[PATCH v2 0/1] target/i386: add support for LAM in CPUID enumeration

[PATCH v2 0/1] target/i386: add support for LAM in CPUID enumeration

[Binbin Wu]

Linear-address masking (LAM) [1], modifies the checking that is applied to
*64-bit* linear addresses, allowing software to use of the untranslated address
bits for metadata and masks the metadata bits before using them as linear 
addresses to access memory.

When the feature is virtualized and exposed to guest, it can be used for efficient
address sanitizers (ASAN) implementation and for optimizations in JITs and virtual
machines.

[1] Intel ISE https://cdrdv2.intel.com/v1/dl/getContent/671368
    Chapter Linear Address Masking (LAM)

---
Changelog
v2:
- Add macro of CPUID definition for LAM bit

v1:
- https://lists.gnu.org/archive/html/qemu-devel/2023-02/msg08036.html


Robert Hoo (1):
  target/i386: add support for LAM in CPUID enumeration

 target/i386/cpu.c | 2 +-
 target/i386/cpu.h | 2 ++
 2 files changed, 3 insertions(+), 1 deletion(-)


base-commit: 51bdb0b57a2d9e84d6915fbae7b5d76c8820cf3c
-- 
2.25.1

[PATCH v2 1/1] target/i386: add support for LAM in CPUID enumeration

[Binbin Wu]

From: Robert Hoo <robert.hu@linux.intel.com>

Linear Address Masking (LAM) is a new Intel CPU feature, which allows software
to use of the untranslated address bits for metadata.

The bit definition:
CPUID.(EAX=7,ECX=1):EAX[26]

Add CPUID definition for LAM.

More info can be found in Intel ISE Chapter "LINEAR ADDRESS MASKING (LAM)"
https://cdrdv2.intel.com/v1/dl/getContent/671368

Signed-off-by: Robert Hoo <robert.hu@linux.intel.com>
Co-developed-by: Binbin Wu <binbin.wu@linux.intel.com>
Signed-off-by: Binbin Wu <binbin.wu@linux.intel.com>
---
 target/i386/cpu.c | 2 +-
 target/i386/cpu.h | 2 ++
 2 files changed, 3 insertions(+), 1 deletion(-)

diff --git a/target/i386/cpu.c b/target/i386/cpu.c
index 1242bd541a..f4436b3657 100644
--- a/target/i386/cpu.c
+++ b/target/i386/cpu.c
@@ -881,7 +881,7 @@ FeatureWordInfo feature_word_info[FEATURE_WORDS] = {
             "fsrc", NULL, NULL, NULL,
             NULL, NULL, NULL, NULL,
             NULL, "amx-fp16", NULL, "avx-ifma",
-            NULL, NULL, NULL, NULL,
+            NULL, NULL, "lam", NULL,
             NULL, NULL, NULL, NULL,
         },
         .cpuid = {
diff --git a/target/i386/cpu.h b/target/i386/cpu.h
index 7201a71de8..eb800ba2e2 100644
--- a/target/i386/cpu.h
+++ b/target/i386/cpu.h
@@ -924,6 +924,8 @@ uint64_t x86_cpu_get_supported_feature_word(FeatureWord w,
 #define CPUID_7_1_EAX_AMX_FP16          (1U << 21)
 /* Support for VPMADD52[H,L]UQ */
 #define CPUID_7_1_EAX_AVX_IFMA          (1U << 23)
+/* Linear Address Masking */
+#define CPUID_7_1_EAX_LAM               (1U << 26)
 
 /* Support for VPDPB[SU,UU,SS]D[,S] */
 #define CPUID_7_1_EDX_AVX_VNNI_INT8     (1U << 4)
-- 
2.25.1

Re: [PATCH v2 1/1] target/i386: add support for LAM in CPUID enumeration

[Xiaoyao Li]

On 5/31/2023 9:32 AM, Binbin Wu wrote:
> From: Robert Hoo <robert.hu@linux.intel.com>
> 
> Linear Address Masking (LAM) is a new Intel CPU feature, which allows software
> to use of the untranslated address bits for metadata.
> 
> The bit definition:
> CPUID.(EAX=7,ECX=1):EAX[26]
> 
> Add CPUID definition for LAM.
> 
> More info can be found in Intel ISE Chapter "LINEAR ADDRESS MASKING (LAM)"
> https://cdrdv2.intel.com/v1/dl/getContent/671368

LAM defines new bits in CR3 and CR4. I think it needs corresponding 
support in QEMU as well.

