From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1752243Ab0HZNYN (ORCPT ); Thu, 26 Aug 2010 09:24:13 -0400 Received: from relay2.sgi.com ([192.48.179.30]:38423 "EHLO relay.sgi.com" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S1751910Ab0HZNVZ (ORCPT ); Thu, 26 Aug 2010 09:21:25 -0400 Message-Id: <20100826132124.721900943@sgi.com> User-Agent: quilt/0.47-1 Date: Thu, 26 Aug 2010 08:19:48 -0500 From: steiner@sgi.com To: akpm@linux-foundation.org, linux-kernel@vger.kernel.org Subject: [Patch 11/25] GRU - interrupt fix for processors without core 0 References: <20100826131937.108920216@sgi.com> Content-Disposition: inline; filename=uv_gru_interrupt_0 Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org From: Jack Steiner GRU interrupts for the kernel context are targeted to core 0 of the lowest socket on the blade. Fix an endcase for 6-socket processors that do not have a APICID for core 0. (There are potentially 8 cores in the socket. A 6 core socket has 2 cores disabled. If core 0 was disabled, the old code did not work). Signed-off-by: Jack Steiner --- drivers/misc/sgi-gru/grumain.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) Index: linux/drivers/misc/sgi-gru/grumain.c =================================================================== --- linux.orig/drivers/misc/sgi-gru/grumain.c 2010-06-09 08:11:43.740083860 -0500 +++ linux/drivers/misc/sgi-gru/grumain.c 2010-06-09 08:11:44.128107324 -0500 @@ -618,7 +618,7 @@ void gru_load_context(struct gru_thread_ cch->unmap_enable = 1; cch->tfm_done_bit_enable = 1; cch->cb_int_enable = 1; - cch->tlb_int_select = 0; /* For now, ints go to cpu 0 */ + cch->tlb_int_select = uv_cpu_core_number(0); /* For now, ints go to cpu 0 */ } else { cch->unmap_enable = 0; cch->tfm_done_bit_enable = 0;