PFS3AIO v3.1 Final

[apex]

Hmmm, I setup 3 partitions using PDS (50445303) identifier on my A2091 booting kickstart 3.1 using HDInsttools. If I change now to Kickstart 1.3, Amiga seems to do something from HD and then I see a box no DOS partition. Changing again to Kickstart 3.1, I see again my partition and everything is fine.

Any ideas?

[Sinphaltimus]

Originally Posted by Toni Wilen

http://www.winuae.net/files/b/pfs3aio_v3.1_test1.zip

- Fixed softlink support, they have always been broken (Piru). I have never used softlinks so I have no idea what happened previously.
- TD64/NSD check also validates returned io_Actual value.
- ACTION_INFO InfoData returns fake block size to work around overflow in programs (including WB) that use 32-bit "in use * 100 / total blocks" formula to calculate size. Returned block size depends on partition size. (This is only visual change, "real" block size in DosEnvec is unchanged). UAE already did this with very large drives without side-effects.

TODO: If I manage to hook up filesystem loading to new UAE debugger, I'll try to find out if there is any obvious ways to reduce memory usage when running on low memory systems.

Is this eventually going to be the new "final"? I've been using it for new installations recently or should I use the final listed in post #1? I'm not having any issues to report with this v3.1_test1.

[AmigaWolf]

Thanks i just updated it to PFS3AIO v3.0 Final.

[Toni Wilen]

Originally Posted by Sinphaltimus

Is this eventually going to be the new "final"? I've been using it for new installations recently or should I use the final listed in post #1? I'm not having any issues to report with this v3.1_test1.

"Small" memory support probably gets included but I am still not sure how to choose between modes automatically. Available memory size? Less than 1M? Less than 2M?

Originally Posted by apex

Hmmm, I setup 3 partitions using PDS (50445303) identifier on my A2091 booting kickstart 3.1 using HDInsttools. If I change now to Kickstart 1.3, Amiga seems to do something from HD and then I see a box no DOS partition. Changing again to Kickstart 3.1, I see again my partition and everything is fine.

I don't know, perhaps HDInsttools does something weird. Attach RDB backup (at least OS3.9 hdtoolbox has menu option) and check A2091 ROM version.

Originally Posted by E-Penguin

Is that 30k per partition? That might start to add up for some people

Yes but using more than 2 partitions is not a good idea

[amiwolf]

Originally Posted by Toni Wilen

"Small" memory support probably gets included but I am still not sure how to choose between modes automatically. Available memory size? Less than 1M? Less than 2M?

How about "Small" mode if no FAST-RAM detected?

[Toni Wilen]

Originally Posted by amiwolf

How about "Small" mode if no FAST-RAM detected?

I don't think 2M Chip is low ram enough and 0.5M+0.5M isn't high ram enough in my opinion.

[kolla]

I would leave it to the user to decide what is "small", for example by specifying sub 30 number of buffers or something.

[Toni Wilen]

Originally Posted by kolla

I would leave it to the user to decide what is "small", for example by specifying sub 30 number of buffers or something.

I am thinking of situation where "small" disk gets moved to large memory/fast system where it would be nice to have automatic small mode disable, for faster disk accesses (probably mainly affects writes).

What if small buffer value will enable it but only if available fast ram is less than 4M?

[kolla]

Sounds fine with me

[cyberhead97]

I have done some tests with pfs3aio v2.3 some time ago on unexpanded A600 and A1200 chip mem only. From my tests of pfs3 without fast mem, pfs3 eats 200-300kb chip ram per partition depending on buffer tweaks etc.

For me this is acceptable as I dont expect much of an unexpanded Amiga. The important part is that pfs3 works and I can access the partitions for e.g. copying or editing files. As Toni mentions partitions should be kept to a max of 2. 3 partitions will eat almost all of an A600 chip men.

To make use of an unexpanded A600 or A1200 for running WHDLoad demos or games, I have had great success with FFS v45.16. This only uses 20-30kb per partition and works well for such a setup.

