Intel AVX512 FP16 instructions use maps 3, 5 and 6. Maps 5 and 6 use 3 bits
in the EVEX.mmm field (0b101, 0b110). Map 5 is for instructions that were FP32
in map 1 (0Fxx). Map 6 is for instructions that were FP32 in map 2 (0F38xx).
There are some exceptions to this rule. Some things in map 1 (0Fxx) with imm8
operands predated our current conventions; those instructions moved to map 3.
FP32 things in map 3 (0F3Axx) found new opcodes in map3 for FP16 because map3
is very sparsely populated. Most of the FP16 instructions share opcodes and
prefix (EVEX.pp) bits with the related FP32 operations.
Intel AVX512 FP16 instructions has new displacements scaling rules, please refer
to the public software developer manual for detail information.
gas/
2021-08-05 Igor Tsimbalist <igor.v.tsimbalist@intel.com>
H.J. Lu <hongjiu.lu@intel.com>
Wei Xiao <wei3.xiao@intel.com>
Lili Cui <lili.cui@intel.com>
* config/tc-i386.c (struct Broadcast_Operation): Adjust comment.
(cpu_arch): Add .avx512_fp16.
(cpu_noarch): Add noavx512_fp16.
(pte): Add evexmap5 and evexmap6.
(build_evex_prefix): Handle EVEXMAP5 and EVEXMAP6.
(check_VecOperations): Handle {1to32}.
(check_VecOperands): Handle CheckRegNumb.
(check_word_reg): Handle Toqword.
(i386_error): Add invalid_dest_and_src_register_set.
(match_template): Handle invalid_dest_and_src_register_set.
* doc/c-i386.texi: Document avx512_fp16, noavx512_fp16.
opcodes/
2021-08-05 Igor Tsimbalist <igor.v.tsimbalist@intel.com>
H.J. Lu <hongjiu.lu@intel.com>
Wei Xiao <wei3.xiao@intel.com>
Lili Cui <lili.cui@intel.com>
* i386-dis.c (EXwScalarS): New.
(EXxh): Ditto.
(EXxhc): Ditto.
(EXxmmqh): Ditto.
(EXxmmqdh): Ditto.
(EXEvexXwb): Ditto.
(DistinctDest_Fixup): Ditto.
(enum): Add xh_mode, evex_half_bcst_xmmqh_mode, evex_half_bcst_xmmqdh_mode
and w_swap_mode.
(enum): Add PREFIX_EVEX_0F3A08_W_0, PREFIX_EVEX_0F3A0A_W_0,
PREFIX_EVEX_0F3A26, PREFIX_EVEX_0F3A27, PREFIX_EVEX_0F3A56,
PREFIX_EVEX_0F3A57, PREFIX_EVEX_0F3A66, PREFIX_EVEX_0F3A67,
PREFIX_EVEX_0F3AC2, PREFIX_EVEX_MAP5_10, PREFIX_EVEX_MAP5_11,
PREFIX_EVEX_MAP5_1D, PREFIX_EVEX_MAP5_2A, PREFIX_EVEX_MAP5_2C,
PREFIX_EVEX_MAP5_2D, PREFIX_EVEX_MAP5_2E, PREFIX_EVEX_MAP5_2F,
PREFIX_EVEX_MAP5_51, PREFIX_EVEX_MAP5_58, PREFIX_EVEX_MAP5_59,
PREFIX_EVEX_MAP5_5A_W_0, PREFIX_EVEX_MAP5_5A_W_1,
PREFIX_EVEX_MAP5_5B_W_0, PREFIX_EVEX_MAP5_5B_W_1,
PREFIX_EVEX_MAP5_5C, PREFIX_EVEX_MAP5_5D, PREFIX_EVEX_MAP5_5E,
PREFIX_EVEX_MAP5_5F, PREFIX_EVEX_MAP5_78, PREFIX_EVEX_MAP5_79,
PREFIX_EVEX_MAP5_7A, PREFIX_EVEX_MAP5_7B, PREFIX_EVEX_MAP5_7C,
PREFIX_EVEX_MAP5_7D_W_0, PREFIX_EVEX_MAP6_13, PREFIX_EVEX_MAP6_56,
PREFIX_EVEX_MAP6_57, PREFIX_EVEX_MAP6_D6, PREFIX_EVEX_MAP6_D7
(enum): Add EVEX_MAP5 and EVEX_MAP6.
