[Intel-gfx] [PATCH 1/2] drm/i915/gen9: Add framework to whitelist specific GPU registers
Mika Kuoppala
mika.kuoppala at linux.intel.com
Tue Oct 6 02:23:53 PDT 2015
Arun Siluvery <arun.siluvery at linux.intel.com> writes:
> Some of the HW registers are privileged and cannot be written to from a
> non-privileged batch buffers coming from userspace unless they are on whitelist.
> Userspace need write access to them to implement mainly preemption related WA and
> also some general WA.
>
> The reason for using this approach is, the register bits that control preemption
> granularity at the HW level are not context save/restored; so even if we set these
> bits always in kernel they are going to change once the context is switched out.
> We can consider making them non-privileged by default but these registers also
> contain other chicken bits which should not be allowed to be modified.
>
> In the later revisions controlling bits are save/restored at context level but
> in the existing revisions these are exported via other debug registers and should
> be on the whitelist. This patch adds changes to provide HW with a list of registers
> to be whitelisted. HW checks this list during execution and provides access accordingly.
>
> HW imposes a limit on the number of registers on whitelist and it is per-engine.
> At this point we are only enabling whitelist for RCS and we don't foresee any
> requirement for other engines.
>
> The registers to be whitelisted are added using generic workaround list mechanism,
> even these are only enablers for userspace workarounds. But by sharing this
> mechanism we get some test assets without additional cost (Mika).
>
> Cc: Mika Kuoppala <mika.kuoppala at intel.com>
> Cc: Nick Hoath <nicholas.hoath at intel.com>
> Signed-off-by: Arun Siluvery <arun.siluvery at linux.intel.com>
> ---
> drivers/gpu/drm/i915/i915_drv.h | 9 ++++++++-
> drivers/gpu/drm/i915/i915_reg.h | 2 ++
> drivers/gpu/drm/i915/intel_ringbuffer.c | 17 +++++++++++++++++
> 3 files changed, 27 insertions(+), 1 deletion(-)
>
> diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
> index 9c10270..44fbd1d 100644
> --- a/drivers/gpu/drm/i915/i915_drv.h
> +++ b/drivers/gpu/drm/i915/i915_drv.h
> @@ -1674,11 +1674,18 @@ struct i915_wa_reg {
> u32 mask;
> };
>
> -#define I915_MAX_WA_REGS 16
> +/*
> + * RING_MAX_NONPRIV_SLOTS is per-engine but at this point we are only
> + * allowing it for RCS as we don't foresee any requirement of having
> + * a whitelist for other engines. When it is really required for
> + * other engines then the limit need to be increased.
> + */
> +#define I915_MAX_WA_REGS (16 + RING_MAX_NONPRIV_SLOTS)
>
> struct i915_workarounds {
> struct i915_wa_reg reg[I915_MAX_WA_REGS];
> u32 count;
> + u32 hw_whitelist_index[I915_NUM_RINGS];
> };
>
I have plans to allow multiple workaround lists. So we could then
have dedicated one for whitelists too, simplifying this.
Reviewed-by: Mika Kuoppala <mika.kuoppala at intel.com>
> struct i915_virtual_gpu {
> diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
> index 05c8621..797c74e 100644
> --- a/drivers/gpu/drm/i915/i915_reg.h
> +++ b/drivers/gpu/drm/i915/i915_reg.h
> @@ -1578,6 +1578,8 @@ enum skl_disp_power_wells {
> #define RING_WAIT_I8XX (1<<0) /* gen2, PRBx_HEAD */
> #define RING_WAIT (1<<11) /* gen3+, PRBx_CTL */
> #define RING_WAIT_SEMAPHORE (1<<10) /* gen6+ */
> +#define RING_FORCE_TO_NONPRIV(base) ((base)+0x4D0)
> +#define RING_MAX_NONPRIV_SLOTS 12
>
> #define GEN7_TLB_RD_ADDR 0x4700
>
> diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.c b/drivers/gpu/drm/i915/intel_ringbuffer.c
> index c82c74c..64b2754 100644
> --- a/drivers/gpu/drm/i915/intel_ringbuffer.c
> +++ b/drivers/gpu/drm/i915/intel_ringbuffer.c
> @@ -800,6 +800,22 @@ static int wa_add(struct drm_i915_private *dev_priv,
>
> #define WA_WRITE(addr, val) WA_REG(addr, 0xffffffff, val)
>
> +static inline int wa_ring_whitelist_reg(struct intel_engine_cs *ring,
> + uint32_t reg_addr)
> +{
> + struct drm_i915_private *dev_priv = ring->dev->dev_private;
> + struct i915_workarounds *wa = &dev_priv->workarounds;
> + const uint32_t index = wa->hw_whitelist_index[ring->id];
> +
> + if (WARN_ON(index >= RING_MAX_NONPRIV_SLOTS))
> + return -EINVAL;
> +
> + WA_WRITE(RING_FORCE_TO_NONPRIV(ring->mmio_base) + index * 4, reg_addr);
> + wa->hw_whitelist_index[ring->id]++;
> +
> + return 0;
> +}
> +
> static int gen8_init_workarounds(struct intel_engine_cs *ring)
> {
> struct drm_device *dev = ring->dev;
> @@ -1094,6 +1110,7 @@ int init_workarounds_ring(struct intel_engine_cs *ring)
> WARN_ON(ring->id != RCS);
>
> dev_priv->workarounds.count = 0;
> + dev_priv->workarounds.hw_whitelist_index[ring->id] = 0;
>
> if (IS_BROADWELL(dev))
> return bdw_init_workarounds(ring);
> --
> 1.9.1
>
> _______________________________________________
> Intel-gfx mailing list
> Intel-gfx at lists.freedesktop.org
> http://lists.freedesktop.org/mailman/listinfo/intel-gfx
More information about the Intel-gfx
mailing list