[PATCH 1/2] drm/amdgpu/sriov: For sriov runtime, use kiq to do invalidate tlb
Christian König
ckoenig.leichtzumerken at gmail.com
Wed Aug 15 13:09:15 UTC 2018
Am 15.08.2018 um 14:54 schrieb Emily Deng:
> To avoid the tlb flush not interrupted by world switch, use kiq and one
> command to do tlb invalidate.
>
> v2:
> Add firmware version checking.
>
> v3:
> Refine the code, and move the firmware
> checking into gfx_v9_0_ring_emit_reg_write_reg_wait.
>
> v4:
> Change the name amdgpu_kiq_invalidate_tlb to amdgpu_kiq_reg_write_reg_wait.
> Remove the in_interrupt.
> Refine code.
>
> SWDEV-161497
>
> Signed-off-by: Emily Deng <Emily.Deng at amd.com>
> ---
> drivers/gpu/drm/amd/amdgpu/amdgpu.h | 4 +++
> drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c | 3 --
> drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c | 15 +++++++-
> drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c | 60 ++++++++++++++++++++++++++++++++
> 4 files changed, 78 insertions(+), 4 deletions(-)
>
> diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
> index 6265b88..19ef7711 100644
> --- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h
> +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
> @@ -212,6 +212,10 @@ enum amdgpu_kiq_irq {
> AMDGPU_CP_KIQ_IRQ_LAST
> };
>
> +#define MAX_KIQ_REG_WAIT 5000 /* in usecs, 5ms */
> +#define MAX_KIQ_REG_BAILOUT_INTERVAL 5 /* in msecs, 5ms */
> +#define MAX_KIQ_REG_TRY 20
> +
> int amdgpu_device_ip_set_clockgating_state(void *dev,
> enum amd_ip_block_type block_type,
> enum amd_clockgating_state state);
> diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c
> index 21adb1b6..3885636 100644
> --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c
> +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c
> @@ -22,9 +22,6 @@
> */
>
> #include "amdgpu.h"
> -#define MAX_KIQ_REG_WAIT 5000 /* in usecs, 5ms */
> -#define MAX_KIQ_REG_BAILOUT_INTERVAL 5 /* in msecs, 5ms */
> -#define MAX_KIQ_REG_TRY 20
>
> uint64_t amdgpu_csa_vaddr(struct amdgpu_device *adev)
> {
> diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
> index 76d979e..e010166 100644
> --- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
> +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
> @@ -4348,8 +4348,21 @@ static void gfx_v9_0_ring_emit_reg_write_reg_wait(struct amdgpu_ring *ring,
> uint32_t ref, uint32_t mask)
> {
> int usepfp = (ring->funcs->type == AMDGPU_RING_TYPE_GFX);
> + struct amdgpu_device *adev = ring->adev;
> + bool fw_version_ok = false;
>
> - if (amdgpu_sriov_vf(ring->adev))
> + if (ring->funcs->type == AMDGPU_RING_TYPE_GFX) {
> + if ((adev->gfx.me_fw_version >= 0x0000009c) && (adev->gfx.me_feature_version >= 42)
> + && (adev->gfx.pfp_fw_version >= 0x000000b1) && (adev->gfx.pfp_feature_version >= 42))
> + fw_version_ok = true;
The indentation here is wrong and the && belongs at the end of the line.
> + } else {
> + if ((adev->gfx.mec_fw_version >= 0x00000193) && (adev->gfx.mec_feature_version >= 42))
> + fw_version_ok = true;
> + }
> +
> + fw_version_ok = (adev->asic_type == CHIP_VEGA10) ? fw_version_ok : false;
> +
> + if (amdgpu_sriov_vf(adev) && fw_version_ok)
Please squash the second patch into this one. The goal is to use the
same code path on both SRIOV and bare metal.
I would rather separate our this change into a patch and then make the
TLB invalidation the second patch.
