There are some great patches in Bugzilla at the moment. A couple of
infamous forward reference bugs
that have been hurting Phobos development. Also, all known compiler
segfaults have patches in this list.
The first block of patches is the most important.

3935 opBinary is instantiated with "="
190 Cannot forward reference typedef/alias in default value for
function parameter
  also fixes
  3979 Order-of-compilation and forward reference errors
4652 Compiler hangs on template with zero-length tuple and another argument
   also fixes
   4676 Overload resolution rejects valid code when mixing variadics,
non-variadics
4564 ICE on undefined variable in foreach over 0 .. undef
3493 Segfault(cast.c) Forward reference with type inference, D1 only.

2511 Covariant return type doesn't work with circular import
2716 Confusion of auto and scope as the class attribute
4278 allow inlining of super calls (undo limitations of bug3500's fix)
2477 Trailing comma in array literal sometimes accepted, sometimes not

4291 Pure functions cannot access mixed in variables
4177 __ctfe can't be used in pure functions

<Haven't tested this one as it's Linux only, but looks impressive>
4620 C++ constructor and template mangling, C++ ABI patch
 it should also fix this next one, though I have included a patch for
3046 anyway
 3046 Segfault with C++ static variable (Linux only)

Don Clugston wrote:
> There are some great patches in Bugzilla at the moment.

[...]

>
> 2511 Covariant return type doesn't work with circular import
>
> 

The patch for this fails test23.d.

That wraps it up for this list, except for 2511. Thanks everyone who submitted these patches.

Don Clugston wrote:
> There are some great patches in Bugzilla at the moment. A couple of
> infamous forward reference bugs
> that have been hurting Phobos development. Also, all known compiler
> segfaults have patches in this list.
> The first block of patches is the most important.
>
> 3935 opBinary is instantiated with "="
> 190 Cannot forward reference typedef/alias in default value for
> function parameter
>   also fixes
>   3979 Order-of-compilation and forward reference errors
> 4652 Compiler hangs on template with zero-length tuple and another argument
>    also fixes
>    4676 Overload resolution rejects valid code when mixing variadics,
> non-variadics
> 4564 ICE on undefined variable in foreach over 0 .. undef
> 3493 Segfault(cast.c) Forward reference with type inference, D1 only.
>
> 2511 Covariant return type doesn't work with circular import
> 2716 Confusion of auto and scope as the class attribute
> 4278 allow inlining of super calls (undo limitations of bug3500's fix)
> 2477 Trailing comma in array literal sometimes accepted, sometimes not
>
> 4291 Pure functions cannot access mixed in variables
> 4177 __ctfe can't be used in pure functions
>
> <Haven't tested this one as it's Linux only, but looks impressive>
> 4620 C++ constructor and template mangling, C++ ABI patch
>  it should also fix this next one, though I have included a patch for
> 3046 anyway
>  3046 Segfault with C++ static variable (Linux only)
>
> 

Hi,

I have added a fix to the patch for 2511, that makes your test case compile. Have you tried it? I thought you'd see the stream of bugzilla changes from the bugs-mailing list...

