Fairly self-explanatory, we can leverage the existing IR functions for
the purpose of these instructions.
In the integer case, we can just insert function pointers
into an array and index it, given all comparison primitives exist
already for the integer side of things.
Fairly self-explanatory, we can leverage the existing IR functions for
the purpose of these instructions.
In the integer case, we can just insert function pointers
into an array and index it, given all comparison primitives exist
already for the integer side of things.
Here we increase the similarity between the A64 and A32 front-ends in terms of their
page_table handling code. In this commit, we:
* Reserve and use r14 as a register to store the page_table pointer.
* Align the code to be more similar in structure.
* Add a conf member to A32EmitContext.
* Remove scratch argument from EmitVAddrLookup.
* Use pext/pdep where not previously used
* Limit pext/pdep to non-AMD platforms due to slowness on AMD
* Use imul/and as alternatives for AMD and non-BMI2 platforms
* VFP instructions were not getting fuzzed due to matching coprocessor instructions (as invalid instructions)
* Fix VPOP writeback for doubles when (imm8 & 1) == 1
* Do not accidentally fuzz unimplemented unconditional instructions
The guest program often accesses the NZCV flags directly much less
often than we need to use them for jumps and other such uses.
Therefore, we store our flags in cpsr_nzcv in a x64-friendly format.
This allows for a reduction in conditional jump related code.
Several issues:
1. Several terminal instructions did not stop at the end of a single-step block
2. x64 backend for the A32 frontend sometimes polluted upper_location_descriptor with the single-stepping flag
We also introduce the enable_optimizations parameter to the A32 frontend.
9bdd1596c Update version
d151562bd Fix punctuation in changelog
346500e70 Fix gcc version check
a434a8f77 Update changelog
9eb47d951 Fix markup
51c58a56b Bump version
3fc33f627 Update changelog
2e32db5b9 Update changelog
c1ce6e01f Update changelog
1c3c80dc1 Update changelog
d1d653d89 Implement the L specifier
73c843748 Follow naming conventions
e588b02b1 Fix posix-mock-test
1a62711d0 Reduce binary size
5b0288158 Merge branch 'master' of github.com:fmtlib/fmt
a133187a8 Update changelog
80ce222ca Fix wide print overload (#1609)
770a94ede Use FMT_THROW macro where applicable
2864e8432 Update readme and add compatibility option
21a295c27 Undo comment change
96c68afe6 Fix -Wsign-conversion warnings
664dd88e3 Enable FMT_STRING() use with types other than string literals
69779b4ed Fix handling of small precision in general format
01a172c96 Add .vs to .gitignore
08ca40ea9 Detect /utf-8 in MSVC
dd97f4920 Improve exception safety in dynamic_format_arg_store
295116948 Move FMT_USE_FLOAT and friends to fmt/format.h
d3e668418 Allow disabling floating point support (#1590)
52d0e1bbe Don't use properties when setting FMT_LIB_NAME
5d32ccfc3 Add back missing OUTPUT_NAME in target properties. (#1598)
3cf619de5 Simplify dynamic_format_arg_store
2559983e7 Color formatting fixed for wide strings (fixes issue #1594) (#1596)
026f99178 Simplify dynamic store
9f70fc3e7 Minor tweaks for dynamic_format_arg_store
6012dc9ab Dynamic arguments storage. Implementation of enhancement from issue #1170. (#1584)
85050aa2e Ability to join elements of std::initializer_list was added
ff486a72a Allow leading zeros in precision (#1579)
678341275 Deprecate fmt::char8_t
6f01b6ebb Fix a typo in CMake config: STRINGS -> STRING
61c5a5160 Fix handling of empty tuples (#1588)
02bfd8a9a Add FMT_HAS_CPP14_ATTRIBUTE / FMT_HAS_CPP17_ATTRIBUTE to test for language-specific attributes.
3c24052cf Workaround 'cannot call member function without object' error on gcc 4.9
f72a905eb Fix handling of volatile enums
941d5e147 Workaround broken fallthrough attribute in the PGI compiler (#1583)
ee2b828b9 Tweak a comment
5bb885665 Workaround for broken [[deprecated]] in PGI compiler (#1581)
1c0c59d4a Fix empty debug postfix
b1adaa988 Remove gcc 4.4 workaround
48e8d0ebe set_doc -> set_verbose
ce0097915 Cleanup CMake config
db4a6cfbf is_static_compiled_format -> is_compiled_format
29a1ea795 Fix clang -Wdisabled-macro-expansion warning from FMT_STRING_IMPL.
8a06ca84c Fix ambiguous overloads of format & format_to
153f753bd Deprecate undocumented _u suffix
eafd07986 Improve width computation
0c6919ec7 Make FMT_DEBUG_POSTFIX a cache variable (#1566)
197a5c372 Apply clang-format
68742e1d8 Fix clang -Wsign-conversion warning in grisu_count_digits. (#1573)
1e8493196 Make compile-time checks in format_to handle references
58e6c84f5 Fix simple -Wsign-conversion cases. (#1571)
75a4525e5 Move FMT_CLANG_VERSION definition to core.h (#1568)
6ccb2e241 Add FMT_NORETURN to assert_fail prototype. (#1569)
bed134a4a Tentative fix for default template param in friend error
b2d3a86ec Make FMT_ASSERT work in constexpr on clang 4.0.1
13d82e32b Don't use internal GTest API
2161a73f2 Fix FMT_FORMAT_AS const specifier position (#1554)
e00997b00 improved use of find (#1560)
0415cf235 add const begin and end overload to buffer (#1553)
3bafd0749 Fix to_string docs
f733882b5 Remove misleading FMT_USE_WINDOWS_H
dc22360c3 Workaround broken UDL templates in GCC < 6.4
1f1b50707 Make formatter override implicit conversion to a C string
24924128e Fix a link error in gcc8 (#1548)
c54cd7180 only modify CMAKE_RUNTIME_OUTPUT_DIRECTORY if it is not already set
43e9b29e5 Only use compiler features if available
b55ea5870 string_view::char_type -> value_type (#1539)
4098970db Update README.rst
314e15001 Fix symbol visibility on Linux when compiling with -fvisibility=hidden (#1535)
f499b393d Apply coding conventions
6c30f4144 Configure fmt.pc library name correctly.
