public inbox for gcc-bugs@sourceware.org
help / color / mirror / Atom feed
* [Bug libstdc++/54307] New: [4.7 regression] increases in memory usage by some C++11 (and C++03) standard containers
@ 2012-08-17 21:43 matt at use dot net
2012-08-17 22:59 ` [Bug libstdc++/54307] " hjl.tools at gmail dot com
` (4 more replies)
0 siblings, 5 replies; 6+ messages in thread
From: matt at use dot net @ 2012-08-17 21:43 UTC (permalink / raw)
To: gcc-bugs
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=54307
Bug #: 54307
Summary: [4.7 regression] increases in memory usage by some
C++11 (and C++03) standard containers
Classification: Unclassified
Product: gcc
Version: 4.7.1
Status: UNCONFIRMED
Severity: major
Priority: P3
Component: libstdc++
AssignedTo: unassigned@gcc.gnu.org
ReportedBy: matt@use.net
Between 4.6 and 4.7.x, the memory used by a few standard containers has
increased, resulting in some OOM and performance issues on an amd64 application
I work on.
4.7:
list<int> size = 24
unordered_map<int,int> size = 64
unordered_set<int,int> size = 64
4.6:
list<int> size = 16
unordered_map<int,int> size = 56
unordered_set<int,int> size = 56
comparing to VC11:
list<int> size = 16
unordered_map<int,int> size = 64
unordered_set<int,int> size = 64
compile commandline is just -std=c++0x. testcase program below:
#include <vector>
#include <array>
#include <deque>
#include <forward_list>
#include <list>
#include <queue>
#include <stack>
#include <tuple>
#include <map>
#include <set>
#include <hash_map>
#include <hash_set>
#include <unordered_map>
#include <unordered_set>
#include <string>
using namespace std;
int main(void)
{
vector<int> v; printf("vector<int> size\t = %d\n", sizeof(v));
array<int,5> a; printf("array<int,5> size\t = %d\n", sizeof(a));
deque<int> d; printf("deque<int> size\t = %d\n", sizeof(d));
forward_list<int> f; printf("forward_list<int> size\t = %d\n", sizeof(f));
list<int> l; printf("list<int> size\t = %d\n", sizeof(l));
queue<int> q; printf("queue<int> size\t = %d\n", sizeof(q));
stack<int> s; printf("stack<int> size\t = %d\n", sizeof(s));
tuple<int,int,int> t; printf("tuple<int,int,int> size\t = %d\n",
sizeof(t));
map<int,int> m; printf("map<int,int> size\t = %d\n", sizeof(m));
set<int> set; printf("set<int> size\t = %d\n", sizeof(set));
unordered_map<int,int> um; printf("unordered_map<int,int> size\t = %d\n",
sizeof(um));
unordered_set<int> us; printf("unordered_set<int,int> size\t = %d\n",
sizeof(us));
string str; printf("string size\t = %d\n", sizeof(string));
return 0;
}
^ permalink raw reply [flat|nested] 6+ messages in thread
* [Bug libstdc++/54307] [4.7 regression] increases in memory usage by some C++11 (and C++03) standard containers
2012-08-17 21:43 [Bug libstdc++/54307] New: [4.7 regression] increases in memory usage by some C++11 (and C++03) standard containers matt at use dot net
@ 2012-08-17 22:59 ` hjl.tools at gmail dot com
2012-08-18 0:05 ` paolo.carlini at oracle dot com
` (3 subsequent siblings)
4 siblings, 0 replies; 6+ messages in thread
From: hjl.tools at gmail dot com @ 2012-08-17 22:59 UTC (permalink / raw)
To: gcc-bugs
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=54307
H.J. Lu <hjl.tools at gmail dot com> changed:
What |Removed |Added
----------------------------------------------------------------------------
Status|UNCONFIRMED |WAITING
Last reconfirmed| |2012-08-17
Target Milestone|--- |4.7.2
Ever Confirmed|0 |1
--- Comment #1 from H.J. Lu <hjl.tools at gmail dot com> 2012-08-17 22:59:26 UTC ---
I believe it has been fixed on 7.2 branch by
http://gcc.gnu.org/ml/gcc-cvs/2012-07/msg00090.html
Can you verified it?