(enum): Add EVEX_W_MAP5_5A, EVEX_W_MAP5_5B,
EVEX_W_MAP5_78_P_0, EVEX_W_MAP5_78_P_2, EVEX_W_MAP5_79_P_0,
EVEX_W_MAP5_79_P_2, EVEX_W_MAP5_7A_P_2, EVEX_W_MAP5_7A_P_3,
EVEX_W_MAP5_7B_P_2, EVEX_W_MAP5_7C_P_0, EVEX_W_MAP5_7C_P_2,
EVEX_W_MAP5_7D, EVEX_W_MAP6_13_P_0, EVEX_W_MAP6_13_P_2,
(get_valid_dis386): Properly handle new instructions.
(intel_operand_size): Handle new modes.
(OP_E_memory): Ditto.
(OP_EX): Ditto.
* i386-dis-evex.h: Updated for AVX512_FP16.
* i386-dis-evex-mod.h: Updated for AVX512_FP16.
* i386-dis-evex-prefix.h: Updated for AVX512_FP16.
* i386-dis-evex-reg.h : Updated for AVX512_FP16.
* i386-dis-evex-w.h : Updated for AVX512_FP16.
* i386-gen.c (cpu_flag_init): Add CPU_AVX512_FP16_FLAGS,
and CPU_ANY_AVX512_FP16_FLAGS. Update CPU_ANY_AVX512F_FLAGS
and CPU_ANY_AVX512BW_FLAGS.
(cpu_flags): Add CpuAVX512_FP16.
(opcode_modifiers): Add DistinctDest.
* i386-opc.h (enum): (AVX512_FP16): New.
(i386_opcode_modifier): Add reqdistinctreg.
(i386_cpu_flags): Add cpuavx512_fp16.
(EVEXMAP5): Defined as a macro.
(EVEXMAP6): Ditto.
* i386-opc.tbl: Add Intel AVX512_FP16 instructions.
* i386-init.h: Regenerated.
* i386-tbl.h: Ditto.
This way not only the overall (source) table size shrinks by quite a
bit and the risk of related templates going out of sync with one another
gets lowered, but also (dis)similarities between neighboring templates
become easier to spot.
Note that for certain SSE2AVX templates this results in benign attribute
changes:
- LDMXCSR and STMXCSR: NoAVX gets set,
- MOVMSKPS, PMOVMSKB, PEXTR{B,W} (register destination), and PINSR{B,W}
(register source): IgnoreSize and NoRex64 get set,
- CVT{DQ,PS}2PD, CVTSD2SS, MOVMSKPD, MOVDDUP, PMOV{S,Z}X{BW,WD,DQ}, and
ROUNDSD: NoRex64 gets set,
- CVTSS2SD, INSERTPS, PEXTRW (memory destination), PINSRW (memory
source), and PMOV{S,Z}X{BD,WQ,BQ}: IgnoreSize gets set.
Similarly the "normal" (non-SSE2AVX)
- non-64-bit CVTS{I,S}2SD forms get NoRex64 set,
- CMP{EQ,ORD,NEQ,UNORD}{P,S}{S,D} forms get C set,
all again in a benign way.
The remaining differences in the generated table are due to re-ordering
of entries in the course of being folded into templates.
The table entries are more natural to read (and slightly shorter) when
the prefixes, like is the case for VEX/XOP/EVEX-encoded entries, are
specified as part of the opcode. This is particularly noticable for
side-by-side legacy and SSE2AVX entries.
