Rietveld Code Review Tool
Help | Bug tracker | Discussion group | Source code | Sign in
(1216)

Issue 7322081: code review 7322081: cmd/gc: replace x*8 by x<<3 etc (Closed)

Can't Edit
Can't Publish+Mail
Start Review
Created:
11 years, 1 month ago by rsc
Modified:
11 years, 1 month ago
CC:
ken2, golang-dev
Visibility:
Public.

Description

cmd/gc: replace x*8 by x<<3 etc Fixes issue 4199.

Patch Set 1 #

Patch Set 2 : diff -r 44896f1c8641 https://code.google.com/p/go/ #

Patch Set 3 : diff -r 44896f1c8641 https://code.google.com/p/go/ #

Unified diffs Side-by-side diffs Delta from patch set Stats (+71 lines, -1 line) Patch
M src/cmd/gc/walk.c View 1 4 chunks +71 lines, -1 line 0 comments Download

Messages

Total messages: 9
rsc
Hello ken2 (cc: golang-dev@googlegroups.com), I'd like you to review this change to https://code.google.com/p/go/
11 years, 1 month ago (2013-02-14 19:53:58 UTC) #1
rsc
*** Submitted as https://code.google.com/p/go/source/detail?r=1c18d925eb4d *** cmd/gc: replace x*8 by x<<3 etc Fixes issue 4199. R=ken2 ...
11 years, 1 month ago (2013-02-14 19:54:04 UTC) #2
dfc
Yay. Thanks Russ. On 15 Feb 2013 06:54, <rsc@golang.org> wrote: > *** Submitted as > ...
11 years, 1 month ago (2013-02-14 19:56:06 UTC) #3
mtj1
Great! On Thu, Feb 14, 2013 at 11:56 AM, Dave Cheney <dave@cheney.net> wrote: > Yay. ...
11 years, 1 month ago (2013-02-14 20:53:36 UTC) #4
dsymonds
Is there a plan to strength reduces some other multiplications? For instance, x*5 is x<<2|x. ...
11 years, 1 month ago (2013-02-14 22:30:27 UTC) #5
mtj1
You mean x<<2+x On Feb 14, 2013 2:30 PM, "David Symonds" <dsymonds@golang.org> wrote: > Is ...
11 years, 1 month ago (2013-02-14 22:40:32 UTC) #6
remyoudompheng
On 2013/2/14 David Symonds <dsymonds@golang.org> wrote: > Is there a plan to strength reduces some ...
11 years, 1 month ago (2013-02-14 22:42:34 UTC) #7
dsymonds
On Fri, Feb 15, 2013 at 9:40 AM, Michael Jones <mtj@google.com> wrote: > You mean ...
11 years, 1 month ago (2013-02-14 22:50:16 UTC) #8
ken3
11 years, 1 month ago (2013-02-15 03:04:29 UTC) #9
Message was sent while issue was closed.
lgtm

the plan9 code generator replaces
mul by some known constants
with shift/add/sub. the table was
generated by a super-opt.

on most modern cpus mul is pretty
fast, so the economies have to be
recalculated, but the tables there will
give the best shift/add answers.
Sign in to reply to this message.

Powered by Google App Engine
RSS Feeds Recent Issues | This issue
This is Rietveld f62528b