You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
@@ -30,6 +30,8 @@ In Visual Studio 2019 version 16.8 and later versions, the **`/fpcvt`** compiler
30
30
31
31
For Visual Studio 2019, the default behavior for x64 targets is consistent with **`/fpcvt:BC`** unless **`/arch:AVX512`** is specified. Usually, the behavior for x86 targets is consistent with **`/fpcvt:IA`**, except under **`/arch:IA32`**, **`/arch:SSE`**, or sometimes where the result of a function call is directly converted to an unsigned integer. Use of **`/fpcvt`** overrides the default, so all conversions are handled consistently on either target. The behavior of conversions for ARM and ARM64 targets isn't consistent with either **`/fpcvt:BC`** or **`/fpcvt:IA`**.
32
32
33
+
Starting with Visual Studio 2022, the default behavior for x64 targets is consistent with `/fpcvt:BC` even when `/arch:AVX512` is specified. To use AVX512 conversion behavior with `/arch:AVX512`, specify `/fpcvt:IA`.
34
+
33
35
Standard C++ specifies that if a truncated floating-point value is exactly representable in an integer type, it must have that value when converted to that type. Otherwise, any behavior at all is allowed. Both **`/fpcvt`** options conform with Standard C++. The only difference is in what values are returned for invalid source values.
34
36
35
37
The **`/fpcvt:IA`** option causes any invalid conversion to return a single *sentinel* value, which is the destination value farthest from zero. For conversion to signed types, the sentinel is the minimum value for that type. Unsigned types use the maximum value. Floating-point operations may return a Not-a-Number (NaN) value to indicate an invalid operation. That indicator isn't an option for conversion to integer types, which don't have NaN values. The sentinel is used as a proxy for a NaN value, although it can also be the result of a valid conversion.
| `/MT` or `/MTd` | *`clang_rt.asan_dynamic-{arch}`*, *`clang_rt.asan_static_runtime_thunk-{arch}`* | *`clang_rt.asan_dynamic-{arch}`*
132
-
| `/MD` or `/MDd` | *`clang_rt.asan_dynamic-{arch}`*, *`clang_rt.asan_dynamic_runtime_thunk-{arch}`* | *`clang_rt.asan_dynamic-{arch}`*
131
+
| `/MT` or `/MTd` | *`clang_rt.asan_dynamic-{arch}.lib`*, *`/wholearchive:clang_rt.asan_static_runtime_thunk-{arch}.lib`* | *`clang_rt.asan_dynamic-{arch}.dll`*
132
+
| `/MD` or `/MDd` | *`clang_rt.asan_dynamic-{arch}.lib`*, *`/wholearchive:clang_rt.asan_dynamic_runtime_thunk-{arch}.lib`* | *`clang_rt.asan_dynamic-{arch}.dll`*
133
133
134
-
The linker option [`/INFERASANLIBS:NO`](../build/reference/inferasanlibs.md) prevents the linker from linking a *`clang_rt.asan*`* library file from the default location. Add the library path in your build scripts if you use this option. Otherwise, the linker reports an unresolved external symbol error.
134
+
The linker option [`/INFERASANLIBS:NO`](../build/reference/inferasanlibs.md) prevents the linker from linking a *`clang_rt.asan*`* library file from the default location. Add the library path in your build scripts if you use this option. Otherwise, the linker reports an unresolved external symbol error. The runtime thunk libraries **must** be linked with the `/wholearchive` option applied.
135
135
136
136
**Previous Versions**
137
137
@@ -140,12 +140,12 @@ Prior to Visual Studio 17.7 Preview 3, statically linked (**`/MT`** or **`/MTd`*
0 commit comments