> Signed-off-by: Robert Hoo <robert.hu@linux.intel.com>
> Co-developed-by: Binbin Wu <binbin.wu@linux.intel.com>
> Signed-off-by: Binbin Wu <binbin.wu@linux.intel.com>
> ---
>   target/i386/cpu.c | 2 +-
>   target/i386/cpu.h | 2 ++
>   2 files changed, 3 insertions(+), 1 deletion(-)
> 
> diff --git a/target/i386/cpu.c b/target/i386/cpu.c
> index 1242bd541a..f4436b3657 100644
> --- a/target/i386/cpu.c
> +++ b/target/i386/cpu.c
> @@ -881,7 +881,7 @@ FeatureWordInfo feature_word_info[FEATURE_WORDS] = {
>               "fsrc", NULL, NULL, NULL,
>               NULL, NULL, NULL, NULL,
>               NULL, "amx-fp16", NULL, "avx-ifma",
> -            NULL, NULL, NULL, NULL,
> +            NULL, NULL, "lam", NULL,
>               NULL, NULL, NULL, NULL,
>           },
>           .cpuid = {
> diff --git a/target/i386/cpu.h b/target/i386/cpu.h
> index 7201a71de8..eb800ba2e2 100644
> --- a/target/i386/cpu.h
> +++ b/target/i386/cpu.h
> @@ -924,6 +924,8 @@ uint64_t x86_cpu_get_supported_feature_word(FeatureWord w,
>   #define CPUID_7_1_EAX_AMX_FP16          (1U << 21)
>   /* Support for VPMADD52[H,L]UQ */
>   #define CPUID_7_1_EAX_AVX_IFMA          (1U << 23)
> +/* Linear Address Masking */
> +#define CPUID_7_1_EAX_LAM               (1U << 26)
>   
>   /* Support for VPDPB[SU,UU,SS]D[,S] */
>   #define CPUID_7_1_EDX_AVX_VNNI_INT8     (1U << 4)

Re: [PATCH v2 1/1] target/i386: add support for LAM in CPUID enumeration

[Binbin Wu]

On 5/31/2023 11:45 AM, Xiaoyao Li wrote:
> On 5/31/2023 9:32 AM, Binbin Wu wrote:
>> From: Robert Hoo <robert.hu@linux.intel.com>
>>
>> Linear Address Masking (LAM) is a new Intel CPU feature, which allows 
>> software
>> to use of the untranslated address bits for metadata.
>>
>> The bit definition:
>> CPUID.(EAX=7,ECX=1):EAX[26]
>>
>> Add CPUID definition for LAM.
>>
>> More info can be found in Intel ISE Chapter "LINEAR ADDRESS MASKING 
>> (LAM)"
>> https://cdrdv2.intel.com/v1/dl/getContent/671368
>
> LAM defines new bits in CR3 and CR4. I think it needs corresponding 
> support in QEMU as well.

In QEMU, there are several callers call cpu_x86_update_{cr3,cr4}().

* target/i386/tcg
   If there is no objection, LAM feature will not be supported for TCG 
of target-i386.
   LAM CPUID bit will not be added to TCG_7_1_EAX_FEATURES.
   helper_write_crN() and helper_vmrun() check CR4 reserved bit before 
calling cpu_x86_update_cr4(),
   i.e. CR4 LAM bit is not allowed to be set in TCG.
   helper_write_crN() and helper_vmrun() check max physcial address bits 
before calling cpu_x86_update_cr3(),
   no change needed, i.e. CR3 LAM bits are not allowed to be set in TCG.
   About CR4 reserved bits, although QEMU code only uses 
cr4_reserved_bits() in target/i386/tcg,
   still want to do the following changes:
diff --git a/target/i386/cpu.h b/target/i386/cpu.h
index eb800ba2e2..3946fe5393 100644
--- a/target/i386/cpu.h
+++ b/target/i386/cpu.h
@@ -261,6 +261,7 @@ typedef enum X86Seg {
  #define CR4_SMAP_MASK   (1U << 21)
  #define CR4_PKE_MASK   (1U << 22)
  #define CR4_PKS_MASK   (1U << 24)
+#define CR4_LAM_SUP_MASK (1U << 28)

