FUSE Passthrough Support May Land For Linux 6.9 To Help Boost I/O Performance

[guglovich]

Hopefully this will hurt the sluggish GVFS and MTP.

[trapexit]

Hopefully this will hurt the sluggish GVFS and MTP.

You mean "help"? It won't. This only helps with overlay/union filesystems.

[bob l'eponge]

If a read fails on a remote filesystem or write results in ENOSPC there is no mechanism to push that info to the FUSE server and do something about it. I think many users will be fine with that feature tradeoff though.

Other than a kernel error you mean ? Can't you un-passthrough in that case ? Since this is supposed to be exceptional case, I *guess* it's not a problem if you have to re-do the operation without the passthrough mode.

[trapexit]

Other than a kernel error you mean ? Can't you un-passthrough in that case ? Since this is supposed to be exceptional case, I *guess* it's not a problem if you have to re-do the operation without the passthrough mode.

As far as I understand the way the feature is implemented you can't "un-passthrough" it. Say you setup the passthrough on open and then a write returns with ENOSPC. The current API does not return that error back to the FUSE server. The server is bypassed for all RW. I believe I'll get a release but not the error. So I can't catch the situation like I do now and do something about it.