1acb73f97 Fix formatting std::chrono::duration types to wide strings (#1533)
09a13244c Disallow passing non-string-literals to FMT_STRING
419db8baa Fix length computation of constexpr C strings
9fc4161f5 fix interal compiler error when building with mingw
25d6916b3 Fix so can work without locale defined
0b2eb6501 Add locale example
fd1cabe46 Workaround a bogus MSVC warning
a844d7ab8 Add namespaces
47d396809 Add more examples
7800173eb Update fill docs
b4218aa0f Test invalid fill
8a3a8177d Bump version
e5f2f8ce7 Add variable-width fill support (#1109)
75765bfad Avoid unnecessary unsigned overflows (#1515)
9bd9738da Remove static and simplify names
bd5f903f2 Add a locale example
06e437fd9 Move docs to the proper place
1bd4f54fa update format
11cc2903e re-fix url link
b124e3e8e fix url link
ffd5f3469 Correct display format
0f0e5ddf5 Add vcpkg installation instructions
1f110702a Remove redundant braces
4ccbe4b5f Avoid namespace clash for fmt
40638a75b Use C++11 compatible std::is_same operations
c8dd9cc99 Use type_identity to block unnecessary template argument deduction (thanks Tim Song)
4bbe57ceb Work-around for nvcc
55b613005 Use C++11-compatible operations
ae3ea156e Fix for older versions of intel compiler
77165fdf8 Use FMT_NOEXCEPT instead of noexcept directly
65ac626c5 Improve join docs
cd0b3f969 check if _SECURE_SCL is defined not equal to 0
cef1e4354 Optimize grisu_gen_digits
0201c8db2 Restructure float_format
9e3f3e8cf Fix handling of output iterators in format_to_n (#1506)
aa07c5765 Move vprint_mojibake to the internal namespace
a73d89e9c Catch invalid uses of fmt::arg
cb8e7caf7 Convert 'char8_t' character sequences to 'char' sequences
b3fd0005d Suppress a bogus -Wdouble-promotion warning
7b478f9de Simplify example
c85efef31 More showpoint fixes and tests (#1498)
455a7c078 Clarify lifetime of basic_format_args
674c326d7 Update syntax.rst
061a9897f Update syntax.rst
d2d1c9c56 warning C4267: 'argument': conversion from 'size_t' to 'DWORD', possible loss of data
b6e19e595 Update apidoc
f219dcd59 Add fmt::bytes
dea7fde8b Deprecate u8string_view
5390e29d4 Enable mojibake
9f6434dcd Improve UTF-8 handling on Windows
dac9a7f99 Improve UTF-8 handling on Windows
3ca9533f3 Flatten forward
7eec036d9 Improve UTF-8 support
e6b37b4af Handle block boundaries in utf8_to_utf16
8cf4c5206 Apply clang-format
74532c23a Make type a scoped enum
b308159be Make round_direction a scoped enum
162995fed Add os.h to docs
8b41362a0 Add trailing decimal point if # is specified (#1476)
1b1c70108 trailing_zeros -> showpoint
d7e72a09e Simplify FMT_STRING_IMPL
2201890d7 Apply clang-format and update inclusion guards
6100ed4bb Eliminate NVCC NVidia compiler emits unreachable code warnings
1afe201ae Handle block boundaries in utf8_to_utf16
cd2b99032 Chore(readme): use https (#1481)
9acf89fef Mitigate MSVC issue with min/max macros (#1480)
9ea42fb26 Rename posix-test to os-test
da2569827 posix.cc -> os.cc
35959a31d Move OS-specific APIs to a separate header
ec2463c90 Implement utf8_to_utf16 using utf8_decode
0012917f6 Add a UTF-8 decoder
9e450911f Give an error on precision overflow
068d20bc3 Avoid shadowing warnings in FMT_STRING
a99fbe67b Apply a typo fix retroactively
adbed11ed Fix a typo
8ab1c5c6e Squelch MSVC warning exporting subclasses of runtime_error (fix for PR #1433) (#1470)
a770009fc Improve error reporting
598e6042d warning C4468: 'fallthrough': attribute must be followed by a case label or a default label
e09814dc9 Merge branch 'master' of github.com:fmtlib/fmt
b272fb360 Extend FMT_FALLTHROUGH compatibily to gcc and clang pre-C++17 (#1469)
f94b7364b Update version
7abec071b Update changelog
b7eb8c892 Prepare for the next release
ae7c50185 Reintroduce sprintf_format for ABI compatibility
9f2e7edae Fix handling of types convertible to std::string_view
fd52de0c6 Add FMT_CUDA_TEST CMake option to enable cuda-test
f675cb887 Remove redundant cast
73a16b827 Fix handling of int128_t in format-impl-test (#1461)
72879db40 Clean-up sign-conversion warnings in public headers
d3aa0c3a2 Clean-up sign-conversion warnings in test code
31de9a1b8 Revert "Clean-up sign-conversion warnings in test code"
227bfe62d Clean-up sign-conversion warnings in test code
95dfdc6cc Update README.rst
5916ff63c Update README.rst
1ab80aa92 Fix handling of types with custom formatters that are convertible to std::string_view
4f4d87661 Remove '%' from the docs
f443bd3ba Ditch decimal_formatter (#1363)
1219b65f2 Relax fallthrough attribute detection
071794ec6 Update version
d22e4ad85 Remove trailing comma
983806b0c Update changelog
02af5beb8 Bump version and update changelog
123e7f7fc Revert #1433 because of build failures (#1450)
168460f02 Remove TYPES
a64f60c84 Remove unneeded FMT_API.
1a599117d Export assert_fail with FMT_API. This fixes dll build.
