Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1757195AbZFCJlB (ORCPT ); Wed, 3 Jun 2009 05:41:01 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1753382AbZFCJkm (ORCPT ); Wed, 3 Jun 2009 05:40:42 -0400 Received: from bilbo.ozlabs.org ([203.10.76.25]:40309 "EHLO bilbo.ozlabs.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751586AbZFCJkl (ORCPT ); Wed, 3 Jun 2009 05:40:41 -0400 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Message-ID: <18982.17586.666132.90983@cargo.ozlabs.ibm.com> Date: Wed, 3 Jun 2009 19:38:58 +1000 From: Paul Mackerras To: Ingo Molnar Cc: Peter Zijlstra , linux-kernel@vger.kernel.org Subject: [PATCH 1/2] perf_counter: powerpc: Fix event alternative code generation on POWER5/5+ X-Mailer: VM 8.0.12 under 22.2.1 (i486-pc-linux-gnu) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 2545 Lines: 70 Commit ef923214 ("perf_counter: powerpc: use u64 for event codes internally") introduced a bug where the return value from function find_alternative_bdecode gets put into a u64 variable and later tested to see if it is < 0. The effect is that we get extra, bogus event code alternatives on POWER5 and POWER5+, leading to error messages such as "oops compute_mmcr failed" being printed and counters not counting properly. This fixes it by using s64 for the return type of find_alternative_bdecode and for the local variable that the caller puts the value in. It also makes the event argument a u64 on POWER5+ for consistency. Signed-off-by: Paul Mackerras --- arch/powerpc/kernel/power5+-pmu.c | 4 ++-- arch/powerpc/kernel/power5-pmu.c | 4 ++-- 2 files changed, 4 insertions(+), 4 deletions(-) diff --git a/arch/powerpc/kernel/power5+-pmu.c b/arch/powerpc/kernel/power5+-pmu.c index c6cdfc1..8471e3c 100644 --- a/arch/powerpc/kernel/power5+-pmu.c +++ b/arch/powerpc/kernel/power5+-pmu.c @@ -242,7 +242,7 @@ static const unsigned char bytedecode_alternatives[4][4] = { * event code for those that do, or -1 otherwise. This also handles * alternative PCMSEL values for add events. */ -static int find_alternative_bdecode(unsigned int event) +static s64 find_alternative_bdecode(u64 event) { int pmc, altpmc, pp, j; @@ -277,7 +277,7 @@ static int power5p_get_alternatives(u64 event, unsigned int flags, u64 alt[]) { int i, j, nalt = 1; int nlim; - u64 ae; + s64 ae; alt[0] = event; nalt = 1; diff --git a/arch/powerpc/kernel/power5-pmu.c b/arch/powerpc/kernel/power5-pmu.c index d534496..1b44c5f 100644 --- a/arch/powerpc/kernel/power5-pmu.c +++ b/arch/powerpc/kernel/power5-pmu.c @@ -250,7 +250,7 @@ static const unsigned char bytedecode_alternatives[4][4] = { * PMCSEL values on other counters. This returns the alternative * event code for those that do, or -1 otherwise. */ -static u64 find_alternative_bdecode(u64 event) +static s64 find_alternative_bdecode(u64 event) { int pmc, altpmc, pp, j; @@ -272,7 +272,7 @@ static u64 find_alternative_bdecode(u64 event) static int power5_get_alternatives(u64 event, unsigned int flags, u64 alt[]) { int i, j, nalt = 1; - u64 ae; + s64 ae; alt[0] = event; nalt = 1; -- 1.6.0.4 -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/