Messages in this thread | | | From | Joakim Tjernlund <> | Subject | Re: [PATCH] jffs2: implement mount option to configure endianness | Date | Wed, 7 Nov 2018 09:16:39 +0000 |
| |
On Wed, 2018-11-07 at 10:05 +0100, David Woodhouse wrote: > > On Tue, 2018-11-06 at 13:49 -0800, Nikunj Kela wrote: > > This patch allows the endianness of the JFSS2 filesystem to be > > specified by mount option 'force_endian=big|little|native'. If > > endianness is not specified, it defaults to 'native' endianness > > thus retaining the existing behavior. > > > > Some architectures benefit from having a single known endianness > > of JFFS2 filesystem (for data, not executables) independent of the > > endianness of the processor (ARM processors can be switched to either > > endianness at run-time). > > > > We have boards that are shipped with BE jffs2 and BE kernel. We are > > now migrating to LE kernel. This mount option helps us in mounting > > BE jffs2 on LE kernel. > > Thanks for implementing this. JFFS2 has often been very sensitive to > performance at mount time — I'd love to see how this affects the time > taken to mount a large file system. If it's significant, we may want a > config option to make this conditional?
Yes, this may slow things down. I am not sure I agree with the impl. either. Could one not make cpu_to_je_X/jeX_to_cpu a function ptr which is set to a func. with the correct endian?
That way it should also be easy to have a config option for endian: Native/BE/LE or Dynamic
Jocke
| |