Add TSX-NI related instructions and new instructions to
x86-opcode-map.txt according to the Intel(R) 64 and IA-32
Architectures Software Developer's Manual Vol2C (June, 2013).
This also includes below updates.
- Fix a typo of MWAIT (the lack of (11B)).
- Change NOP Ev to prefetchw Ev
- Add CRC32 new prefix style (66&F2)
- Add ADCX, ADOX, RDSEED, CLAC and STAC instructions
Signed-off-by: Masami Hiramatsu <[email protected]>
Cc: Thomas Gleixner <[email protected]>
Cc: Ingo Molnar <[email protected]>
Cc: "H. Peter Anvin" <[email protected]>
---
arch/x86/lib/x86-opcode-map.txt | 42 +++++++++++++++++++---------------
arch/x86/tools/gen-insn-attr-x86.awk | 4 ++-
2 files changed, 27 insertions(+), 19 deletions(-)
diff --git a/arch/x86/lib/x86-opcode-map.txt b/arch/x86/lib/x86-opcode-map.txt
index 5d7e51f..533a85e 100644
--- a/arch/x86/lib/x86-opcode-map.txt
+++ b/arch/x86/lib/x86-opcode-map.txt
@@ -1,10 +1,8 @@
# x86 Opcode Maps
#
# This is (mostly) based on following documentations.
-# - Intel(R) 64 and IA-32 Architectures Software Developer's Manual Vol.2
-# (#325383-040US, October 2011)
-# - Intel(R) Advanced Vector Extensions Programming Reference
-# (#319433-011,JUNE 2011).
+# - Intel(R) 64 and IA-32 Architectures Software Developer's Manual Vol.2C
+# (#326018-047US, June 2013)
#
#<Opcode maps>
# Table: table-name
@@ -29,6 +27,7 @@
# - (F3): the last prefix is 0xF3
# - (F2): the last prefix is 0xF2
# - (!F3) : the last prefix is not 0xF3 (including non-last prefix case)
+# - (66&F2): Both 0x66 and 0xF2 prefixes are specified.
Table: one byte opcode
Referrer:
@@ -246,8 +245,8 @@ c2: RETN Iw (f64)
c3: RETN
c4: LES Gz,Mp (i64) | VEX+2byte (Prefix)
c5: LDS Gz,Mp (i64) | VEX+1byte (Prefix)
-c6: Grp11 Eb,Ib (1A)
-c7: Grp11 Ev,Iz (1A)
+c6: Grp11A Eb,Ib (1A)
+c7: Grp11B Ev,Iz (1A)
c8: ENTER Iw,Ib
c9: LEAVE (d64)
ca: RETF Iw
@@ -293,8 +292,8 @@ ef: OUT DX,eAX
# 0xf0 - 0xff
f0: LOCK (Prefix)
f1:
-f2: REPNE (Prefix)
-f3: REP/REPE (Prefix)
+f2: REPNE (Prefix) | XACQUIRE (Prefix)
+f3: REP/REPE (Prefix) | XRELEASE (Prefix)
f4: HLT
f5: CMC
f6: Grp3_1 Eb (1A)
@@ -326,7 +325,8 @@ AVXcode: 1
0a:
0b: UD2 (1B)
0c:
-0d: NOP Ev | GrpP
+# AMD's prefetch group. Intel supports prefetchw(/1) only.
