTL); + if (!amdgpu_sriov_vf(adev)) { + u32 reg_offset = (type == AMDGPU_SDMA_IRQ_INSTANCE0) ? + sdma_v5_0_get_reg_offset(adev, 0, mmSDMA0_CNTL) : + sdma_v5_0_get_reg_offset(adev, 1, mmSDMA0_CNTL); - sdma_cntl = RREG32(reg_offset); - sdma_cntl = REG_SET_FIELD(sdma_cntl, SDMA0_CNTL, TRAP_ENABLE, - state == AMDGPU_IRQ_STATE_ENABLE ? 1 : 0); - WREG32(reg_offset, sdma_cntl); + sdma_cntl = RREG32(reg_offset); + sdma_cntl = REG_SET_FIELD(sdma_cntl, SDMA0_CNTL, TRAP_ENABLE, + state == AMDGPU_IRQ_STATE_ENABLE ? 1 : 0); + WREG32(reg_offset, sdma_cntl); + } return 0; } @@ -1355,7 +1592,25 @@ static int sdma_v5_0_process_trap_irq(struct amdgpu_device *adev, struct amdgpu_irq_src *source, struct amdgpu_iv_entry *entry) { + uint32_t mes_queue_id = entry->src_data[0]; + DRM_DEBUG("IH: SDMA trap\n"); + + if (adev->enable_mes && (mes_queue_id & AMDGPU_FENCE_MES_QUEUE_FLAG)) { + struct amdgpu_mes_queue *queue; + + mes_queue_id &= AMDGPU_FENCE_MES_QUEUE_ID_MASK; + + spin_lock(&adev->mes.queue_id_lock); + queue = idr_find(&adev->mes.queue_id_idr, mes_queue_id); + if (queue) { + DRM_DEBUG("process smda queue id = %d\n", mes_queue_id); + amdgpu_fence_process(queue->ring); + } + spin_unlock(&adev->mes.queue_id_lock); + return 0; + } + switch (entry->client_id) { case SOC15_IH_CLIENTID_SDMA0: switch (entry->ring_id) { @@ -1470,12 +1725,14 @@ static int sdma_v5_0_set_clockgating_state(void *handle, if (amdgpu_sriov_vf(adev)) return 0; - switch (adev->asic_type) { - case CHIP_NAVI10: + switch (adev->ip_versions[SDMA0_HWIP][0]) { + case IP_VERSION(5, 0, 0): + case IP_VERSION(5, 0, 2): + case IP_VERSION(5, 0, 5): sdma_v5_0_update_medium_grain_clock_gating(adev, - state == AMD_CG_STATE_GATE ? true : false); + state == AMD_CG_STATE_GATE); sdma_v5_0_update_medium_grain_light_sleep(adev, - state == AMD_CG_STATE_GATE ? true : false); + state == AMD_CG_STATE_GATE); break; default: break; @@ -1490,7 +1747,7 @@ static int sdma_v5_0_set_powergating_state(void *handle, return 0; } -static void sdma_v5_0_get_clockgating_state(void *handle, u32 *flags) +static void sdma_v5_0_get_clockgating_state(void *handle, u64 *flags) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; int data; @@ -1532,7 +1789,8 @@ static const struct amdgpu_ring_funcs sdma_v5_0_ring_funcs = { .align_mask = 0xf, .nop = SDMA_PKT_NOP_HEADER_OP(SDMA_OP_NOP), .support_64bit_ptrs = true, - .vmhub = AMDGPU_GFXHUB, + .secure_submission_supported = true, + .vmhub = AMDGPU_GFXHUB_0, .get_rptr = sdma_v5_0_ring_get_rptr, .get_wptr = sdma_v5_0_ring_get_wptr, .set_wptr = sdma_v5_0_ring_set_wptr, @@ -1543,10 +1801,11 @@ static const struct amdgpu_ring_funcs sdma_v5_0_ring_funcs = { 6 + /* sdma_v5_0_ring_emit_pipeline_sync */ /* sdma_v5_0_ring_emit_vm_flush */ SOC15_FLUSH_GPU_TLB_NUM_WREG * 3 + - SOC15_FLUSH_GPU_TLB_NUM_REG_WAIT * 6 + + SOC15_FLUSH_GPU_TLB_NUM_REG_WAIT * 6 * 2 + 10 + 10 + 10, /* sdma_v5_0_ring_emit_fence x3 for user fence, vm fence */ - .emit_ib_size = 7 + 6, /* sdma_v5_0_ring_emit_ib */ + .emit_ib_size = 5 + 7 + 6, /* sdma_v5_0_ring_emit_ib */ .emit_ib = sdma_v5_0_ring_emit_ib, + .emit_mem_sync = sdma_v5_0_ring_emit_mem_sync, .emit_fence = sdma_v5_0_ring_emit_fence, .emit_pipeline_sync = sdma_v5_0_ring_emit_pipeline_sync, .emit_vm_flush = sdma_v5_0_ring_emit_vm_flush, @@ -1557,6 +1816,7 @@ static const struct amdgpu_ring_funcs sdma_v5_0_ring_funcs = { .pad_ib = sdma_v5_0_ring_pad_ib, .emit_wreg = sdma_v5_0_ring_emit_wreg, .emit_reg_wait = sdma_v5_0_ring_emit_reg_wait, + .emit_reg_write_reg_wait = sdma_v5_0_ring_emit_reg_write_reg_wait, .init_cond_exec = sdma_v5_0_ring_init_cond_exec, .patch_cond_exec = sdma_v5_0_ring_patch_cond_exec, .preempt_ib = sdma_v5_0_ring_preempt_ib, @@ -1583,7 +1843,8 @@ static const struct amdgpu_irq_src_funcs sdma_v5_0_illegal_inst_irq_funcs = { static void sdma_v5_0_set_irq_funcs(struct amdgpu_device *adev) { - adev->sdma.trap_irq.num_types = AMDGPU_SDMA_IRQ_LAST; + adev->sdma.trap_irq.num_types = AMDGPU_SDMA_IRQ_INSTANCE0 + + adev->sdma.num_instances; adev->sdma.trap_irq.funcs = &sdma_v5_0_trap_irq_funcs; adev->sdma.illegal_inst_irq.funcs = &sdma_v5_0_illegal_inst_irq_funcs; } @@ -1591,10 +1852,11 @@ static void sdma_v5_0_set_irq_funcs(struct amdgpu_device *adev) /** * sdma_v5_0_emit_copy_buffer - copy buffer using the sDMA engine * - * @ring: amdgpu_ring structure holding ring information + * @ib: indirect buffer to copy to * @src_offset: src GPU address * @dst_offset: dst GPU address * @byte_count: number of bytes to xfer + * @tmz: if a secure copy should be used * * Copy GPU buffers using the DMA engine (NAVI10). * Used by the amdgpu ttm implementation to move pages if @@ -1603,10 +1865,12 @@ static void sdma_v5_0_set_irq_funcs(struct amdgpu_device *adev) static void sdma_v5_0_emit_copy_buffer(struct amdgpu_ib *ib, uint64_t src_offset, uint64_t dst_offset, - uint32_t byte_count) + uint32_t byte_count, + bool tmz) { ib->ptr[ib->length_dw++] = SDMA_PKT_HEADER_OP(SDMA_OP_COPY) | - SDMA_PKT_HEADER_SUB_OP(SDMA_SUBOP_COPY_LINEAR); + SDMA_PKT_HEADER_SUB_OP(SDMA_SUBOP_COPY_LINEAR) | + SDMA_PKT_COPY_LINEAR_HEADER_TMZ(tmz ? 1 : 0); ib->ptr[ib->length_dw++] = byte_count - 1; ib->ptr[ib->length_dw++] = 0; /* src/dst endian swap */ ib->ptr[ib->length_dw++] = lower_32_bits(src_offset); @@ -1618,7 +1882,7 @@ static void sdma_v5_0_emit_copy_buffer(struct amdgpu_ib *ib, /** * sdma_v5_0_emit_fill_buffer - fill buffer using the sDMA engine * - * @ring: amdgpu_ring structure holding ring information + * @ib: indirect buffer to fill * @src_data: value to write to buffer * @dst_offset: dst GPU address * @byte_count: number of bytes to xfer @@ -1664,17 +1928,15 @@ static const struct amdgpu_vm_pte_funcs sdma_v5_0_vm_pte_funcs = { static void sdma_v5_0_set_vm_pte_funcs(struct amdgpu_device *adev) { - struct drm_gpu_scheduler *sched; unsigned i; if (adev->vm_manager.vm_pte_funcs == NULL) { adev->vm_manager.vm_pte_funcs = &sdma_v5_0_vm_pte_funcs; for (i = 0; i < adev->sdma.num_instances; i++) { - sched = &adev->sdma.instance[i].ring.sched; - adev->vm_manager.vm_pte_rqs[i] = - &sched->sched_rq[DRM_SCHED_PRIORITY_KERNEL]; + adev->vm_manager.vm_pte_scheds[i] = + &adev->sdma.instance[i].ring.sched; } - adev->vm_manager.vm_pte_num_rqs = adev->sdma.num_instances; + adev->vm_manager.vm_pte_num_scheds = adev->sdma.num_instances; } } |