Re: [PATCH v3 4/6] iommu/io-pgtable-arm: add support for non-strict mode

[Robin Murphy]

On 2018-07-26 8:20 AM, Leizhen (ThunderTown) wrote:
> On 2018/7/25 6:25, Robin Murphy wrote:
> > On 2018-07-12 7:18 AM, Zhen Lei wrote:
> > > To support the non-strict mode, now we only tlbi and sync for the strict
> > > mode. But for the non-leaf case, always follow strict mode.
> > >
> > > Use the lowest bit of the iova parameter to pass the strict mode:
> > > 0, IOMMU_STRICT;
> > > 1, IOMMU_NON_STRICT;
> > > Treat 0 as IOMMU_STRICT, so that the unmap operation can compatible with
> > > other IOMMUs which still use strict mode.
> > >
> > > Signed-off-by: Zhen Lei <thunder.leizhen@xxxxxxxxxx>
> > > ---
> > >    drivers/iommu/io-pgtable-arm.c | 23 ++++++++++++++---------
> > >    1 file changed, 14 insertions(+), 9 deletions(-)
> > >
> > > diff --git a/drivers/iommu/io-pgtable-arm.c b/drivers/iommu/io-pgtable-arm.c
> > > index 010a254..9234db3 100644
> > > --- a/drivers/iommu/io-pgtable-arm.c
> > > +++ b/drivers/iommu/io-pgtable-arm.c
> > > @@ -292,7 +292,7 @@ static void __arm_lpae_set_pte(arm_lpae_iopte *ptep, arm_lpae_iopte pte,
> > >      static size_t __arm_lpae_unmap(struct arm_lpae_io_pgtable *data,
> > >                       unsigned long iova, size_t size, int lvl,
> > > -                   arm_lpae_iopte *ptep);
> > > +                   arm_lpae_iopte *ptep, int strict);
> > >      static void __arm_lpae_init_pte(struct arm_lpae_io_pgtable *data,
> > >                    phys_addr_t paddr, arm_lpae_iopte prot,
> > > @@ -334,7 +334,7 @@ static int arm_lpae_init_pte(struct arm_lpae_io_pgtable *data,
> > >            size_t sz = ARM_LPAE_BLOCK_SIZE(lvl, data);
> > >              tblp = ptep - ARM_LPAE_LVL_IDX(iova, lvl, data);
> > > -        if (WARN_ON(__arm_lpae_unmap(data, iova, sz, lvl, tblp) != sz))
> > > +        if (WARN_ON(__arm_lpae_unmap(data, iova, sz, lvl, tblp, IOMMU_STRICT) != sz))
> > >                return -EINVAL;
> > >        }
> > >    @@ -531,7 +531,7 @@ static void arm_lpae_free_pgtable(struct io_pgtable *iop)
> > >    static size_t arm_lpae_split_blk_unmap(struct arm_lpae_io_pgtable *data,
> > >                           unsigned long iova, size_t size,
> > >                           arm_lpae_iopte blk_pte, int lvl,
> > > -                       arm_lpae_iopte *ptep)
> > > +                       arm_lpae_iopte *ptep, int strict)
> >
> > DMA code should never ever be splitting blocks anyway, and frankly the TLB maintenance here is dodgy enough (since we can't reasonably do break-before make as VMSA says we should) that I *really* don't want to introduce any possibility of making it more asynchronous. I'd much rather just hard-code the expectation of strict == true for this.
>
> OK, I will hard-code strict=true for it.
>
> But since it never ever be happened, why did not give a warning at the beginning?

Because DMA code is not the only caller of iommu_map/unmap. It's 
perfectly legal in the IOMMU API to partially unmap a previous mapping 
such that a block entry needs to be split. The DMA API, however, is a 
lot more constrined, and thus by construction the iommu-dma layer will 
never generate a block-splitting iommu_unmap() except as a result of 
illegal DMA API usage, and we obviously do not need to optimise for that 
(you will get a warning about mismatched unmaps under dma-debug, but 
it's a bit too expensive to police in the general case).

Robin.

> > >    {
> > >        struct io_pgtable_cfg *cfg = &data->iop.cfg;
> > >        arm_lpae_iopte pte, *tablep;
> > > @@ -576,15 +576,18 @@ static size_t arm_lpae_split_blk_unmap(struct arm_lpae_io_pgtable *data,
> > >        }
> > >          if (unmap_idx < 0)
> > > -        return __arm_lpae_unmap(data, iova, size, lvl, tablep);
> > > +        return __arm_lpae_unmap(data, iova, size, lvl, tablep, strict);
> > >          io_pgtable_tlb_add_flush(&data->iop, iova, size, size, true);
> > > +    if (!strict)
> > > +        io_pgtable_tlb_sync(&data->iop);
> > > +
> > >        return size;
> > >    }
> > >      static size_t __arm_lpae_unmap(struct arm_lpae_io_pgtable *data,
> > >                       unsigned long iova, size_t size, int lvl,
> > > -                   arm_lpae_iopte *ptep)
> > > +                   arm_lpae_iopte *ptep, int strict)
> > >    {
> > >        arm_lpae_iopte pte;
> > >        struct io_pgtable *iop = &data->iop;
> > > @@ -609,7 +612,7 @@ static size_t __arm_lpae_unmap(struct arm_lpae_io_pgtable *data,
> > >                io_pgtable_tlb_sync(iop);
> > >                ptep = iopte_deref(pte, data);
> > >                __arm_lpae_free_pgtable(data, lvl + 1, ptep);
> > > -        } else {
> > > +        } else if (strict) {
> > >                io_pgtable_tlb_add_flush(iop, iova, size, size, true);
> > >            }
> > >    @@ -620,25 +623,27 @@ static size_t __arm_lpae_unmap(struct arm_lpae_io_pgtable *data,
> > >             * minus the part we want to unmap
> > >             */
> > >            return arm_lpae_split_blk_unmap(data, iova, size, pte,
> > > -                        lvl + 1, ptep);
> > > +                        lvl + 1, ptep, strict);
> > >        }
> > >          /* Keep on walkin' */
> > >        ptep = iopte_deref(pte, data);
> > > -    return __arm_lpae_unmap(data, iova, size, lvl + 1, ptep);
> > > +    return __arm_lpae_unmap(data, iova, size, lvl + 1, ptep, strict);
> > >    }
> > >      static size_t arm_lpae_unmap(struct io_pgtable_ops *ops, unsigned long iova,
> > >                     size_t size)
> > >    {
> > > +    int strict = ((iova & IOMMU_STRICT_MODE_MASK) == IOMMU_STRICT);
> > >        struct arm_lpae_io_pgtable *data = io_pgtable_ops_to_data(ops);
> > >        arm_lpae_iopte *ptep = data->pgd;
> > >        int lvl = ARM_LPAE_START_LVL(data);
> > >    +    iova &= ~IOMMU_STRICT_MODE_MASK;
> > >        if (WARN_ON(iova >= (1ULL << data->iop.cfg.ias)))
> > >            return 0;
> > >    -    return __arm_lpae_unmap(data, iova, size, lvl, ptep);
> > > +    return __arm_lpae_unmap(data, iova, size, lvl, ptep, strict);
> > >    }
> > >      static phys_addr_t arm_lpae_iova_to_phys(struct io_pgtable_ops *ops,
> >
> > .