DOS 6.22 ignores the number_of_sectors_per_fat entry in the boot block, but
insists on the minimal amount. Since I miscalculated that value, it did
not match the expectations of DOS.
Signed-off-by: Johannes Schindelin <[EMAIL PROTECTED]>
---
> If you will find the solution to the MS-DOS problem, i'd be
> very greatful if you will share the patch/workaround.
Well, here you are. It was not the second FAT which was missing,
but DOS looking for direntries where VVFAT put some sectors
of the second FAT...
block-vvfat.c | 2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
applies-to: 883ade6f323513393835edeb321b52a98cf56430
80721ce5b55e71e5ae15145e4ca91faa00b70001
diff --git a/block-vvfat.c b/block-vvfat.c
index a7ad868..4d33095 100644
--- a/block-vvfat.c
+++ b/block-vvfat.c
@@ -710,7 +710,7 @@ static int init_directory(BDRVVVFATState
memset(&(s->first_sectors[0]),0,0x40*0x200);
/* TODO: if FAT32, this is probably wrong */
- s->sectors_per_fat=0xfc;
+ s->sectors_per_fat=0xec;
s->sectors_per_cluster=0x10;
s->cluster_size=s->sectors_per_cluster*0x200;
s->cluster=malloc(s->cluster_size);
---
0.99.9.GIT
_______________________________________________
Qemu-devel mailing list
Qemu-devel@nongnu.org
http://lists.nongnu.org/mailman/listinfo/qemu-devel
