From patchwork Tue Jul 12 14:42:39 2016 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Anthony PERARD X-Patchwork-Id: 9225545 Return-Path: Received: from mail.wl.linuxfoundation.org (pdx-wl-mail.web.codeaurora.org [172.30.200.125]) by pdx-korg-patchwork.web.codeaurora.org (Postfix) with ESMTP id CBDC96086B for ; Tue, 12 Jul 2016 14:49:11 +0000 (UTC) Received: from mail.wl.linuxfoundation.org (localhost [127.0.0.1]) by mail.wl.linuxfoundation.org (Postfix) with ESMTP id BBA9A27D29 for ; Tue, 12 Jul 2016 14:49:11 +0000 (UTC) Received: by mail.wl.linuxfoundation.org (Postfix, from userid 486) id B04D927DCD; Tue, 12 Jul 2016 14:49:11 +0000 (UTC) X-Spam-Checker-Version: SpamAssassin 3.3.1 (2010-03-16) on pdx-wl-mail.web.codeaurora.org X-Spam-Level: X-Spam-Status: No, score=-4.2 required=2.0 tests=BAYES_00, RCVD_IN_DNSWL_MED autolearn=ham version=3.3.1 Received: from lists.xenproject.org (lists.xenproject.org [192.237.175.120]) (using TLSv1.2 with cipher AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by mail.wl.linuxfoundation.org (Postfix) with ESMTPS id CE37C27D29 for ; Tue, 12 Jul 2016 14:49:10 +0000 (UTC) Received: from localhost ([127.0.0.1] helo=lists.xenproject.org) by lists.xenproject.org with esmtp (Exim 4.84_2) (envelope-from ) id 1bMyxJ-000456-Ic; Tue, 12 Jul 2016 14:46:37 +0000 Received: from mail6.bemta6.messagelabs.com ([85.158.143.247]) by lists.xenproject.org with esmtp (Exim 4.84_2) (envelope-from ) id 1bMyxI-0003tK-3O for xen-devel@lists.xen.org; Tue, 12 Jul 2016 14:46:36 +0000 Received: from [85.158.143.35] by server-3.bemta-6.messagelabs.com id BC/FC-22092-BC205875; Tue, 12 Jul 2016 14:46:35 +0000 X-Brightmail-Tracker: H4sIAAAAAAAAA+NgFprIIsWRWlGSWpSXmKPExsXitHRDpO5pptZ wg20TxSyWfFzM4sDocXT3b6YAxijWzLyk/IoE1oy2r8fYCmbFVmxa28DWwNjm1sXIySEh4C/x r3crK4jNJmAgsWL6VUYQW0RAVmJ11xz2LkYuDmaBdYwSP2fvYgNJCAv4Smzv3g/WwCKgKnFu6 0d2EJtXwEbiRvc7FoihchIN5+8zg9icArYSba1PwWwhoJq1S2exQthqEjcWLmOB6BWUODnzCZ jNLCAhcfDFC2aIOdwSt09PZZ7AyDcLSdksJGULGJlWMaoXpxaVpRbpGuolFWWmZ5TkJmbm6Bo amOnlphYXJ6an5iQmFesl5+duYgQGFQMQ7GDc+dzpEKMkB5OSKG8zc0u4EF9SfkplRmJxRnxR aU5q8SFGGQ4OJQleP8bWcCHBotT01Iq0zBxgeMOkJTh4lER47UHSvMUFibnFmekQqVOMilLiv JogCQGQREZpHlwbLKYuMcpKCfMyAh0ixFOQWpSbWYIq/4pRnINRSZh3OsgUnsy8Erjpr4AWMw EtrnVoBllckoiQkmpglOVZb7mbu86B8yafrGo8T//JDVIZF02nK+lOtHVXWvytS9dk5SyWg/u y5orWPDD21b+SHbr0hl/F9V1HzzM6SCinbFtnLLzR6afB5d/F8+09AmKmshaf+cfhGZLrsrRX KsWCoV5eIepy+NEkqSNN80KOdkY+kvfZL3P896W0y1MZ/x8KKfBWYinOSDTUYi4qTgQAqJ8k5 qQCAAA= X-Env-Sender: prvs=994f31c19=anthony.perard@citrix.com X-Msg-Ref: server-8.tower-21.messagelabs.com!1468334791!23436962!3 X-Originating-IP: [66.165.176.89] X-SpamReason: No, hits=0.0 required=7.0 tests=sa_preprocessor: VHJ1c3RlZCBJUDogNjYuMTY1LjE3Ni44OSA9PiAyMDMwMDc=\n, received_headers: No Received headers X-StarScan-Received: X-StarScan-Version: 8.77; banners=-,-,- X-VirusChecked: Checked Received: (qmail 64672 invoked from network); 12 Jul 2016 14:46:34 -0000 Received: from smtp.citrix.com (HELO SMTP.CITRIX.COM) (66.165.176.89) by server-8.tower-21.messagelabs.com with RC4-SHA encrypted SMTP; 12 Jul 2016 14:46:34 -0000 X-IronPort-AV: E=Sophos;i="5.28,352,1464652800"; d="scan'208";a="365726518" From: Anthony PERARD To: Date: Tue, 12 Jul 2016 15:42:39 +0100 Message-ID: <20160712144251.558-3-anthony.perard@citrix.com> X-Mailer: git-send-email 2.9.0 In-Reply-To: <20160712144251.558-1-anthony.perard@citrix.com> References: <20160712144251.558-1-anthony.perard@citrix.com> MIME-Version: 1.0 X-DLP: MIA2 Cc: Anthony PERARD , Wei Liu , boris.ostrovsky@oracle.com, Ian Jackson , roger.pau@citrix.com Subject: [Xen-devel] [PATCH v6 02/14] libxc: Prepare a start info structure for hvmloader X-BeenThere: xen-devel@lists.xen.org X-Mailman-Version: 2.1.18 Precedence: list List-Id: Xen developer discussion List-Unsubscribe: , List-Post: List-Help: List-Subscribe: , Errors-To: xen-devel-bounces@lists.xen.org Sender: "Xen-devel" X-Virus-Scanned: ClamAV using ClamSMTP ... and load BIOS/UEFI firmware into guest memory. This adds a new firmware module, system_firmware_module. It is loaded in the guest memory and final location is provided to hvmloader via the hvm_start_info struct. This patch create the hvm_start_info struct for HVM guest that have a device model, so this is now common code with HVM guest without device model. Signed-off-by: Anthony PERARD Acked-by: Wei Liu --- CC: boris.ostrovsky@oracle.com CC: roger.pau@citrix.com Changes in V6: - change description (for verbose output) of the allocation of hvm_start_info from "HVMlite start info" to "HVM start info". Changes in V5: - in alloc_magic_pages_hvm, check dom->device_model only once instead of twice (fold second if into previous else) - rework add_module_to_list to make it easier to read - also comment about the intended memory layout of start_info and the modules - in bootlate_hvm(), drop start_page and use start_info as they point to the same address - rename xc_dom_image.bios_module to xc_dom_image.system_firmware_module - rename module name to "firmware" (was "bios") Changes in V4: - change title to suggest the change of beavior - remove code to load acpi tables (dsdt) - Update public/xen.h about hvm_start_info available on other HVM guest in %ebx. Changes in V3: - rename acpi_table_module to full_acpi_module. - factorise module loading, using new function to load existing optinal module, this should not change anything - should now use the same code to loads modules as for HVMlite VMs. this avoid duplication of code. - no more generic cmdline with a list of modules, each module have its name in the module specific cmdline. - scope change for common code between hvmlite and hvmloader --- tools/libxc/include/xc_dom.h | 3 + tools/libxc/xc_dom_hvmloader.c | 3 + tools/libxc/xc_dom_x86.c | 152 +++++++++++++++++++++++++++++------------ xen/include/public/xen.h | 2 +- 4 files changed, 116 insertions(+), 44 deletions(-) diff --git a/tools/libxc/include/xc_dom.h b/tools/libxc/include/xc_dom.h index 6cb10c4..0629971 100644 --- a/tools/libxc/include/xc_dom.h +++ b/tools/libxc/include/xc_dom.h @@ -209,6 +209,9 @@ struct xc_dom_image { /* If unset disables the setup of the IOREQ pages. */ bool device_model; + /* BIOS/Firmware passed to HVMLOADER */ + struct xc_hvm_firmware_module system_firmware_module; + /* Extra ACPI tables passed to HVMLOADER */ struct xc_hvm_firmware_module acpi_module; diff --git a/tools/libxc/xc_dom_hvmloader.c b/tools/libxc/xc_dom_hvmloader.c index 8150b74..6eb8516 100644 --- a/tools/libxc/xc_dom_hvmloader.c +++ b/tools/libxc/xc_dom_hvmloader.c @@ -169,6 +169,9 @@ static int modules_init(struct xc_dom_image *dom) { int rc; + rc = module_init_one(dom, &dom->system_firmware_module, + "System Firmware module"); + if ( rc ) goto err; rc = module_init_one(dom, &dom->acpi_module, "ACPI module"); if ( rc ) goto err; rc = module_init_one(dom, &dom->smbios_module, "SMBIOS module"); diff --git a/tools/libxc/xc_dom_x86.c b/tools/libxc/xc_dom_x86.c index 021f8a8..bc2dbb2 100644 --- a/tools/libxc/xc_dom_x86.c +++ b/tools/libxc/xc_dom_x86.c @@ -69,6 +69,9 @@ #define round_up(addr, mask) ((addr) | (mask)) #define