From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-out.m-online.net (mail-out.m-online.net [212.18.0.9]) by ozlabs.org (Postfix) with ESMTP id 47C17681C2 for ; Mon, 21 Nov 2005 07:30:58 +1100 (EST) To: Kylo Ginsberg From: Wolfgang Denk Mime-version: 1.0 Content-type: text/plain; charset=ISO-8859-1 In-reply-to: Your message of "Sun, 20 Nov 2005 10:58:26 PST." <61cc712d0511201058p367df361i7e46591336918712@mail.gmail.com> Date: Sun, 20 Nov 2005 21:30:56 +0100 Sender: wd@denx.de Message-Id: <20051120203056.11EAB353F54@atlas.denx.de> Cc: linuxppc-embedded@ozlabs.org Subject: Re: Memtest-SDRAM List-Id: Linux on Embedded PowerPC Developers Mail List List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , In message <61cc712d0511201058p367df361i7e46591336918712@mail.gmail.com> you wrote: > > > A simple test is: if Linux is running without problems under load > > than there are no problems with your SDRAM. > > Yes, but determining the proper load to test with can be tricky. Ummm.. I disagree. Avtually it's pretty simple. Boot the system with root file system mounted over NFS. The compile a Linux kernel natively on the board. This is doing all you want: stressing network (lots of DMA), stressing the memory (lots of context switches and heavy RAM use), etc. If your system survives this, you can be 98% sure that it will pass any other test as well. > E.g., I had a subtle memory timing issue that only turned up under > very peculiar circumstances under linux; this was very difficult to Of course there are really weird cases, but fortunaltey these are rare exceptions. Best regards, Wolfgang Denk -- Software Engineering: Embedded and Realtime Systems, Embedded Linux Phone: (+49)-8142-66989-10 Fax: (+49)-8142-66989-80 Email: wd@denx.de Committee, n.: A group of men who individually can do nothing but as a group decide that nothing can be done. - Fred Allen