Re: [mpir-devel] Test results

2010-01-07 Thread Case Vanhorsen
On Thu, Jan 7, 2010 at 7:05 AM, Bill Hart wrote: > Sorry, I mean it is needed by make install. So perhaps it is the > Makefile in the top directory which needs this added. > I actually had to edit the Makefile in the doc directory. It was trying to create the info sub-directory. Everything seems t

Re: [mpir-devel] Test results

2010-01-07 Thread Bill Hart
Sorry, I mean it is needed by make install. So perhaps it is the Makefile in the top directory which needs this added. Bill. 2010/1/7 Bill Hart : > Oh, that is a problem caused by autotools. Probably to fix it, you can > edit the Makefile in the test directory. You'll probably find a line > that

Re: [mpir-devel] Test results

2010-01-07 Thread Bill Hart
Oh, that is a problem caused by autotools. Probably to fix it, you can edit the Makefile in the test directory. You'll probably find a line that says: mkdir...@mkdir_p@ You need to replace this with a command to run mkdir. On my system it is: MKDIR_P=/usr/bin/mkdir -p I don't know why this woul

Re: [mpir-devel] Test results

2010-01-07 Thread Case Vanhorsen
On Thu, Jan 7, 2010 at 6:37 AM, Case Vanhorsen wrote: > On Thu, Jan 7, 2010 at 4:53 AM, Bill Hart wrote: >> Yes, it was fixed. It definitely shouldn't do this. >> >> 2010/1/7 Case Vanhorsen : >>> Hello, >>> >>> The current trunk (r2579) is passing all my 64-bit Windows stress tests. >>> >>> But m

Re: [mpir-devel] Test results

2010-01-07 Thread Case Vanhorsen
On Thu, Jan 7, 2010 at 4:53 AM, Bill Hart wrote: > Yes, it was fixed. It definitely shouldn't do this. > > 2010/1/7 Case Vanhorsen : >> Hello, >> >> The current trunk (r2579) is passing all my 64-bit Windows stress tests. >> >> But mpz_nextprime is still returning too many composites. For example,

Re: [mpir-devel] Test results

2010-01-07 Thread Bill Hart
Yes, it was fixed. It definitely shouldn't do this. 2010/1/7 Case Vanhorsen : > Hello, > > The current trunk (r2579) is passing all my 64-bit Windows stress tests. > > But mpz_nextprime is still returning too many composites. For example, > gmpy.next_prime(3270400) is returning mpz(3270403). > > I

[mpir-devel] Re: Test results

2010-01-07 Thread Cactus
On Jan 7, 7:11 am, Case Vanhorsen wrote: > Hello, > > The current trunk (r2579) is passing all my 64-bit Windows stress tests. > > But mpz_nextprime is still returning too many composites. For example, > gmpy.next_prime(3270400) is returning mpz(3270403). > > I thought this was fixed earlier? H