[uae] Re: Will there be a upcoming E-UAE release?
- From: "Keith G. Robertson-Turner" <uae-freelists@xxxxxxxxxx>
- To: uae@xxxxxxxxxxxxx
- Date: Mon, 24 Nov 2008 20:30:01 +0000
Verily I say unto thee, that Andrew 'Truck' Holland spake thusly:
>> 4. use fdisk to create an AFFS RDB ???
>
> gnu parted, rather than fdisk or cfdisk.
I just had a look.
parted Work.hdf mklabel amiga
Works OK.
But then...
parted Work.hdf mkpartfs UDH2 affs3 0 100
Does nothing, but...
parted Work.hdf mkpart UDH2 0 100
Creates an unformated partition.
The resulting hardfile is still unrecognised by uae/HDToolbox/format.
Performing a hex dump of the two hardfiles, shows that the working one
(Workbench.hdf) contains the string "RDSK" at offset 0, whereas the non
-working one (Work.hdf) has that string at offset 400 (hex), so it seems
parted is not creating the partition table in the proper location, for
some reason.
Here's the relevant part of the logfile:
# Working hardfile (Workbench.hdf)
Mounting uaehf.device 0 (0) (size=
536870912):
RDB: 'UDH1' dostype=444F5303
BlockSize: 512, Surfaces: 1, SectorsPerBlock 1
SectorsPerTrack: 272, Reserved: 2, LowCyl 3, HighCyl 3854
Buffers: 30, BufMemType: 00000000, MaxTransfer: 00ffffff, BootPri: 0
RDB: RDB filesystem 444F5303 version 40.1
RDB: 444F5303 in FileSystem.resouce version 40.1
RDB: fs in FileSystem.resource is newer or same, ignoring RDB filesystem
# Linux filesystem
Mounting uaehf.device 1 (0):
FS: mounted virtual unit HD1 (/home/kgr/Documents)
# Non-working hardfile (Work.hdf)
Mounting uaehf.device 2 (0) (size=
104857600):
RDB: 'UDH2' dostype=4C4E5800
BlockSize: 512, Surfaces: 4, SectorsPerBlock 1
SectorsPerTrack: 32, Reserved: 2, LowCyl 3, HighCyl 1525
Buffers: 30, BufMemType: 00000000, MaxTransfer: 7fffffff, BootPri: 0
FS: Docs starting..
PTEST @ 100290a4
UAEHF: READ DEFECT DATA
UAEHF: READ DEFECT DATA
UAEHF: READ DEFECT DATA
>> hardfile2=rw,HD0:$(FILE_PATH)/Amiga/Workbench.hdf,0,0,0,512,-1,
>> hardfile=rw,0,0,0,512,$(FILE_PATH)/Amiga/Workbench.hdf
>>
>> I'm not entirely clear if both these lines are necessary, or if one
>> is a legacy left over from a previous version of the program.
>
> The documentation indicates that one is legacy, yes. Also, Hi-toro's
> generated configurations contain only the hardfile2 line.
Ah OK, thanks for clearing that up. I take it I can just remove the line
that begins with "hardfile=" then?
>> hardfile2=rw,HD1:$(FILE_PATH)/Work.img,0,0,0,512,-1,
^^^
Typo corrected.
>> hardfile=rw,0,0,0,512,$(FILE_PATH)/Work.hdf
>>
>> But when I ran uae, HDToolbox could only see HD0; HD1 was missing.
>
> Yes, as it didn't have the RDB. Try gnu parted; Well that's what I'm
> going to do right as soon as I have a chance to test this at home.
See above. I just can't get this to work.
>> Clearly this process is non-trivial, and needs some serious work to
>> be actually usable.
>
> Well, depends. First of all, the biggest issue is the mkfs.affs item
I be happy enough just being able to create an unformatted hardfile that
can be formatted in the host. As it is, nothing I do results in a device
that's recognized by uae/HDToolBox/format.
> I'll write my bits up as soon as I've done them.
Thanks.
--
Regards,
Keith G. Robertson-Turner
Other related posts:
