Samuel Thibault, le dim. 28 mai 2023 23:35:21 +0200, a ecrit: > Sergey Bugaev, le dim. 28 mai 2023 13:32:12 +0300, a ecrit: > > On Sun, May 28, 2023 at 04:34 AM Samuel Thibault > > <samuel.thiba...@gnu.org> wrote: > > > And rumpdisk as well :) > > > > > > [ 2.9500050] wd3 at atabus5 drive 0 > > > [ 2.9500050] wd3: <QEMU HARDDISK> > > > [ 2.9500050] wd3: drive supports 16-sector PIO transfers, LBA48 > > > addressing > > > [ 2.9500050] wd3: 50001 MB, 101589 cyl, 16 head, 63 sec, 512 bytes/sect > > > x 102402048 sectors > > > [ 2.9600050] wd3: drive supports PIO mode 4, DMA mode 2, Ultra-DMA mode > > > 5 (Ultra/100), NCQ (32 tags) > > > [ 2.9600050] wd3(ahcisata0:3:0): using PIO mode 4, DMA mode 2, > > > Ultra-DMA mode 5 (Ultra/100) (using DMA), NCQ (31 tags) > > > > > > and it seems to be working :D > > > > Yay, this is so cool! You could make an attempt to boot off the rumpdisk > > then? > > Yes, it does work
It is however very unstable, I keep getting various crashes in various processes, most often apparently related to TLS access. Possibly the addition of a process in the I/O path is triggering much more context switches and triggers bugs in fs reloading. Samuel
