Sponsored Content
Special Forums UNIX and Linux Applications High Performance Computing Memory Barriers for (Ubuntu) Linux (i686) Post 302430462 by Corona688 on Thursday 17th of June 2010 05:25:35 PM
Old 06-17-2010
P.S. On a two-core single-CPU system, the overhead of XCHG vs LOCK XCHG with five seperate processes:

Code:
$ ./a.out & ./a.out & ./a.out & ./a.out & ./a.out &
12225 !Lock     time = 0 M 8 S 657 ms 205 us = 0.116 Mops/s
12229 !Lock     time = 0 M 8 S 801 ms 676 us = 0.114 Mops/s
12227 !Lock     time = 0 M 8 S 896 ms 459 us = 0.112 Mops/s
12228 !Lock     time = 0 M 8 S 958 ms 739 us = 0.112 Mops/s
12226 !Lock     time = 0 M 9 S 157 ms 723 us = 0.109 Mops/s
12228 Lock      time = 0 M 8 S 610 ms 749 us = 0.116 Mops/s
12227 Lock      time = 0 M 8 S 719 ms 860 us = 0.115 Mops/s
12225 Lock      time = 0 M 9 S 49 ms 622 us = 0.111 Mops/s
12226 Lock      time = 0 M 8 S 608 ms 304 us = 0.116 Mops/s
12229 Lock      time = 0 M 9 S 48 ms 352 us = 0.111 Mops/s

The code is a million loops of this:
Code:
                        "LOOP1:                 \n"
                        "       xchg    %ebx, a \n"
                        "       xchg    %ebx, a \n"
                        "       xchg    %ebx, a \n"
                        "       xchg    %ebx, a \n"
                        "       xchg    %ebx, a \n"
                        "       loop    LOOP1   \n"

except once with LOCK XCHG, one with just XCHG. No significant difference.

---------- Post updated at 03:25 PM ---------- Previous update was at 03:14 PM ----------

Quote:
Originally Posted by gorga
I originally used it with existing thread-pools, but found I needed more control over the allocation of "tasks" to "cores"
How so?
Quote:
hence I'm making my own.
I don't see how using a different structure excludes pthreads. You wanted to avoid pthreads since it used atomic ops, and are prepared to use atomic ops instead? It's best to write portably if possible anyway.

Last edited by Corona688; 06-17-2010 at 06:42 PM..
 

4 More Discussions You Might Find Interesting

1. UNIX for Dummies Questions & Answers

Memory-waste in Ubuntu/Debian?

I have 512 mem on this laptop, though 'top' tells me I only have 380. However, Ubuntu is using 288 mb of memory, when I only have 3 terminals, running lynx, vim(for this file) and (of course) top. Considering it I have lynx running a 600 page txt file, which of course would eat some memory but 300?... (0 Replies)
Discussion started by: riwa
0 Replies

2. Linux

i686, x86 64, ppc

Hi, i am quite new to linux. I am interested in fedora linux distro. Fedora Project I dont know which one to choose, either i686, x86 64 or ppc. I prefer a live cd, coz its easy to use. And what is the difference between "Fedora Desktop Live Media" and "Fedora KDE Live Media". (3 Replies)
Discussion started by: superblacksmith
3 Replies

3. Programming

Getting the total virtual memory for ubuntu in c++

Hi guys , i need to get the total virtual memory in ubuntu but i need to write a C++ code for that, any idea on how to go about doing it? any references? or website that i can refer to ? (6 Replies)
Discussion started by: xiaojesus
6 Replies

4. Ubuntu

XP and Linux (Ubuntu) on same disk, Can I install Ubuntu on not-yet partitioned portion of disk?

My PC (Esprimo, 3 yeas old) has one hard drive having 2 partitions C: (80 GB NTFS, XP) and D: (120 GB NTFS, empty) and and a 200 MB area that yet is not-partitioned. I would like to try Ubuntu and to install Ubuntu on the not-partitioned area . The idea is to have the possibility to run... (7 Replies)
Discussion started by: C.Weidemann
7 Replies
LOCK(2) 							System Calls Manual							   LOCK(2)

NAME
lockinit, lock, canlock, unlock - shared memory spin lock SYNOPSIS
#include <lock.h> void lockinit(void); void lock(Lock *lk); int canlock(Lock *lk); void unlock(Lock *lk); /* Alef only */ adt Lock { void lock(*Lock); void unlock(*Lock); int canlock(*Lock); }; adt QLock { void lock(*Lock); void unlock(*Lock); int canlock(*Lock); }; adt RWlock { void Rlock(*RWlock); void Runlock(*RWlock); void Wlock(*RWlock); void Wunlock(*RWlock); }; adt Ref { int inc(*Ref); int dec(*Ref); int ref(*Ref); }; DESCRIPTION
These routines are used by processes sharing memory to synchronize using spin locks. Lockinit must be called before the first use of the other routines. Lock blocks until the lock has been obtained. Canlock is non-blocking. It tries to obtain a lock and returns a non-zero value if it was successful, 0 otherwise. Unlock releases a lock. Alef Alef locks have similar functionality, but no special initialization is required. The ADT Lock has functions lock, unlock, and canlock, just like locks in C. QLocks have the same interface but are not spin locks; instead if the lock is taken QLock.lock will suspend execu- tion of the calling task until it is released. Although Locks are the more primitive lock, their use is discouraged and even erroneous for most purposes. For example, Locks cannot syn- chronize between tasks in the same proc. Use QLocks instead. RWlocks manage access to a data structure that has distinct readers and writers. RWlock.Rlock grants read access; RWlock.Runlock releases it. RWlock.Wlock grants write access; RWlock.Wunlock releases it. There may be any number of simultaneous readers, but only one writer. Moreover, if write access is granted no one may have read access until write access is released. Refs manage reference counters. Ref.inc increments the counter and returns the old value; Ref.dec decrements the counter and returns the new value. Ref.ref returns the current value. SOURCE
/sys/src/liblock SEE ALSO
rfork in fork(2) LOCK(2)
All times are GMT -4. The time now is 02:15 PM.
Unix & Linux Forums Content Copyright 1993-2022. All Rights Reserved.
Privacy Policy