Rainer

Walter Bright wrote:
> That wraps it up for this list, except for 2511. Thanks everyone who submitted these patches.
>
> Don Clugston wrote:
>> There are some great patches in Bugzilla at the moment. A couple of
>> infamous forward reference bugs
>> that have been hurting Phobos development. Also, all known compiler
>> segfaults have patches in this list.
>> The first block of patches is the most important.
>>
>> 3935 opBinary is instantiated with "="
>> 190 Cannot forward reference typedef/alias in default value for
>> function parameter
>>   also fixes
>>   3979 Order-of-compilation and forward reference errors
>> 4652 Compiler hangs on template with zero-length tuple and another
>> argument
>>    also fixes
>>    4676 Overload resolution rejects valid code when mixing variadics,
>> non-variadics
>> 4564 ICE on undefined variable in foreach over 0 .. undef
>> 3493 Segfault(cast.c) Forward reference with type inference, D1 only.
>>
>> 2511 Covariant return type doesn't work with circular import
>> 2716 Confusion of auto and scope as the class attribute
>> 4278 allow inlining of super calls (undo limitations of bug3500's fix)
>> 2477 Trailing comma in array literal sometimes accepted, sometimes not
>>
>> 4291 Pure functions cannot access mixed in variables
>> 4177 __ctfe can't be used in pure functions
>>
>> <Haven't tested this one as it's Linux only, but looks impressive>
>> 4620 C++ constructor and template mangling, C++ ABI patch
>>  it should also fix this next one, though I have included a patch for
>> 3046 anyway
>>  3046 Segfault with C++ static variable (Linux only)
>>
>> 
> _______________________________________________
> dmd-internals mailing list
> dmd-internals at puremagic.com
> http://lists.puremagic.com/mailman/listinfo/dmd-internals
>

No, haven't tried it yet.

Rainer Schuetze wrote:
> Hi,
>
> I have added a fix to the patch for 2511, that makes your test case compile. Have you tried it? I thought you'd see the stream of bugzilla changes from the bugs-mailing list...
>
> Rainer
>
> Walter Bright wrote:
>> That wraps it up for this list, except for 2511. Thanks everyone who submitted these patches.
>>
>> Don Clugston wrote:
>>> There are some great patches in Bugzilla at the moment. A couple of
>>> infamous forward reference bugs
>>> that have been hurting Phobos development. Also, all known compiler
>>> segfaults have patches in this list.
>>> The first block of patches is the most important.
>>>
>>> 3935 opBinary is instantiated with "="
>>> 190 Cannot forward reference typedef/alias in default value for
>>> function parameter
>>>   also fixes
>>>   3979 Order-of-compilation and forward reference errors
>>> 4652 Compiler hangs on template with zero-length tuple and another
>>> argument
>>>    also fixes
>>>    4676 Overload resolution rejects valid code when mixing variadics,
>>> non-variadics
>>> 4564 ICE on undefined variable in foreach over 0 .. undef
>>> 3493 Segfault(cast.c) Forward reference with type inference, D1 only.
>>>
>>> 2511 Covariant return type doesn't work with circular import
>>> 2716 Confusion of auto and scope as the class attribute
>>> 4278 allow inlining of super calls (undo limitations of bug3500's fix)
>>> 2477 Trailing comma in array literal sometimes accepted, sometimes not
>>>
>>> 4291 Pure functions cannot access mixed in variables
>>> 4177 __ctfe can't be used in pure functions
>>>
>>> <Haven't tested this one as it's Linux only, but looks impressive>
>>> 4620 C++ constructor and template mangling, C++ ABI patch
>>>  it should also fix this next one, though I have included a patch for
>>> 3046 anyway
>>>  3046 Segfault with C++ static variable (Linux only)
>>>
>>> 
>> _______________________________________________
>> dmd-internals mailing list
>> dmd-internals at puremagic.com
>> http://lists.puremagic.com/mailman/listinfo/dmd-internals
>>
>
> _______________________________________________
> dmd-internals mailing list
> dmd-internals at puremagic.com
> http://lists.puremagic.com/mailman/listinfo/dmd-internals
>
>

Now it fails elsewhere, see bugzilla 2511 for test case.