+0d: GrpP
0e: FEMMS
# 3DNow! uses the last imm byte as opcode extension.
0f: 3DNow! Pq,Qq,Ib
@@ -729,12 +729,12 @@ dc: VAESENC Vdq,Hdq,Wdq (66),(v1)
dd: VAESENCLAST Vdq,Hdq,Wdq (66),(v1)
de: VAESDEC Vdq,Hdq,Wdq (66),(v1)
df: VAESDECLAST Vdq,Hdq,Wdq (66),(v1)
-f0: MOVBE Gy,My | MOVBE Gw,Mw (66) | CRC32 Gd,Eb (F2)
-f1: MOVBE My,Gy | MOVBE Mw,Gw (66) | CRC32 Gd,Ey (F2)
+f0: MOVBE Gy,My | MOVBE Gw,Mw (66) | CRC32 Gd,Eb (F2) | CRC32 Gd,Eb (66&F2)
+f1: MOVBE My,Gy | MOVBE Mw,Gw (66) | CRC32 Gd,Ey (F2) | CRC32 Gd,Ew (66&F2)
f2: ANDN Gy,By,Ey (v)
f3: Grp17 (1A)
f5: BZHI Gy,Ey,By (v) | PEXT Gy,By,Ey (F3),(v) | PDEP Gy,By,Ey (F2),(v)
-f6: MULX By,Gy,rDX,Ey (F2),(v)
+f6: ADCX Gy,Ey (66) | ADOX Gy,Ey (F3) | MULX By,Gy,rDX,Ey (F2),(v)
f7: BEXTR Gy,Ey,By (v) | SHLX Gy,Ey,By (66),(v) | SARX Gy,Ey,By (F3),(v) | SHRX Gy,Ey,By (F2),(v)
EndTable
@@ -861,8 +861,8 @@ EndTable
GrpTable: Grp7
0: SGDT Ms | VMCALL (001),(11B) | VMLAUNCH (010),(11B) | VMRESUME (011),(11B) | VMXOFF (100),(11B)
-1: SIDT Ms | MONITOR (000),(11B) | MWAIT (001)
-2: LGDT Ms | XGETBV (000),(11B) | XSETBV (001),(11B) | VMFUNC (100),(11B)
+1: SIDT Ms | MONITOR (000),(11B) | MWAIT (001),(11B) | CLAC (010),(11B) | STAC (011),(11B)
+2: LGDT Ms | XGETBV (000),(11B) | XSETBV (001),(11B) | VMFUNC (100),(11B) | XEND (101)(11B) | XTEST (110)(11B)
3: LIDT Ms
4: SMSW Mw/Rv
5:
@@ -880,15 +880,21 @@ EndTable
GrpTable: Grp9
1: CMPXCHG8B/16B Mq/Mdq
6: VMPTRLD Mq | VMCLEAR Mq (66) | VMXON Mq (F3) | RDRAND Rv (11B)
-7: VMPTRST Mq | VMPTRST Mq (F3)
+7: VMPTRST Mq | VMPTRST Mq (F3) | RDSEED Rv (11B)
EndTable
GrpTable: Grp10
EndTable
-GrpTable: Grp11
-# Note: the operands are given by group opcode
-0: MOV
+# Grp11A and Grp11B are expressed as Grp11 in Intel SDM
+GrpTable: Grp11A
+0: MOV Eb,Ib
+7: XABORT Ib (000),(11B)
+EndTable
+
+GrpTable: Grp11B
+0: MOV Eb,Iz
+7: XBEGIN Jz (000),(11B)
EndTable
GrpTable: Grp12
diff --git a/arch/x86/tools/gen-insn-attr-x86.awk b/arch/x86/tools/gen-insn-attr-x86.awk
index e6773dc..093a892 100644
--- a/arch/x86/tools/gen-insn-attr-x86.awk
+++ b/arch/x86/tools/gen-insn-attr-x86.awk
@@ -68,7 +68,7 @@ BEGIN {
lprefix1_expr = "\\((66|!F3)\\)"
lprefix2_expr = "\\(F3\\)"
- lprefix3_expr = "\\((F2|!F3)\\)"
+ lprefix3_expr = "\\((F2|!F3|66\\&F2)\\)"
lprefix_expr = "\\((66|F2|F3)\\)"
max_lprefix = 4
@@ -83,6 +83,8 @@ BEGIN {
prefix_num["Operand-Size"] = "INAT_PFX_OPNDSZ"
prefix_num["REPNE"] = "INAT_PFX_REPNE"
prefix_num["REP/REPE"] = "INAT_PFX_REPE"
+ prefix_num["XACQUIRE"] = "INAT_PFX_REPNE"
+ prefix_num["XRELEASE"] = "INAT_PFX_REPE"
prefix_num["LOCK"] = "INAT_PFX_LOCK"
prefix_num["SEG=CS"] = "INAT_PFX_CS"
prefix_num["SEG=DS"] = "INAT_PFX_DS"
On 08/06/2013 12:37 AM, Masami Hiramatsu wrote:
> Add TSX-NI related instructions and new instructions to
> x86-opcode-map.txt according to the Intel(R) 64 and IA-32
> Architectures Software Developer's Manual Vol2C (June, 2013).
