--- Comment #12 from paolo at gcc dot gnu dot org 2006-08-05 10:11 ---
Subject: Bug 28587
Author: paolo
Date: Sat Aug 5 10:11:13 2006
New Revision: 115947
URL: http://gcc.gnu.org/viewcvs?root=gccview=revrev=115947
Log:
2006-08-05 Paolo Carlini [EMAIL PROTECTED]
PR
--- Comment #13 from pcarlini at suse dot de 2006-08-05 10:12 ---
Fixed.
--
pcarlini at suse dot de changed:
What|Removed |Added
Status|ASSIGNED
--- Comment #14 from gdr at integrable-solutions dot net 2006-08-05 23:26
---
Subject: Re: vectorbool is extremely slow (900x slower than it should be)
pcarlini at suse dot de [EMAIL PROTECTED] writes:
| Fixed.
Thanks Paolo!
-- Gaby
--
--- Comment #15 from sabre at nondot dot org 2006-08-05 23:30 ---
Thanks a *lot* Paolo! It works great now.
-Chris
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=28587
--- Comment #1 from pinskia at gcc dot gnu dot org 2006-08-03 18:23 ---
This is most likely due to the fact the C++98 standard special cases bool. Yes
there is a defect against the standard about that and I think it was already
going to change.
--
--- Comment #2 from sabre at nondot dot org 2006-08-03 18:31 ---
Andrew, I'm well aware that vectorbool stores things in compact form. If you
read my example, it's clear that I understand that.
Even with the current algorithm used by vectorbool, it should not use
std::fill internally,
--- Comment #3 from pcarlini at suse dot de 2006-08-03 19:16 ---
Hi. I think this specific issue can be (almost) fixed rather easily, by making
fill_insert smarter, that is slowly setting some bits of the previous and next
underlying unsigned long and then proceeding one unsigned long
--- Comment #4 from sabre at nondot dot org 2006-08-03 19:21 ---
the people actively working in the C++ Standard Commitee strongly dislike
vectorbool for many reasons, and probably it will be deprecated in C++03 and
replacement added. That implies, in turn, that it's not so easy to
--- Comment #5 from pcarlini at suse dot de 2006-08-03 19:28 ---
(In reply to comment #4)
Fair enough. My impression was that this was because std::vectorbool is
not a container and that the specialization doesn't act like the parent
container. My (possibly flawed) understanding
--- Comment #6 from gdr at integrable-solutions dot net 2006-08-03 21:13
---
Subject: Re: vectorbool is extremely slow (900x slower than it should be)
pcarlini at suse dot de [EMAIL PROTECTED] writes:
| By the way, I think Andrew has a point, maybe not clearly stated, in that all
|
--- Comment #7 from pcarlini at suse dot de 2006-08-03 22:12 ---
(In reply to comment #6)
some committee members dislike vectorbool for various reasons; but
I don't see a chance it is going to change.
Really? Or that comment of yours is just the effect of my typo C++03 for C++0x?
I
--- Comment #8 from gdr at integrable-solutions dot net 2006-08-03 23:01
---
Subject: Re: vectorbool is extremely slow (900x slower than it should be)
pcarlini at suse dot de [EMAIL PROTECTED] writes:
| (In reply to comment #6)
| some committee members dislike vectorbool for
--- Comment #9 from pcarlini at suse dot de 2006-08-04 00:19 ---
(In reply to comment #8)
| (In reply to comment #6)
| some committee members dislike vectorbool for various reasons; but
| I don't see a chance it is going to change.
|
| Really?
well, that is my opinion. To
--
pcarlini at suse dot de changed:
What|Removed |Added
AssignedTo|unassigned at gcc dot gnu |pcarlini at suse dot de
|dot org |
--- Comment #10 from pcarlini at suse dot de 2006-08-04 02:23 ---
Created an attachment (id=12013)
-- (http://gcc.gnu.org/bugzilla/attachment.cgi?id=12013action=view)
Minimal fix
Can you test a bit the attached together with your code, both from the
correctness and the performance
--- Comment #11 from pcarlini at suse dot de 2006-08-04 02:25 ---
Created an attachment (id=12014)
-- (http://gcc.gnu.org/bugzilla/attachment.cgi?id=12014action=view)
Minimal fix
Can you test a bit the attached together with your code, both from the
correctness and the performance
16 matches
Mail list logo
