diff options
| -rw-r--r-- | docs/LangRef.html | 134 |
1 files changed, 123 insertions, 11 deletions
diff --git a/docs/LangRef.html b/docs/LangRef.html index 35ad94d..cfc1c7d 100644 --- a/docs/LangRef.html +++ b/docs/LangRef.html @@ -3970,7 +3970,7 @@ IfUnequal: <h5>Syntax:</h5> <pre> - <result> = fadd <ty> <op1>, <op2> <i>; yields {ty}:result</i> + <result> = fadd [fast-math flags]* <ty> <op1>, <op2> <i>; yields {ty}:result</i> </pre> <h5>Overview:</h5> @@ -3982,7 +3982,28 @@ IfUnequal: floating point values. Both arguments must have identical types.</p> <h5>Semantics:</h5> -<p>The value produced is the floating point sum of the two operands.</p> +<p>The value produced is the floating point sum of the two operands. This + instruction can also take any number of fast-math flags, which are + optimization hints to enable otherwise unsafe floating point + optimizations:</p> +<ol> + + <li><tt>nnan</tt>: No NaNs - Allow optimizations to assume the arguments and + result are not NaN. Such optimizations are required to retain defined behavior + over NaNs, but the value of the result is undefined.</li> + + <li><tt>ninf</tt>: No Inf - Allow optimizations to assume the arguments and + result are not +/-Inf. Such optimizations are required to retain defined + behavior over +/-Inf, but the value of the result is undefined.</li> + + <li><tt>nsz</tt>: No Signed Zeros: Allow optimizations to treat the + sign of a zero argument or result as insignificant. </li> + + <li><tt>fast</tt>: Allow algebraically equivalent transformations that may + dramatically change results in floating point (e.g. reassociate). This flag + implies all the others.</li> + +</ol> <h5>Example:</h5> <pre> @@ -4052,7 +4073,7 @@ IfUnequal: <h5>Syntax:</h5> <pre> - <result> = fsub <ty> <op1>, <op2> <i>; yields {ty}:result</i> + <result> = fsub [fast-math flags]* <ty> <op1>, <op2> <i>; yields {ty}:result</i> </pre> <h5>Overview:</h5> @@ -4069,7 +4090,28 @@ IfUnequal: floating point values. Both arguments must have identical types.</p> <h5>Semantics:</h5> -<p>The value produced is the floating point difference of the two operands.</p> +<p>The value produced is the floating point difference of the two operands. This + instruction can also take any number of fast-math flags, which are + optimization hints to enable otherwise unsafe floating point + optimizations:</p> +<ol> + + <li><tt>nnan</tt>: No NaNs - Allow optimizations to assume the arguments and + result are not NaN. Such optimizations are required to retain defined behavior + over NaNs, but the value of the result is undefined.</li> + + <li><tt>ninf</tt>: No Inf - Allow optimizations to assume the arguments and + result are not +/-Inf. Such optimizations are required to retain defined + behavior over +/-Inf, but the value of the result is undefined.</li> + + <li><tt>nsz</tt>: No Signed Zeros: Allow optimizations to treat the + sign of a zero argument or result as insignificant. </li> + + <li><tt>fast</tt>: Allow algebraically equivalent transformations that may + dramatically change results in floating point (e.g. reassociate). This flag + implies all the others.</li> + +</ol> <h5>Example:</h5> <pre> @@ -4138,7 +4180,7 @@ IfUnequal: <h5>Syntax:</h5> <pre> - <result> = fmul <ty> <op1>, <op2> <i>; yields {ty}:result</i> + <result> = fmul [fast-math flags]* <ty> <op1>, <op2> <i>; yields {ty}:result</i> </pre> <h5>Overview:</h5> @@ -4150,7 +4192,28 @@ IfUnequal: floating point values. Both arguments must have identical types.</p> <h5>Semantics:</h5> -<p>The value produced is the floating point product of the two operands.</p> +<p>The value produced is the floating point product of the two operands. This + instruction can also take any number of fast-math flags, which are + optimization hints to enable otherwise unsafe floating point + optimizations:</p> +<ol> + + <li><tt>nnan</tt>: No NaNs - Allow optimizations to assume the arguments and + result are not NaN. Such optimizations are required to retain defined behavior + over NaNs, but the value of the result is undefined.