b160123e3 Update ChangeLog.rst
598158856 Fix compilation with MinGW
8bbe76af3 Add a missing decimal point in exponent notation with trailing zeros
4ca6821e8 Update version
7111a1eb9 Bump version
ae00bbdc9 Update changelog
e71e07d9f Update changlog
0184df702 Update docs
1cbae6e9b Put vprint declarations in one place
159f89e2b Fixing installation directory of '*.dll' files on Windows
4b120b68a Clean up includes
186b225d9 Update changlog
4cbf4888e Update changelog
e31f2b3d0 Update changelog
62da1db62 Avoid wchar_t instantiations
3bc28fcc6 Squelch MSVC warning exporting subclasses of runtime_error
3c05fa46c Update changelog
ba6e330fd digits -> num_bits
6037b3cae Fix dangling else problem in FMT_ASSERT
fafb03fa6 Fix handling of fallback_uintptr
2f9acd183 Remove dependency on <cassert>
aaf829bfb Fix fallback pointer formatting on big endian, take 2
b994a0ab1 Fix handling of missing fraction in snprintf_float
bb205d940 Fix fallback pointer formatting on big endian
ef7369ce9 Update docs
40e4c227d Update changelog
ea54b21e7 Remove invalid noexcept annotation
9cbf4b087 Fix -Wconversion warnings
1200a34e1 Update changelog
9c7e2a6c6 Add missing newline
34e921f6f Update docs
c3be0f593 Refactor floating-point formatting
c68703c9f float_spec -> float_specs
9a21728b0 Remove gen_digits_params
3de36e934 Enable -Wswitch-enum in CI
4afb39bc2 Update README.rst
7ffa62db1 Fix precision handling in snprintf_float
0d07db123 Fix handling of streamable and convertible to string types
d19ed6716 Fix hexfloat buffer reallocation
99b6e928d Fix handling of types with deleted rvalue conversion to string (#1421)
57cd3f72e Update comment
111fc127f Remove fp::operator-
6003ec3f2 Simplify Grisu implementation
8877a6772 Instantiate snprintf_float
75fff1db6 Minor cleanup
28d7191c2 Don't print trailing zero with fixed, precision=0, and showpoint (#1417)
43271ba8e Handle null terminator at the end of the buffer
63a9f3fcd fix bad oss fuzz link in the oss-fuzz badge
4cf59ce73 Integrate Grisu and sprintf digit generators
7395472dd Refactor floating-point formatting
9108b25da Merge branch 'float'
4d366c68b Merge branch 'master' of github.com:fmtlib/fmt
ded1e7679 Refactor floating point formatting
c7edd8e57 Cleanup FP formatting
75108a56f Don't print % for nan and inf
3e1f70fe0 Merge write_fp into write
125fc5e52 Update comment
6793ffc1d Update README.rst
f4fcc5fd2 Update README.rst
4de41aa65 Move basic_writer::write_fp to where it belongs
404a880bd Make parse_arg_id more readable
092d2dc7b Merge safe-duration-cast.h into chrono.h
093e55421 Remove redundant qualification
d0696b0aa warning C4456: declaration of 'num_digits' hides previous local declaration
66d7746bb Use grisu for exponent notation
e9bff7881 Don't parse % unless FMT_DEPRECATED_PERCENT is set
57b6f2966 Deprecate the fmt macro
d79493e5e Remove Grisu2
78842ce0d test: add default constructor for a const value
5420bcce2 Make % an opt-in to improve compatibility with std::format
56a2e2075 Refactor float spec parsing
ed117baa4 Replace bool with float_format and add exponential
f26446290 Move float_spec_handler to internal namespace and update asserts
7e1cb3237 Fix indentation
f67783d7e Clarify that numeric alignment is deprecated
1c6d85f7b Apply coding conventions to examples
4a1da44f9 Apply coding conventions to examples
080b6899d Tweak the docs
c01ec54fd Document and clean basic_format_parse_context
b0c2ab93f Bump version
9b7fe2a4a Don't use POSIX API on UWP
c58b7d9c2 Use overridden locale in ostream
ceff9b0b2 Tweak the docs
3dc8639f8 [docs] Added conda
dcde089b4 Improve POSIX API detection
2145a7bdc Move has_formatter into the public fmt namespace. (#1407)
52ae134f8 Remove broken CI config
0d6dd0cc6 Correct basic_string_view from string ctor
1f918159e [clang-tidy] Replace deprecated C headers
6868f888b [clang-tidy] Add missing override
87cd545a1 [clang-tidy] Replace {} with = default
12f9437e2 [clang-tidy] Use auto
bb0c8bfea [clang-tidy] Add noexcept where move is used
e6e829890 [clang-tidy] Add parentheses to macro arguments
0f0848e4f [clang-tidy] Use braced init list
a1fb5c733 [clang-tidy] Changes suffixes to uppercase
8a411c2bc [clang-tidy] Turn deleted function to public
0047dc10a Mark apidoc as rst
263cdef8a Merge branch 'master' of github.com:fmtlib/fmt
d4ca54253 Update docs
5bb7b28e1 Document members
1409dfe76 Try fix CI
f1559e1d5 Use grouping() from locale for specifier 'n'
ffd05e65e basic_parse_context -> basic_format_parse_context per standard and document
0889856d6 Fix UTF-8 truncation
d6eede9e0 Remove redundant ctor
213e09644 Workaround X11 madness (#1388)
6bfc9af8c Add double support to compile
3487f1b9c Always inline grisu_gen_digits and disable grisu2 by default
791294d17 Apply get_cached_power optimization by jk-jeon
8e700619b Simplify format_handler
58c6f8c7f Make unsigned-integer-overflow sanitizer happy (#1377)
40414b344 Don't emit trailing zeros in exponential notation (#1376)
b7a157401 Simplify grisu_writer
7aa58c30b Simplify NVCC checks
8e9bffa98 clang-format
ce4d87acd Remove obsolete comment and clang-format
21acc2af4 Fix more Visual Studio 2019 pedantic warnings (#1371)
00669427d Patch compiler error when building using nvcc
d39ebf3ff Optimize counting
6498bc6d3 Simplify grisu_writer
a967dcbe2 Improve handling of signs
8498bc97d Initialize all the things
e2ea94067 Handle assymetric boundaries
2bc5585ff Fix computing lower boundaries for smallest normalized double
bb728a572 packed_arg_bitsize -> packed_arg_bits and remove packed_arg_mask
36d1390e6 Implement round half to even
599e0aef4 Support single precision floats in grisu formatting
91f7619cc Fix Visual Studio 2019 pedantic warning C4334: '<<': result of 32-bit shift implicitly converted to 64 bits (was 64-bit shift intended?)
