On 2009-02-02 18:37:08 -0600, Gabriel Dos Reis wrote:
On Mon, Feb 2, 2009 at 6:08 PM, Vincent Lefevre <vincent@vinc17.org> wrote:
On 2009-02-02 22:30:03 +0100, Roberto Bagnara wrote:
#if 1 // CHECKME: the test seems to speed up the GMP computation. const Coefficient& y_i = y[i]; if (y_i != 0) sub_mul_assign(x_i, y_i, normalized_x_k); #else sub_mul_assign(x_i, y[i], normalized_x_k); #endif // 1
Couldn't this be due to C++ overhead? Have you tried in C?
That would be surprising.
Yes, but can anyone guess any non-surprising reason? :) BTW, Roberto, are you sure that the speed-up is due to the test itself and not an arbitrary side effect because the code has just changed? -- Vincent Lefèvre <vincent@vinc17.org> - Web: <http://www.vinc17.org/> 100% accessible validated (X)HTML - Blog: <http://www.vinc17.org/blog/> Work: CR INRIA - computer arithmetic / Arenaire project (LIP, ENS-Lyon)