From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:42225) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1afSaS-0004Tm-2C for qemu-devel@nongnu.org; Mon, 14 Mar 2016 09:31:14 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1afSaM-0001U5-Ay for qemu-devel@nongnu.org; Mon, 14 Mar 2016 09:31:08 -0400 Received: from mail-ob0-x243.google.com ([2607:f8b0:4003:c01::243]:33489) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1afSaM-0001TZ-5b for qemu-devel@nongnu.org; Mon, 14 Mar 2016 09:31:02 -0400 Received: by mail-ob0-x243.google.com with SMTP id hc1so14181009obb.0 for ; Mon, 14 Mar 2016 06:31:01 -0700 (PDT) Sender: Corey Minyard References: <1456340356-17147-1-git-send-email-minyard@acm.org> <1456340356-17147-3-git-send-email-minyard@acm.org> <20160313160206-mutt-send-email-mst@redhat.com> From: Corey Minyard Message-ID: <56E6BD13.6000303@acm.org> Date: Mon, 14 Mar 2016 20:30:59 +0700 MIME-Version: 1.0 In-Reply-To: <20160313160206-mutt-send-email-mst@redhat.com> Content-Type: text/plain; charset=windows-1252; format=flowed Content-Transfer-Encoding: 7bit Subject: Re: [Qemu-devel] [PATCH 2/5] pc: Postpone SMBIOS table installation to post machine init Reply-To: minyard@acm.org List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: "Michael S. Tsirkin" Cc: Igor Mammedov , Corey Minyard , qemu-devel@nongnu.org, Paolo Bonzini On 03/13/2016 09:03 PM, Michael S. Tsirkin wrote: > On Wed, Feb 24, 2016 at 12:59:13PM -0600, minyard@acm.org wrote: >> From: Corey Minyard >> >> This is the same place that the ACPI SSDT table gets added, so that >> devices can add themselves to the SMBIOS table. >> >> Signed-off-by: Corey Minyard > This changes the order of fw cfg files, which > is guest visible. > Need to make it depend on either machine type Do you mean add a new machine type, like adding a 2.7 version of pc-i440fx and pc-q35? > or the presence of ipmi somehow. You could pre-scan for ipmi options from early in the pc code. Kind of hackish. One other option would be to just have ACPI support for IPMI. Older OSes might not work; I'm not sure how important that is. Does anyone have a preference here? To me, adding a new machine type looks easiest, but I'm not sure of the effects on users. -corey >> --- >> hw/i386/pc.c | 11 ++++++----- >> 1 file changed, 6 insertions(+), 5 deletions(-) >> >> diff --git a/hw/i386/pc.c b/hw/i386/pc.c >> index 0aeefd2..da8fc76 100644 >> --- a/hw/i386/pc.c >> +++ b/hw/i386/pc.c >> @@ -778,8 +778,6 @@ static FWCfgState *bochs_bios_init(AddressSpace *as) >> acpi_tables, acpi_tables_len); >> fw_cfg_add_i32(fw_cfg, FW_CFG_IRQ0_OVERRIDE, kvm_allows_irq0_override()); >> >> - pc_build_smbios(fw_cfg); >> - >> fw_cfg_add_bytes(fw_cfg, FW_CFG_E820_TABLE, >> &e820_reserve, sizeof(e820_reserve)); >> fw_cfg_add_file(fw_cfg, "etc/e820", e820_table, >> @@ -1161,6 +1159,7 @@ void pc_machine_done(Notifier *notifier, void *data) >> { >> PCMachineState *pcms = container_of(notifier, >> PCMachineState, machine_done); >> + FWCfgState *fw_cfg = pcms->fw_cfg; >> PCIBus *bus = pcms->bus; >> >> if (bus) { >> @@ -1172,15 +1171,17 @@ void pc_machine_done(Notifier *notifier, void *data) >> extra_hosts++; >> } >> } >> - if (extra_hosts && pcms->fw_cfg) { >> + if (extra_hosts && fw_cfg) { >> uint64_t *val = g_malloc(sizeof(*val)); >> *val = cpu_to_le64(extra_hosts); >> - fw_cfg_add_file(pcms->fw_cfg, >> - "etc/extra-pci-roots", val, sizeof(*val)); >> + fw_cfg_add_file(fw_cfg, "etc/extra-pci-roots", val, sizeof(*val)); >> } >> } >> >> acpi_setup(); >> + if (fw_cfg) { >> + pc_build_smbios(fw_cfg); >> + } >> } >> >> void pc_guest_info_init(PCMachineState *pcms) >> -- >> 2.5.0