I promised to write an MPIR Digest detailing development about every
three weeks, but didn't get off to a good start. I've been obsessing
over another computational project I've been working on for some weeks
and haven't found the time. Here is the next (belated) article in the
series:

* New assembly code for left and right shift on x86_64
* New assembly functions mpn_lshift2 and mpn_rshift2
* Fixed a memory leak in lehmer gcd code
* New generic mullow and mulhi functions
* Update to latest version of yasm assembler
* Did some work to merge David Harvey's division code
* First attempt at writing sb_divappr_q needed by above code (probably
doesn't even compile yet, let alone work)
* Some updating of 32 bit Windows support for AMD64, p3 and p4
* Fix a cpu misidentification on BSD
* Fix a BSD install problem
* Remove the requirement to type make install-gmpcompat
* Fixed a make try warning on solaris
* Fix make distclean to clean up properly after a fat binary build
* Update make check to test both static and dynamic library functions
where the code actually differs
* Fixed a bug in make distcheck
* Made library version numbers x.y.z instead of x.y when switching
minor version number (in line with GMP)
* Added a script for setting all version numbers automatically when
doing a release
* Fixed bug in mpf_eq recently reported on the GMP list
* New function mpn_neg_n
* Added Robert Gerbicz's new mpz factorial code and tuning cutoff
* New mpn_mulmod_2expp1 function
* New mpn_mulmod_2expm1 function
* Benchmark for mpn functions
* Remove very old deprecated functions
* New k8 mpn_lshiftc assembler function
* Remove mpbsd support from MPIR
* Remove ancient ansi2knr conversion
* New core2/penryn mpn_store assembly code
* New nehalem mpn_store assembly code
* New core2/penryn copyi assembly code
* Added k8/k10 32 bit optimised assembly code
* New mpz_next_probab_prime_p function
* Optimised assembly code for add_err1_n for k8/k10
* Optimised assembly code for sub_err1_n for k8/k10
* Clean up of Windows batch build system
* New core2/penryn addlsh_n assembly code
* Macro functions inclsh1 and declsh1
* Fix make try so it can profile macro functions
* New nehalem addlsh_n assembly code
* New assembly code for k8/k10 addlsh_n, sublsh_n and carry in variants
* New k10_2 path for optimised Phenom II assembly code
* New assembly code for k8/k10 inclsh_n, declsh_n

The major contributors to the code above are Jason Moxham, Brian
Gladman, Robert Gerbicz and me.

Bill.

--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups 
"mpir-devel" group.
To post to this group, send email to [email protected]
To unsubscribe from this group, send email to 
[email protected]
For more options, visit this group at 
http://groups.google.com/group/mpir-devel?hl=en
-~----------~----~----~----~------~----~------~--~---

Reply via email to