From mboxrd@z Thu Jan 1 00:00:00 1970 From: "Abu M. Muttalib" Subject: RE: Commenting out out_of_memory() function in __alloc_pages() Date: Tue, 11 Jul 2006 20:27:07 +0530 Message-ID: MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit In-Reply-To: <20060707095441.GA3913@lnx-holt.americas.sgi.com> Sender: owner-linux-mm@kvack.org Return-Path: To: Robin Holt Cc: Alan Cox , nickpiggin@yahoo.com.au, Robert Hancock , chase.venters@clientec.com, kernelnewbies@nl.linux.org, linux-newbie@vger.kernel.org, linux-kernel@vger.kernel.org, linux-mm List-ID: The funniest part is that with memset commented out_of_memory observed, contrary to my expectation. I don't know why. It shouldn't have. I am running the application on an ARM target. Regards, Abu. -----Original Message----- From: Robin Holt [mailto:holt@sgi.com] Sent: Friday, July 07, 2006 3:25 PM To: Abu M. Muttalib Cc: kernelnewbies@nl.linux.org; linux-newbie@vger.kernel.org; linux-kernel@vger.kernel.org; linux-mm Subject: Re: Commenting out out_of_memory() function in __alloc_pages() I am not sure about x86, but on ia64, you would be very hard pressed for this application to actually run you out of memory. With the memset commented out, you would be allocating vmas, etc, but you would not be actually putting pages behind those virtual addresses. Thanks, Robin --------------------------- test1.c ---------------------------------- #include #include main() { char* buff; int count; count=0; while(1) { printf("\nOOM Test: Counter = %d", count); buff = (char*) malloc(1024); // memset(buff,'\0',1024); count++; if (buff==NULL) { printf("\nOOM Test: Memory allocation error"); } } } -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to majordomo@kvack.org. For more info on Linux MM, see: http://www.linux-mm.org/ . Don't email: email@kvack.org