c4dc6bef2 Apply clang-format
646966e97 Reduce bigint capacity
a5abe5d95 Handle negative exponent and nonnegative power
1cbc5fa6c Handle negative exponent and rename value/pow10 to numerator/denominator
f7a5748fd Partially implement (FPP)^2
0e94b931a Fix a linkage error introduced by #1360 (#1362)
5e58eb97b Implement add_compare
3a15ea3ea Rename write_double to write_fp
b87ac4d84 Distinguish float from double
a927dda9b Use words for packed constants
dd11d4584 Encode types using 5 bits
b55551f90 Implement more comparison operators
96f91428c Add defaulted copy and move operations to format_error and system_error (#1347)
b732f28c0 Deduplicate color vformat and vprint
a82c1dc6d use memory_buffer to make color print behave atomic #1348 (#1351)
2730e9018 Fix compile error in printf with gcc9 (#1354)
e4d6d9d7c Implement divmod
a1079e9fd Fix undefined in format-test (#1349)
b66bb6b71 Fix undefined in core-test and printf-test (#1345)
b60114533 Implement more bigint operations
c41cea8b1 Initial implementation of square
0c7650373 Fix handling of types convertible to std::string_view
057101370 Repoint one more Python 2 link to Python 3
d2c9276fc let README point to python 3 instead of 2
0fc7bd157 Fix ambiguity for types with dodgy conversions
b4f1988c4 Provide overload for `fmt::join` that handles `std::tuple`s
4b8f8fac9 Update README.rst
3b2fc033d Making CUDA test work with CMAKE_MSVC_RUNTIME_LIBRARY
ac59d9f3a workaround VS2019 Update 3 compiler bug. (#1328)
8f27ce4d8 add test for multiple compilation types
89b0c71fa fix name clash in header-only mode
f6a783ad2 fix `get`
ccc8f5db0 Disable integral operator<< (#1316)
20fdb88a1 Remove redundant and nonportable test (#1313)
f29901097 Don't use const char* overload of operator<< (#1309)
758446c80 Fix a warning (#1319) and simplify code
f7aedc5fc Fix shared build on Solaris
840a817ed add oss fuzz badge
79c923ba2 fmt.pc.in: Fix for cross compilation
5dc577c06 Update ChangeLog.rst
c6d1a94a9 Fix fallback_format (#1306)
0656045d0 Fix format overload that takes text_style (#1305)
c85ae23c7 Add max_value
b3bf66576 Implement multiplication and part of assignment from pow of 10
0887887e2 Implement left shift
ac71d853b Refactor normalize and clean up
6649b8e0c value -> bigit
56b5c192a Add a bigint stub and reenable grisu
b2f0b6e44 Tweak comment
3d9f3c163 Integrate new format string compilation
19547d514 Update index.rst
972ffd315 Update README.rst
2ed412fa3 Update README.rst
df4dcf2ec Fixed vs 2015 warning about unused variable when grisu is off
f7a4b4ab9 Make numeric alignment optional
611cf0b3c Format octal 0 as 0
58a8f2f53 Add ccache to the list of projects
1882b9687 Reduce the numer of ifdefs with an empty (u)int128_t fallback
6de0454b4 Add support for built-in __int128 when available
16e3c48bb Move definition of FMT_USE_INT128 to core.h
8ce5f680f Rename internal::is_integral and internal::is_arithmetic
2fd8f9ec8 Initial implementation of optimal compile-time formatter generation
fe642d764 Clarify use of the core API in header-only mode (#1296)
a128b5b2c Simplify format string compilation
466128de0 Remove unused code and refactor
22e98a5b6 Make compile work with user-defined types
f18a3f36a Remove string_view_metadata
7cad33563 Refactor format string compilation
e1ab6bc00 Simplify format string compilation
24a88545d Add -Werror to tests
422e7b9d7 Fix compile-time checks for user-defined types (#1292)
d1dd9d532 Document floating-point n specifier (#1291)
9a56a608e Fix warnings (#1288)
c76957565 FixedEnum -> StrongEnum and make it a regression test
200ee6f10 Fix minor clang-tidy warnings
bcd9b9331 Map not int enum to correct underlying_type (#1286)
345ba07f1 Add a CUDA test
9e2490be4 Rename precision parameter
3f75e2b69 Make buffer_range public and update custom formatting docs (#1281)
744302add Workaround broken [[deprecated]] in Intel compiler (#1273)
f5556225a Eliminate shadow variable warning
ad3c7855e Fix typos.
7512a55aa Update version
9bd2f1f9a Temporarily disable Grisu
b9815cf04 Simplify part_counter
fe00cddde Move part counter to the namespace scope
1a7d172dc which_value -> kind
006c2546f Get rid of basic_prepared_format
4ce006fb6 Simplify format string compilation
e2e557e27 Remove preparator
a5f470eb1 Test and fix compiled format_to_n and formatted_size
4070c1d80 PrepareTest -> CompileTest
f8b59251c Remove unused preparator
643aa4c8c Simplify format string compilation
ffe29a92f Get rid of internal::compile
e94d72366 Fix compiled format string version of format_to_n
a9337395a Remove unused variable
562a9f499 Remove parts_container_concept_check
b257c56e5 Fix size types
1488df339 Bump version
b918e3ff8 Fix docopt usage
23b04ca6d Update usage instructions
5d5918a74 Update usage instructions
413d97b33 Fix formatting in changelog
790fd90f4 Update changelog
c0890be77 Update changelog
5a4b24613 Update the default floating-point format docs (#1270)
45dc3486f Update license
5a353fa8c Draft license
c84d227da Update changelog
b2d4ca154 Workaround an issue with std::filesystem::path being an infinitely deep range (#1268)
2aae6b120 Fix doc for importing header only target with CMake
9f09b8eed Fix a warning
2bb8120d9 Fix a warning
c1e97392b Fix warnings
4e99e09bb Fix warnings
1607a0187 Suppress a warning
e8219952c Restrict fmt::compile to literal strings
544b53733 Update changelog
83c13a1fe Update changelog
2546dafec Update changelog
02c04b173 Update changelog
5360ab0b5 Fix iteration over named arguments (#1168)
b615eca96 Update changelog
22a42c0b0 Cleanup compile implementation
c63624ed4 Update changelog
211d31240 Don't terminate on I/O errors in report_error
ef55e74e0 Update changelog
612669d34 Update changelog
ba63ac8c1 Make format_to a non-member
3fe49163b Revert "Make format_to a non-member"
a5bd3ddb2 Make format_to a non-member
3df0ea34e Remove unnecessary overloads
436acf348 Make formatted_size & format_to_n non-members
823128049 Make format a non-member
1dfa2591b Make operator _a constexpr
6f2dd30e3 Simplify udl_formatter return type
431d0f85c Use basic_string_view in non-template UDL
ad71f5a70 Use forwarding references in UDL template
d0f2f3b81 Update usage.rst
eac2796ce Fix locale detection
ee3625746 Internalize format_part
bbf0bada3 Simplify compile
3730b4f03 Cleanup compile implementation
25ff2efc0 Update changelog
9393fe26f prepare -> compile
b48ffc14a Update changelog
3268531bc Workaround a broken C locale on Windows
8bd59ec93 Use fputws for outputting wide strings
1235f0a24 Fix typo in usage.rst
4fb73d114 Update README.rst
f9ceefb0f Fix a warning and is_negative usage
6bcc3fd21 Fix warnings
6a497e1d0 Fix warnings
e9b9b0aef clang-format
ec24342b2 Fix more warnings
230b24944 Fix sign conversion warnings
cadd92d69 Fix#1232.
