April 26, 2026 at 4:32 PM, "Milos Nikic" <[email protected] mailto:[email protected]?to=%22Milos%20Nikic%22%20%3Cnikic.milos%40gmail.com%3E > wrote:
> > To add I would love to hear if there are any issues on real hardware so that > we could address them promptly. > Regards, > Milos > > On Sun, Apr 26, 2026, 1:31 PM Milos Nikic <[email protected] > mailto:[email protected] > wrote: > > > > > Hello Joshua, > > I have been using it since the first patch was sent (that was why I was > > discovering so many issues at first). > > > > And I got asked by 3 separate people on how to install it and turn it on > > etc since a while (after v5 so about 6 weeks ago). > > > > I didn't get any more bug reports after V6. > > > > But I got to say that I don't know who is using real hardware, I know don't > > (for now) and that is the only thing I can say for sure I'd be happy to try it on real hardware, but you might need to really help me figure out how to get it all set up. I suppose that I need to set up your development environment. You wrote a script for that at some point... > > > > Kind regards, > > Milos > > > > On Sun, Apr 26, 2026, 12:53 PM <[email protected] > > mailto:[email protected] > wrote: > > > > > > > > April 1, 2026 at 12:17 AM, "Milos Nikic" <[email protected] > > > mailto:[email protected] mailto:[email protected] > > > mailto:[email protected] > > > ?to=%22Milos%20Nikic%22%20%3Cnikic.milos%40gmail.com http://40gmail.com/ > > > %3E > wrote: > > > > > > > > > > > Hi Samuel, > > > > Just dropping a quick ping on this v5 thread now that it's been a > > > > couple of weeks. > > > > > > > > Since sending it, I’ve (and some other people) continued stress-testing > > > > it and using it daily on top of all the patches that came through in > > > > between. > > > > > > May I ask how many people are using the beta "ext3fs" on the Hurd? Do we > > > have any real hardware testers? > > > > > > > It works great but there is an assert that blows up on unjournaled > > > > mounts in v5 (as mentioned previously), and some compiler warnings when > > > > compiled on 32-bit and a rare deadlock that only triggers when the ring > > > > buffer reaches 100% saturation under synthetic I/O. > > > > > > > I have the fixes for all of that ready in my local tree. They don't > > > > change the architecture at all! > > > > To make your eventual review as easy as possible, how would you like me > > > > to handle them? > > > > > > > > Should I send a v6 now with those fixes included, or hold onto them > > > > until you've had a chance to look at the core v5 architecture? > > > > > > > > Best, > > > > Milos > > > > > > > > > >
