DDE28

Well, the show is tomorrow. Have fun everybody who is going.

Meanwhile, any news on what DDE28 adds/changes? Does objasm no longer attempt to parse UAL/ARMv7 instructions when building code for (circa) ARMv4 in non-UAL mode? :-P

The examples should all build this time around :P

Just downloaded and installed DDE28. The installation is a bit messy.
The Readme recommends deleting, or moving aside the current version and installing afresh. This hits at least 6 directories, some of which are used by other software, eg Documents.

It is difficult to see what has changed as directories and files have been restamped.
Is there a log of changes available? I have looked on the ROOL site but have not succeded in finding anything.

Just an observation – not a complaint.

Mods for Nerds

Comparing my versions of stuff with the provided changelog:

• AMU
  • Failure to read a time stamp, for example because the underlying filing system
    reported an error, no longer cause an immediate error. The error will only be
    reported if the dependency is required by a target. This avoids a problem where
    the makefile generates the name using macros, but the macros are unset for some
    targets, and are therefore invalid filenames.

• CC
  • Module relocation offsets are now annotated correctly in the disassembly produced by calling the compiler with switches -zM -S at the same time.
  • Following an inlined memset() of structures where a loop is deemed necessary, and the structure size can’t be expressed as an ARM immediate, any subsequent use of a pointer to the structure is now correctly based.
  • Fixed a bug where a mask operation comprising a shift left followed by a shift right, used as an array index (which needs another shift left to compute the byte offset) was incorrectly optimised.
  • The scheduler will now track LDM instructions which are indirectly reading the stack via another base register. This ensures the instructions aren’t scheduled to a point before the write of the data which they want to load.
  • Integer expressions of the form a = b - (c * d) will now produce the MLS instruction when architecture 6T2, or greater, is selected.
  • The inline assembler similarly now accepts MLS. During transformation if the target architecture doesn’t support it, a MUL/SUB pair will be substituted.
  • Fixed a CSE bug where an inlined function whose arguments comprised a string that was short enough to fit in the literal pool, and where that same string had pointer arithmetic performed on it, the resulting ADR instruction output would have the top 16 bits of the instruction set to 0xFFFF, which is an illegal opcode.
  • Fixed a CSE bug where the 2nd result from a built in function (such as div/rem) was needed and where the function could be evaluated at compile time. The compiler would instead keep track of the 1st result so a later use of the result would incorrectly substitute some earlier value. Note that run time evaluated functions were not affected.
  • The -cpu switch now recognises additional architectures 2a/3g/5tewmmx/5tewmmx2/6z for consistency with ObjAsm. They are otherwise treated as synonyms for 2/3/5tej/5tej/6k respectively.

(and… still uses FPE maths for floats instead of VFP…)

• CMHG
  • -nothing-

• DDT
  • When run in a fully paletted 256 colour mode, DDT’s window furniture is now rendered correctly rather than in blue/yellow.
  • Support for single stepping ARMv6T2’s MLS instruction added.

• link
  • -nothing-

• objasm
  • The mnemonic for CRC32 has been corrected to match the ARM ARM and the above tables, previously the assembler was looking for a syntax from a draft specification.

I’ve not looked at the build process, headers, documentation, examples, libraries, or the other tools such as libfile or make…

Is there a log of changes available? I have looked on the ROOL site but have not succeded in finding anything.

Changelogs are in Documents.DDE.ChangeLogs.

Not a file-by-file changelog however, which would be very atypical for any software I know.

Thanks Rick for the summary – as you suggest – mods for nerds – not a lot to see, move along please.

And thanks for the pointer to the change logs. I still haven’t got used to thinking of Acorn C/C++ as DDE.

Thanks Rick for the summary

The show news has less gobbledegook:

As if that wasn’t enough to delight and amaze, we’ll also be launching DDE 28, the official development tools for RISC OS. Aside from new features and bug fixes this edition includes electronic copies of the Desktop Development Tools box set – handy for quickly searching and clicking between books, and also if you’re short on shelf space for the physical copies.

Aside from new features and bug fixes this edition includes electronic copies of the Desktop Development Tools box set

Not sure what this means. If it refers to manuals then there is not much difference from the previous release DDE27.

not a lot to see, move along please.

It’s a compiler suite. What did you expect? Green icons?

Not sure what this means. If it refers to manuals then there is not much difference from the previous release DDE27.

I’d need to turn the RISC OS machine on to check, but did DDE27 only have the pre-update manuals included in PDF format?

It’s a compiler suite. What did you expect?

Easy rounded corners?
Seriously, I suspect you may find a diff on various macros will turn up a few under the bonnet changes/fixes.

The latest documentation has been available with the DDE for at least the last few years. I think that sentence was there to serve as a reminder, not an indicator of a new feature (it can be broken down into three components; “new features”, “bug fixes”, “electronic copies of manuals”)

I suspect you may find a diff on various macros will turn up a few under the bonnet changes/fixes.

That’s sort of what I was hinting at – that the changes would be mainly low level nerdy stuff rather than, say, prettier icons or indeed rounded corners… Bug fixes to the tools, various architectural enhancements etc…

If it refers to manuals then there is not much difference from the previous release DDE27.

Using my advanced powers of actually opening the documents in Adobe Acrobat I can see

while I’ve not compared them page by page, they’re fatter if nothing else.

Issue 4, Apr 2015, 452 pages

Gee, and I thought it was bad that my Pi didn’t switch to winter time by itself¹, somebody’s clock is two and a half years out! ;-)

¹ Using CET/CEST hacked into UK territory, not related to the other timezone issues…I don’t think?

So the DDE28 versions are the same as the printed ones. It’s odd that 27 included the previous editions; I’d never noticed that before.

Does DDE28 have an error flag to suppress the error ‘Error: declaration with no effect’ you get with code like

struct
{
    int a;
    void* b;
};

union
{
    int a;
    void* b;
}

I know that this isn’t c99 but working around it is messy.

Uh… If there’s no name, how do you refer to it?

You just refer to the elements directly. The example above would fail because there are 2 a variables and 2 b variables at the same depth. You see it used like this.

typedef struct
{
    struct 
    {
        int a, b;
    };
    union 
    {
        int c; 
        void* d;
    };
} test_t;

int main(void)
{
    test_t t;
    t.a = 1;
    t.b = 2;
    t.c = 3;
    t.d = NULL;
    return 0;
}

It’s just used to reduce the length of the path to a variable.

I haven’t seen it said explicitly, but having the changelogs along with DDE28 is useless unless a person owns DDE28.

Yes, it is up to marketing to make you buy it or upgrade. Only in the latter case are the changelogs relevant. So it’s up to the marketing guys to explain why DDE28 is better than DDE27. The wheels are in motion.

Even in this case the change logs are incomplete. I posted parts of the text file change logs above, but you will note that Sprow mentions a change that doesn’t appear to be noted in the DDE release info. I suspect he has access to the compiler sources so will be able to see the version by version CVS changes, which are inaccessible to us. We just get the edited highlights. ;-)

At any rate, some info on why DDE28 is better than DDE27 is, as you say, just basic marketing.

I just wish it was open sourced and free to download to encourage developers. I have zero incentive to upgrade from DDE24 when I have to play full price for it again and I’ve stopped using DDE24 because the code output doesn’t appear to be fully ARMv7 compatible.

I tend to agree but don’t you at least qualify for the “Half price DDE Download”? If not, it would seem reasonable to offer existing owners generous, non time-limited discounts for upgrades.

There’s a six month time restriction on the “half price” offer, but I don’t believe updates come out within that timeframe, so it’s a bit cheeky really.