41076d1de Use `FMT_THROW` in fmt/color.h.
d561cb47a Opt-in macro to enforce use of compile-time format strings
699fe8e71 Remove const qualification in compile-time checks
da2d33f1f Update changelog
af4734fd1 Fix warnings
a3a74672a Update signatures
c11e68305 Update signatures
dd8cc8b0b Disallow passing views as lvalues
f6f0415b8 typedef -> using
c92dc3746 typedef -> using
e3f20d3e1 Remove gcc 4.4 workaround and use proper alias templates (#940)
b43881232 Clean up is_output_iterator
bc15e3700 typedef -> using
cf5ebf205 Cleanup
9c20e72de Cleanup
79209598f core_format_specs -> sprintf_specs
5488d0b53 spec -> specs
e4f84ee1c Refactor format_specs for #1109 and #940
8e0dcd20b Remove old deprecated functions
1d3e3d8c0 Make the 'n' format specifier work with grisu disabled
bc628f8d4 Fix EXPECT_DEBUG_DEATH_IF_SUPPORTED (#1214)
edd13fcc1 Fix small number rounding with fixed precision in grisu
6a031347e Remove unnecessary qualification
bc14c6ee2 Use the decimal point from locale
476f25cd8 Remove ancient gcc workaround
c9d5a08ed Add the 'n' float format specifier
f487ddbdf thousands_sep -> add_thousands_sep
d8fd1699b Make data names follow naming conventions
c286ffc88 int_traits -> uint32_or_64_t
1289782f0 Get rid of add_thousands_sep
2249f5571 Simplify thousands separator handling and cleanup
e76446958 Cleanup and remove deprecated visit
de37de912 Remove deprecated color API
df1a3a141 Remove null_terminating_iterator
78dec87a4 typedef -> using
ded0a3bb3 Internalize undocumented basic_writer
83174f2a1 Refactor ranges
c2e84ee9c Fix FormatTest.StrError on Solaris
2711cb167 Apply clang format and other minor formatting tweaks
ab0ba8a9d Don't emit decimal point if there are no trailing digits (#1210)
bd3fd3bfd clang-format
260c11590 Fix formatting of 0.0 with (#1210)
9d97201ed add oss-fuzz support
037b84f21 Fix warnings
e37ee419c Export exceptions destructors
e3488fcae Fix Warning
d5d586561 Move strtod_l check to CMake since it's very system-specific
29ef7d31e Fix a warning
af83192d7 Fix warnings
6952732b6 Remove misplaced checks
b97e5d8c2 Remove deprecated fmt/time.h
572b077db Fix warnings
72e519a4b Add formatter<std::string_view>
635e01fe7 Update <format> to the current wording
0e72c9804 Make undocumented output_range internal
f13906f40 back_insert_range<internal::buffer> -> buffer_range
22ddd4b98 Simplify feature checks
bb827341e Add FMT_FUNC where necessary
64c54703d Cleanup
1c3197b8d Fix warnings
fd2292f13 Cleanup checked iterators
4912cff65 Fix handling of mapped types in compile checks (#1200)
463984383 Cleanup
e29708ee5 uintptr -> fallback_uintptr to avoid confusion with uintptr_t
f03a6c532 Cleanup
6c3d584e6 Simplify warning suppression
12f468388 Make iterator_t an alias template
874d6727e Remove workarounds for pre-C++11 compilers
a9940192f Fix warnings
34b541835 Fix warnings
cbbee1b38 Fix handling of hexfloat
92a44db11 Fix warnings
d05d42751 Remove old msvc workaround from arg_formatter_base and fix warning
d32fe0f3f Fix hadling of nullptr
e5422db4b Fix handling of uintptr_t
9d7b64a25 More cleanup
5e293bd97 Remove unnecessary qualification
4a502d980 Add remove_reference_t
d384cdd39 Make is_contiguous_back_insert_iterator internal
32544b610 More cleanup
39f522a13 get_types -> encode_types
d7d2bebf9 Remove redundant typename
9427f15be Fixed issue with formatting to an array of chars
87fbc6f75 get_type -> mapped_type_constant
a48daa60e Remove remove_volatile
afdbbac75 Cleanup
e33fe14f5 Simplify visit
e895da2ec Clean up value
e1a67b528 Remove TODO
a291f07e1 Clean up argument mapping
5d9100fa2 Move char8_t to fmt/core.h
4faadff0a Add preliminary user-defined type support
5d4873359 Clean up value construction
0f0b42861 Cleanup
209db68b2 Get rid of FMT_CONSTEXPR11
cb4c59495 Deprecate convert_to_int
40779749a is_formattable -> has_formatter
b3cf8613b Make formatter specializations override implicit conversions
3fdba0492 Reduce the number of nontrivial formatter instantiations
f5f3ffac5 Merge string make_value overloads
a38b99a18 Fix a regression in named argument handling introduced by prepare
aa31028b2 char_t_impl -> char_t
0787d6974 Simplify SFINAE
bae00aa8d Simplify is_string
388bb389e integral_constant -> bool_constant
7e39c7e6f Put stringy stuff together
3eff8f94d Update docs
469a4bbd3 Use enable_if_t
d2ee5f240 Merge tests
c264e641e Add conditional_t for pre-C++14
4aa0dc578 Fix docs
064ce6b6c Specialize is_char for character types
7893d8539 Clarify why we cannot have nice things
5bafcb437 Add comments about things broken in MSVC
153024255 Fix signature in the docs
eddb84cfc Fix formatting of exotic characters
7e42c65bb Document a more useful to_string_view overload
1e6e87cb7 Update docs
0c6a6e025 Get rid of the FILE* hack and reword apidocs
1653244c6 Fix compilation issue on VS2019 (#1186) (#1191)
d54e64b3c Make buffer_context an alias template
ec6651087 Remove old is_constructible workarounds and replace typedefs with using
4d4b8c238 FMT_CHAR -> char_t
56d2b9110 Install git
89d6c959b Fix cmake link
76ef39fc5 Update vagrant config
af2c73772 Implement parsing of string_views
9df0e2d1f Implement string parsing
5b7bbf885 Revert "FMT_CHAR -> char_t"
4c650057a FMT_CHAR -> char_t
2833c76f2 Move char_t to template param to reduce symbol size
67feef558 Make enable_if_t more std-like and move to fmt namespace
78daa50ff Fix handling of chrono durations with minimal signed rep
87e4ea290 Fix a warning
c56b17029 Add msvc2019 to CI
c929684e3 inlines count_code_points(basic_string_view<char8_t)
f57227a14 FMT_ENABLE_IF -> enable_if_t
634f707f2 Simplify char_t and remove msvc2013 from CI
406e632bd result_of -> invoke_result_t
49f78a427 Demacrify
637bf3c6d Workaround a bug in clang-format
8302c2f33 fmt::internal::declval -> std::declval
d07cc2026 FMT_EXPLICIT -> explicit, FMT_NULL -> nullptr
4a7966c77 Drop gcc 4.4 from CI
9b3c24b99 One weird trick to simplify docs; doxygen hates it
30bce6c14 Fix a few chrono formatting corner cases (#1178)
e5512c5d5 Use static_assert instead of SFINAE in arg(...)
