[ also adding Harishankar et al. to Cc, for the patch see : https://patchwork.kernel.org/project/netdevbpf/patch/20250815140510.1287598-1-nandakumar@xxxxxxxxxxxxxxxx/ ] On 8/15/25 9:10 PM, Eduard Zingerman wrote:
On Fri, 2025-08-15 at 19:35 +0530, Nandakumar Edamana wrote:This commit addresses a challenge explained in an open question ("How can we incorporate correlation in unknown bits across partial products?") left by Harishankar et al. in their paper: https://arxiv.org/abs/2105.05398 When LSB(a) is uncertain, we know for sure that it is either 0 or 1, from which we could find two possible partial products and take a union. Experiment shows that applying this technique in long multiplication improves the precision in a significant number of cases (at the cost of losing precision in a relatively lower number of cases). This commit also removes the value-mask decomposition technique employed by Harishankar et al., as its direct incorporation did not result in any improvements for the new algorithm. Signed-off-by: Nandakumar Edamana <nandakumar@xxxxxxxxxxxxxxxx> ---Hi Nandakumar, Could you please provide a selftest demonstrating a difference in behavior? What technique did you employ to estimate the number of cases when precision is improved vs worsened? If this is some kind of a program doing randomized testing, could you please share a link to it? Thanks, Eduard