round_pg_up(addr) (((addr) + PAGE_SIZE_X86 - 1) & ~(PAGE_SIZE_X86 - 1)) +#define HVMLOADER_MODULE_MAX_COUNT 1 +#define HVMLOADER_MODULE_NAME_SIZE 10 + struct xc_dom_params { unsigned levels; xen_vaddr_t vaddr_mask; @@ -590,6 +593,7 @@ static int alloc_magic_pages_hvm(struct xc_dom_image *dom) xen_pfn_t special_array[X86_HVM_NR_SPECIAL_PAGES]; xen_pfn_t ioreq_server_array[NR_IOREQ_SERVER_PAGES]; xc_interface *xch = dom->xch; + size_t start_info_size = sizeof(struct hvm_start_info); /* Allocate and clear special pages. */ for ( i = 0; i < X86_HVM_NR_SPECIAL_PAGES; i++ ) @@ -624,8 +628,6 @@ static int alloc_magic_pages_hvm(struct xc_dom_image *dom) if ( !dom->device_model ) { - size_t start_info_size = sizeof(struct hvm_start_info); - if ( dom->cmdline ) { dom->cmdline_size = ROUNDUP(strlen(dom->cmdline) + 1, 8); @@ -635,17 +637,18 @@ static int alloc_magic_pages_hvm(struct xc_dom_image *dom) /* Limited to one module. */ if ( dom->ramdisk_blob ) start_info_size += sizeof(struct hvm_modlist_entry); - - rc = xc_dom_alloc_segment(dom, &dom->start_info_seg, - "HVMlite start info", 0, start_info_size); - if ( rc != 0 ) - { - DOMPRINTF("Unable to reserve memory for the start info"); - goto out; - } } else { + start_info_size += + sizeof(struct hvm_modlist_entry) * HVMLOADER_MODULE_MAX_COUNT; + /* + * Add extra space to write modules name. + * The HVMLOADER_MODULE_NAME_SIZE accounts for NUL byte. + */ + start_info_size += + HVMLOADER_MODULE_NAME_SIZE * HVMLOADER_MODULE_MAX_COUNT; + /* * Allocate and clear additional ioreq server pages. The default * server will use the IOREQ and BUFIOREQ special pages above. @@ -672,6 +675,14 @@ static int alloc_magic_pages_hvm(struct xc_dom_image *dom) NR_IOREQ_SERVER_PAGES); } + rc = xc_dom_alloc_segment(dom, &dom->start_info_seg, + "HVM start info", 0, start_info_size); + if ( rc != 0 ) + { + DOMPRINTF("Unable to reserve memory for the start info"); + goto out; + } + /* * Identity-map page table is required for running with CR0.PG=0 when * using Intel EPT. Create a 32-bit non-PAE page directory of superpages. @@ -1689,42 +1700,89 @@ static int alloc_pgtables_hvm(struct xc_dom_image *dom) return 0; } +/* + * The memory layout of the start_info page and the modules, and where the + * addresses are stored: + * + * /----------------------------------\ + * | struct hvm_start_info | + * +----------------------------------+ <- start_info->modlist_paddr + * | struct hvm_modlist_entry[0] | + * +----------------------------------+ + * | struct hvm_modlist_entry[1] | + * +----------------------------------+ <- modlist[0].cmdline_paddr + * | cmdline of module 0 | + * | char[HVMLOADER_MODULE_NAME_SIZE] | + * +----------------------------------+ <- modlist[1].cmdline_paddr + * | cmdline of module 1 | + * +----------------------------------+ + */ +static void add_module_to_list(struct xc_dom_image *dom, + struct xc_hvm_firmware_module *module, + const char *name, + struct hvm_modlist_entry *modlist, + struct hvm_start_info *start_info) +{ + uint32_t index = start_info->nr_modules; + void *modules_cmdline_start = modlist + HVMLOADER_MODULE_MAX_COUNT; + uint64_t modlist_paddr = (dom->start_info_seg.pfn << PAGE_SHIFT) + + ((uintptr_t)modlist - (uintptr_t)start_info); + uint64_t modules_cmdline_paddr = modlist_paddr + + sizeof(struct hvm_modlist_entry) * HVMLOADER_MODULE_MAX_COUNT; + + if ( module->length == 0 ) + return; + + assert(start_info->nr_modules < HVMLOADER_MODULE_MAX_COUNT); + assert(strnlen(name, HVMLOADER_MODULE_NAME_SIZE) + < HVMLOADER_MODULE_NAME_SIZE); + + modlist[index].paddr = module->guest_addr_out; + modlist[index].size = module->length; + + strncpy(modules_cmdline_start + HVMLOADER_MODULE_NAME_SIZE * index, + name, HVMLOADER_MODULE_NAME_SIZE); + modlist[index].cmdline_paddr = + modules_cmdline_paddr + HVMLOADER_MODULE_NAME_SIZE * index; + + start_info->nr_modules++; +} + static int bootlate_hvm(struct xc_dom_image *dom) { uint32_t domid = dom->guest_domid; xc_interface *xch = dom->xch; + struct hvm_start_info *start_info; + size_t start_info_size; + struct hvm_modlist_entry *modlist; - if ( !dom->device_model ) - { - struct hvm_start_info *start_info; - size_t start_info_size; - void *start_page; - - start_info_size = sizeof(*start_info) + dom->cmdline_size; - if ( dom->ramdisk_blob ) - start_info_size += sizeof(struct hvm_modlist_entry); + start_info_size = sizeof(*start_info) + dom->cmdline_size; + if ( dom->ramdisk_blob ) + start_info_size += sizeof(struct hvm_modlist_entry); - if ( start_info_size > - dom->start_info_seg.pages << XC_DOM_PAGE_SHIFT(dom) ) - { - DOMPRINTF("Trying to map beyond start_info_seg"); - return -1; - } + if ( start_info_size > + dom->start_info_seg.pages << XC_DOM_PAGE_SHIFT(dom) ) + { + DOMPRINTF("Trying to map beyond start_info_seg"); + return -1; + } - start_page = xc_map_foreign_range(xch, domid, start_info_size, - PROT_READ | PROT_WRITE, - dom->start_info_seg.pfn); - if ( start_page == NULL ) - { - DOMPRINTF("Unable to map HVM start info page"); - return -1; - } + start_info = xc_map_foreign_range(xch, domid, start_info_size, + PROT_READ | PROT_WRITE, + dom->start_info_seg.pfn); + if ( start_info == NULL ) + { + DOMPRINTF("Unable to map HVM start info page"); + return -1; + } - start_info = start_page; + modlist = (void*)(start_info + 1) + dom->cmdline_size; + if ( !dom->device_model ) + { if ( dom->cmdline ) { - char *cmdline = start_page + sizeof(*start_info); + char *cmdline = (void*)(start_info + 1); strncpy(cmdline, dom->cmdline, dom->cmdline_size); start_info->cmdline_paddr = (dom->start_info_seg.pfn << PAGE_SHIFT) + @@ -1733,22 +1791,30 @@ static int bootlate_hvm(struct xc_dom_image *dom) if ( dom->ramdisk_blob ) { - struct hvm_modlist_entry *modlist = - start_page + sizeof(*start_info) + dom->cmdline_size; modlist[0].paddr = dom->ramdisk_seg.vstart - dom->parms.virt_base; modlist[0].size = dom->ramdisk_seg.vend - dom->ramdisk_seg.vstart; - start_info->modlist_paddr = (dom->start_info_seg.pfn << PAGE_SHIFT) + - ((uintptr_t)modlist - (uintptr_t)start_info); start_info->nr_modules = 1; } - - start_info->magic = XEN_HVM_START_MAGIC_VALUE; - - munmap(start_page, start_info_size); } else { + add_module_to_list(dom, &dom->system_firmware_module, "firmware", + modlist, start_info); + } + + if ( start_info->nr_modules ) + { + start_info->modlist_paddr = (dom->start_info_seg.pfn << PAGE_SHIFT) + + ((uintptr_t)modlist - (uintptr_t)start_info); + } + + start_info->magic = XEN_HVM_START_MAGIC_VALUE; + + munmap(start_info, start_info_size); + + if ( dom->device_model ) + { void *hvm_info_page; if ( (hvm_info_page = xc_map_foreign_range( diff --git a/xen/include/public/xen.h b/xen/include/public/xen.h index b9e5e0f..427e74f 100644 --- a/xen/include/public/xen.h +++ b/xen/include/public/xen.h @@ -821,7 +821,7 @@ struct start_info { typedef struct start_info start_info_t; /* - * Start of day structure passed to PVH guests in %ebx. + * Start of day structure passed to PVH guests and to HVM guests in %ebx. * * NOTE: nothing will be loaded at physical address 0, so a 0 value in any * of the address fields should be treated as not present.