From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from sc8-sf-mx1-b.sourceforge.net ([10.3.1.11] helo=sc8-sf-mx1.sourceforge.net) by sc8-sf-list1.sourceforge.net with esmtp (Exim 4.30) id 1DcoTA-00030k-9D for user-mode-linux-devel@lists.sourceforge.net; Mon, 30 May 2005 10:54:32 -0700 Received: from zcars04f.nortelnetworks.com ([47.129.242.57]) by sc8-sf-mx1.sourceforge.net with esmtp (Exim 4.41) id 1DcoT5-0005vz-Vb for user-mode-linux-devel@lists.sourceforge.net; Mon, 30 May 2005 10:54:31 -0700 Received: from zrtpd0jn.us.nortel.com (zrtpd0jn.us.nortel.com [47.140.202.35]) id j4UHsJP16343 for ; Mon, 30 May 2005 13:54:19 -0400 (EDT) Message-ID: <29D15BBCA340DA4D8146D38B4924FC7A105057@zcarhxm0.corp.nortel.com> From: "Majid Salame" MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----_=_NextPart_001_01C56540.97E6B129" Subject: [uml-devel] Slower system call with SMP Sender: user-mode-linux-devel-admin@lists.sourceforge.net Errors-To: user-mode-linux-devel-admin@lists.sourceforge.net List-Unsubscribe: , List-Id: The user-mode Linux development list List-Post: List-Help: List-Subscribe: , List-Archive: Date: Mon, 30 May 2005 13:54:15 -0400 To: user-mode-linux-devel@lists.sourceforge.net This message is in MIME format. Since your mail reader does not understand this format, some or all of this message may not be legible. ------_=_NextPart_001_01C56540.97E6B129 Content-Type: text/plain The cost of a system call it seems to be higher when Running a RHE host with dual CPU (SMP turned on) and the image is patched with skas patch. Doe that make sense ? Doing the same test by disabling SMP on the host, and results are much faster. NOTE SMP is turned off for the UML. Thanks, Majid. ------_=_NextPart_001_01C56540.97E6B129 Content-Type: text/html Slower system call with SMP

The cost of a system call it seems to be higher when
Running a RHE host with dual CPU (SMP turned on) and the image is patched with skas patch.

Doe that make sense ? Doing the same test by disabling SMP on the host, and results are much faster.
NOTE SMP is turned off for the UML.

Thanks,
Majid.