  #define CR4_RESERVED_MASK \
  (~(target_ulong)(CR4_VME_MASK | CR4_PVI_MASK | CR4_TSD_MASK \
@@ -269,7 +270,8 @@ typedef enum X86Seg {
                  | CR4_OSFXSR_MASK | CR4_OSXMMEXCPT_MASK | CR4_UMIP_MASK \
                  | CR4_LA57_MASK \
                  | CR4_FSGSBASE_MASK | CR4_PCIDE_MASK | CR4_OSXSAVE_MASK \
-                | CR4_SMEP_MASK | CR4_SMAP_MASK | CR4_PKE_MASK | 
CR4_PKS_MASK))
+                | CR4_SMEP_MASK | CR4_SMAP_MASK | CR4_PKE_MASK | 
CR4_PKS_MASK \
+                | CR4_LAM_SUP_MASK))

  #define DR6_BD          (1 << 13)
  #define DR6_BS          (1 << 14)
@@ -2469,6 +2471,9 @@ static inline uint64_t 
cr4_reserved_bits(CPUX86State *env)
      if (!(env->features[FEAT_7_0_ECX] & CPUID_7_0_ECX_PKS)) {
          reserved_bits |= CR4_PKS_MASK;
      }
+    if (!(env->features[FEAT_7_1_EAX] & CPUID_7_1_EAX_LAM)) {
+        reserved_bits |= CR4_LAM_SUP_MASK;
+    }
      return reserved_bits;
  }


* target/i386/gdbstub
   x86_cpu_gdb_write_register() will call cpu_x86_update_cr3() to update 
cr3/cr4.
   Allow gdb to set the LAM bit(s) to CR3/CR4, if vcpu doesn't support 
LAM, set sregs will fail in KVM.


* target/i386/cpu
   x86_cpu_reset_hold() will call cpu_x86_update_cr4() to reset cr4, it 
should be OK.


>
>> Signed-off-by: Robert Hoo <robert.hu@linux.intel.com>
>> Co-developed-by: Binbin Wu <binbin.wu@linux.intel.com>
>> Signed-off-by: Binbin Wu <binbin.wu@linux.intel.com>
>> ---
>>   target/i386/cpu.c | 2 +-
>>   target/i386/cpu.h | 2 ++
>>   2 files changed, 3 insertions(+), 1 deletion(-)
>>
>> diff --git a/target/i386/cpu.c b/target/i386/cpu.c
>> index 1242bd541a..f4436b3657 100644
>> --- a/target/i386/cpu.c
>> +++ b/target/i386/cpu.c
>> @@ -881,7 +881,7 @@ FeatureWordInfo feature_word_info[FEATURE_WORDS] = {
>>               "fsrc", NULL, NULL, NULL,
>>               NULL, NULL, NULL, NULL,
>>               NULL, "amx-fp16", NULL, "avx-ifma",
>> -            NULL, NULL, NULL, NULL,
>> +            NULL, NULL, "lam", NULL,
>>               NULL, NULL, NULL, NULL,
>>           },
>>           .cpuid = {
>> diff --git a/target/i386/cpu.h b/target/i386/cpu.h
>> index 7201a71de8..eb800ba2e2 100644
>> --- a/target/i386/cpu.h
>> +++ b/target/i386/cpu.h
>> @@ -924,6 +924,8 @@ uint64_t 
>> x86_cpu_get_supported_feature_word(FeatureWord w,
>>   #define CPUID_7_1_EAX_AMX_FP16          (1U << 21)
>>   /* Support for VPMADD52[H,L]UQ */
>>   #define CPUID_7_1_EAX_AVX_IFMA          (1U << 23)
>> +/* Linear Address Masking */
>> +#define CPUID_7_1_EAX_LAM               (1U << 26)
>>     /* Support for VPDPB[SU,UU,SS]D[,S] */
>>   #define CPUID_7_1_EDX_AVX_VNNI_INT8     (1U << 4)
>