From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from atlrel7.hp.com (atlrel7.hp.com [156.153.255.213]) by dsl2.external.hp.com (Postfix) with ESMTP id BDF704838 for ; Sun, 9 Feb 2003 00:41:32 -0700 (MST) Received: from udlkern.fc.hp.com (udlkern.fc.hp.com [15.1.52.48]) by atlrel7.hp.com (Postfix) with ESMTP id 273981C01B82 for ; Sun, 9 Feb 2003 02:41:32 -0500 (EST) Date: Sun, 9 Feb 2003 00:40:14 -0700 (MST) From: John Marvin Message-Id: <200302090740.AAA17008@udlkern.fc.hp.com> To: parisc-linux@lists.parisc-linux.org Cc: grundler@dsl2.external.hp.com Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Subject: [parisc-linux] re: [parisc-linux-cvs] linux grundler Sender: parisc-linux-admin@lists.parisc-linux.org Errors-To: parisc-linux-admin@lists.parisc-linux.org List-Help: List-Post: List-Subscribe: , List-Id: parisc-linux developers list List-Unsubscribe: , List-Archive: > o moved disable_sr_hash() from SMP to common code path so all > CPU's (including monarch) have this disabled. Huh? You didn't seriously think the monarch still had sr hashing enabled did you? Believe me, shared memory wouldn't work at all if that was the case. This change is wrong. init_per_cpu() is called too early for the monarch. disable_sr_hashing relies on the boot cpu data to be initialized to determine which code should be used to disable the sr hashing. It might work for PA2.0, since it is the default case, but you probably broke all of the other processors. disable_sr_hashing() is called for the monarch in mm/init.c in setup_bootmem(). John Marvin jsm@fc.hp.com