> This also includes below updates.
> - Fix a typo of MWAIT (the lack of (11B)).
> - Change NOP Ev to prefetchw Ev
> - Add CRC32 new prefix style (66&F2)
> - Add ADCX, ADOX, RDSEED, CLAC and STAC instructions
>
> Signed-off-by: Masami Hiramatsu <[email protected]>
> Cc: Thomas Gleixner <[email protected]>
> Cc: Ingo Molnar <[email protected]>
> Cc: "H. Peter Anvin" <[email protected]>
Thanks, applied!
You may want to also look at:
http://software.intel.com/en-us/intel-isa-extensions
... specifically ...
http://software.intel.com/sites/default/files/319433-015.pdf
For a *lot* of new instructions coming down the pipe.
-hpa
Commit-ID: 3e21bb092d07e6d394e6d754057d4ff2d363d318
Gitweb: http://git.kernel.org/tip/3e21bb092d07e6d394e6d754057d4ff2d363d318
Author: Masami Hiramatsu <[email protected]>
AuthorDate: Tue, 6 Aug 2013 16:37:50 +0900
Committer: H. Peter Anvin <[email protected]>
CommitDate: Tue, 6 Aug 2013 08:08:47 -0700
x86, insn: Add new opcodes as of June, 2013
Add TSX-NI related instructions and new instructions to
x86-opcode-map.txt according to the Intel(R) 64 and IA-32
Architectures Software Developer's Manual Vol2C (June, 2013).
This also includes below updates.
- Fix a typo of MWAIT (the lack of (11B)).
- Change NOP Ev to prefetchw Ev
- Add CRC32 new prefix style (66&F2)
- Add ADCX, ADOX, RDSEED, CLAC and STAC instructions
Signed-off-by: Masami Hiramatsu <[email protected]>
Link: http://lkml.kernel.org/r/[email protected]
Cc: Thomas Gleixner <[email protected]>
Cc: Ingo Molnar <[email protected]>
Cc: "H. Peter Anvin" <[email protected]>
Signed-off-by: H. Peter Anvin <[email protected]>
---
arch/x86/lib/x86-opcode-map.txt | 42 ++++++++++++++++++++----------------
arch/x86/tools/gen-insn-attr-x86.awk | 4 +++-
2 files changed, 27 insertions(+), 19 deletions(-)
diff --git a/arch/x86/lib/x86-opcode-map.txt b/arch/x86/lib/x86-opcode-map.txt
index 5d7e51f..533a85e 100644
--- a/arch/x86/lib/x86-opcode-map.txt
+++ b/arch/x86/lib/x86-opcode-map.txt
@@ -1,10 +1,8 @@
# x86 Opcode Maps
#
# This is (mostly) based on following documentations.
-# - Intel(R) 64 and IA-32 Architectures Software Developer's Manual Vol.2
-# (#325383-040US, October 2011)
-# - Intel(R) Advanced Vector Extensions Programming Reference
-# (#319433-011,JUNE 2011).
+# - Intel(R) 64 and IA-32 Architectures Software Developer's Manual Vol.2C
+# (#326018-047US, June 2013)
#
#<Opcode maps>
# Table: table-name
@@ -29,6 +27,7 @@
# - (F3): the last prefix is 0xF3
# - (F2): the last prefix is 0xF2
# - (!F3) : the last prefix is not 0xF3 (including non-last prefix case)
+# - (66&F2): Both 0x66 and 0xF2 prefixes are specified.
Table: one byte opcode
Referrer:
@@ -246,8 +245,8 @@ c2: RETN Iw (f64)
c3: RETN
c4: LES Gz,Mp (i64) | VEX+2byte (Prefix)
c5: LDS Gz,Mp (i64) | VEX+1byte (Prefix)
-c6: Grp11 Eb,Ib (1A)
-c7: Grp11 Ev,Iz (1A)
+c6: Grp11A Eb,Ib (1A)
+c7: Grp11B Ev,Iz (1A)
c8: ENTER Iw,Ib
c9: LEAVE (d64)
ca: RETF Iw
@@ -293,8 +292,8 @@ ef: OUT DX,eAX
# 0xf0 - 0xff
f0: LOCK (Prefix)
f1:
-f2: REPNE (Prefix)
-f3: REP/REPE (Prefix)
+f2: REPNE (Prefix) | XACQUIRE (Prefix)
+f3: REP/REPE (Prefix) | XRELEASE (Prefix)
f4: HLT
f5: CMC
f6: Grp3_1 Eb (1A)
@@ -326,7 +325,8 @@ AVXcode: 1
0a:
0b: UD2 (1B)
0c:
-0d: NOP Ev | GrpP
+# AMD's prefetch group. Intel supports prefetchw(/1) only.