Rainer Schuetze wrote:
> Hi,
>
> I have added a fix to the patch for 2511, that makes your test case compile. Have you tried it? I thought you'd see the stream of bugzilla changes from the bugs-mailing list...
>
> Rainer
>
> Walter Bright wrote:
>> That wraps it up for this list, except for 2511. Thanks everyone who submitted these patches.
>>
>> Don Clugston wrote:
>>> There are some great patches in Bugzilla at the moment. A couple of
>>> infamous forward reference bugs
>>> that have been hurting Phobos development. Also, all known compiler
>>> segfaults have patches in this list.
>>> The first block of patches is the most important.
>>>
>>> 3935 opBinary is instantiated with "="
>>> 190 Cannot forward reference typedef/alias in default value for
>>> function parameter
>>>   also fixes
>>>   3979 Order-of-compilation and forward reference errors
>>> 4652 Compiler hangs on template with zero-length tuple and another
>>> argument
>>>    also fixes
>>>    4676 Overload resolution rejects valid code when mixing variadics,
>>> non-variadics
>>> 4564 ICE on undefined variable in foreach over 0 .. undef
>>> 3493 Segfault(cast.c) Forward reference with type inference, D1 only.
>>>
>>> 2511 Covariant return type doesn't work with circular import
>>> 2716 Confusion of auto and scope as the class attribute
>>> 4278 allow inlining of super calls (undo limitations of bug3500's fix)
>>> 2477 Trailing comma in array literal sometimes accepted, sometimes not
>>>
>>> 4291 Pure functions cannot access mixed in variables
>>> 4177 __ctfe can't be used in pure functions
>>>
>>> <Haven't tested this one as it's Linux only, but looks impressive>
>>> 4620 C++ constructor and template mangling, C++ ABI patch
>>>  it should also fix this next one, though I have included a patch for
>>> 3046 anyway
>>>  3046 Segfault with C++ static variable (Linux only)
>>>
>>> 
>> _______________________________________________
>> dmd-internals mailing list
>> dmd-internals at puremagic.com
>> http://lists.puremagic.com/mailman/listinfo/dmd-internals
>>
>
> _______________________________________________
> dmd-internals mailing list
> dmd-internals at puremagic.com
> http://lists.puremagic.com/mailman/listinfo/dmd-internals
>
>

Hmm, I don't see a difference to the previous testcase. Unfortunately, dsource seems down, so I cannot get the latest dmd revision.

Rainer


Walter Bright wrote:
> Now it fails elsewhere, see bugzilla 2511 for test case.
>
> Rainer Schuetze wrote:
>> Hi,
>>
>> I have added a fix to the patch for 2511, that makes your test case compile. Have you tried it? I thought you'd see the stream of bugzilla changes from the bugs-mailing list...
>>
>> Rainer
>>
>> Walter Bright wrote:
>>> That wraps it up for this list, except for 2511. Thanks everyone who submitted these patches.
>>>
>>> Don Clugston wrote:
>>>> There are some great patches in Bugzilla at the moment. A couple of
>>>> infamous forward reference bugs
>>>> that have been hurting Phobos development. Also, all known compiler
>>>> segfaults have patches in this list.
>>>> The first block of patches is the most important.
>>>>
>>>> 3935 opBinary is instantiated with "="
>>>> 190 Cannot forward reference typedef/alias in default value for
>>>> function parameter
>>>>   also fixes
>>>>   3979 Order-of-compilation and forward reference errors
>>>> 4652 Compiler hangs on template with zero-length tuple and another
>>>> argument
>>>>    also fixes
>>>>    4676 Overload resolution rejects valid code when mixing variadics,
>>>> non-variadics
>>>> 4564 ICE on undefined variable in foreach over 0 .. undef
>>>> 3493 Segfault(cast.c) Forward reference with type inference, D1 only.
>>>>
>>>> 2511 Covariant return type doesn't work with circular import
>>>> 2716 Confusion of auto and scope as the class attribute
>>>> 4278 allow inlining of super calls (undo limitations of bug3500's fix)
>>>> 2477 Trailing comma in array literal sometimes accepted, sometimes not
>>>>
>>>> 4291 Pure functions cannot access mixed in variables
>>>> 4177 __ctfe can't be used in pure functions
>>>>
>>>> <Haven't tested this one as it's Linux only, but looks impressive>
>>>> 4620 C++ constructor and template mangling, C++ ABI patch
>>>>  it should also fix this next one, though I have included a patch for
>>>> 3046 anyway
>>>>  3046 Segfault with C++ static variable (Linux only)
>>>>
>>>> 
>>> _______________________________________________
>>> dmd-internals mailing list
>>> dmd-internals at puremagic.com
>>> http://lists.puremagic.com/mailman/listinfo/dmd-internals
>>>
>>
>> _______________________________________________
>> dmd-internals mailing list
>> dmd-internals at puremagic.com
>> http://lists.puremagic.com/mailman/listinfo/dmd-internals
>>
>>
> _______________________________________________
> dmd-internals mailing list
> dmd-internals at puremagic.com
> http://lists.puremagic.com/mailman/listinfo/dmd-internals
>

