09.01.2013 Views

AMD x86-64 Architecture Programmer's Manual, Volume 4, 128-Bit ...

AMD x86-64 Architecture Programmer's Manual, Volume 4, 128-Bit ...

AMD x86-64 Architecture Programmer's Manual, Volume 4, 128-Bit ...

SHOW MORE
SHOW LESS

You also want an ePaper? Increase the reach of your titles

YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.

26568—Rev. 3.05—September 2003 <strong>AMD</strong><strong>64</strong> Technology<br />

CVTTPD2PI Convert Packed Double-Precision Floating-Point<br />

to Packed Doubleword Integers, Truncated<br />

Converts two packed double-precision floating-point values in an XMM register or a<br />

<strong>128</strong>-bit memory location to two packed 32-bit signed integer values and writes the<br />

converted values in an MMX register.<br />

Mnemonic Opcode Description<br />

CVTTPD2PI mmx, xmm/mem<strong>128</strong> 66 0F 2C /r Converts packed double-precision floating-point values in an<br />

XMM register or <strong>128</strong>-bit memory location to packed doubleword<br />

integer values in the destination MMX register. Inexact results are<br />

truncated.<br />

If the result of the conversion is an inexact value, the value is truncated (rounded<br />

toward zero). If the floating-point value is a NaN, infinity, or if the result of the<br />

conversion is larger than the maximum signed doubleword (–2 31 to +2 31 –1), the<br />

instruction returns the 32-bit indefinite integer value (8000_0000h) when the invalidoperation<br />

exception (IE) is masked.<br />

Related Instructions<br />

CVTDQ2PD, CVTPD2DQ, CVTPD2PI, CVTPI2PD, CVTSD2SI, CVTSI2SD,<br />

CVTTPD2DQ, CVTTSD2SI<br />

rFLAGS Affected<br />

None<br />

mmx<br />

63 32 31 0<br />

xmm/mem<strong>128</strong><br />

127 <strong>64</strong> 63 0<br />

convert<br />

convert<br />

cvttpd2pi.eps<br />

CVTTPD2PI 87

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!