On Thu, Feb 22, 2024 at 10:31:53AM +0100, Daniel Wagner wrote: > On Wed, Feb 21, 2024 at 10:32:05AM -0800, Luis Chamberlain wrote: > > > One discussion point I'd like to add is > > > > > > - running blktest against real hardare/target > > > > We've resolved this in fstests with canonicalizing device symlinks, and > > through kdevops its possible to even use PCIe passthrough onto a guest > > using dynamic kconfig (ie, specific to the host). > > > > It should be possible to do that in blktests too, but the dynamic > > kconfig thing is outside of scope, but this is a long winded way of > > suggestin that if we extend blktests to add a canonon-similar device > > function, then since kdevops supports blktests you get that pcie > > passthrough for free too. > > I should have been more precise here, I was trying to say supporting > real fabrics targets. blktests already has some logic for PCI targets > with $TEST_DEV but I haven't really looked into this part yet. Do fabric targets have a symlink which remains static? Luis
