Displaying 3 results from an estimated 3 matches for "basecase".
Did you mean:
usecase
2010 Jun 11
4
[LLVMdev] Bignum development
Hi all,
After searching for a decent compiler backend for ages (google
sometimes isn't helpful), I recently stumbled upon LLVM. Woot!!
I work on bignum arithmetic (I'm a professional mathematician) and
have recently decided to switch from developing GPL'd bignum code to
BSD licensed code. (See http://www.mpir.org/ which I contributed to
for a while - a fork of GMP).
Please bear with
2010 Jun 11
3
[LLVMdev] Bignum development
...hen adding. On AMD x86_64 we get 2.5 cycles per limb in direct
> assembly for addmul_1 (actually achieving the maximum retirement rate
> for macro ops on AMD), but more like 10 cycles per limb with C. Even
> the assembly version of addmul_1 is not efficient enough for a
> multiplication basecase though, where you essentially need the whole
> basecase multiplication done in assembly!
>
> It is this assembly basecase which I would like to be able to code
> directly in LLVM assembly.
Hmm... might be interesting, but you'll likely get much better
practical results by just
>...
2010 Jun 12
0
[LLVMdev] Bignum development
...MD x86_64 we get 2.5 cycles per limb in direct
>> assembly for addmul_1 (actually achieving the maximum retirement rate
>> for macro ops on AMD), but more like 10 cycles per limb with C. Even
>> the assembly version of addmul_1 is not efficient enough for a
>> multiplication basecase though, where you essentially need the whole
>> basecase multiplication done in assembly!
>>
>> It is this assembly basecase which I would like to be able to code
>> directly in LLVM assembly.
>
> Hmm... might be interesting, but you'll likely get much better
>...