Putting pfs3aio in custom burned Kickstart rom could also reduce chip mem without having tried that.

[klx300r]

Originally Posted by supaduper

Just updated my RDB with Toni`s Ver 19 pfs3aio and its all good

Any noobs who want to do this for partitions less then 5gb in size , just extract pfs3aio from Toni`s download then put it into L: directory.
Then fire up HDtoolbox, then >, Partition Drive >select your first partition then> tick Advanced Options, >Update > input L: pfs3aio into dialog box > Click OK ( if successful it will change to Ver19 in the window) > Click OK > Save changes to drive.

Your done..enjoy

ok so just to be super clear this is a stand along version? i/e we don't need to install nay previous version

[Leandro Jardim]

Originally Posted by kolla

I would leave it to the user to decide what is "small", for example by specifying sub 30 number of buffers or something.

Originally Posted by Toni Wilen

I am thinking of situation where "small" disk gets moved to large memory/fast system where it would be nice to have automatic small mode disable, for faster disk accesses (probably mainly affects writes).

What if small buffer value will enable it but only if available fast ram is less than 4M?

I think that experienced users will want to specify explicitily what mode they want.

I prefer the use of another dostype in the same way Toni has done before when him incorporated the PFS and PFSDS together. I think it's more logical, following the lines of thinking of AmigaDOS.

I don't know what most users think, but I have the feel that the act of changing modes when depending of the buffer count or of the available memory size is a sort of a hack.

But I trust Toni very much, and I don't want to insult him, so maybe I could be wrong.

[malko]

Originally Posted by Toni Wilen

I am thinking of situation where "small" disk gets moved to large memory/fast system where it would be nice to have automatic small mode disable, for faster disk accesses (probably mainly affects writes).

What if small buffer value will enable it but only if available fast ram is less than 4M?

-1 .
For me "automatic" change of a config is a no go !
As far as i know, you never implemented any "automatic" things like that, did you ?

[Gilloo]

My tests: format and copy 800MB of various files
The machine: A2000 2MBchip 4MBfast KS 40 (Workbench 3.1)
The disk: HP v165W 8GB (mini thumb key)(one partition created with HDInstTools from the last official PFS3 package)
PFS3 18.2 (pfs3aio may 2014) => issue, index -1 during the copy
PFS3 19.0 (pfs3aio http://aminet.net/package/disk/misc/pfs3aio) charming copy !!

Also works on KS 34 (WB 1.3) format and copy only, not the HDInstTool update phase.

Thanks Toni, you have fixed a big bug that apparently occurs only on massive copy, or when copying a large amount of little files. It wasn't occur on update or individual copy

[supaduper]

Originally Posted by klx300r

ok so just to be super clear this is a stand along version? i/e we don't need to install nay previous version

No

But as always do any changes to a copy of your system first ( I use a VHD image in WinUae to try any updates before committing it to a hard drive) to save any heartache

[Toni Wilen]

Originally Posted by malko

-1 .
For me "automatic" change of a config is a no go !
As far as i know, you never implemented any "automatic" things like that, did you ?

Why would you want (special case) "slow/small memory" behavior when you more than enough RAM?

EDIT: I don't want to add separate version or dostypes because it is guaranteed to cause confusion ("why this version is slower than previous"). Thats why I want it to be mostly automatic, requiring at least small enough buffer value that no normal configuration uses.

[E-Penguin]

Negative buffer sizes to indicated small mode, perhaps?

-30 => small mode, 30 buffers.

side note; I am unable to write "buffers" without typing "buggers" the first few times. Is it just me?

[kolla]

I don't think HDToolBox etc allow negative number of buffers, lol

And I think it makes sense to use buffers for this, as buffers are already "magic" with PFS, even before pfs3aio.

[E-Penguin]

Of course it can. You can just count backwards from 65535 (=-1), assuming it's a 16-bit field. If you can't do 2's complement in your head then you have no business using computers, quite frankly...

[spudje]

E-Penguin, I think you will be very dissapointed then to know, that the majority of people in the world using computers cannot...