Rockbox mail archive

• Next message: Steve: "PIC-based remote (son of Remocon)"
• Previous message: gerald_clark_at_mindspring.com: "Re: disk write testing"
• In reply to: Greg Haerr: "Re: disk write testing"
• Next in thread: Björn Stenberg: "Re: disk write testing"
• Messages sorted by: [ date ] [ thread ] [ subject ] [ author ] [ attachment ]

Greg Haerr wrote:
> I'll show you by submitting a patch that greatly cleans up the madness
> in file.c

Sounds good. I'm curious to see how you've solved it.

> Lseek shouldn't be mucking around performing data file i/o, other than
> chasing down the cluster chain (which doesn't work, see below).

Actually it does work. You seem to be reading old code.

> Why is seeksector decremented as the first statement?

Quoting the source:

    if (seeksector) {
        /* we need to find the sector BEFORE the requested, since
           the file struct stores the last accessed sector */
        seeksector--;
 
This behaviour was changed from how the fat code used to work. Before write was added, the fat_file struct stored nextsector. Now it stores lastsector.

The reason for the change is that we cannot store nextsector when writing, since it would mean allocating a new cluster before we know if we need one.

-- 
Björn

• Next message: Steve: "PIC-based remote (son of Remocon)"
• Previous message: gerald_clark_at_mindspring.com: "Re: disk write testing"
• In reply to: Greg Haerr: "Re: disk write testing"
• Next in thread: Björn Stenberg: "Re: disk write testing"
• Messages sorted by: [ date ] [ thread ] [ subject ] [ author ] [ attachment ]