[PATCH v3 2/4] iommu/arm-smmu: Workaround for Marvell Armada-AP806 SoC erratum #582743

Tomasz Nowicki tn at semihalf.com
Fri Jul 3 07:24:26 EDT 2020



On 03.07.2020 11:03, Robin Murphy wrote:
> On 2020-07-02 21:16, Tomasz Nowicki wrote:
>> From: Hanna Hawa <hannah at marvell.com>
>>
>> Due to erratum #582743, the Marvell Armada-AP806 can't access 64bit to
>> ARM SMMUv2 registers.
>>
>> Provide implementation relevant hooks:
>> - split the writeq/readq to two accesses of writel/readl.
>> - mask the MMU_IDR2.PTFSv8 fields to not use AArch64 format (but
>> only AARCH32_L) since with AArch64 format 32 bits access is not 
>> supported.
>>
>> Note that separate writes/reads to 2 is not problem regards to
>> atomicity, because the driver use the readq/writeq while initialize
>> the SMMU, report for SMMU fault, and use spinlock in one
>> case (iova_to_phys).
> 
> The comment about the spinlock seems to be out of date, and TBH that 
> whole sentence is a bit unclear - how about something like:
> 
> "Note that most 64-bit registers like TTBRn can be accessed as two 
> 32-bit halves without issue, and AArch32 format ensures that the 
> register writes which must be atomic (for TLBI etc.) need only be 32-bit."
> 
>> Signed-off-by: Hanna Hawa <hannah at marvell.com>
>> Signed-off-by: Gregory CLEMENT <gregory.clement at bootlin.com>
>> Signed-off-by: Tomasz Nowicki <tn at semihalf.com>
>> ---
>>   Documentation/arm64/silicon-errata.rst |  3 ++
>>   drivers/iommu/arm-smmu-impl.c          | 52 ++++++++++++++++++++++++++
>>   2 files changed, 55 insertions(+)
>>
>> diff --git a/Documentation/arm64/silicon-errata.rst 
>> b/Documentation/arm64/silicon-errata.rst
>> index 936cf2a59ca4..157214d3abe1 100644
>> --- a/Documentation/arm64/silicon-errata.rst
>> +++ b/Documentation/arm64/silicon-errata.rst
>> @@ -125,6 +125,9 @@ stable kernels.
>>   | Cavium         | ThunderX2 Core  | #219            | 
>> CAVIUM_TX2_ERRATUM_219      |
>>   
>> +----------------+-----------------+-----------------+-----------------------------+ 
>>
>>   
>> +----------------+-----------------+-----------------+-----------------------------+ 
>>
>> +| Marvell        | ARM-MMU-500     | #582743         | 
>> N/A                         |
>> ++----------------+-----------------+-----------------+-----------------------------+ 
>>
>> ++----------------+-----------------+-----------------+-----------------------------+ 
>>
>>   | Freescale/NXP  | LS2080A/LS1043A | A-008585        | 
>> FSL_ERRATUM_A008585         |
>>   
>> +----------------+-----------------+-----------------+-----------------------------+ 
>>
>>   
>> +----------------+-----------------+-----------------+-----------------------------+ 
>>
>> diff --git a/drivers/iommu/arm-smmu-impl.c 
>> b/drivers/iommu/arm-smmu-impl.c
>> index c75b9d957b70..c1fc5e1b8193 100644
>> --- a/drivers/iommu/arm-smmu-impl.c
>> +++ b/drivers/iommu/arm-smmu-impl.c
>> @@ -147,6 +147,53 @@ static const struct arm_smmu_impl arm_mmu500_impl 
>> = {
>>       .reset = arm_mmu500_reset,
>>   };
>> +static u64 mrvl_mmu500_readq(struct arm_smmu_device *smmu, int page, 
>> int off)
>> +{
>> +    u64 val;
>> +
>> +    /*
>> +     * Marvell Armada-AP806 erratum #582743.
>> +     * Split all the readq to double readl
>> +     */
>> +    val = (u64)readl_relaxed(arm_smmu_page(smmu, page) + off + 4) << 32;
>> +    val |= readl_relaxed(arm_smmu_page(smmu, page) + off);
> 
> Even though io-64-nonatomic-hi-lo.h doesn't override readq() etc. for 
> 64-bit builds, you can still use hi_lo_readq_relaxed() directly.
> 
>> +
>> +    return val;
>> +}
>> +
>> +static void mrvl_mmu500_writeq(struct arm_smmu_device *smmu, int 
>> page, int off,
>> +                   u64 val)
>> +{
>> +    /*
>> +     * Marvell Armada-AP806 erratum #582743.
>> +     * Split all the writeq to double writel
>> +     */
>> +    writel_relaxed(upper_32_bits(val), arm_smmu_page(smmu, page) + 
>> off + 4);
>> +    writel_relaxed(lower_32_bits(val), arm_smmu_page(smmu, page) + off);
> 
> Similarly, hi_lo_writeq_relaxed().
> 
>> +}
>> +
>> +static u32 mrvl_mmu500_cfg_id2_fixup(u32 id)
>> +{
>> +
>> +    /*
>> +     * Armada-AP806 erratum #582743.
>> +     * Hide the SMMU_IDR2.PTFSv8 fields to sidestep the AArch64
>> +     * formats altogether and allow using 32 bits access on the
>> +     * interconnect.
>> +     */
>> +    id &= ~(ARM_SMMU_ID2_PTFS_4K | ARM_SMMU_ID2_PTFS_16K |
>> +        ARM_SMMU_ID2_PTFS_64K);
>> +
>> +    return id;
>> +}
>> +
>> +static const struct arm_smmu_impl mrvl_mmu500_impl = {
>> +    .read_reg64 = mrvl_mmu500_readq,
>> +    .write_reg64 = mrvl_mmu500_writeq,
>> +    .cfg_id2_fixup = mrvl_mmu500_cfg_id2_fixup,
>> +    .reset = arm_mmu500_reset,
>> +};
>> +
>>   struct arm_smmu_device *arm_smmu_impl_init(struct arm_smmu_device 
>> *smmu)
>>   {
>> @@ -160,6 +207,11 @@ struct arm_smmu_device *arm_smmu_impl_init(struct 
>> arm_smmu_device *smmu)
>>        */
>>       switch (smmu->model) {
>>       case ARM_MMU500:
>> +        if (of_device_is_compatible(smmu->dev->of_node,
> 
> Nit: there's a local "np" variable now.
> 
>> +                        "marvell,ap806-smmu-500")) {
>> +            smmu->impl = &mrvl_mmu500_impl;
>> +            return smmu;
>> +        }
> 
> Please put this with the other integration checks below the switch 
> statement. Yes, it means we'll end up assigning smmu->impl twice for 
> this particular case, but that's the intended pattern.
> 

Thanks, all above comments do make sense and will be fixed in next spin.

Thanks,
Tomasz



More information about the linux-arm-kernel mailing list