From patchwork Wed Jan 9 18:53:49 2013 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Eduardo Habkost X-Patchwork-Id: 1954021 Return-Path: X-Original-To: patchwork-kvm@patchwork.kernel.org Delivered-To: patchwork-process-083081@patchwork1.kernel.org Received: from vger.kernel.org (vger.kernel.org [209.132.180.67]) by patchwork1.kernel.org (Postfix) with ESMTP id F3B893FDDA for ; Wed, 9 Jan 2013 18:53:22 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S932458Ab3AISxT (ORCPT ); Wed, 9 Jan 2013 13:53:19 -0500 Received: from mx1.redhat.com ([209.132.183.28]:21276 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S932441Ab3AISxK (ORCPT ); Wed, 9 Jan 2013 13:53:10 -0500 Received: from int-mx11.intmail.prod.int.phx2.redhat.com (int-mx11.intmail.prod.int.phx2.redhat.com [10.5.11.24]) by mx1.redhat.com (8.14.4/8.14.4) with ESMTP id r09IqSKq015989 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=OK); Wed, 9 Jan 2013 13:52:28 -0500 Received: from blackpad.lan.raisama.net (vpn1-7-24.gru2.redhat.com [10.97.7.24]) by int-mx11.intmail.prod.int.phx2.redhat.com (8.14.4/8.14.4) with ESMTP id r09IqRsa011533; Wed, 9 Jan 2013 13:52:27 -0500 Received: by blackpad.lan.raisama.net (Postfix, from userid 500) id 71525203D3E; Wed, 9 Jan 2013 16:54:05 -0200 (BRST) From: Eduardo Habkost To: qemu-devel@nongnu.org Cc: =?UTF-8?q?Andreas=20F=C3=A4rber?= , Igor Mammedov , kvm@vger.kernel.org Subject: [RFC 09/12] pc: Set fw_cfg data based on APIC ID calculation Date: Wed, 9 Jan 2013 16:53:49 -0200 Message-Id: <1357757632-1950-10-git-send-email-ehabkost@redhat.com> In-Reply-To: <1357757632-1950-1-git-send-email-ehabkost@redhat.com> References: <1357757632-1950-1-git-send-email-ehabkost@redhat.com> X-Scanned-By: MIMEDefang 2.68 on 10.5.11.24 Sender: kvm-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: kvm@vger.kernel.org This changes FW_CFG_MAX_CPUS and FW_CFG_NUMA to use apic_id_for_cpu(), so the NUMA table can be based on the APIC IDs, instead of CPU index (SeaBIOS knows nothing about CPU indexes, just APIC IDs). Signed-off-by: Eduardo Habkost --- Changes v2: - Get PC object as argument - Add more detailed comments explaining the reason for FW_CFG_MAX_CPUS not being simply 'max_cpus' Changes v3: - Use PCInitArgs instead of PC object Changes v4: - Don't use PCInitArgs, just add the necessary data for apic_id_limit() as argument - Rename function to pc_apic_id_limit() - Rename max_apic_id to apic_id_limit --- hw/pc.c | 44 +++++++++++++++++++++++++++++++++++++------- 1 file changed, 37 insertions(+), 7 deletions(-) diff --git a/hw/pc.c b/hw/pc.c index beb816b..38ae807 100644 --- a/hw/pc.c +++ b/hw/pc.c @@ -541,6 +541,18 @@ int e820_add_entry(uint64_t address, uint64_t length, uint32_t type) return index; } +/* Calculates the limit to CPU APIC ID values + * + * This function returns the limit for the APIC ID value, so that all + * CPU APIC IDs are < pc_apic_id_limit(). + * + * This is used for FW_CFG_MAX_CPUS. See comments on bochs_bios_init(). + */ +static unsigned int pc_apic_id_limit(unsigned int max_cpus) +{ + return apic_id_for_cpu(max_cpus - 1) + 1; +} + static void *bochs_bios_init(void) { void *fw_cfg; @@ -548,9 +560,24 @@ static void *bochs_bios_init(void) size_t smbios_len; uint64_t *numa_fw_cfg; int i, j; + unsigned int apic_id_limit = pc_apic_id_limit(max_cpus); fw_cfg = fw_cfg_init(BIOS_CFG_IOPORT, BIOS_CFG_IOPORT + 1, 0, 0); - fw_cfg_add_i16(fw_cfg, FW_CFG_MAX_CPUS, (uint16_t)max_cpus); + /* FW_CFG_MAX_CPUS is a bit confusing/problematic on x86: + * + * SeaBIOS needs FW_CFG_MAX_CPUS for CPU hotplug, but the CPU hotplug + * QEMU<->SeaBIOS interface is not based on the "CPU index", but on the APIC + * ID of hotplugged CPUs[1]. This means that FW_CFG_MAX_CPUS is not the + * "maximum number of CPUs", but the "limit to the APIC ID values SeaBIOS + * may see". + * + * So, this means we must not use max_cpus, here, but the maximum possible + * APIC ID value, plus one. + * + * [1] The only kind of "CPU identifier" used between SeaBIOS and QEMU is + * the APIC ID, not the "CPU index" + */ + fw_cfg_add_i16(fw_cfg, FW_CFG_MAX_CPUS, (uint16_t)apic_id_limit); fw_cfg_add_i32(fw_cfg, FW_CFG_ID, 1); fw_cfg_add_i64(fw_cfg, FW_CFG_RAM_SIZE, (uint64_t)ram_size); fw_cfg_add_bytes(fw_cfg, FW_CFG_ACPI_TABLES, (uint8_t *)acpi_tables, @@ -570,21 +597,24 @@ static void *bochs_bios_init(void) * of nodes, one word for each VCPU->node and one word for each node to * hold the amount of memory. */ - numa_fw_cfg = g_malloc0((1 + max_cpus + nb_numa_nodes) * 8); + numa_fw_cfg = g_malloc0((1 + apic_id_limit + nb_numa_nodes) * 8); numa_fw_cfg[0] = cpu_to_le64(nb_numa_nodes); - for (i = 0; i < max_cpus; i++) { + unsigned int cpu_idx; + for (cpu_idx = 0; cpu_idx < max_cpus; cpu_idx++) { + unsigned int apic_id = apic_id_for_cpu(cpu_idx); + assert(apic_id < apic_id_limit); for (j = 0; j < nb_numa_nodes; j++) { - if (test_bit(i, node_cpumask[j])) { - numa_fw_cfg[i + 1] = cpu_to_le64(j); + if (test_bit(cpu_idx, node_cpumask[j])) { + numa_fw_cfg[apic_id + 1] = cpu_to_le64(j); break; } } } for (i = 0; i < nb_numa_nodes; i++) { - numa_fw_cfg[max_cpus + 1 + i] = cpu_to_le64(node_mem[i]); + numa_fw_cfg[apic_id_limit + 1 + i] = cpu_to_le64(node_mem[i]); } fw_cfg_add_bytes(fw_cfg, FW_CFG_NUMA, (uint8_t *)numa_fw_cfg, - (1 + max_cpus + nb_numa_nodes) * 8); + (1 + apic_id_limit + nb_numa_nodes) * 8); return fw_cfg; }