------_=_NextPart_001_01C56540.97E6B129-- ------------------------------------------------------- This SF.Net email is sponsored by Yahoo. Introducing Yahoo! Search Developer Network - Create apps using Yahoo! Search APIs Find out how you can build Yahoo! directly into your own Applications - visit http://developer.yahoo.net/?fr=offad-ysdn-ostg-q22005 _______________________________________________ User-mode-linux-devel mailing list User-mode-linux-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/user-mode-linux-devel From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from sc8-sf-mx1-b.sourceforge.net ([10.3.1.11] helo=sc8-sf-mx1.sourceforge.net) by sc8-sf-list1.sourceforge.net with esmtp (Exim 4.30) id 1Dcp74-0004uQ-Dq for user-mode-linux-devel@lists.sourceforge.net; Mon, 30 May 2005 11:35:46 -0700 Received: from nepa.nlc.no ([195.159.31.6] ident=qmailr) by sc8-sf-mx1.sourceforge.net with smtp (Exim 4.41) id 1Dcp72-0003N8-EJ for user-mode-linux-devel@lists.sourceforge.net; Mon, 30 May 2005 11:35:46 -0700 Message-ID: <38187.82.164.126.70.1117478138.squirrel@82.164.126.70> In-Reply-To: <29D15BBCA340DA4D8146D38B4924FC7A105057@zcarhxm0.corp.nortel.com> References: <29D15BBCA340DA4D8146D38B4924FC7A105057@zcarhxm0.corp.nortel.com> Subject: Re: [uml-devel] Slower system call with SMP From: stian@nixia.no MIME-Version: 1.0 Content-Type: text/plain;charset=UTF-8 Content-Transfer-Encoding: 8bit Sender: user-mode-linux-devel-admin@lists.sourceforge.net Errors-To: user-mode-linux-devel-admin@lists.sourceforge.net List-Unsubscribe: , List-Id: The user-mode Linux development list List-Post: List-Help: List-Subscribe: , List-Archive: Date: Mon, 30 May 2005 20:35:38 +0200 (CEST) To: Majid Salame Cc: user-mode-linux-devel@lists.sourceforge.net > Doe that make sense ? Doing the same test by disabling SMP on the host, > and > results are much faster. > NOTE SMP is turned off for the UML. This is a shot from the hip, but UML uses multiple processes for it's buisness. When host is SMP, these can be mapped to different cpu's, and when interprocess stuff is to occure, cpu's have to sync. Also, some few kernel things is slower on SMP, while user-space things that are not related, and multithreaded apps run faster. Stian ------------------------------------------------------- This SF.Net email is sponsored by Yahoo. Introducing Yahoo! Search Developer Network - Create apps using Yahoo! Search APIs Find out how you can build Yahoo! directly into your own Applications - visit http://developer.yahoo.net/?fr=offad-ysdn-ostg-q22005 _______________________________________________ User-mode-linux-devel mailing list User-mode-linux-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/user-mode-linux-devel From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from sc8-sf-mx1-b.sourceforge.net ([10.3.1.11] helo=sc8-sf-mx1.sourceforge.net) by sc8-sf-list1.sourceforge.net with esmtp (Exim 4.30) id 1DcpTj-00065z-VZ for user-mode-linux-devel@lists.sourceforge.net; Mon, 30 May 2005 11:59:11 -0700 Received: from zrtps0kn.nortelnetworks.com ([47.140.192.55]) by sc8-sf-mx1.sourceforge.net with esmtp (Exim 4.41) id 1DcpTi-0006cl-3j for user-mode-linux-devel@lists.sourceforge.net; Mon, 30 May 2005 11:59:11 -0700 Message-ID: <29D15BBCA340DA4D8146D38B4924FC7A105059@zcarhxm0.corp.nortel.com> From: "Majid Salame" Subject: RE: [uml-devel] Slower system call with SMP MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----_=_NextPart_001_01C56549.A26CA839" Sender: user-mode-linux-devel-admin@lists.sourceforge.net Errors-To: user-mode-linux-devel-admin@lists.sourceforge.net List-Unsubscribe: , List-Id: The user-mode Linux development list List-Post: List-Help: List-Subscribe: , List-Archive: Date: Mon, 30 May 2005 14:58:58 -0400 To: "'stian@nixia.no'" Cc: user-mode-linux-devel@lists.sourceforge.net This message is in MIME format. Since your mail reader does not understand this format, some or all of this message may not be legible. ------_=_NextPart_001_01C56549.A26CA839 Content-Type: text/plain > UML uses multiple processes for it's buisness. With Skas UML is just one process as it appears on the Host, or do you mean something else. >When host is > SMP, these can be mapped to different cpu's, and when > interprocess stuff is to occure, cpu's have to sync. > I saw the syscall about 300% slower, it does not make sense due to interprocess. IT should not be that slow, unless missing UML implementation > Also, some few kernel things is slower on SMP, while > user-space things that are not related, and multithreaded > apps run faster. > > Stian > > ------_=_NextPart_001_01C56549.A26CA839 Content-Type: text/html RE: [uml-devel] Slower system call with SMP

> UML uses multiple processes for it's buisness.
With Skas UML is just one process as it appears on the Host, or do you mean something else.

>When host is
> SMP, these can be mapped to different cpu's, and when
> interprocess stuff is to occure, cpu's have to sync.
>
I saw the syscall about 300% slower, it does not make sense due to interprocess.
IT should not be that slow, unless missing UML implementation

> Also, some few kernel things is slower on SMP, while
> user-space things that are not related, and multithreaded
> apps run faster.
>
> Stian
>
>

