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
HelpText<"Disable compaction pass in finalizer. This pass is used to decide whether to use a compacted, i.e. shorter encoding of machine instructions wherever ISA allows.">;
36
38
37
39
def emit_debug : PlainFlag<"g">,
38
40
HelpText<"Enable generation of debug information and enables kernel debug">;
HelpText<"Allows an application to specify that single precision floating-point divide (x/y and 1/x) and sqrt used in the program source are correctly rounded.">;
HelpText<"-ze-opt-regular-grf-kernel <string> tells IGC to use regular GRF size if kernel name contains <string >regardless of module-level options.">;
HelpText<"When this flag is present, it indicates that any OpenCL buffers can be more than 4GB in size. If it is absent, all buffers are not more than 4GB in size.">;
HelpText<"Controls how single precision and double precision denormalized numbers are handled. If specified as a build option, the single precision denormalized numbers may be flushed to zero; double precision denormalized numbers may also be flushed to zero if the optional extension for double precision is supported">;
76
86
77
-
def strict_aliasing : ClFlag<"strict-aliasing">;
87
+
def strict_aliasing : ClFlag<"strict-aliasing">,
88
+
HelpText<"Allows the compiler to assume the strictest aliasing rules.">;
78
89
79
-
def mad_enable : ClFlag<"mad-enable">;
90
+
def mad_enable : ClFlag<"mad-enable">,
91
+
HelpText<"Allow a * b + c to be replaced by a mad instruction.">;
80
92
81
-
def no_signed_zeros : ClFlag<"no-signed-zeros">;
93
+
def no_signed_zeros : ClFlag<"no-signed-zeros">,
94
+
HelpText<"Allow optimizations for floating-point arithmetic that ignore the signedness of zero.">;
HelpText<"Allow optimizations for floating-point arithmetic that (a) assume that arguments and results are valid, (b) may violate the IEEE 754 standard, (c) assume relaxed OpenCL numerical compliance requirements as defined in the unsafe math optimization section of the OpenCL C or OpenCL SPIR-V Environment specifications, and (d) may violate edge case behavior in the OpenCL C or OpenCL SPIR-V Environment specifications.">;
HelpText<"Allow optimizations for floating-point arithmetic that assume that arguments and results are not NaNs, +Inf, -Inf. This option may violate the OpenCL numerical compliance requirements for single precision and double precision floating-point, as well as edge case behavior.">;
HelpText<"Sets the optimization options -cl-finite-math-only and -cl-unsafe-math-optimizations. This option causes the preprocessor macro __FAST_RELAXED_MATH__ to be defined in the OpenCL program.">;
88
104
89
105
def match_sincospi : ClFlag<"match-sincospi">;
90
106
91
-
def oclfe_w : PlainFlag<"w">;
107
+
def oclfe_w : PlainFlag<"w">,
108
+
HelpText<"Remove all warning messages.">;
92
109
93
-
def oclfe_werror : PlainFlag<"Werror">;
110
+
def oclfe_werror : PlainFlag<"Werror">,
111
+
HelpText<"Treat every warning as an error.">;
94
112
95
-
def kernel_arg_info : ClFlag<"kernel-arg-info">;
113
+
def kernel_arg_info : ClFlag<"kernel-arg-info">,
114
+
HelpText<"Allow the compiler to store information about the arguments of a kernel(s) in the program executable. The argument information stored includes the argument name, its type, the address space and access qualifiers used.">;
96
115
97
-
def oclfe_x : PlainSeparate<"x">;
116
+
def oclfe_x : PlainSeparate<"x">,
117
+
HelpText<"Manualy provide type of file. Takes only spir or spir64 as argument.">;
98
118
99
-
def oclfe_D : PlainJoinedOrSeparate<"D">;
119
+
def oclfe_D : PlainJoinedOrSeparate<"D">,
120
+
HelpText<"Manually define macros.">;
100
121
101
-
def oclfe_I : PlainJoinedOrSeparate<"I">;
122
+
def oclfe_I : PlainJoinedOrSeparate<"I">,
123
+
HelpText<"Add directory to the list of directories which will be searched for header files."> ;
HelpText<"OpenCL's global IDs are assumed to start from the origin at global offsets (offset_x, offset_y, offset_z) When this flag is present, it indicates that the global offsets are (0,0,0).">;
HelpText<"When this flag is present, it indicates that any OpenCL buffers can be more than 4GB in size. If it is absent, all buffers are not more than 4GB in size.">;
HelpText<"This flag, together with *[-cl-intel|-ze-opt]-greater-than-4GB-buffer-required* is used to convert stateless memory accesses, called messages or load/store, into stateful ones. The OpenCL runtime can create a surface whose base is either *buffer_base* or *buffer_base + buffer_offset*, based on whether *buffer_offset* is used.">;
HelpText<"For any load and store (aka message) whose address = *ptrArg + offset*, where *ptrArg* is a kernel pointer argument, offset is assumed to be non-negative if this flag is present.">;
HelpText<"Overrides the current number of threads value defined by the user's command line option for the entire module or the compiler choice by heuristics.">;
HelpText<"When this flag is present, it guarantees that size of private memory allocated per thread can not be less then the given value. Constraint: <SIZE> >= 0.">;
HelpText<"When this flag is present, it guarantees that size of scratch space private memory allocated per thread can not be less then the given value. Constraint: <SIZE> >= 0.">;
0 commit comments