afc571aed Document join and relax its compiler requirements
ad360a62b add gcc 8 c++17 release build
3cf12d7b1 add gcc 8 c++17 build
bb254d146 Disable std-format-test by default
291ba837f Remove wrong compile-time checks from printf (#1173)
01c631af9 Implement unsigned and long long parsing
4fcd4a4bd Reuse parse_format_string in scan
2346779d6 More documentation fixes
afc1a74a6 Minor documentation fixes
ad0eade47 Remove MSVC nonsense
28c187bcd Capitalize titles
91bb3aaf0 fmtlib.net -> fmt.dev
5e7bdf1b9 Clean up vagrant config
570453f27 Add a vagrant config for testing gcc 4.4
ef6282fc4 Fix gcc 4.4 build
e3e470bb6 Remove deprecated format_decimal
67179dbc2 Remove deprecated format_decimal
a5ffa735d Fix gcc 4.4 build
5ee080463 Experiment with scan API
25b72fc4c Move <format> to tests not to confuse users
d179ec5f8 Simplify Grisu
2a9e8b52d Fix advance_to() and begin() using iterator (#1159)
2c77562b1 Fix ambiguous formatter specialization in fmt/ranges.h (#1123)
98b377529 Add support for exotic string_view iterators (#1156)
b488df6cf Fix Grisu3 stopping conditions
f4dfd6e30 Suppress all clang-target-msvc test warning in CMake and other misc fixes (#1151)
a6e8ed15c Disable UDL templates on GCC 9 by default (#1148)
de5da5091 Fix formatting of extreme durations (#1154)
ea2976e6d Move internal::uintptr_t test to format-impl-test (#1152)
77d6036cd Fix unexpected trailing decimal point (#1153)
ccc318e80 Update README.rst
4c8efd694 Update README.rst
d22d11b5f Update README.rst
e9bab6d02 Improve handling of large durations
f52c09f92 Fix format_to_n docs in 5.1.0
118d8bccc Fix compilation error under MSVC 19.21 (#1140)
6828d549e Add FMT_ENABLE_IF_T
3fd134be0 Move test_count_digits to format-impl-test and disable gtest warnings (#1147)
ca7c1f89d Fixed a compile error under MSVC.
ca978b3d2 Fix handling of nan durations
c1d430e61 Improve handling of negative durations
38a85502e Use the same rep type for seconds to prevent overflow
241414028 Eliminate shadowed variable warnings from gcc-7.2
29c10fbf6 Fix DLL visibility of explicit instantiation "declaration" of internal::basic_data<void> in header format.h and the explicit instantiation "definition" in format.cc (#1134)
4a4d72f91 Fix handling of invalid string in chrono formatting
b3cc9c056 Merge remote-tracking branch 'upstream/master' into invalidcolons
2e3352fd0 provoke assertion
4c721e3a2 Fix chrono formatting with invalid argument id (#1132)
8d8ea21c6 Partially implement Grisu3
40a797564 Remove trailing zeros
cb46397df Fix typo
134904c88 Re-enabled constexpr tests
bd516e342 Convert negative precision to zero in printf (#1127)
5efb24dd2 Add specialization test
946498cfb Fix handling of zero precision
6b2086391 fmt::ptr: support unique_ptr and shared_ptr.
d306585a3 Don't inject internal names into std (#1120)
544b92793 Don't detect C compiler
dc94010fa Remove char_traits (#1117)
397e8dd9d clang-format
2b415b7af Restructure printf_arg_formatter to make it customizable
5d755d0a4 Fix handling of volatile char (#1115)
bade46aae Optimize grisu using uint128_t
41fbaeb3b Add <format> test
8bc0adb9b Get rid of obsolete cmake stuff
1763d0e7a Add MongoDB to "Projects using this library"
f569c1ba2 doc fix: time -> chrono
ccd70f59e Workaround bogus unreachable warnings in MSVC
a4969ebe0 Link to fmt(...) docs from index
a6ad29aa3 Update <format>
52eb3fe27 Update <format>
09e2ac5e4 Update <format>
df4ea0c76 Update <format>
718f60acc Fix shadowing warning (#1105) and clang format
aeb5ad3ce Enable [[noreturn]] some.
280839548 basic_buffer -> buffer
6e37c2003 Use compile features cxx_auto_type cxx_variadic_templates instead of cxx_std_11
3de3d76a3 Add compile features for cmake 3.8+
07d5a86a7 Fix warnings
ab1474ef6 Workaround segfault in doxygen and apply clang-format
918ab77ba Try fixing doc build
735b1fadc Only update key on Linux
3c531b735 Fix expired key error when installing with apt
f10a7e2e4 Test exotic pointer formatting
bd8177177 Add support for platforms without uintptr_t
0302927f5 Optimize pointer formatting
bb6842ba3 Simplify to_string
b23c8633f Detect presence of uintptr_t
b588d7f35 Fix a couple of deprecated things.
