Branch: refs/heads/master
Home:   https://github.com/D-Programming-Language/phobos

Commit: 9ecf947290f1149e1b062a75dafd30326697fe00
    https://github.com/D-Programming-Language/phobos/commit/9ecf947290f1149e1b062a75dafd30326697fe00
Author: Don Clugston <dclugston at googlemail.com>
Date:   2011-02-23 (Wed, 23 Feb 2011)

Changed paths:
  M std/bigint.d
  M std/internal/math/biguintcore.d

Log Message:
-----------
5568 A problem with BigInt modulus

This is quite subtle. The recursive division algorithm has a possibility of one
bit of overflow in the second recursive call, in cases where the top quarter of
the dividend is identical to the top half of the divisor.
The description of the recursive division algorithm in "Modern Computer
Arithmetic" 0.5.9 is rather vague about this (which is why I missed it
originally, though the test case does cause assert failure in biguint).
Unfortunately it requires fairly substantial changes to the function, which
makes the algorithm a lot less elegant. But I have managed to implement it
without any change to the basic division algorithm (which is very fast).


Commit: 9fa7393417015257342401860ea4a2fa4f55db41
    https://github.com/D-Programming-Language/phobos/commit/9fa7393417015257342401860ea4a2fa4f55db41
Author: Walter Bright <walter at walterbright.com>
Date:   2011-02-23 (Wed, 23 Feb 2011)

Changed paths:
  M std/bigint.d
  M std/internal/math/biguintcore.d

Log Message:
-----------
Merge branch 'bigint5568' of https://github.com/donc/phobos into donc-bigint5568

Looks like the biguintcore unit tests are pissed off, across the board (well, I'm assuming freebsd will eventually show the same problem, it's not caught up yet).

On 2/23/2011 1:14 PM, noreply at github.com wrote:
> Branch: refs/heads/master Home:   https://github.com/D-Programming-Language/phobos
> 
> Commit: 9ecf947290f1149e1b062a75dafd30326697fe00 https://github.com/D-Programming-Language/phobos/commit/9ecf947290f1149e1b062a75dafd30326697fe00 Author: Don Clugston <dclugston at googlemail.com> Date:   2011-02-23 (Wed, 23 Feb 2011)
> 
> Changed paths: M std/bigint.d M std/internal/math/biguintcore.d
> 
> Log Message: ----------- 5568 A problem with BigInt modulus
> 
> This is quite subtle. The recursive division algorithm has a possibility of one bit of overflow in the second recursive call, in cases where the top quarter of the dividend is identical to the top half of the divisor. The description of the recursive division algorithm in "Modern Computer Arithmetic" 0.5.9 is rather vague about this (which is why I missed it originally, though the test case does cause assert failure in biguint). Unfortunately it requires fairly substantial changes to the function, which makes the algorithm a lot less elegant. But I have managed to implement it without any change to the basic division algorithm (which is very fast).
> 
> 
> Commit: 9fa7393417015257342401860ea4a2fa4f55db41 https://github.com/D-Programming-Language/phobos/commit/9fa7393417015257342401860ea4a2fa4f55db41 Author: Walter Bright <walter at walterbright.com> Date:   2011-02-23 (Wed, 23 Feb 2011)
> 
> Changed paths: M std/bigint.d M std/internal/math/biguintcore.d
> 
> Log Message: ----------- Merge branch 'bigint5568' of https://github.com/donc/phobos into donc-bigint5568
> 
> 
> _______________________________________________ phobos mailing list phobos at puremagic.com http://lists.puremagic.com/mailman/listinfo/phobos

It's that bloddy -release in the makefile has crept in again into my branch (and maybe into Walter's as well). Sorry, I'm not going to have ANY time for D until next week. Please revert the changes, or comment out the failing tests I added to std.bigint.


On 23 February 2011 22:41, Brad Roberts <braddr at puremagic.com> wrote:
> Looks like the biguintcore unit tests are pissed off, across the board (well, I'm assuming freebsd will eventually show the same problem, it's not caught up yet).
>
> On 2/23/2011 1:14 PM, noreply at github.com wrote:
>> Branch: refs/heads/master Home: ? https://github.com/D-Programming-Language/phobos
>>
>> Commit: 9ecf947290f1149e1b062a75dafd30326697fe00 https://github.com/D-Programming-Language/phobos/commit/9ecf947290f1149e1b062a75dafd30326697fe00 Author: Don Clugston <dclugston at googlemail.com> Date: ? 2011-02-23 (Wed, 23 Feb 2011)
>>
>> Changed paths: M std/bigint.d M std/internal/math/biguintcore.d
>>
>> Log Message: ----------- 5568 A problem with BigInt modulus
>>
>> This is quite subtle. The recursive division algorithm has a possibility of one bit of overflow in the second recursive call, in cases where the top quarter of the dividend is identical to the top half of the divisor. The description of the recursive division algorithm in "Modern Computer Arithmetic" 0.5.9 is rather vague about this (which is why I missed it originally, though the test case does cause assert failure in biguint). Unfortunately it requires fairly substantial changes to the function, which makes the algorithm a lot less elegant. But I have managed to implement it without any change to the basic division algorithm (which is very fast).
>>
>>
>> Commit: 9fa7393417015257342401860ea4a2fa4f55db41 https://github.com/D-Programming-Language/phobos/commit/9fa7393417015257342401860ea4a2fa4f55db41 Author: Walter Bright <walter at walterbright.com> Date: ? 2011-02-23 (Wed, 23 Feb 2011)
>>
>> Changed paths: M std/bigint.d M std/internal/math/biguintcore.d
>>
>> Log Message: ----------- Merge branch 'bigint5568' of https://github.com/donc/phobos into donc-bigint5568
>>
>>
>> _______________________________________________ phobos mailing list phobos at puremagic.com http://lists.puremagic.com/mailman/listinfo/phobos
>
> _______________________________________________
> phobos mailing list
> phobos at puremagic.com
> http://lists.puremagic.com/mailman/listinfo/phobos
>