[Info-vax] VAX Macro to C conversion

[John Reagan]

On Monday, June 24, 2019 at 3:01:08 PM UTC-4, Simon Clubley wrote:
> On 2019-06-24, Dave Froble <davef at tsoft-inc.com> wrote:
> > On 6/24/2019 1:24 PM, Simon Clubley wrote:
> >> On 2019-06-24, Michael Moroney <moroney at world.std.spaamtrap.com> wrote:
> >>> =?UTF-8?Q?Jan-Erik_S=c3=b6derholm?= <jan-erik.soderholm at telia.com> writes:
> >>>
> >>>> Den 2019-06-24 kl. 15:59, skrev Michael Moroney:
> >>>>> Simon Clubley <clubley at remove_me.eisner.decus.org-Earth.UFP> writes:
> >>>>>
> >>>>>> How much work would be involved in a rewrite of the shadowing code in C ?
> >>>>>
> >>>>> $ dir/size/grand xf72_resd$:[shadowing*.src]*.mar;
> >>>>>
> >>>>> Grand total of 12 directories, 55 files, 9878 blocks.
> >>>>> $
> >>>>>
> >>
> >> Thanks.
> >>
> >>>
> >>>> And how many "Records searched:" from:
> >>>
> >>>> $ sea/stat/win=0 xf72_resd$:[shadowing*.src]*.mar; qwertyasdfgh
> >>>
> >>>> ?
> >>>
> >>> $ sea/stat/win=0 xf72_resd$:[shadowing*.src]*.mar; qwertyasdfgh
> >>>
> >>> Files searched:                55       Buffered I/O count:       217
> >>> Records searched:          159718       Direct I/O count:         130
> >>> Characters searched:      4639981       Page faults:               24
> >>> Records matched:                0       Elapsed CPU time:  0 00:00:00.45
> >>> Lines printed:                  0       Elapsed time:      0 00:00:00.97
> >>> %SEARCH-I-NOMATCHES, no strings matched
> >>> $
> >>>
> >>>
> >>>> So the new file system will not support shadowing from the start?
> >>>
> >>> The new file system will support disks > 2TB.  Shadowing doesn't, and cannot
> >>> without a lot of work.
> >>
> >> Will shadowing work on the new file system if disk sizes are kept
> >> at 2TB or smaller ?
> >>
> >> If so, is it viable to convert the signed values to unsigned values
> >> to at least get the maximum disk size supported up to 4TB ?
> >
> > I'd guess, in such an environment, in Macro-32, that signed intergers 
> > are not used.  But what do I know.
> >
> 
> Actually, there has traditionally been a problem with signed integers
> used in some data structures where unsigned integers would be better.
> 

There has been a discussion about using signed integers in the LLVM source with a limited exceptions for unsigned when discussion bitfields.

For the junior language lawyers in the crowd, go watch

https://www.youtube.com/watch?v=wvtFGa6XJDU

If that 6 minutes wasn't enough, then go spend an entire hour with

https://www.youtube.com/watch?v=yG1OZ69H_-o

(skip to 33:26 for a real fun bug)

I'll respond to the OP's questions next.