From patchwork Fri May 26 02:24:38 2017 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Haozhong Zhang X-Patchwork-Id: 9749623 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 DBCD76032C for ; Fri, 26 May 2017 02:26:06 +0000 (UTC) Received: from mail.wl.linuxfoundation.org (localhost [127.0.0.1]) by mail.wl.linuxfoundation.org (Postfix) with ESMTP id CF0E027F4B for ; Fri, 26 May 2017 02:26:06 +0000 (UTC) Received: by mail.wl.linuxfoundation.org (Postfix, from userid 486) id C3ED728380; Fri, 26 May 2017 02:26:06 +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=-6.9 required=2.0 tests=BAYES_00,RCVD_IN_DNSWL_HI autolearn=ham version=3.3.1 Received: from lists.gnu.org (lists.gnu.org [208.118.235.17]) (using TLSv1 with cipher AES256-SHA (256/256 bits)) (No client certificate requested) by mail.wl.linuxfoundation.org (Postfix) with ESMTPS id EDBAC27F4B for ; Fri, 26 May 2017 02:26:05 +0000 (UTC) Received: from localhost ([::1]:34572 helo=lists.gnu.org) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1dE4x3-0001rY-3I for patchwork-qemu-devel@patchwork.kernel.org; Thu, 25 May 2017 22:26:05 -0400 Received: from eggs.gnu.org ([2001:4830:134:3::10]:49184) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1dE4w1-0001ok-Mv for qemu-devel@nongnu.org; Thu, 25 May 2017 22:25:05 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1dE4w0-0002IP-CU for qemu-devel@nongnu.org; Thu, 25 May 2017 22:25:01 -0400 Received: from mga02.intel.com ([134.134.136.20]:56309) by eggs.gnu.org with esmtps (TLS1.0:DHE_RSA_AES_256_CBC_SHA1:32) (Exim 4.71) (envelope-from ) id 1dE4w0-0002Hy-0v for qemu-devel@nongnu.org; Thu, 25 May 2017 22:25:00 -0400 Received: from fmsmga005.fm.intel.com ([10.253.24.32]) by orsmga101.jf.intel.com with ESMTP/TLS/DHE-RSA-AES256-GCM-SHA384; 25 May 2017 19:24:58 -0700 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="5.38,394,1491289200"; d="scan'208";a="106767504" Received: from hz-desktop.sh.intel.com (HELO localhost) ([10.239.159.149]) by fmsmga005.fm.intel.com with ESMTP; 25 May 2017 19:24:56 -0700 From: Haozhong Zhang To: qemu-devel@nongnu.org Date: Fri, 26 May 2017 10:24:38 +0800 Message-Id: <20170526022438.17478-2-haozhong.zhang@intel.com> X-Mailer: git-send-email 2.11.0 In-Reply-To: <20170526022438.17478-1-haozhong.zhang@intel.com> References: <20170526022438.17478-1-haozhong.zhang@intel.com> X-detected-operating-system: by eggs.gnu.org: Genre and OS details not recognized. X-Received-From: 134.134.136.20 Subject: [Qemu-devel] [PATCH 2/2] hostmem-file: add an attribute 'align' to set its alignment X-BeenThere: qemu-devel@nongnu.org X-Mailman-Version: 2.1.21 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Cc: Haozhong Zhang , Xiao Guangrong , Eduardo Habkost , Peter Crosthwaite , Stefan Hajnoczi , Igor Mammedov , Paolo Bonzini , Dan Williams , Richard Henderson Errors-To: qemu-devel-bounces+patchwork-qemu-devel=patchwork.kernel.org@nongnu.org Sender: "Qemu-devel" X-Virus-Scanned: ClamAV using ClamSMTP file_ram_alloc() currently maps the backend file via mmap to a virtual address aligned to the value returned by qemu_fd_getpagesize(). When a DAX device (e.g. /dev/dax0.0) is used as the backend file, its kernel mmap implementation may require an alignment larger than what qemu_fd_get_pagesize() returns (e.g. 2MB vs. 4KB), and mmap may fail. This commit adds an attribute 'align' to hostmem-file, so that users can specify a proper alignment that satisfies the kernel requirement. If 'align' is not specified or is 0, the value returned by qemu_fd_get_pagesize() will be used as before. Signed-off-by: Haozhong Zhang --- Cc: Eduardo Habkost Cc: Igor Mammedov Cc: Paolo Bonzini Cc: Peter Crosthwaite Cc: Richard Henderson Cc: Xiao Guangrong Cc: Stefan Hajnoczi Cc: Dan Williams --- backends/hostmem-file.c | 41 ++++++++++++++++++++++++++++++++++++++++- exec.c | 8 +++++++- include/exec/memory.h | 2 ++ memory.c | 2 ++ numa.c | 2 +- 5 files changed, 52 insertions(+), 3 deletions(-) diff --git a/backends/hostmem-file.c b/backends/hostmem-file.c index fc4ef46d11..d44fb41b55 100644 --- a/backends/hostmem-file.c +++ b/backends/hostmem-file.c @@ -33,6 +33,7 @@ struct HostMemoryBackendFile { bool share; char *mem_path; + uint64_t align; }; static void @@ -57,7 +58,7 @@ file_backend_memory_alloc(HostMemoryBackend *backend, Error **errp) path = object_get_canonical_path(OBJECT(backend)); memory_region_init_ram_from_file(&backend->mr, OBJECT(backend), path, - backend->size, fb->share, + backend->size, fb->align, fb->share, fb->mem_path, errp); g_free(path); } @@ -104,6 +105,40 @@ static void file_memory_backend_set_share(Object *o, bool value, Error **errp) } static void +file_memory_backend_get_align(Object *o, Visitor *v, const char *name, + void *opaque, Error **errp) +{ + HostMemoryBackendFile *fb = MEMORY_BACKEND_FILE(o); + uint64_t val = fb->align; + + visit_type_size(v, name, &val, errp); +} + +static void +file_memory_backend_set_align(Object *o, Visitor *v, const char *name, + void *opaque, Error **errp) +{ + HostMemoryBackend *backend = MEMORY_BACKEND(o); + HostMemoryBackendFile *fb = MEMORY_BACKEND_FILE(o); + Error *local_err = NULL; + uint64_t val; + + if (host_memory_backend_mr_inited(backend)) { + error_setg(&local_err, "cannot change property value"); + goto out; + } + + visit_type_size(v, name, &val, &local_err); + if (local_err) { + goto out; + } + fb->align = val; + + out: + error_propagate(errp, local_err); +} + +static void file_backend_class_init(ObjectClass *oc, void *data) { HostMemoryBackendClass *bc = MEMORY_BACKEND_CLASS(oc); @@ -116,6 +151,10 @@ file_backend_class_init(ObjectClass *oc, void *data) object_class_property_add_str(oc, "mem-path", get_mem_path, set_mem_path, &error_abort); + object_class_property_add(oc, "align", "int", + file_memory_backend_get_align, + file_memory_backend_set_align, + NULL, NULL, &error_abort); } static void file_backend_instance_finalize(Object *o) diff --git a/exec.c b/exec.c index ff16f04f2b..5bb62e2e98 100644 --- a/exec.c +++ b/exec.c @@ -1549,7 +1549,13 @@ static void *file_ram_alloc(RAMBlock *block, } block->page_size = qemu_fd_getpagesize(fd); - block->mr->align = block->page_size; + if (block->mr->align % block->page_size) { + error_setg(errp, "alignment 0x%" PRIx64 " must be " + "multiple of page size 0x%" PRIx64, + block->mr->align, block->page_size); + goto error; + } + block->mr->align = MAX(block->page_size, block->mr->align); #if defined(__s390x__) if (kvm_enabled()) { block->mr->align = MAX(block->mr->align, QEMU_VMALLOC_ALIGN); diff --git a/include/exec/memory.h b/include/exec/memory.h index 99e0f54d86..05d3d0da3b 100644 --- a/include/exec/memory.h +++ b/include/exec/memory.h @@ -441,6 +441,7 @@ void memory_region_init_resizeable_ram(MemoryRegion *mr, * @name: Region name, becomes part of RAMBlock name used in migration stream * must be unique within any device * @size: size of the region. + * @align: alignment of the region. * @share: %true if memory must be mmaped with the MAP_SHARED flag * @path: the path in which to allocate the RAM. * @errp: pointer to Error*, to store an error if it happens. @@ -449,6 +450,7 @@ void memory_region_init_ram_from_file(MemoryRegion *mr, struct Object *owner, const char *name, uint64_t size, + uint64_t align, bool share, const char *path, Error **errp); diff --git a/memory.c b/memory.c index b727f5ec0e..5165b9aa08 100644 --- a/memory.c +++ b/memory.c @@ -1386,6 +1386,7 @@ void memory_region_init_ram_from_file(MemoryRegion *mr, struct Object *owner, const char *name, uint64_t size, + uint64_t align, bool share, const char *path, Error **errp) @@ -1394,6 +1395,7 @@ void memory_region_init_ram_from_file(MemoryRegion *mr, mr->ram = true; mr->terminates = true; mr->destructor = memory_region_destructor_ram; + mr->align = align; mr->ram_block = qemu_ram_alloc_from_file(size, mr, share, path, errp); mr->dirty_log_mask = tcg_enabled() ? (1 << DIRTY_MEMORY_CODE) : 0; } diff --git a/numa.c b/numa.c index ca731455e9..39a25aa1d2 100644 --- a/numa.c +++ b/numa.c @@ -541,7 +541,7 @@ static void allocate_system_memory_nonnuma(MemoryRegion *mr, Object *owner, if (mem_path) { #ifdef __linux__ Error *err = NULL; - memory_region_init_ram_from_file(mr, owner, name, ram_size, false, + memory_region_init_ram_from_file(mr, owner, name, ram_size, 0, false, mem_path, &err); if (err) { error_report_err(err);