------_=_NextPart_001_01C56549.A26CA839-- ------------------------------------------------------- This SF.Net email is sponsored by Yahoo. Introducing Yahoo! Search Developer Network - Create apps using Yahoo! Search APIs Find out how you can build Yahoo! directly into your own Applications - visit http://developer.yahoo.net/?fr=offad-ysdn-ostg-q22005 _______________________________________________ User-mode-linux-devel mailing list User-mode-linux-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/user-mode-linux-devel From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from sc8-sf-mx2-b.sourceforge.net ([10.3.1.12] helo=sc8-sf-mx2.sourceforge.net) by sc8-sf-list1.sourceforge.net with esmtp (Exim 4.30) id 1Dd7OG-0001HS-74 for user-mode-linux-devel@lists.sourceforge.net; Tue, 31 May 2005 07:06:44 -0700 Received: from lakshmi.addtoit.com ([198.99.130.6] helo=lakshmi.solana.com) by sc8-sf-mx2.sourceforge.net with esmtp (Exim 4.41) id 1Dd7OE-0004MR-B9 for user-mode-linux-devel@lists.sourceforge.net; Tue, 31 May 2005 07:06:44 -0700 From: Jeff Dike Subject: Re: [uml-devel] Slower system call with SMP Message-ID: <20050531135731.GA9909@ccure.user-mode-linux.org> References: <29D15BBCA340DA4D8146D38B4924FC7A105057@zcarhxm0.corp.nortel.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <29D15BBCA340DA4D8146D38B4924FC7A105057@zcarhxm0.corp.nortel.com> Sender: user-mode-linux-devel-admin@lists.sourceforge.net Errors-To: user-mode-linux-devel-admin@lists.sourceforge.net List-Unsubscribe: , List-Id: The user-mode Linux development list List-Post: List-Help: List-Subscribe: , List-Archive: Date: Tue, 31 May 2005 09:57:31 -0400 To: Majid Salame Cc: user-mode-linux-devel@lists.sourceforge.net > The cost of a system call it seems to be higher when > Running a RHE host with dual CPU (SMP turned on) and the image is patched > with skas patch. > > Doe that make sense ? Doing the same test by disabling SMP on the host, and > results are much faster. What exactly was the test? A getpid loop? Jeff ------------------------------------------------------- This SF.Net email is sponsored by Yahoo. Introducing Yahoo! Search Developer Network - Create apps using Yahoo! Search APIs Find out how you can build Yahoo! directly into your own Applications - visit http://developer.yahoo.net/?fr=offad-ysdn-ostg-q22005 _______________________________________________ User-mode-linux-devel mailing list User-mode-linux-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/user-mode-linux-devel From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from sc8-sf-mx1-b.sourceforge.net ([10.3.1.11] helo=sc8-sf-mx1.sourceforge.net) by sc8-sf-list1.sourceforge.net with esmtp (Exim 4.30) id 1Dd7w2-0003PG-99 for user-mode-linux-devel@lists.sourceforge.net; Tue, 31 May 2005 07:41:38 -0700 Received: from zrtps0kp.nortelnetworks.com ([47.140.192.56]) by sc8-sf-mx1.sourceforge.net with esmtp (Exim 4.41) id 1Dd7w0-0002H5-Hg for user-mode-linux-devel@lists.sourceforge.net; Tue, 31 May 2005 07:41:38 -0700 Message-ID: <29D15BBCA340DA4D8146D38B4924FC7A10505B@zcarhxm0.corp.nortel.com> From: "Majid Salame" Subject: RE: [uml-devel] Slower system call with SMP MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----_=_NextPart_001_01C565EE.D661107C" Sender: user-mode-linux-devel-admin@lists.sourceforge.net Errors-To: user-mode-linux-devel-admin@lists.sourceforge.net List-Unsubscribe: , List-Id: The user-mode Linux development list List-Post: List-Help: List-Subscribe: , List-Archive: Date: Tue, 31 May 2005 10:41:32 -0400 To: 'Jeff Dike' Cc: user-mode-linux-devel@lists.sourceforge.net This message is in MIME format. Since your mail reader does not understand this format, some or all of this message may not be legible. ------_=_NextPart_001_01C565EE.D661107C Content-Type: text/plain Yes. And I am trying fork stress test. Majid. > -----Original Message----- > From: Jeff Dike [mailto:jdike@addtoit.com] > Sent: Tuesday, May 31, 2005 9:58 AM > To: Salame, Majid [CAR:2V62:EXCH] > Cc: user-mode-linux-devel@lists.sourceforge.net > Subject: Re: [uml-devel] Slower system call with SMP > > > > The cost of a system call it seems to be higher when > > Running a RHE host with dual CPU (SMP turned on) and the image is > > patched with skas patch. > > > > Doe that make sense ? Doing the same test by disabling SMP on the > > host, and results are much faster. > > What exactly was the test? A getpid loop? > > Jeff > > ------_=_NextPart_001_01C565EE.D661107C Content-Type: text/html Content-Transfer-Encoding: quoted-printable RE: [uml-devel] Slower system call with SMP

Yes. And I am trying fork stress test.

Majid.

