qemu-devel
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [RFC PATCH v2 02/11] hw/arm/smmuv3: Update translation config to hol


From: Eric Auger
Subject: Re: [RFC PATCH v2 02/11] hw/arm/smmuv3: Update translation config to hold stage-2
Date: Fri, 17 Mar 2023 12:37:11 +0100
User-agent: Mozilla/5.0 (X11; Linux x86_64; rv:102.0) Gecko/20100101 Thunderbird/102.5.0

Hi Mostafa,

On 2/26/23 23:06, Mostafa Saleh wrote:
> In preparation for adding stage-2 support, add a S2 config
> struct(SMMUS2Cfg), composed of the following fields and embedded in
> the main SMMUTransCfg:
>  -tsz: Input range
>  -sl0: start level of translation
>  -affd: AF fault disable
>  -granule_sz: Granule page shift
>  -vmid: VMID
>  -vttb: PA of translation table
>  -oas: Output address size
>
> They will be used in the next patches in stage-2 address translation.
>
> No functional change intended.
>
> Signed-off-by: Mostafa Saleh <smostafa@google.com>
> ---
> Changes in v2:
> -Add oas
> ---
>  include/hw/arm/smmu-common.h | 11 +++++++++++
>  1 file changed, 11 insertions(+)
>
> diff --git a/include/hw/arm/smmu-common.h b/include/hw/arm/smmu-common.h
> index 9fcff26357..2deead08d6 100644
> --- a/include/hw/arm/smmu-common.h
> +++ b/include/hw/arm/smmu-common.h
> @@ -58,6 +58,16 @@ typedef struct SMMUTLBEntry {
>      uint8_t granule;
>  } SMMUTLBEntry;
>  
> +typedef struct SMMUS2Cfg {
> +    uint8_t tsz;            /* Input range */
nit: Size of IPA input region. Called t0sz
> +    uint8_t sl0;            /* Start level of translation */
> +    bool affd;              /* AF Fault Disable */
> +    uint8_t granule_sz;     /* Granule page shift */
> +    uint8_t oas;            /* Output address size */
called s2ps. if you don't want to rename you may add this in the comment.

I am suprised to not see S2R which would match S1 record_faults.

Also without further reading we can wonder wheter the parent
iotlb_hits/misses also record S2 events?

I think we need to be/make clear what fields of the original S1 parent
struct also are relevant for the S2 only case.
Maybe tag them with a specific comment S1-only or S1 | S2. It may be
cleaner to introduce a union and common fields in the parent struct though.

Thanks

Eric
> +    uint16_t vmid;          /* Virtual machine ID */
> +    uint64_t vttb;          /* PA of translation table */
> +} SMMUS2Cfg;
> +
>  /*
>   * Generic structure populated by derived SMMU devices
>   * after decoding the configuration information and used as
> @@ -77,6 +87,7 @@ typedef struct SMMUTransCfg {
>      SMMUTransTableInfo tt[2];
>      uint32_t iotlb_hits;       /* counts IOTLB hits for this asid */
>      uint32_t iotlb_misses;     /* counts IOTLB misses for this asid */
> +    struct SMMUS2Cfg s2cfg;
>  } SMMUTransCfg;
>  
>  typedef struct SMMUDevice {




reply via email to

[Prev in Thread] Current Thread [Next in Thread]