^ permalink raw reply [flat|nested] 6+ messages in thread
* [Bug libstdc++/54307] [4.7 regression] increases in memory usage by some C++11 (and C++03) standard containers
2012-08-17 21:43 [Bug libstdc++/54307] New: [4.7 regression] increases in memory usage by some C++11 (and C++03) standard containers matt at use dot net
2012-08-17 22:59 ` [Bug libstdc++/54307] " hjl.tools at gmail dot com
@ 2012-08-18 0:05 ` paolo.carlini at oracle dot com
2012-08-21 17:27 ` matt at use dot net
` (2 subsequent siblings)
4 siblings, 0 replies; 6+ messages in thread
From: paolo.carlini at oracle dot com @ 2012-08-18 0:05 UTC (permalink / raw)
To: gcc-bugs
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=54307
--- Comment #2 from Paolo Carlini <paolo.carlini at oracle dot com> 2012-08-18 00:05:37 UTC ---
Note that the unordered containers have been profoundly reworked to conform to
C++11, the size isn't going to change.
^ permalink raw reply [flat|nested] 6+ messages in thread
* [Bug libstdc++/54307] [4.7 regression] increases in memory usage by some C++11 (and C++03) standard containers
2012-08-17 21:43 [Bug libstdc++/54307] New: [4.7 regression] increases in memory usage by some C++11 (and C++03) standard containers matt at use dot net
2012-08-17 22:59 ` [Bug libstdc++/54307] " hjl.tools at gmail dot com
2012-08-18 0:05 ` paolo.carlini at oracle dot com
@ 2012-08-21 17:27 ` matt at use dot net
2012-08-21 17:35 ` paolo.carlini at oracle dot com
2012-09-07 10:35 ` rguenth at gcc dot gnu.org
4 siblings, 0 replies; 6+ messages in thread
From: matt at use dot net @ 2012-08-21 17:27 UTC (permalink / raw)
To: gcc-bugs
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=54307
--- Comment #3 from Matt Hargett <matt at use dot net> 2012-08-21 17:26:55 UTC ---
Paolo, what about list<int>? Does VC11 achieve the size GCC 4.6 has by not
being compliant somehow?
^ permalink raw reply [flat|nested] 6+ messages in thread
* [Bug libstdc++/54307] [4.7 regression] increases in memory usage by some C++11 (and C++03) standard containers
2012-08-17 21:43 [Bug libstdc++/54307] New: [4.7 regression] increases in memory usage by some C++11 (and C++03) standard containers matt at use dot net
` (2 preceding siblings ...)
2012-08-21 17:27 ` matt at use dot net
@ 2012-08-21 17:35 ` paolo.carlini at oracle dot com
2012-09-07 10:35 ` rguenth at gcc dot gnu.org
4 siblings, 0 replies; 6+ messages in thread
From: paolo.carlini at oracle dot com @ 2012-08-21 17:35 UTC (permalink / raw)
To: gcc-bugs
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=54307
--- Comment #4 from Paolo Carlini <paolo.carlini at oracle dot com> 2012-08-21 17:35:25 UTC ---
I have no idea what they are doing in their implementation, there are of course
trade offs. When we decide to globally break the ABI to implement a C++11
Standard Conforming std:list we mean to add a data member to provide an O(1)
list<>::size(), that is what we already "experimented" in 4.7.0. Unless of
course over the next months somebody suggests something else which goes well
together with our other design choices.
^ permalink raw reply [flat|nested] 6+ messages in thread
* [Bug libstdc++/54307] [4.7 regression] increases in memory usage by some C++11 (and C++03) standard containers
2012-08-17 21:43 [Bug libstdc++/54307] New: [4.7 regression] increases in memory usage by some C++11 (and C++03) standard containers matt at use dot net
` (3 preceding siblings ...)
2012-08-21 17:35 ` paolo.carlini at oracle dot com
@ 2012-09-07 10:35 ` rguenth at gcc dot gnu.org
4 siblings, 0 replies; 6+ messages in thread
From: rguenth at gcc dot gnu.org @ 2012-09-07 10:35 UTC (permalink / raw)
To: gcc-bugs
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=54307
Richard Guenther <rguenth at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
Status|WAITING |RESOLVED
Resolution| |FIXED
--- Comment #5 from Richard Guenther <rguenth at gcc dot gnu.org> 2012-09-07 10:35:04 UTC ---
I believe this change has been reverted for 4.7.2.
^ permalink raw reply [flat|nested] 6+ messages in thread
end of thread, other threads:[~2012-09-07 10:35 UTC | newest]
Thread overview: 6+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2012-08-17 21:43 [Bug libstdc++/54307] New: [4.7 regression] increases in memory usage by some C++11 (and C++03) standard containers matt at use dot net
2012-08-17 22:59 ` [Bug libstdc++/54307] " hjl.tools at gmail dot com
2012-08-18 0:05 ` paolo.carlini at oracle dot com
2012-08-21 17:27 ` matt at use dot net
2012-08-21 17:35 ` paolo.carlini at oracle dot com
2012-09-07 10:35 ` rguenth at gcc dot gnu.org
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).