+0d: GrpP
0e: FEMMS
# 3DNow! uses the last imm byte as opcode extension.
0f: 3DNow! Pq,Qq,Ib
@@ -729,12 +729,12 @@ dc: VAESENC Vdq,Hdq,Wdq (66),(v1)
dd: VAESENCLAST Vdq,Hdq,Wdq (66),(v1)
de: VAESDEC Vdq,Hdq,Wdq (66),(v1)
df: VAESDECLAST Vdq,Hdq,Wdq (66),(v1)
-f0: MOVBE Gy,My | MOVBE Gw,Mw (66) | CRC32 Gd,Eb (F2)
-f1: MOVBE My,Gy | MOVBE Mw,Gw (66) | CRC32 Gd,Ey (F2)
+f0: MOVBE Gy,My | MOVBE Gw,Mw (66) | CRC32 Gd,Eb (F2) | CRC32 Gd,Eb (66&F2)
+f1: MOVBE My,Gy | MOVBE Mw,Gw (66) | CRC32 Gd,Ey (F2) | CRC32 Gd,Ew (66&F2)
f2: ANDN Gy,By,Ey (v)
f3: Grp17 (1A)
f5: BZHI Gy,Ey,By (v) | PEXT Gy,By,Ey (F3),(v) | PDEP Gy,By,Ey (F2),(v)
-f6: MULX By,Gy,rDX,Ey (F2),(v)
+f6: ADCX Gy,Ey (66) | ADOX Gy,Ey (F3) | MULX By,Gy,rDX,Ey (F2),(v)
f7: BEXTR Gy,Ey,By (v) | SHLX Gy,Ey,By (66),(v) | SARX Gy,Ey,By (F3),(v) | SHRX Gy,Ey,By (F2),(v)
EndTable
@@ -861,8 +861,8 @@ EndTable
GrpTable: Grp7
0: SGDT Ms | VMCALL (001),(11B) | VMLAUNCH (010),(11B) | VMRESUME (011),(11B) | VMXOFF (100),(11B)
-1: SIDT Ms | MONITOR (000),(11B) | MWAIT (001)
-2: LGDT Ms | XGETBV (000),(11B) | XSETBV (001),(11B) | VMFUNC (100),(11B)
+1: SIDT Ms | MONITOR (000),(11B) | MWAIT (001),(11B) | CLAC (010),(11B) | STAC (011),(11B)
+2: LGDT Ms | XGETBV (000),(11B) | XSETBV (001),(11B) | VMFUNC (100),(11B) | XEND (101)(11B) | XTEST (110)(11B)
3: LIDT Ms
4: SMSW Mw/Rv
5:
@@ -880,15 +880,21 @@ EndTable
GrpTable: Grp9
1: CMPXCHG8B/16B Mq/Mdq
6: VMPTRLD Mq | VMCLEAR Mq (66) | VMXON Mq (F3) | RDRAND Rv (11B)
-7: VMPTRST Mq | VMPTRST Mq (F3)
+7: VMPTRST Mq | VMPTRST Mq (F3) | RDSEED Rv (11B)
EndTable
GrpTable: Grp10
EndTable
-GrpTable: Grp11
-# Note: the operands are given by group opcode
-0: MOV
+# Grp11A and Grp11B are expressed as Grp11 in Intel SDM
+GrpTable: Grp11A
+0: MOV Eb,Ib
+7: XABORT Ib (000),(11B)
+EndTable
+
+GrpTable: Grp11B
+0: MOV Eb,Iz
+7: XBEGIN Jz (000),(11B)
EndTable
GrpTable: Grp12
diff --git a/arch/x86/tools/gen-insn-attr-x86.awk b/arch/x86/tools/gen-insn-attr-x86.awk
index e6773dc..093a892 100644
--- a/arch/x86/tools/gen-insn-attr-x86.awk
+++ b/arch/x86/tools/gen-insn-attr-x86.awk
@@ -68,7 +68,7 @@ BEGIN {
lprefix1_expr = "\\((66|!F3)\\)"
lprefix2_expr = "\\(F3\\)"
- lprefix3_expr = "\\((F2|!F3)\\)"
+ lprefix3_expr = "\\((F2|!F3|66\\&F2)\\)"
lprefix_expr = "\\((66|F2|F3)\\)"
max_lprefix = 4
@@ -83,6 +83,8 @@ BEGIN {
prefix_num["Operand-Size"] = "INAT_PFX_OPNDSZ"
prefix_num["REPNE"] = "INAT_PFX_REPNE"
prefix_num["REP/REPE"] = "INAT_PFX_REPE"
+ prefix_num["XACQUIRE"] = "INAT_PFX_REPNE"
+ prefix_num["XRELEASE"] = "INAT_PFX_REPE"
prefix_num["LOCK"] = "INAT_PFX_LOCK"
prefix_num["SEG=CS"] = "INAT_PFX_CS"
prefix_num["SEG=DS"] = "INAT_PFX_DS"
(2013/08/07 0:16), H. Peter Anvin wrote:
> On 08/06/2013 12:37 AM, Masami Hiramatsu wrote:
>> Add TSX-NI related instructions and new instructions to
