Re: [PATCH 1/1] nvme: always enable multipath
From: Keith Busch
Date: Fri Nov 22 2024 - 13:16:38 EST
On Fri, Nov 22, 2024 at 12:52:51PM -0500, John Meneghini wrote:
> On 11/22/24 10:39, Keith Busch wrote:
> > Anyway, we only use the one physical function, so they're certainly not
> > multipath devices here. We disable the CONFIG option because while the
> > nvme multipath IO overhead is low, it's not zero.
>
> Do we know what the performance impact is? Has anybody measured the difference?
It doesn't benefit single path devices, so any cost for it is too high.
It's clearly not free.