An implication is that we now need to use "unsigned long long" for the
initially parsed opcode in i386-gen. I don't expect this to be an issue.
Just like is already done for VEX/XOP/EVEX encoded insns, record the
encoding space information in the respective opcode modifier field. Do
this again without changing the source table, but rather by deriving the
values from their existing source representation.
In the majority of cases we can easily determine the length from the
encoding, irrespective of whether a prefix is specified there as well.
We further don't even need to record the value in the table entries, as
it's easy enough to determine it (without any guesswork, unless an insn
with major opcode 00 appeared that requires a 2nd opcode byte to be
specified explicitly) when installing the chosen template for further
processing.
Should an encoding appear which
- has a major opcode byte of 66, F3, or F2,
- requires a 2nd opcode byte to be specified explicitly,
- doesn't have a mandatory prefix
we'd need to convert all templates presently encoding a mandatory prefix
this way to the Prefix_0X<nn> model to eliminate the respective guessing
i386-gen does.
Just like is already done for legacy encoded insns, record the mandatory
prefix information in the respective opcode modifier field. Do this
without changing the source table, but rather by deriving the values from
their existing source representation.
This is in preparation of opcode_length going away as a field in the
templates. Identify pseudo prefixes by a base opcode of zero instead:
No real prefix has an opcode of zero. This at the same time allows
dropping a curious special case from i386-gen.
Since most attributes are identical for all pseudo prefixes, take the
opportunity and also template them.
Commit 8b65b8953a ("x86: Remove the prefix byte from non-VEX/EVEX
base_opcode") used the opcodeprefix field for two distinct purposes. In
preparation of having VEX/XOP/EVEX and non-VEX templates become similar
in the representatioon of both encoding space and opcode prefixes, split
the field to have a separate one holding an insn's opcode space.
RepPrefixOk, HLEPrefixOk, and NoTrackPrefixOk can't be specified
together, so can share an enum-like field. IsLockable can be inferred
from HLE setting and hence only needs specifying when neither of them
is present.
Having this count explicitly in the table is redundant and (even if just
slightly) disturbs clarity. Infer the count from the number of operands
actually found.
Also convert the "no operands" indicator from "{ 0 }" to just "{}", as
that (now) ends up being easier to parse.
Rename VexOpcode to OpcodePrefix so that OpcodePrefix can be used for
regular encoding prefix.
gas/
* config/tc-i386.c (build_vex_prefix): Replace vexopcode with
opcodeprefix.
(build_evex_prefix): Likewise.
(is_any_vex_encoding): Don't check vexopcode.
(output_insn): Handle opcodeprefix.
opcodes/
* i386-gen.c (opcode_modifiers): Replace VexOpcode with
OpcodePrefix.
* i386-opc.h (VexOpcode): Renamed to ...
(OpcodePrefix): This.
(PREFIX_NONE): New.
(PREFIX_0X66): Likewise.
(PREFIX_0XF2): Likewise.
(PREFIX_0XF3): Likewise.
* i386-opc.tbl (Prefix_0X66): New.
(Prefix_0XF2): Likewise.
(Prefix_0XF3): Likewise.
Replace VexOpcode= with OpcodePrefix=. Use Prefix_0X66 on xorpd.
Use Prefix_0XF3 on cvtdq2pd. Use Prefix_0XF2 on cvtpd2dq.
* i386-tbl.h: Regenerated.
We check register-only source operand to decide if two source operands of
VEX encoded instructions should be swapped. But source operands in AMX
instructions with two source operands swapped are all register-only
operand. Add SwapSources to indicate two source operands should be
swapped.
gas/
* config/tc-i386.c (build_modrm_byte): Check vexswapsources to
swap two source operands.
opcodes/
* i386-gen.c (opcode_modifiers): Add VexSwapSources.
* i386-opc.h (VexSwapSources): New.
(i386_opcode_modifier): Add vexswapsources.