> gfx_v9_0_wait_reg_mem(ring, usepfp, 0, 1, reg0, reg1,
> ref, mask, 0x20);
> else
> diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
> index ed467de..9726c7e 100644
> --- a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
> +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
> @@ -311,6 +311,59 @@ static uint32_t gmc_v9_0_get_invalidate_req(unsigned int vmid)
> return req;
> }
>
> +signed long amdgpu_kiq_reg_write_reg_wait(struct amdgpu_device *adev, struct amdgpu_vmhub *hub,
> + unsigned eng, u32 req, uint32_t vmid)
That is still specialized, e.g. you assume that you work with a VMHUB
here instead of specifying the registers, req and value separately.
> +{
> + signed long r, cnt = 0;
> + unsigned long flags;
> + uint32_t seq;
> + struct amdgpu_kiq *kiq = &adev->gfx.kiq;
> + struct amdgpu_ring *ring = &kiq->ring;
> +
> + if (!ring->ready) {
> + return -EINVAL;
> + }
Please drop the {} here.
Christian.
> +
> + spin_lock_irqsave(&kiq->ring_lock, flags);
> +
> + amdgpu_ring_alloc(ring, 32);
> + amdgpu_ring_emit_reg_write_reg_wait(ring, hub->vm_inv_eng0_req + eng,
> + hub->vm_inv_eng0_ack + eng,
> + req, 1 << vmid);
> + amdgpu_fence_emit_polling(ring, &seq);
> + amdgpu_ring_commit(ring);
> + spin_unlock_irqrestore(&kiq->ring_lock, flags);
> +
> + r = amdgpu_fence_wait_polling(ring, seq, MAX_KIQ_REG_WAIT);
> +
> + /* don't wait anymore for gpu reset case because this way may
> + * block gpu_recover() routine forever, e.g. this virt_kiq_rreg
> + * is triggered in TTM and ttm_bo_lock_delayed_workqueue() will
> + * never return if we keep waiting in virt_kiq_rreg, which cause
> + * gpu_recover() hang there.
> + *
> + * also don't wait anymore for IRQ context
> + * */
> + if (r < 1 && (adev->in_gpu_reset || in_interrupt()))
> + goto failed_kiq;
> +
> + might_sleep();
> +
> + while (r < 1 && cnt++ < MAX_KIQ_REG_TRY) {
> + msleep(MAX_KIQ_REG_BAILOUT_INTERVAL);
> + r = amdgpu_fence_wait_polling(ring, seq, MAX_KIQ_REG_WAIT);
> + }
> +
> + if (cnt > MAX_KIQ_REG_TRY)
> + goto failed_kiq;
> +
> + return 0;
> +
> +failed_kiq:
> + pr_err("failed to invalidate tlb with kiq\n");
> + return r;
> +}
> +
> /*
> * GART
> * VMID 0 is the physical GPU addresses as used by the kernel.
> @@ -332,6 +385,7 @@ static void gmc_v9_0_flush_gpu_tlb(struct amdgpu_device *adev,
> /* Use register 17 for GART */
> const unsigned eng = 17;
> unsigned i, j;
> + int r;
>
> spin_lock(&adev->gmc.invalidate_lock);
>
> @@ -339,6 +393,12 @@ static void gmc_v9_0_flush_gpu_tlb(struct amdgpu_device *adev,
> struct amdgpu_vmhub *hub = &adev->vmhub[i];
> u32 tmp = gmc_v9_0_get_invalidate_req(vmid);
>
> + if (amdgpu_sriov_vf(adev)) {
> + r = amdgpu_kiq_reg_write_reg_wait(adev, hub, eng, tmp, vmid);
> + if (!r)
> + continue;
> + }
> +
> WREG32_NO_KIQ(hub->vm_inv_eng0_req + eng, tmp);
>
> /* Busy wait for ACK.*/
More information about the amd-gfx
mailing list