On 2012-09-05 20:32, Nick Sabalausky wrote:

> Meh, it's not physically *possible* to write software that doesn't
> infringe on some software patent or another. If they're worried about
> infringing on some patent, they may as well just stop writing code
> altogether.

Yeah, I agree.

-- 
/Jacob Carlborg

Jacob Carlborg wrote:
> On 2012-09-05 14:32, Piotr Szturmaj wrote:
>> angel wrote:
>>> Check out LDC web-page ...
>>> In github we see they're up to date - merging 2.060
>>
>> What's the status of SEH on Windows?
>
> It's not looking good, at least not for 32bit. Borland has some kind of
> patent on something related to SEH. Due to this, the LLVM/Clang
> developers won't implement SEH for Windows.

AFAIK FreePascal, which is somewhat compatible with Borland's Delphi, has SEH support on Windows. I wonder how they solved patent issues.

On 05/09/12 19:32, Nick Sabalausky wrote:
> Meh, it's not physically *possible* to write software that doesn't
> infringe on some software patent or another. If they're worried about
> infringing on some patent, they may as well just stop writing code
> altogether.

Yes, but there's a difference between writing software knowing that you're probably going to come up against an infringement problem somewhere, versus writing software that will infringe a known patent.

On 05/09/12 20:12, Jacob Carlborg wrote:
> On 2012-09-05 14:32, Piotr Szturmaj wrote:
>> angel wrote:
>>> Check out LDC web-page ...
>>> In github we see they're up to date - merging 2.060
>>
>> What's the status of SEH on Windows?
>
> It's not looking good, at least not for 32bit. Borland has some kind of
> patent on something related to SEH. Due to this, the LLVM/Clang
> developers won't implement SEH for Windows.

That sounds paranoid to me. I believe the patent is essentially a workaround for the absence of thread-local variables on 16-bit Windows.

In D we have working thread-local variables even on Windows versions that don't support them, thanks to Rainer's brilliant work. Which gives very many ways of implementing exception handling on Windows.

On 2012-09-13 16:40, Don Clugston wrote:

> That sounds paranoid to me. I believe the patent is essentially a
> workaround for the absence of thread-local variables on 16-bit Windows.

Yeah, I know.

> In D we have working thread-local variables even on Windows versions
> that don't support them, thanks to Rainer's brilliant work. Which gives
> very many ways of implementing exception handling on Windows.

So DMD uses its own implementation of SEH? Is it compatible with the one used by Microsoft?

-- 
/Jacob Carlborg

On Sep 13, 2012, at 11:21 AM, Jacob Carlborg <doob@me.com> wrote:

> On 2012-09-13 16:40, Don Clugston wrote:
> 
>> In D we have working thread-local variables even on Windows versions that don't support them, thanks to Rainer's brilliant work. Which gives very many ways of implementing exception handling on Windows.
> 
> So DMD uses its own implementation of SEH? Is it compatible with the one used by Microsoft?

Digital Mars has a redistribution license for MSVC libraries, so perhaps it sidesteps the issue?

On Wednesday, 5 September 2012 at 14:42:25 UTC, Russel Winder wrote:
> It appears that the LDC website is in need of an overhaul.

No kidding. I thought the project was completely dead. Updating a web page is usually very easy, strange ...

-rt

On Fri, 2012-09-14 at 00:52 +0200, Rob T wrote:
> On Wednesday, 5 September 2012 at 14:42:25 UTC, Russel Winder wrote:
> > It appears that the LDC website is in need of an overhaul.
> 
> No kidding. I thought the project was completely dead. Updating a web page is usually very easy, strange ...

Rather usefully, LDC seems far from dead, it builds and works very well. It just seems that LDC-related effort only relates to the Git codebase repository! ;-)

-- 
Russel. ============================================================================= Dr Russel Winder      t: +44 20 7585 2200   voip: sip:russel.winder@ekiga.net 41 Buckmaster Road    m: +44 7770 465 077   xmpp: russel@winder.org.uk London SW11 1EN, UK   w: www.russel.org.uk  skype: russel_winder

On 2012-09-13 16:40, Don Clugston wrote:

> That sounds paranoid to me. I believe the patent is essentially a
> workaround for the absence of thread-local variables on 16-bit Windows.

BTW, regardless if it's under a patent or not we can't force them to implement SEH in LLVM.

The description of the patent just says that it's about a system for compilers that allow support runtime exception handling.

-- 
/Jacob Carlborg

On Fri, 2012-09-14 at 08:25 +0200, Jacob Carlborg wrote:
> On 2012-09-13 16:40, Don Clugston wrote:
> 
> > That sounds paranoid to me. I believe the patent is essentially a workaround for the absence of thread-local variables on 16-bit Windows.
> 
> BTW, regardless if it's under a patent or not we can't force them to implement SEH in LLVM.
> 
> The description of the patent just says that it's about a system for compilers that allow support runtime exception handling.

Presumably the patent (if it exists and is enforceable) only applies in the USA. So for the rest of the world there is absolutely no need to worry about it. Only people in the USA have to worry. Which means one version for in the USA and one version for everywhere else?

-- 
Russel. ============================================================================= Dr Russel Winder      t: +44 20 7585 2200   voip: sip:russel.winder@ekiga.net 41 Buckmaster Road    m: +44 7770 465 077   xmpp: russel@winder.org.uk London SW11 1EN, UK   w: www.russel.org.uk  skype: russel_winder