jones_supa writes: Thiago Macieira of Intel is coordinating a proposal to enable SSE2 by default for all floating point mathematics for the 32-bit builds of the Qt toolkit beginning with the Qt 5.3 release. This likely change is due to the V4 JIT engine already generating FP math for Qt's QML only using SSE x86 registers. On Linux the GCC flags -msse2 -mfpmath=sse would be used during compilation, and MSVS2012 generates SSE2 code by default. In Qt there exists also a small handful of SSE3 improvements, which will still be runtime checked. After the change the developers assume that anyone using Qt 5.3+ will basically be on an x86 CPU from 2004 or later — basically the Pentium 4 and Pentium M era or AMD Athlon 64. Rare users of older hardware can still switch to a source-based build of Qt where they can manually disable the SSE2+ support. The Streaming SIMD Extensions is an instruction set extension for x86 processors which allows doing faster floating point mathematics over the classic x87 math coprocessor.
I judge a religion as being good or bad based on whether its adherents
become better people as a result of practicing it.
- Joe Mullally, computer salesman