The 32-bit MXCSR register contains control and status information for SSE and AVX SIMD floating-point operations.
| Property | Description |
|---|---|
| FlushToZero | Bit 15 (FTZ) of the MXCSR register enables the flush-to-zero mode, which controls the masked response to a SIMD floating-point underflow condition. |
| RoundingControlMask | Mask for rounding control bits. |
| RoundToNearest | Rounded result is the closest to the infinitely precise result. If two values are equally close, the result is the even value (that is, the one with the least-significant bit of zero). Default. |
| RoundDown | Rounded result is closest to but no greater than the infinitely precise result. |
| RoundUp | Rounded result is closest to but no less than the infinitely precise result. |
| RoundTowardZero | Rounded result is closest to but no greater in absolute value than the infinitely precise result. |
| PrecisionMask | Bits 7 through 12 provide individual mask bits for the SIMD floating-point exceptions. An exception type is masked if the corresponding mask bit is set, and it is unmasked if the bit is clear. These mask bits are set upon a power-up or reset. This causes all SIMD floating-point exceptions to be initially masked. |
| UnderflowMask | Bits 7 through 12 provide individual mask bits for the SIMD floating-point exceptions. An exception type is masked if the corresponding mask bit is set, and it is unmasked if the bit is clear. These mask bits are set upon a power-up or reset. This causes all SIMD floating-point exceptions to be initially masked. |
| OverflowMask | Bits 7 through 12 provide individual mask bits for the SIMD floating-point exceptions. An exception type is masked if the corresponding mask bit is set, and it is unmasked if the bit is clear. These mask bits are set upon a power-up or reset. This causes all SIMD floating-point exceptions to be initially masked. |
| DivideByZeroMask | Bits 7 through 12 provide individual mask bits for the SIMD floating-point exceptions. An exception type is masked if the corresponding mask bit is set, and it is unmasked if the bit is clear. These mask bits are set upon a power-up or reset. This causes all SIMD floating-point exceptions to be initially masked. |
| DenormalOperationMask | Bits 7 through 12 provide individual mask bits for the SIMD floating-point exceptions. An exception type is masked if the corresponding mask bit is set, and it is unmasked if the bit is clear. These mask bits are set upon a power-up or reset. This causes all SIMD floating-point exceptions to be initially masked. |
| InvalidOperationMask | Bits 7 through 12 provide individual mask bits for the SIMD floating-point exceptions. An exception type is masked if the corresponding mask bit is set, and it is unmasked if the bit is clear. These mask bits are set upon a power-up or reset. This causes all SIMD floating-point exceptions to be initially masked. |
| ExceptionMask | Combine all bits for exception masking into one mask for convenience. |
| DenormalsAreZeroes | Bit 6 (DAZ) of the MXCSR register enables the denormals-are-zeros mode, which controls the processor’s response to a SIMD floating-point denormal operand condition. |
| PrecisionFlag | Bits 0 through 5 of the MXCSR register indicate whether a SIMD floating-point exception has been detected. They are "sticky" flags. That is, after a flag is set, it remains set until explicitly cleared. To clear these flags, use the LDMXCSR or the FXRSTOR instruction to write zeroes to them. |
| UnderflowFlag | Bits 0 through 5 of the MXCSR register indicate whether a SIMD floating-point exception has been detected. They are "sticky" flags. That is, after a flag is set, it remains set until explicitly cleared. To clear these flags, use the LDMXCSR or the FXRSTOR instruction to write zeroes to them. |
| OverflowFlag | Bits 0 through 5 of the MXCSR register indicate whether a SIMD floating-point exception has been detected. They are "sticky" flags. That is, after a flag is set, it remains set until explicitly cleared. To clear these flags, use the LDMXCSR or the FXRSTOR instruction to write zeroes to them. |
| DivideByZeroFlag | Bits 0 through 5 of the MXCSR register indicate whether a SIMD floating-point exception has been detected. They are "sticky" flags. That is, after a flag is set, it remains set until explicitly cleared. To clear these flags, use the LDMXCSR or the FXRSTOR instruction to write zeroes to them. |
| DenormalFlag | Bits 0 through 5 of the MXCSR register indicate whether a SIMD floating-point exception has been detected. They are "sticky" flags. That is, after a flag is set, it remains set until explicitly cleared. To clear these flags, use the LDMXCSR or the FXRSTOR instruction to write zeroes to them. |
| InvalidOperationFlag | Bits 0 through 5 of the MXCSR register indicate whether a SIMD floating-point exception has been detected. They are "sticky" flags. That is, after a flag is set, it remains set until explicitly cleared. To clear these flags, use the LDMXCSR or the FXRSTOR instruction to write zeroes to them. |
| FlagMask | Combines all bits for flags into one mask for convenience. |