[U-Boot] [PATCH] mtest: Fix end address of increment/decrement test

[Peter Tyser <ptyser@xes-inc.com>]

Hi Wolfgang,

> > As far as the output, my vote would be to align the end address to a
> > 32-bit address and add 3.  eg assuming a starting address of 1000 and
> > ending addresses of:
> > 0x1ffc - output: Testing 00001000 ... 00001fff
> > 0x1ffd - output: Testing 00001000 ... 00001fff
> > 0x1ffe - output: Testing 00001000 ... 00001fff
> > 0x1fff - output: Testing 00001000 ... 00001fff
> > 0x2000 - output: Testing 00001000 ... 00002003
> > 0x2001 - output: Testing 00001000 ... 00002003
> 
> No, please do not implement such automatic alignment; it may be useful
> for some cases, but it may as well hurt, for example if you
> intentionally want to run mtest with misalignment, like giving both
> odd start and end addresses.

I didn't express it well, but what I was getting at was that the
"Testing X .. Y" would ideally state exactly what is being tested.
Unaligned addresses would still be allowed.  I think right now the end
address is always automatically aligned to the same alignment as the
start address though, so the current output is very misleading.

eg:
mw.l 0x1000 0x12345678 0x1000
mtest 0x1003 0x1ffc 1 1
Testing 00001003 ... 00001ffc:
...
md 0x1000 10; md 0x1ff0 10
00001000: 12345600 00000000 00000000 00000000    .4V.............
00001010: 00000000 00000000 00000000 00000000    ................
00001020: 00000000 00000000 00000000 00000000    ................
00001030: 00000000 00000000 00000000 00000000    ................
00001ff0: 00000000 00000000 00000000 00000078    ...............x
00002000: 12345678 12345678 12345678 12345678    .4Vx.4Vx.4Vx.4Vx
00002010: 12345678 12345678 12345678 12345678    .4Vx.4Vx.4Vx.4Vx
00002020: 12345678 12345678 12345678 12345678    .4Vx.4Vx.4Vx.4Vx

You can see the starting alignment was respected, but the ending
alignment was truncated to be 32-bit aligned to the starting address.
In the above example, I think it would be nice to see "Testing
00001003 ... 00001ffe".  Or some other way such that the user knows that
their input wasn't executed to a T; their end address was truncated.

Best,
Peter