Patch forthcoming.
Created attachment 226574 [details] the patch
Comment on attachment 226574 [details] the patch r=me
Comment on attachment 226574 [details] the patch It seems to me that the "minimum" in computeNumberOfWorkerThreads() is not really the minimum after the subtraction of 1 in Options.h. Could computeNumberOfWorkerThreads() always have a true minimum of 1 and pass 1 or 7 as the maximum? Or subtract the 1 in computeNumberOfWorkerThreads. Seems a little better to have all the min/max calculation in one place, computeNumberOfWorkerThreads.
Comment on attachment 226574 [details] the patch Didn't mean to change the r+ from Geoff.
(In reply to comment #3) > (From update of attachment 226574 [details]) > It seems to me that the "minimum" in computeNumberOfWorkerThreads() is not really the minimum after the subtraction of 1 in Options.h. Could computeNumberOfWorkerThreads() always have a true minimum of 1 and pass 1 or 7 as the maximum? Or subtract the 1 in computeNumberOfWorkerThreads. Seems a little better to have all the min/max calculation in one place, computeNumberOfWorkerThreads. Yeah it's weird. It's the result of sharing this function with the GC. The GC wants to know the number of worker threads including the main thread while the JIT wants to know the number of worker threads excluding the main thread. The minus-1 is the exclusion of the main thread.
Landed in http://trac.webkit.org/changeset/165553