Date: Tue, 31 Jul 2018 13:18:08 -0700 From: bob prohaska <fbsd@www.zefox.net> To: Mark Millard <marklmi@yahoo.com> Cc: freebsd-arm@freebsd.org, bob prohaska <fbsd@www.zefox.net> Subject: Re: RPI3 swap experiments Message-ID: <20180731201808.GE94742@www.zefox.net> In-Reply-To: <93D7A41F-3AFA-48FE-A82E-FC4AA76A355C@yahoo.com> References: <4ED9B658-A5A8-4BA6-9412-EBB7150B4B66@yahoo.com> <20180723190257.GA47869@www.zefox.net> <76BCFCB9-1071-4557-9FDE-017444ADBF42@yahoo.com> <20180725232453.GA57716@www.zefox.net> <20180731054712.GA92917@www.zefox.net> <d17be735-4d7e-73b6-4af1-a64470bc9e32@sentry.org> <20180731153531.GA94742@www.zefox.net> <908FB299-07CF-4E88-9C18-298CA357AD01@yahoo.com> <20180731181923.GC94742@www.zefox.net> <93D7A41F-3AFA-48FE-A82E-FC4AA76A355C@yahoo.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On Tue, Jul 31, 2018 at 11:54:41AM -0700, Mark Millard wrote: > > You seem to be indicating "one swap device" worked even with > multiple swap partitions on the device, at least for the USB > example. Yes, in both "all on USB" and "all on microSD" configurations, but it's not symmetric: One pair of storage devices will run -j4 buildworld with all swap on microSD but not with all swap on USB, the other pair runs -j4 buildworld with all swap on USB but not all swap on microSD. The microSD cards are marked the same, the USB devices are marked USB3.0 versus USB3.1 and the 3.1 devices is reported to be "slower" by some reviewers. That's the pair running now. > But you also have the example of a microsSD which got > the OOMA activity with only one swap device in use (but multiple > swap partitions on that device). > Yes. The storage pair containing the USB 3.0 flash drive will run -j4 buildworld with all swap on USB but will not with all swap on microSD. It also won't run with mixed swap, which was a surprise. My hope was to demonstrate that interleaving was somehow the culprit, but that backfired, since microSD alone failed too. I'm planning to switch back to the USB3.0 setup, just to eliminate one variable (out of far too many!) That storage pair has the most flexible swap partitioning and presumably better performance. Thanks for reading, bob prohaska
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20180731201808.GE94742>