> On Wed, Feb 15, 2012 at 02:40:37PM +0000, Al Viro wrote: > > FWIW, having looked at that code again, I'm definitely in favour of > dumping > the ENDIANNESS and MMI config options; the latter is not time-critical at > all > - removed both options The MMI option was more meant to make that specific extra functionality more present to the kernel (compile) user. - removed the endian_swap mount option (the reported option) I agree it shall be just sufficient to silently mount the fs. Different endianness is not of any big importance for using it. Sounds to me more like something for any userland tools if someone desires more information about low-level fs-structure stuff. - fixed a page adressing bug in long_filename() Long filename blocks are not always 512 bytes. Yes, the longfilename structure itself is limited to 512bytes, but that's a result of the smallest supported blocksize to be 512 bytes. Longfile limit is 510 + 2 bytes length field = longfile record. However, in a 1k blocksize filesystem, it will still eat up a whole block. - fixed broken mmi superblock handling, broken by the switch to pagemap, root_node etc. stuff ;) - Documentation file added That should give a quick overview on how the qnx6fs bits and pieces fit in the whole context. Feedback very welcome! (as always) I hope that someone with coding skills after reading that introduction, should be able to fiddle out the rest by reading the driver source. - Lost sb fixed Active sb-bh is now released in put_super() - not before. - dir_longfilename() return codes cleaned up - qnx6fs.h cleaned up - switched from for-loops to memcpy()'s in super_mmi() and iput() to copy data from one struct to the other - became friend with git and created the attached patch with it (hope the commit format is now proper) + wanted to say "Thank you!" for the absolut fantastic work you've done on the pagemap stuff! Patch here: http://a6.ontika.net/patches/0001-fs-initial-qnx6fs-addition.patch.gz It's against vfs.fs git tree. Kai -- To unsubscribe from this list: send the line "unsubscribe linux-fsdevel" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
