Hi Eric

On 2025/12/2 23:19, Eric Auger wrote:
Hi Tao,

On 10/12/25 5:06 PM, Tao Tang wrote:
As the first step in implementing secure page table walks, this patch
introduces the logic to decode security-related attributes from various
SMMU structures.

The NSCFG bits from the Context Descriptor are now decoded and stored.
These bits control the security attribute of the starting-level
translation table, which is crucial for managing secure and non-secure
memory accesses.

The SMMU_S_IDR1.SEL2 bit is read to determine if Secure stage 2
translations are supported. This capability is cached in the
SMMUTransCfg structure for the page table walker's use.

Finally, new macros (PTE_NS, PTE_NSTABLE) are added to prepare for
extracting attributes from page and table descriptors. To improve
clarity, these different attribute bits are organized into distinct
subsections in the header file.

Signed-off-by: Tao Tang <[email protected]>
---
  hw/arm/smmu-internal.h       | 16 ++++++++++++++--
  hw/arm/smmuv3-internal.h     |  2 ++
  hw/arm/smmuv3.c              |  2 ++
  include/hw/arm/smmu-common.h |  3 +++
  4 files changed, 21 insertions(+), 2 deletions(-)

diff --git a/hw/arm/smmu-internal.h b/hw/arm/smmu-internal.h
index d143d296f3..a0454f720d 100644
--- a/hw/arm/smmu-internal.h
+++ b/hw/arm/smmu-internal.h
@@ -58,16 +58,28 @@
      ((level == 3) &&                                                    \
       ((pte & ARM_LPAE_PTE_TYPE_MASK) == ARM_LPAE_L3_PTE_TYPE_PAGE))
+/* Block & page descriptor attributes */
+/* Non-secure bit */
+#define PTE_NS(pte) \
+    (extract64(pte, 5, 1))
+
  /* access permissions */
#define PTE_AP(pte) \
      (extract64(pte, 6, 2))
+/* access flag */
+#define PTE_AF(pte) \
+    (extract64(pte, 10, 1))
+
+
+/* Table descriptor attributes */
  #define PTE_APTABLE(pte) \
      (extract64(pte, 61, 2))
-#define PTE_AF(pte) \
-    (extract64(pte, 10, 1))
+#define PTE_NSTABLE(pte) \
+    (extract64(pte, 63, 1))
+
  /*
   * TODO: At the moment all transactions are considered as privileged (EL1)
   * as IOMMU translation callback does not pass user/priv attributes.
diff --git a/hw/arm/smmuv3-internal.h b/hw/arm/smmuv3-internal.h
index 99fdbcf3f5..1e757af459 100644
--- a/hw/arm/smmuv3-internal.h
+++ b/hw/arm/smmuv3-internal.h
@@ -703,6 +703,8 @@ static inline int oas2bits(int oas_field)
  #define CD_R(x)          extract32((x)->word[1], 13, 1)
  #define CD_A(x)          extract32((x)->word[1], 14, 1)
  #define CD_AARCH64(x)    extract32((x)->word[1], 9 , 1)
+#define CD_NSCFG0(x)     extract32((x)->word[2], 0, 1)
+#define CD_NSCFG1(x)     extract32((x)->word[4], 0, 1)
/**
   * tg2granule - Decodes the CD translation granule size field according
diff --git a/hw/arm/smmuv3.c b/hw/arm/smmuv3.c
index 55f4ad1757..3686056d8e 100644
--- a/hw/arm/smmuv3.c
+++ b/hw/arm/smmuv3.c
@@ -812,6 +812,7 @@ static int decode_cd(SMMUv3State *s, SMMUTransCfg *cfg,
              tt->ttb = CACHED_ENTRY_TO_ADDR(entry, tt->ttb);
          }
+ tt->nscfg = i ? CD_NSCFG1(cd) : CD_NSCFG0(cd);
          tt->had = CD_HAD(cd, i);
          trace_smmuv3_decode_cd_tt(i, tt->tsz, tt->ttb, tt->granule_sz, 
tt->had);
      }
@@ -915,6 +916,7 @@ static SMMUTransCfg *smmuv3_get_config(SMMUDevice *sdev, 
SMMUEventInfo *event,
              cfg = NULL;
              return cfg;
          }
+        cfg->sel2 = FIELD_EX32(s->bank[SMMU_SEC_SID_S].idr[1], S_IDR1, SEL2);
I don't get why we store sel2 in the cfg as it does not vary.

Thanks

Eric

You're absolutely right—caching SEL2 in SMMUTransCfg was unnecessary. I didn’t think it through carefully at the time. I’ll drop that change in the next revision.

Thanks,

Tao


Reply via email to