* i386-opc.tbl: Add VexSwapSources to BMI2 and BMI instructions
with two source operands swapped.
* i386-tbl.h: Regenerated.
Rename VecSIB to SIB to support Intel Advanced Matrix Extensions which
introduces instructions with a mandatory SIB byte which isn't a vector
SIB (VSIB).
gas/
* config/tc-i386.c (check_VecOperands): Replace vecsib with sib.
Replace VecSIB128, VecSIB256 and VecSIB512 with VECSIB128,
VECSIB256 and VECSIB512, respectively.
(build_modrm_byte): Replace vecsib with sib.
opcodes/
* i386-gen.c (opcode_modifiers): Replace VecSIB with SIB.
(VecSIB128): Renamed to ...
(VECSIB128): This.
(VecSIB256): Renamed to ...
(VECSIB256): This.
(VecSIB512): Renamed to ...
(VECSIB512): This.
(VecSIB): Renamed to ...
(SIB): This.
(i386_opcode_modifier): Replace vecsib with sib.
* i386-opc.tbl (VexSIB128): New.
(VecSIB256): Likewise.
(VecSIB512): Likewise.
Replace VecSIB=1, VecSIB=2 and VecSIB=3 with VexSIB128, VecSIB256
and VecSIB512, respectively.
In order to reduce redundancy as well as the chance of things going out
of sync (see a later patch for an example), make the opcode table
generator capable of recognizing and expanding templated templates. Use
the new capability for compacting the general purpose conditional insns.
It is almost entirely redundant with Size64, and the sole case (CRC32)
where direct replacement isn't possible can easily be taken care of in
another way.
AMD ABM has 2 instructions: popcnt and lzcnt. ABM CPUID feature bit has
been reused for lzcnt and a POPCNT CPUID feature bit is added for popcnt
which used to be the part of SSE4.2. This patch removes CpuABM and adds
CpuPOPCNT. It changes ABM to enable both lzcnt and popcnt, changes SSE4.2
to also enable popcnt.
gas/
* config/tc-i386.c (cpu_arch): Add .popcnt.
* doc/c-i386.texi: Remove abm and .abm. Add popcnt and .popcnt.
Add a tab before @samp{.sse4a}.
opcodes/
* i386-gen.c (cpu_flag_init): Replace CpuABM with
CpuLZCNT|CpuPOPCNT. Add CpuPOPCNT to CPU_SSE4_2_FLAGS. Add
CPU_POPCNT_FLAGS.
(cpu_flags): Remove CpuABM. Add CpuPOPCNT.
* i386-opc.h (CpuABM): Removed.
(CpuPOPCNT): New.
(i386_cpu_flags): Remove cpuabm. Add cpupopcnt.
* i386-opc.tbl: Replace CpuABM|CpuSSE4_2 with CpuPOPCNT on
popcnt. Remove CpuABM from lzcnt.
* i386-init.h: Regenerated.
* i386-tbl.h: Likewise.
Since SSE3 is independent of SSE4a, don't disable SSE3 when disabling
SSE4a.
* i386-gen.c (cpu_flag_init): Remove CPU_ANY_SSE3_FLAGS from
CPU_ANY_SSE4A_FLAGS.
commit 7deea9aad8 changed nosse4 to include CpuSSE4a. But AMD SSE4a is
a superset of SSE3 and Intel SSE4 is a superset of SSSE3. Disable Intel
SSE4 shouldn't disable AMD SSE4a. This patch restores nosse4. It also
adds .sse4a and nosse4a.
gas/
* config/tc-i386.c (cpu_arch): Add .sse4a and nosse4a. Restore
nosse4.
* doc/c-i386.texi: Document sse4a and nosse4a.
opcodes/
* i386-gen.c (cpu_flag_init): Add CPU_ANY_SSE4A_FLAGS. Remove
CPU_ANY_SSE4_FLAGS.