>> x86-opcode-map.txt according to the Intel(R) 64 and IA-32
>> Architectures Software Developer's Manual Vol2C (June, 2013).
>> This also includes below updates.
>> - Fix a typo of MWAIT (the lack of (11B)).
>> - Change NOP Ev to prefetchw Ev
>> - Add CRC32 new prefix style (66&F2)
>> - Add ADCX, ADOX, RDSEED, CLAC and STAC instructions
>>
>> Signed-off-by: Masami Hiramatsu <[email protected]>
>> Cc: Thomas Gleixner <[email protected]>
>> Cc: Ingo Molnar <[email protected]>
>> Cc: "H. Peter Anvin" <[email protected]>
>
> Thanks, applied!
>
> You may want to also look at:
>
> http://software.intel.com/en-us/intel-isa-extensions
>
> ... specifically ...
>
> http://software.intel.com/sites/default/files/319433-015.pdf
>
> For a *lot* of new instructions coming down the pipe.
Thanks, Andi also asked me to support them ;)
As far as I can see, that requires more work to support
EVEX decoding. And perhaps, I have to classify new instructions
and operands and introduce new abbreviations/superscripts for
them, because there is no opcode map in the above document.
Would you intel guys have any idea about new opcode map?
Thank you,
--
Masami HIRAMATSU
IT Management Research Dept. Linux Technology Center
Hitachi, Ltd., Yokohama Research Laboratory
E-mail: [email protected]
On 08/06/2013 09:51 PM, Masami Hiramatsu wrote:
>
> Thanks, Andi also asked me to support them ;)
> As far as I can see, that requires more work to support
> EVEX decoding. And perhaps, I have to classify new instructions
> and operands and introduce new abbreviations/superscripts for
> them, because there is no opcode map in the above document.
>
> Would you intel guys have any idea about new opcode map?
>
What is it you need, specifically?
-hpa
(2013/08/07 14:03), H. Peter Anvin wrote:
> On 08/06/2013 09:51 PM, Masami Hiramatsu wrote:
>>
>> Thanks, Andi also asked me to support them ;)
>> As far as I can see, that requires more work to support
>> EVEX decoding. And perhaps, I have to classify new instructions
>> and operands and introduce new abbreviations/superscripts for
>> them, because there is no opcode map in the above document.
>>
>> Would you intel guys have any idea about new opcode map?
>>
>
> What is it you need, specifically?
A new opcode map about AVX-512, which introduces ZMM registers which
has 512-bit width. Perhaps, we need to introduce new code of operand
type for that new large registers, I guess it is Octa-Quadword(oq)? :)
And also, MASK instructions may require new addressing method codes for
opmask registers. Of course I can make new one for those and update
after official opcode map is released, as I did on AVX instructions.
Thank you,
--
Masami HIRAMATSU
IT Management Research Dept. Linux Technology Center
Hitachi, Ltd., Yokohama Research Laboratory
E-mail: [email protected]