«

Hi,

I'm using Boost::uBLAS::mapped_matrix for working with sparse matrices.
At the moment, the maximum dimension of mapped_matrix is 65535x65535
since std::size_t compiles as unsigned int (UInt32).

I'm looking for a way of increasing this limit. The best way I can
think of would be to ensure that size_t uses a bigger (64-bit) type
rather than the current UInt32.

Can I force std::size_t to be defined as unsigned long? How/where?
I'm using VC 2005 (VC on Windows.

Thanks for your support.

Cheers,
Greg

wrote:
> I'm using Boost::uBLAS::mapped_matrix for working with sparse
> matrices. At the moment, the maximum dimension of mapped_matrix is
> 65535x65535 since std::size_t compiles as unsigned int (UInt32).
>
> I'm looking for a way of increasing this limit. The best way I can
> think of would be to ensure that size_t uses a bigger (64-bit) type
> rather than the current UInt32.
>
> Can I force std::size_t to be defined as unsigned long?

Usually that's compiler-dependent. But what would be the point?
'unsigned long' is 32 bits on Windows.

> How/where?
> I'm using VC 2005 (VC on Windows.

You need to ask in 'microsoft.public.vc.language' or other relevant
newsgroup. My suspicion is that you need to go 64 bits.

V
--
Please remove capital 'A's when replying by e-mail
I do not respond to top-posted replies, please don't ask

Can I use template arguments to force the use of std::map<double,
unsigned long> ?


wrote:
> Hi,
>
> I'm using Boost::uBLAS::mapped_matrix for working with sparse matrices.
> At the moment, the maximum dimension of mapped_matrix is 65535x65535
> since std::size_t compiles as unsigned int (UInt32).
>
> I'm looking for a way of increasing this limit. The best way I can
> think of would be to ensure that size_t uses a bigger (64-bit) type
> rather than the current UInt32.
>
> Can I force std::size_t to be defined as unsigned long? How/where?
> I'm using VC 2005 (VC on Windows.
>
> Thanks for your support.
>
> Cheers,
> Greg

Victor Bazarov wrote:
> > Can I force std::size_t to be defined as unsigned long?
>
> Usually that's compiler-dependent. But what would be the point?
> 'unsigned long' is 32 bits on Windows.

Sorry, I meant to say 'unsigned long long' which is equivalent to
__uint64

Victor Bazarov wrote:
> > How/where?
> > I'm using VC 2005 (VC on Windows.
>
> You need to ask in 'microsoft.public.vc.language' or other relevant
> newsgroup. My suspicion is that you need to go 64 bits.

Will try them. If anyone who sees this post who knows how to force
size_t to use __uint64 please post.

Greg

Victor Bazarov wrote:
> > Can I force std::size_t to be defined as unsigned long?
>
> Usually that's compiler-dependent. But what would be the point?
> 'unsigned long' is 32 bits on Windows.

Sorry, I meant to say 'unsigned long long' which is equivalent to
__uint64

Victor Bazarov wrote:
> > How/where?
> > I'm using VC 2005 (VC on Windows.
>
> You need to ask in 'microsoft.public.vc.language' or other relevant
> newsgroup. My suspicion is that you need to go 64 bits.

Will try them. If anyone who sees this post who knows how to force
size_t to use __uint64 please post.

Greg

wrote:
> Victor Bazarov wrote:
>>> Can I force std::size_t to be defined as unsigned long?
>>
>> Usually that's compiler-dependent. But what would be the point?
>> 'unsigned long' is 32 bits on Windows.
>
> Sorry, I meant to say 'unsigned long long' which is equivalent to
> __uint64

There is no "unsigned long long" in C++.

> Victor Bazarov wrote:
>>> How/where?
>>> I'm using VC 2005 (VC on Windows.
>>
>> You need to ask in 'microsoft.public.vc.language' or other relevant
>> newsgroup. My suspicion is that you need to go 64 bits.
>
> Will try them. If anyone who sees this post who knows how to force
> size_t to use __uint64 please post.

Whatever 'size_t' is defined to is governed by the implementation,
not by the programmer. You need to go to the newsgroup that deals
with your implementation.

V
--
Please remove capital 'A's when replying by e-mail
I do not respond to top-posted replies, please don't ask