> -----Original Message-----
> From: Jeff Dike [mailto:jdike@addtoit.com]
> Sent: Tuesday, May 31, 2005 9:58 AM
> To: Salame, Majid [CAR:2V62:EXCH]
> Cc: = user-mode-linux-devel@lists.sourceforge.net
> Subject: Re: [uml-devel] Slower system call = with SMP
>
>
> > The cost of a system call it seems to be = higher when
> > Running a RHE host with dual CPU (SMP = turned on) and the image is
> > patched with skas patch.
> >
> > Doe that make sense ? Doing the same test = by disabling SMP on the
> > host, and results are much faster.
>
> What exactly was the test?  A getpid = loop?
>
>       =         =         =         Jeff
>
>

------_=_NextPart_001_01C565EE.D661107C-- ------------------------------------------------------- This SF.Net email is sponsored by Yahoo. Introducing Yahoo! Search Developer Network - Create apps using Yahoo! Search APIs Find out how you can build Yahoo! directly into your own Applications - visit http://developer.yahoo.net/?fr=offad-ysdn-ostg-q22005 _______________________________________________ User-mode-linux-devel mailing list User-mode-linux-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/user-mode-linux-devel From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from sc8-sf-mx2-b.sourceforge.net ([10.3.1.92] helo=mail.sourceforge.net) by sc8-sf-list1.sourceforge.net with esmtp (Exim 4.30) id 1DxowH-000859-4n for user-mode-linux-devel@lists.sourceforge.net; Wed, 27 Jul 2005 09:39:25 -0700 Received: from smtp005.mail.ukl.yahoo.com ([217.12.11.36]) by mail.sourceforge.net with smtp (Exim 4.44) id 1DxowD-0001IY-HX for user-mode-linux-devel@lists.sourceforge.net; Wed, 27 Jul 2005 09:39:25 -0700 From: Blaisorblade Subject: Re: [uml-devel] Slower system call with SMP References: <29D15BBCA340DA4D8146D38B4924FC7A105059@zcarhxm0.corp.nortel.com> In-Reply-To: <29D15BBCA340DA4D8146D38B4924FC7A105059@zcarhxm0.corp.nortel.com> MIME-Version: 1.0 Content-Disposition: inline Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit Message-Id: <200507271806.50285.blaisorblade@yahoo.it> Sender: user-mode-linux-devel-admin@lists.sourceforge.net Errors-To: user-mode-linux-devel-admin@lists.sourceforge.net List-Unsubscribe: , List-Id: The user-mode Linux development list List-Post: List-Help: List-Subscribe: , List-Archive: Date: Wed, 27 Jul 2005 18:06:50 +0200 To: user-mode-linux-devel@lists.sourceforge.net Cc: Majid Salame , "'stian@nixia.no'" On Monday 30 May 2005 20:58, Majid Salame wrote: > > UML uses multiple processes for it's buisness. > With Skas UML is just one process as it appears on the Host, or do you mean > something else. No, it's at least two (+ possible service threads): one is the thread executing kernel code, which ptraces another thread which executes userspace code. > >When host is > > SMP, these can be mapped to different cpu's, and when > > interprocess stuff is to occure, cpu's have to sync. It's not just interprocess comunication. Those two threads could be running on two different processors (which will happen very likely, I fear, if UML is the only thing running, and can happen even without this - I don't think scheduler optimizations consider ptrace() as something needing to join threads together). > I saw the syscall about 300% slower, it does not make sense due to > interprocess. > IT should not be that slow, unless missing UML implementation I think it could, if the two threads run on different processors. Luckily, you can stop the host kernel from doing that: there should be a control option in /proc/, something like cpus_allowed. Ok, let's assume you talk about SKAS (I'm not expert in TT implementation, should check a couple of things, but you said you use SKAS). Well, it makes sense. In late days I've been studying signal delivery code, and for UML operations, to have a page fault in the guest, what happens is that the guest takes a SIGSEGV. That signal must be intercepted by the UML kernel thread. Sadly, this is largely suboptimal in Linux. In fact, you have, at signal delivery time, to wake up the guest thread (possibly via inter-processor comunication, which is an IPI), which will then have to wake up its tracer (and if that's on a different CPU, that's through an IPI), and then get back to wake up the thread itself. > > Also, some few kernel things is slower on SMP, while > > user-space things that are not related, and multithreaded > > apps run faster. > > Stian -- Inform me of my mistakes, so I can keep imitating Homer Simpson's "Doh!". Paolo Giarrusso, aka Blaisorblade (Skype ID "PaoloGiarrusso", ICQ 215621894) http://www.user-mode-linux.org/~blaisorblade ___________________________________ Yahoo! Mail: gratis 1GB per i messaggi e allegati da 10MB http://mail.yahoo.it ------------------------------------------------------- SF.Net email is sponsored by: Discover Easy Linux Migration Strategies from IBM. Find simple to follow Roadmaps, straightforward articles, informative Webcasts and more! Get everything you need to get up to speed, fast. http://ads.osdn.com/?ad_id=7477&alloc_id=16492&op=click _______________________________________________ User-mode-linux-devel mailing list User-mode-linux-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/user-mode-linux-devel