018d8b57f Remove broken snprintf
1987db663 clang-format
a6d1ad741 handle fwrite results #1098
e979c782d Extend basic_writer with write() method for doubles with optional format_specs argument
9e1531c1e install pkg-config file into libdir
0a66e4cbb Update README.rst
91acfe685 Fix UBSAN warning
b7e6bf967 clang format
da0ea4161 Make compile-time checks work with fallback formatter (#1088)
7ad3015f5 Added missing typename to FMT_STRING. (#1089)
f0b572da0 Update wording test
6d416cf67 Forward declare is_string<FILE*> specializations
b742f622a Create PR template to help agree to the licensing terms (#1083)
02a6f16b5 Update LICENSE.rst
294fd7df9 Remove isinf workaround
17c6900f8 Update docs
0faa968cc Make floating-point formatting locale-independent
bc784d362 Remove isnan workaround
53379dfd0 Don't set CMAKE_BUILD_TYPE if fmt is a subproject (#1081)
76d326a2a Enable grisu for general format
c21c6b8c4 Move enable_if to template params
ec645ca26 Update readme and doxygen config
ae0b0dab9 Remove obsolete TODO
5466a5b41 Document inherited members
9b392a683 Update readme
0fa65cf32 Add example
fdd0149e7 Update readme
e19a95b27 Update readme
ef3927497 Update readme
a7f68dcc0 Update readme
1428b3429 Update readme
97619e27a More fixed precision tests
dd6cc0e6a Merge time-test into chrono-test
a939c7595 Merge fmt/time.h into fmt/chrono.h
17e4b5392 Make chrono formatting work without exceptions (#1062)
a82b3680d More tests and fixed precision fixes
287342dab cmake: default FMT_PKGCONFIG_DIR to a relative path
e28429ee7 Prevent overflow with zero precision
49bbf3c87 Simplify shadowing warning fix
5e5506f83 Update readme
e06523361 Visual Studio 2017: warning C4456: declaration of 'result' hides previous local declaration
ebec00138 Reword licensing part
8daa3c683 Clarify contribution licensing
0d418a8d5 Update and rename CONTRIBUTING.rst to CONTRIBUTING.md
4c66dad8c Refactor digit generation
b1f7cca89 stop -> handler and swap args
f90d33ca1 error_ulp -> error
78c755dc4 Update readme
f23017015 Update readme
ced8aa8c9 Update readme
77d54251f Update benchmark results
1632f72cb Test get_round_direction
8129b9bc4 Test that the library can be compiled witout locales
835087dd0 Add color-test
4523053e6 Fix typos.
8407f4cb2 Round close to zero fixed precision
49d244c06 Don't emit more than precision digits (#1072)
3466d9c84 Don't override fixed formatting depending on exponent (#1072)
93d22dec3 Implement rounding up and clean up FP formatting
d560ddac2 Temporarily disable Grisu for fixed formatting
2d981bb13 Add documentation for '%' format type. (#1071)
d8434baa0 Declare the size of RESET_COLOR.
8f7780a4f Correct comment
2e526a664 Fix handling of output iterator in ranges
79b79f329 Add support for '%' type to output floating point values as a percentage.
287eaab3b Increment output iterator in basic_writer::write for character types (#1056)
a97757736 Use grisu for fixed precision
327d4b6e9 Fixes for some pedantic warnings (#1054)
8af651be3 Implement fixed precision
187bd1b8b Clarify lifetimes of named_arg parameters (#1051)
4e5694fd0 Update arg signature
82c24edcf Workaround a bogus Qt Creator warning
ed138d794 test: assert-test: fix typo in else-branch of EXPECT_DEBUG_DEATH_IF_SUPPORTED
0476a51cb Add Sublime syntax (#1037)
c5aad69f2 Restore deprecated begin()
5b0006476 Make stopping condition configurable in grisu
a44238f2e Improve grisu
83808076e Minor cleanup
9660ea1bf Simplify format string checks
4a9d67636 Gradle 4.10 >> Gradle 5.2
f041f128f Minor cleanup
e4572e5de Update std implementation
442fa1bd4 Decouple format and parse contexts
744e66bb0 Deprecate format_context::parse_context()
d231d68a8 Fix handling of custom context
01f34d0b0 Fix library deprecation warning
9a0a24f90 Test is_streamable with overloaded comma operators
430e6ac9b Protect against overloaded comma operators in decltype
467520e7a Remove unused macro
5a314a528 Eliminate extra copy on floating-point formatting
9989e7f4e Update benchmark results
31510cb43 Fix warnings from Visual Studio
61c9b563c Replace 'std::result_of' by 'std::invoke_result' where possible (#1025)
864b9a220 Correct the comment
153833683 Remove unnecessary checks
22de5a755 Fix warnings from Travis
355eb6d29 Enable grisu for shortest roundtrip (default) formatting
b8d34e0db Fix rounding
e61cac687 Minor grisu improvements
7fbbfed8c Fix warnings caused by usage of deprecated functionality
c3268f4e5 Remove use cases of deprecated functionality
34951f199 Replace comments regarding deprecation with attributes
3f52336e6 Simplify formatter selection
7ca8fc3b1 Fix a bogus warning
5289dd600 Test formatting of special numbers
bf6529f2c Update docs
070061224 Implement 'chrono' formatting specifiers '%Q' and '%q'
06c005b7b Clarify that compile-time checks don't support named arguments
4f6fda558 Add a grisu test stub
9f70b034e Implement precision for floating-point durations.
7cdb1e5e4 Workaround broken is_default_constructible in MSVC
5f1ceebc7 Make formatter<T> override ostream<< for templates (#952)
1b11b000c Update readme
83f052930 Add code from p0645
fdd8e333c Fix compilation with locales disabled (#1011)
dad1eec84 Workaround unimplemented T... on gcc 4.6.x (#1008)
b0cde860a Implement 'snprintf(OutputIt it, size_t n, const S &format, const Args & ... args)' (#917)
e05dfb088 Fix compile errors due to name-hiding and an unused function argument.
dde095fab Revert "Reorder defines"
4a059914a Reorder defines
58b6f8db4 Format the code using clang-format
9a777b9e1 Implemented fmt::prepare()
da55e96f5 Install ninja to fix android build
dc8f8ce4c Fix handling of dynamic width in chrono formatter
f5cc77cea Get rid of 'null_terminating_iterator' in printf.h (#980)
39623a740 Replaced usage of gtest's internal scoped_ptr with unique_ptr.
ae1de3a8d Add support for using text_style in format and vformat directly (#993) * Closes#993
1b8a216dd Improve docs
b3ad759a0 Bump version and correct changelog
git-subtree-dir: externals/fmt
git-subtree-split: 9bdd1596cef1b57b9556f8bef32dc4a32322ef3e
Instead of looking up the page table like:
table[addr >> 12][addr & 0xFFF]
We can use a global offset on the table to query the memory like:
table[addr >> 12][addr]
This saves two instructions on *every* memory access within the recompiler.
