On Fri, Nov 25, 2005 at 09:56:20PM -0500, Kris Kennaway wrote: > > I attach two patches, they realize the above ideas, respectively. > > Note that I didn't test either. > > You probably should do so ;-) This isn't the easiest thing for someone > to test without such corrupted media. Umm... I was hoping that someone in the list will suggest a way for easily creating corrupt files. As I don't have a corrupt media, either :) I picked this when Fuse was panicking upon md attaching from a Fuse based filesystem, cf. http://lists.freebsd.org/pipermail/freebsd-ports/2005-November/027724.html The fuse module didn't implement VOP_READ (it rather used a file level "fo_read" method), hence from the point of the md code all files under a Fuse filesystem seem corrupt. So this was triggered. (Since then I added the VOP_READ, btw.) If I test the proposed fixes with a VOP_READ-less fuse module and they will save me from the panic (and work well on traditional fs-es), will you say it's OK? The fuse module lacks a proper audit from "serious" people, so you could dismiss it as well, saying you can't be sure that the problem is not with Fuse... Btw, if you said yes, that means you can as well easily get at a fine "corrupt media" :) CsabaReceived on Sat Nov 26 2005 - 04:50:02 UTC
This archive was generated by hypermail 2.4.0 : Wed May 19 2021 - 11:38:48 UTC