On Tue, Jun 13, 2017 at 10:43:05AM +0100, Wilco Dijkstra wrote: > Richard Earnshaw (lists) wrote: > > > > Why 1 and not 2? Many processors have 2 fp pipes and forcing this down > > to a sequential stream is not obviously the right thing. > > 1 was faster than 2. Like I said, the reassociation is too aggressive and even > splits multiply-add rather than keeping them. Until reassociation is fixed and > able to split a complex expression into 2 independent chains of similar depth > while keeping multiply-accumulate operations, it is best to set it to 1 for > now. > > > If reassociation is is causing excess spilling, then the right fix for > > that is to look at the pressure model, not hammer the problem away. > > The problem of the GCC scheduler hugely increasing register pressure has > existed for many years with no progress being made. Unless we're willing to > start a project improving this I do not believe there will be a solution any > time > soon. So changing the association width is the best solution for the time > being.
Are there bugs with testcases open in bugzilla for this? Was the regression in "more recent" GCCs that are more aggressive at resassociation noted somewhere? It is hard to take this fix with the justification given, and no idea of what compiler behaviour we're masking with this patch. I don't think I can approve it with what is presented here, particularly so for "generic". Richard's point of why "1" rather than "2" also needs a better answer than given. Thanks, James