Original change by degasus in A64 emitter
Instead of looking up the page table like:
table[addr >> 12][addr & 0xFFF]
We can use a global offset on the table to query the memory like:
table[addr >> 12][addr]
This saves two instructions on *every* memory access within the recompiler.
Thanks at skmp for the idea.
Also solves a performance regression initially introduced by b6e8297e369f2dc4758bafe944e51efb8d1a2552,
primarily due to excessively mismatched load/store sizes causing less than optimal load-to-store forwarding.
Removes unnecessary header dependencies that have accumulated over time
as changes have been made. Lessens the amount of files that need to be
rebuilt when the headers change.
Most of the time when this occurs, it's a bug. Thankfully this isn't the
case. However, we can resolve these cases to make the codebase more
consistent.
This is pretty gross, but until DYNARMIC_NO_BUNDLED_FMT is eliminated,
this fixes the use of it in existing libraries or applications making
use of dynarmic.
UserConfig instances contain up to 16 std::shared_ptr<Coprocessor>
instances. We can std::move here to avoid performing 16 redundant atomic
reference increment and decrement operations.
Mostly inconsequential on x64, but we may as well signify intent.
Nothing in particular prevents these from being constexpr. Do so to make
them consistent with the bulk of other functions in this header that are
constexpr.
Negation in (standard IEEE) floating-point is simply flipping the sign-bit, so this
operation will never be more complex than what is presented here, making
constexpr a reasonable allowance.
The result of GetArg() is returned by value, so this is essentially
still a copy. While the previous code *is* valid, this communicates what
is actually happening a little more explicitly.
Ensures that the header dependency is always satisfied directly, and not
through other project headers. While we're at it, we can qualify the
call with the std:: namespace.
Given the same argument is used inside the condition's body if it's
true, we can just utilize the local to cut out a GetArg() operation.
Avoids redundant internal assertion checking.
Ensures the alternate half-precision state is preserved within the
location descriptors, which will be necessary when implementing the
half-precision extensions for VFP and NEON.
Removes a boost header from the public includes in favor of using the
standard-provided std::variant.
The use of boost in public interfaces is often a dealbreaker for some
people. Given we use std::optional in the header already, we can
transition over to std::variant from boost::variant.
With this removal, this makes all of our dependencies internal to the
library itself.
Similar to the variant within the ARM-mode translator visitor. This will
be used in subsequent changes to implement the hint instructions
introduced in ARMv7.
Now that we fuzz against Unicorn, we aren't just restricted to VFPv2.
VFPv3 and VFPv4 facilities can now be implemented. This renames
constructs mentioning VFPv2 to just refer to VFP.
Implements the ARM-mode variants of the CRC32 instructions introduced
within ARMv8. This is also one of the instruction cases where there is
UNPREDICTABLE behavior that is constrained (we must do one of the
options indicated by the reference manual).
In both documented cases of constrained unpredictable behavior, we treat
the instructions as unpredictable in order to allow library users to
hook the unpredictable exception to provide the intended behavior they
desire.
We also make the arrays static here, as MSVC tends to load the whole
array every time the function is called, instead of storing the data
within rodata.
This also line breaks the elements a little earlier for readability.
With deduction guides, we can eliminate the need to explicitly size the
array. Also newlines the elements based off their relation, making it
slightly nicer to read.
Replaces type aliases of raw integral types with the more type-safe Imm
template, like how the AArch64 frontend has been using it.
This makes the two frontends more consistent with one another.
Avoids potentially dumping boost, fmt, and xbyak targets into a
top-level namespace without any qualification, which can lead to build
errors in projects that already make use of them.
The SetRegister() IR function doesn't allow specifying the PC as a
register. This is a discrepancy that slipped through (my bad). Instead,
we can use BranchWritePC(), like how the other similar PC modifying
locations do it.
This'll reduce the amount of noise necessary in changes implementing
half-precision instructions, as the type can just be prepended to the
switch cases, instead of rewriting the whole if/else branch.
Provides basic implementations of the barrier instruction introduced
within ARMv7. Currently these simply mirror the behavior of the AArch64
equivalents.
Performs a similar tidying up of the Thumb translator, like what was
done with the regular ARM translator to make it consistent with the rest
of the codebase.
The A32 backend (both Thumb and ARM), will likely see more changes to it
in the near future, so this just acts as a "dusting off".
Now that the constructor and destructors have been placed within the cpp
file, we can forward declare the memory pool data structures. Now, a
change to the memory pool code won't ripple across the entirety of the
IR emitter.
While initially done to potentially prevent creating bugs due to C++
having a silly type-promotion mechanism involving types < sizeof(int)
and unsignedness, given that the bulk of these functions' usages
are on exit paths, these can return the correct type to avoid the need
to cast at every usage point.
Prevents potentially inlining allocation code everywhere. While we're at
it, also explicitly delete/default the copy/move constructor/assignment
operators to be explicit about them.
Makes bracing consistent, and variables const where applicable to be
consistent with the rest of the codebase.
In most bracing cases, they'd need to be added to conditionals that
would involve checking stack pointer alignment in the future anyways.
Generally hash functions shouldn't throw exceptions. It's also a
requirement for the standard library-provided hash functions to not
throw exceptions.
An exception to this rule is made for user-defined specializations,
however we can just be consistent with the standard library on this to
allow it to play nicer with it.
While we're at it, we can also make the std::less specializations
noexcpet as well, since they also can't throw.
To a reader this might look particularly strange, given the function
itself has a void return value, but this is actually valid, given the
function in the return statement also has a void return value.
This instead alters it to be a little easier to parse and potentially be
a little less confusing at a glance.
These are cases that are defined as reserved within the ARMv8 reference
manual, so we can handle them as such instead of as unallocated
encodings.
While this doesn't actually change emulated behavior, it does at least
allow the JIT to generate the more appropriate exception.
Strictly speaking, immh being zero is defined as reserved in the ARMv8
reference manual. This was just an error on my part when introducing the
SQSHL immediate scalar variant.
