Re: [PATCH v18 2/4] iommu/arm-smmu: Add a way for implementations to influence SCTLR

[Will Deacon]

On Tue, Nov 03, 2020 at 10:28:13AM -0700, Jordan Crouse wrote:
> On Mon, Nov 02, 2020 at 06:18:45PM +0000, Robin Murphy wrote:
> > On 2020-11-02 17:14, Jordan Crouse wrote:
> > >From: Rob Clark <robdclark@xxxxxxxxxxxx>
> > >
> > >For the Adreno GPU's SMMU, we want SCTLR.HUPCF set to ensure that
> > >pending translations are not terminated on iova fault.  Otherwise
> > >a terminated CP read could hang the GPU by returning invalid
> > >command-stream data.
> > >
> > >Signed-off-by: Rob Clark <robdclark@xxxxxxxxxxxx>
> > >Reviewed-by: Bjorn Andersson <bjorn.andersson@xxxxxxxxxx>
> > >Signed-off-by: Jordan Crouse <jcrouse@xxxxxxxxxxxxxx>
> > >---
> > >
> > >  drivers/iommu/arm/arm-smmu/arm-smmu-qcom.c | 6 ++++++
> > >  drivers/iommu/arm/arm-smmu/arm-smmu.c      | 3 +++
> > >  drivers/iommu/arm/arm-smmu/arm-smmu.h      | 3 +++
> > >  3 files changed, 12 insertions(+)
> > >
> > >diff --git a/drivers/iommu/arm/arm-smmu/arm-smmu-qcom.c b/drivers/iommu/arm/arm-smmu/arm-smmu-qcom.c
> > >index 1e942eed2dfc..0663d7d26908 100644
> > >--- a/drivers/iommu/arm/arm-smmu/arm-smmu-qcom.c
> > >+++ b/drivers/iommu/arm/arm-smmu/arm-smmu-qcom.c
> > >@@ -129,6 +129,12 @@ static int qcom_adreno_smmu_init_context(struct arm_smmu_domain *smmu_domain,
> > >  	    (smmu_domain->cfg.fmt == ARM_SMMU_CTX_FMT_AARCH64))
> > >  		pgtbl_cfg->quirks |= IO_PGTABLE_QUIRK_ARM_TTBR1;
> > >+	/*
> > >+	 * On the GPU device we want to process subsequent transactions after a
> > >+	 * fault to keep the GPU from hanging
> > >+	 */
> > >+	smmu_domain->cfg.sctlr_set |= ARM_SMMU_SCTLR_HUPCF;
> > >+
> > >  	/*
> > >  	 * Initialize private interface with GPU:
> > >  	 */
> > >diff --git a/drivers/iommu/arm/arm-smmu/arm-smmu.c b/drivers/iommu/arm/arm-smmu/arm-smmu.c
> > >index dad7fa86fbd4..1f06ab219819 100644
> > >--- a/drivers/iommu/arm/arm-smmu/arm-smmu.c
> > >+++ b/drivers/iommu/arm/arm-smmu/arm-smmu.c
> > >@@ -617,6 +617,9 @@ void arm_smmu_write_context_bank(struct arm_smmu_device *smmu, int idx)
> > >  	if (IS_ENABLED(CONFIG_CPU_BIG_ENDIAN))
> > >  		reg |= ARM_SMMU_SCTLR_E;
> > >+	reg |= cfg->sctlr_set;
> > >+	reg &= ~cfg->sctlr_clr;
> > 
> > Since we now have a write_s2cr hook, I'm inclined to think that the
> > consistency of a write_sctlr hook that could similarly apply its own
> > arbitrary tweaks would make sense for this. Does anyone have any strong
> > opinions?
> 
> None from me. That would make an eventual stall-on-fault implementation easier
> too.

Sounds like people like this idea, so please can you spin a new version with
that so that I can queue the first three patches for 5.11?

Cheers,

Will