I think I've got it fixed, running the test suite now.

Rainer Schuetze wrote:
>
> Hmm, I don't see a difference to the previous testcase. Unfortunately, dsource seems down, so I cannot get the latest dmd revision.
>
> Rainer
>
>
> Walter Bright wrote:
>> Now it fails elsewhere, see bugzilla 2511 for test case.
>>
>> Rainer Schuetze wrote:
>>> Hi,
>>>
>>> I have added a fix to the patch for 2511, that makes your test case compile. Have you tried it? I thought you'd see the stream of bugzilla changes from the bugs-mailing list...
>>>
>>> Rainer
>>>
>>> Walter Bright wrote:
>>>> That wraps it up for this list, except for 2511. Thanks everyone who submitted these patches.
>>>>
>>>> Don Clugston wrote:
>>>>> There are some great patches in Bugzilla at the moment. A couple of
>>>>> infamous forward reference bugs
>>>>> that have been hurting Phobos development. Also, all known compiler
>>>>> segfaults have patches in this list.
>>>>> The first block of patches is the most important.
>>>>>
>>>>> 3935 opBinary is instantiated with "="
>>>>> 190 Cannot forward reference typedef/alias in default value for
>>>>> function parameter
>>>>>   also fixes
>>>>>   3979 Order-of-compilation and forward reference errors
>>>>> 4652 Compiler hangs on template with zero-length tuple and another
>>>>> argument
>>>>>    also fixes
>>>>>    4676 Overload resolution rejects valid code when mixing variadics,
>>>>> non-variadics
>>>>> 4564 ICE on undefined variable in foreach over 0 .. undef
>>>>> 3493 Segfault(cast.c) Forward reference with type inference, D1 only.
>>>>>
>>>>> 2511 Covariant return type doesn't work with circular import
>>>>> 2716 Confusion of auto and scope as the class attribute
>>>>> 4278 allow inlining of super calls (undo limitations of bug3500's
>>>>> fix)
>>>>> 2477 Trailing comma in array literal sometimes accepted, sometimes
>>>>> not
>>>>>
>>>>> 4291 Pure functions cannot access mixed in variables
>>>>> 4177 __ctfe can't be used in pure functions
>>>>>
>>>>> <Haven't tested this one as it's Linux only, but looks impressive>
>>>>> 4620 C++ constructor and template mangling, C++ ABI patch
>>>>>  it should also fix this next one, though I have included a patch for
>>>>> 3046 anyway
>>>>>  3046 Segfault with C++ static variable (Linux only)
>>>>>
>>>>> 
>>>> _______________________________________________
>>>> dmd-internals mailing list
>>>> dmd-internals at puremagic.com
>>>> http://lists.puremagic.com/mailman/listinfo/dmd-internals
>>>>
>>>
>>> _______________________________________________
>>> dmd-internals mailing list
>>> dmd-internals at puremagic.com
>>> http://lists.puremagic.com/mailman/listinfo/dmd-internals
>>>
>>>
>> _______________________________________________
>> dmd-internals mailing list
>> dmd-internals at puremagic.com
>> http://lists.puremagic.com/mailman/listinfo/dmd-internals
>>
>
> _______________________________________________
> dmd-internals mailing list
> dmd-internals at puremagic.com
> http://lists.puremagic.com/mailman/listinfo/dmd-internals
>
>