From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:60030) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1bKIlm-0006Wf-9F for qemu-devel@nongnu.org; Tue, 05 Jul 2016 01:19:39 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1bKIli-00035I-47 for qemu-devel@nongnu.org; Tue, 05 Jul 2016 01:19:37 -0400 Received: from mx0a-001b2d01.pphosted.com ([148.163.156.1]:49560) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1bKIlh-00034q-SC for qemu-devel@nongnu.org; Tue, 05 Jul 2016 01:19:34 -0400 Received: from pps.filterd (m0098410.ppops.net [127.0.0.1]) by mx0a-001b2d01.pphosted.com (8.16.0.11/8.16.0.11) with SMTP id u655JVi5082972 for ; Tue, 5 Jul 2016 01:19:33 -0400 Received: from e23smtp05.au.ibm.com (e23smtp05.au.ibm.com [202.81.31.147]) by mx0a-001b2d01.pphosted.com with ESMTP id 23xa529jkx-1 (version=TLSv1.2 cipher=AES256-SHA bits=256 verify=NOT) for ; Tue, 05 Jul 2016 01:19:32 -0400 Received: from localhost by e23smtp05.au.ibm.com with IBM ESMTP SMTP Gateway: Authorized Use Only! Violators will be prosecuted for from ; Tue, 5 Jul 2016 15:19:29 +1000 From: Sam Bobroff Date: Tue, 5 Jul 2016 15:19:22 +1000 In-Reply-To: References: <20160608022602.GU9226@voom.fritz.box> In-Reply-To: References: Message-Id: <17cc8e9fb2ab41356e207c6a14d5987754fa4da8.1467695915.git.sam.bobroff@au1.ibm.com> Subject: [Qemu-devel] [PATCH 1/3] spapr: Disable ibm, pa-features HTM bit List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: david@gibson.dropbear.id.au Cc: anton@au1.ibm.com, mikey@neuling.org, aik@ozlabs.ru, mpe@ellerman.id.au, agraf@suse.de, qemu-devel@nongnu.org, paulus@samba.org, qemu-ppc@nongnu.org, linuxppc-dev@lists.ozlabs.org There are a few issues with our handling of the ibm,pa-features HTM bit: - We don't support transactional memory in PR KVM, so don't tell the OS that we do. - In full emulation we have a minimal implementation of HTM that always fails, so for performance reasons lets not tell the OS that we support it either. - In HV KVM mode, we should mirror the host HTM enabled state by checking a KVM capability or looking at the AT_HWCAP2 bit. For now unconditionally disable it by removing HTM from the pa-features bits. It will be re-enabled in a subsequent patch specifically for HV KVM. Signed-off-by: Sam Bobroff --- hw/ppc/spapr.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/hw/ppc/spapr.c b/hw/ppc/spapr.c index 78ebd9e..704aae7 100644 --- a/hw/ppc/spapr.c +++ b/hw/ppc/spapr.c @@ -635,7 +635,7 @@ static void spapr_populate_cpu_dt(CPUState *cs, void *fdt, int offset, 0xf6, 0x1f, 0xc7, 0xc0, 0x80, 0xf0, 0x80, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x80, 0x00, - 0x80, 0x00, 0x80, 0x00, 0x80, 0x00 }; + 0x80, 0x00, 0x80, 0x00, 0x00, 0x00 }; uint8_t *pa_features; size_t pa_size; -- 2.1.0