On Mon, 17 Jun, at 10:46:28AM, Jan Beulich wrote: > To me, all this looks like it is being done on phenomenological basis, > taking a particular build of a particular firmware implementation as > the reference. Imo we shouldn't change the code in this way. This > also applies to the fact that the step is being doubled rather than > e.g. tripled: With it ending up a "again" anyway (see below), what's > the point of avoiding one more of the iterations? > > Generic considerations would result in the increment being at least > 3 * element size; twice the element size assumes that the allocator > would behave in certain ways (like returning the head or tail part of > a larger piece of memory). I have no issue with changing the multiplier. But let's get clarification from Zach as to what exactly is going on here. > I agree that there ought to be an upper limit. But a single retry here > again looks like a tailored solution to a particular observed (mis-) > behavior, rather than something resulting from general considerations. What value would you suggest for the retry? -- Matt Fleming, Intel Open Source Technology Center -- To unsubscribe from this list: send the line "unsubscribe stable" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html