</li> + + <li><tt>ninf</tt>: No Inf - Allow optimizations to assume the arguments and + result are not +/-Inf. Such optimizations are required to retain defined + behavior over +/-Inf, but the value of the result is undefined.</li> + + <li><tt>nsz</tt>: No Signed Zeros: Allow optimizations to treat the + sign of a zero argument or result as insignificant. </li> + + <li><tt>fast</tt>: Allow algebraically equivalent transformations that may + dramatically change results in floating point (e.g. reassociate). This flag + implies all the others.</li> + +</ol> <h5>Example:</h5> <pre> @@ -4252,7 +4315,7 @@ IfUnequal: <h5>Syntax:</h5> <pre> - <result> = fdiv <ty> <op1>, <op2> <i>; yields {ty}:result</i> + <result> = fdiv [fast-math flags]* <ty> <op1>, <op2> <i>; yields {ty}:result</i> </pre> <h5>Overview:</h5> @@ -4264,7 +4327,32 @@ IfUnequal: floating point values. Both arguments must have identical types.</p> <h5>Semantics:</h5> -<p>The value produced is the floating point quotient of the two operands.</p> +<p>The value produced is the floating point quotient of the two operands. This + instruction can also take any number of fast-math flags, which are + optimization hints to enable otherwise unsafe floating point + optimizations:</p> +<ol> + + <li><tt>nnan</tt>: No NaNs - Allow optimizations to assume the arguments and + result are not NaN. Such optimizations are required to retain defined behavior + over NaNs, but the value of the result is undefined.</li> + + <li><tt>ninf</tt>: No Inf - Allow optimizations to assume the arguments and + result are not +/-Inf. Such optimizations are required to retain defined + behavior over +/-Inf, but the value of the result is undefined.</li> + + <li><tt>nsz</tt>: No Signed Zeros: Allow optimizations to treat the + sign of a zero argument or result as insignificant. </li> + + <li><tt>arcp</tt>: Allow Reciprocal: Allow optimizations to use the reciprocal + of an argument rather than perform division. </li> + + <li><tt>fast</tt>: Allow algebraically equivalent transformations that may + dramatically change results in floating point (e.g. reassociate). This flag + implies all the others.</li> + +</ol> +</p> <h5>Example:</h5> <pre> @@ -4371,7 +4459,7 @@ IfUnequal: <h5>Syntax:</h5> <pre> - <result> = frem <ty> <op1>, <op2> <i>; yields {ty}:result</i> + <result> = frem [fast-math flags]* <ty> <op1>, <op2> <i>; yields {ty}:result</i> </pre> <h5>Overview:</h5> @@ -4384,8 +4472,32 @@ IfUnequal: floating point values. Both arguments must have identical types.</p> <h5>Semantics:</h5> -<p>This instruction returns the <i>remainder</i> of a division. The remainder - has the same sign as the dividend.</p> + <p>This instruction returns the <i>remainder</i> of a division. The remainder + has the same sign as the dividend. This instruction can also take any number + of fast-math flags, which are optimization hints to enable otherwise unsafe + floating point optimizations:</p> + +<ol> + + <li><tt>nnan</tt>: No NaNs - Allow optimizations to assume the arguments and + result are not NaN. Such optimizations are required to retain defined behavior + over NaNs, but the value of the result is undefined.</li> + + <li><tt>ninf</tt>: No Inf - Allow optimizations to assume the arguments and + result are not +/-Inf. Such optimizations are required to retain defined + behavior over +/-Inf, but the value of the result is undefined.</li> + + <li><tt>nsz</tt>: No Signed Zeros: Allow optimizations to treat the + sign of a zero argument or result as insignificant. </li> + + <li><tt>arcp</tt>: Allow Reciprocal: Allow optimizations to use the reciprocal + of an argument rather than perform division. </li> + + <li><tt>fast</tt>: Allow algebraically equivalent transformations that may + dramatically change results in floating point (e.g. reassociate). This flag + implies all the others.</li> + +</ol> <h5>Example:</h5> <pre> |