Commit d835a58baa disabled sysenter/sysenter in 64-bit mode by
default. By default, assembler should accept common, Intel64 only
and AMD64 ISAs since there are no conflicts.
gas/
PR gas/25516
* config/tc-i386.c (intel64): Renamed to ...
(isa64): This.
(match_template): Accept Intel64 only instruction by default.
(i386_displacement): Updated.
(md_parse_option): Updated.
* c-i386.texi: Update -mamd64/-mintel64 documentation.
* testsuite/gas/i386/i386.exp: Run x86-64-sysenter. Pass
-mamd64 to x86-64-sysenter-amd.
* testsuite/gas/i386/x86-64-sysenter.d: New file.
opcodes/
PR gas/25516
* i386-gen.c (opcode_modifiers): Replace AMD64 and Intel64
with ISA64.
* i386-opc.h (AMD64): Removed.
(Intel64): Likewose.
(AMD64): New.
(INTEL64): Likewise.
(INTEL64ONLY): Likewise.
(i386_opcode_modifier): Replace amd64 and intel64 with isa64.
* i386-opc.tbl (Amd64): New.
(Intel64): Likewise.
(Intel64Only): Likewise.
Replace AMD64 with Amd64. Update sysenter/sysenter with
Cpu64 and Intel64Only. Remove AMD64 from sysenter/sysenter.
* i386-tbl.h: Regenerated.
In memory operand addressing, which forms of displacement are permitted
besides Disp8 is pretty clearly limited
- outside of 64-bit mode, Disp16 or Disp32 only, depending on address
size (MPX being special in not allowing Disp16),
- in 64-bit mode, Disp32s or Disp64 without address size override, and
solely Disp32 with one.
Adjust assembler and i386-gen to match this, observing that templates
already get adjusted before trying to match them against input depending
on the presence of an address size prefix.
This adjustment logic gets extended to all cases, as certain DispNN
values should also be dropped when there's no such prefix. In fact
behavior of the assembler, perhaps besides the exact diagnostics wording,
should not differ between there being templates applicable to 64-bit and
non-64-bit at the same time, or there being fully separate sets of
templates, with their DispNN settings already reduced accordingly.
This adjustment logic further gets guarded such that there wouldn't be
and Disp<N> conversion based on address size prefix when this prefix
doesn't control the width of the displacement (on branches other than
absolute ones).
These adjustments then also allow folding two MOV templates, which had
been split between 64-bit and non-64-bits variants so far.
Once in this area also
- drop the bogus DispNN from JumpByte templates, leaving just the
correct Disp8 there (compensated by i386_finalize_displacement()
now setting Disp8 on their operands),
- add the missing Disp32S to XBEGIN.
Note that the changes make it necessary to temporarily mark a test as
XFAIL; this will get taken care of by a subsequent patch. The failing
parts are entirely bogus and will get replaced.
... instead of an operand one. Which operand it applies to can be
determined from other operand properties, but as it turns out the only
place it is actually used at doesn't even need further qualification.
EsSeg (a per-operand bit) is used with IsString (a per-insn attribute)
only. Extend the attribute to 2 bits, thus allowing to encode
- not a string insn,
- string insn with neither operand requiring use of %es:,
- string insn with 1st operand requiring use of %es:,
- string insn with 2nd operand requiring use of %es:,
which covers all possible cases, allowing to drop EsSeg.
The (transient) need to comment out the OTUnused #define did uncover an
oversight in the earlier OTMax -> OTNum conversion, which is being taken
care of here.
Drop the remaining instances left in place by commit c3949f432f ("x86:
limit ImmExt abuse), now that we have a way to specify specific GPRs.
Take the opportunity and also introduce proper 16-bit forms of
applicable SVME insns as well as 1-operand forms of CLZERO.
Special register "class" instances can't be combined with one another
(neither in templates nor in register entries), and hence it is not a
good use of resources (memory as well as execution time) to represent
them as individual bits of a bit field.
Furthermore the generalization becoming possible will allow
improvements to the handling of insns accepting only individual
registers as their operands.
