Re: How to challenge a GCC patch?

["Randy Brukardt" <randy@rrsoftware.com>]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 3310 bytes --]

"J-P. Rosen" <rosen@adalog.fr> wrote in message 
news:sjerti$5s1$1@dont-email.me...
> Le 02/10/2021 à 11:34, Randy Brukardt a écrit :
>
>> No, the existing part also had major problems. Many of the terms were 
>> never
>> defined, many possible effects were missing from the various lists of
>> results, and loads of other things as well.
> Huh? ASIS uses the terms of, and as defined in, the LRM. And I never hit a 
> "missing result". Curious to see what you are aluding too.

Go back and read many of the SI99s. I'm not going to waste time doing it 
again.

>>>> The entire model of ASIS doesn't make much sense for static analysis
>>>> purposes, it's way too focused on syntax rather than semantics.
> Yes, it is a description of the syntactic. Where else can you start from?

The semantics, of course. You have a list of entities in a scope. The only 
interesting thing from the source code is the line/position of the 
declaration and/or use. The details of the source are irrelevant (such as 
whether optional keywords are given).

>>> It is a exact image of the program, from which you can derive all
>>> information you need. Some higher level queries are needed, but they can
>>> be provided as secondary queries or added to the standard.
>>
>> That's exactly the problem. You start with the source code, which is way 
>> too
>> low a level for any useful analysis. At most, you want a simple 
>> connection
>> to the source in the semantic information, not trying to preserve every
>> punctuation mark and comment. (Janus/Ada discards all of that stuff as 
>> soon
>> as parsing succeeds.) If you need to refer to the original source, say 
>> for
>> error handling purposes, then do that, but don't waste vast amounts of 
>> space
>> and time trying to keep loads of irrelevant material.
> The liaison to the source is not fundamental to any serious analysis, and 
> anyway you are not required to provide it.

It's a basic part of the ASIS definition, and I fail to see how you could do 
much without it. You have to keep nonsense such as whether someone specified 
"in" or a matching id after "end". A waste of time and space. You also have 
to keep the entire program in a tree form, which is also dubious (Janus/Ada 
certainly does not do that).

>> No sane compiler (validated or not) keeps all of the irrelevant syntactic
>> detail required by ASIS. It ends up getting reconstructed solely for the 
>> use
>> of ASIS, and how a rarely used interface is somehow more reliable escapes
>> me.
> Except for the not-required text interface, I see very few of these 
> "irrelevant syntactic details". OK, you have to keep a boolean to remember 
> if "end" is followed by a name. Big deal...

It is a big deal when multiplied by dozens of such things. And that 
information has to be carried through from the front end -- we discard 
everything possible from the front end, as it cuts the I/O load and 
effectively doubled the speed of the compiler. Maybe doesn't matter as much 
these days, but back then it was a difference between running the ACVC in 5 
days or 10 days.

On the rest, we're going to agree to disagree. Don't expect any support from 
me for doing anything with ASIS in the ARG.

                         Randy.