Number: 483

Date:  3-Apr-84 17':28':50

Submitter: Sannella.PA

Source: JONL.PA

Subject: system software should try to reduce FIXP <-> FLOATP coercions 

Lisp Version: 

Description: '
Date': 18 MAR 84 17':19 PST'
From': JONL.PA'
Subject': Re': Floating arith optimizatio'
To':   kaplan, lispsupport'
cc':   JONL'
...'
'
2) system code, such as on AARITH etc, should not depend upon these'
   sometime compiler/ucode optimizations.  One man calls this a matter of'
   "taste" -- another sees a benchmark speedup of over 50% [yes, the loss'
   in SQRT was enough to slow down Darrel''s benchmark, which he claims is'
   making the  popular rounds in ComputerScience circles,  from 20 secs to'
   30 secs on a Dorado.'
'


Workaround: 

Test Case: 

Edit-By: 

Edit-Date: 

Attn: lisp

Assigned To: 

In/By: 

Disposition: 

System: Language Support

Subsystem: Arithmetic

Machine: 

Disk: 

Microcode Version: 

Memory Size: 

File Server: 

Server Software Version: 

Difficulty: 

Frequency: Everytime

Impact: Moderate

Priority: Unlikely

Status: Wish

Problem Type: Performance

Source Files: