From nobody Mon May 12 22:56:34 2025 Delivered-To: importer2@patchew.org Authentication-Results: mx.zohomail.com; dkim=pass; spf=pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) smtp.mailfrom=qemu-devel-bounces+importer2=patchew.org@nongnu.org ARC-Seal: i=1; a=rsa-sha256; t=1715092649; cv=none; d=zohomail.com; s=zohoarc; b=eOmTnFaXL+ZDBW91PhLvTDCyFSpRtujJckczp40OgfV1bXVsnQ2zgODc6IVTh5bdB5LpQN/I+yxAFQywanxXTSwH2241MzhPxFIFhGybgv0fjjq/m/0SQW1N04iyrCOdcSQxsBZ3cHZ9OC5ixvkq/rdMdmyVmu7Rh52Q2PJBwaw= ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=zohomail.com; s=zohoarc; t=1715092649; h=Content-Transfer-Encoding:Cc:Cc:Date:Date:From:From:In-Reply-To:List-Subscribe:List-Post:List-Id:List-Archive:List-Help:List-Unsubscribe:MIME-Version:Message-ID:References:Sender:Subject:Subject:To:To:Message-Id:Reply-To; bh=ewN5YfXgUpF8Cx+aqolyklTQ+08U+mpSJMGlBwx0Dlg=; b=K+7YWMnXzfq0FEdIZ3IqdzBziSybjrV2PTy5S6L1oB+e7iR9Fi0vemnjdBpdnrdb3vPEyfqcUmahbA6uhHeca/jm3HsyeaY0GwnfSO93Cj4P9p8xH7hfrpTYbZzX8oRaOFRN/4J4ZNQ88z8bPoMmM+eSb+oZW8VO9QpHYf3XMgI= ARC-Authentication-Results: i=1; mx.zohomail.com; dkim=pass; spf=pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) smtp.mailfrom=qemu-devel-bounces+importer2=patchew.org@nongnu.org Return-Path: Received: from lists.gnu.org (lists.gnu.org [209.51.188.17]) by mx.zohomail.com with SMTPS id 1715092649117947.0366958544465; Tue, 7 May 2024 07:37:29 -0700 (PDT) Received: from localhost ([::1] helo=lists1p.gnu.org) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1s4LuM-0006ZB-K5; Tue, 07 May 2024 10:35:06 -0400 Received: from eggs.gnu.org ([2001:470:142:3::10]) by lists.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1s4LuI-0006LA-GM for qemu-devel@nongnu.org; Tue, 07 May 2024 10:35:02 -0400 Received: from mail-pf1-x42f.google.com ([2607:f8b0:4864:20::42f]) by eggs.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_128_GCM_SHA256:128) (Exim 4.90_1) (envelope-from ) id 1s4LuB-0002I0-GU for qemu-devel@nongnu.org; Tue, 07 May 2024 10:35:02 -0400 Received: by mail-pf1-x42f.google.com with SMTP id d2e1a72fcca58-6f4302187c0so4010254b3a.1 for ; Tue, 07 May 2024 07:34:55 -0700 (PDT) Received: from mnissler.ba.rivosinc.com ([64.71.180.162]) by smtp.gmail.com with ESMTPSA id pq10-20020a17090b3d8a00b002af8056917csm11827206pjb.29.2024.05.07.07.34.53 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Tue, 07 May 2024 07:34:53 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=rivosinc-com.20230601.gappssmtp.com; s=20230601; t=1715092494; x=1715697294; darn=nongnu.org; h=content-transfer-encoding:mime-version:references:in-reply-to :message-id:date:subject:cc:to:from:from:to:cc:subject:date :message-id:reply-to; bh=ewN5YfXgUpF8Cx+aqolyklTQ+08U+mpSJMGlBwx0Dlg=; b=Y/M2NRCZJnr3POJNRVDzaQvj7EqH7jU0YmFWm/yVUOjVCnPm63TjE7qHsLU4EmvcfA mDW/HyfmeE3Kd8Tc7aeRU+WIBGaxh8Mn3VvRmr0pYdfEk6im14JFOsPrC/ypnisEeD6Y hp+RlmjMycJOsiGx0QQ1x7cCEyjr5yfnwkaKtEsC9Sm954jjlAuUAKm4U3KlgmOyFZ8c h+qUHVg3njKFqLQYUMJKZHl9pn7SvfTEI+a/ev4CfJvPYeOdkqvk42J4H1h4whpfNsp/ GRF3jmiT5siDKKgyw9bX18ykZX9yd6ts4qvDhzEX4CoScSIDVQIPFx2fc1YfkylZQDEo d72g== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1715092494; x=1715697294; h=content-transfer-encoding:mime-version:references:in-reply-to :message-id:date:subject:cc:to:from:x-gm-message-state:from:to:cc :subject:date:message-id:reply-to; bh=ewN5YfXgUpF8Cx+aqolyklTQ+08U+mpSJMGlBwx0Dlg=; b=bWy9ViS6XUbzb3JiZkgIcLuPS4zC/Qy1bdimf2jWVkK+7HZCU15LrcHWhZlYTXMHsC zf8ptkiCSmQiHk8HDBZNb3A9gUEvNPHJGmEtXyYHYzKLIBM3vwv/Q81dWMbU1sNc2xhs aEPD+Z8Ls79K+QozVlMROYxJIUAecvDASoc1bCkC8kxVUj2tCU/gA0QrPO5gRGtBqXxl w6hDAy+t5bCCH8CU8YGnSHw12LNpDHFy+7amXoEEMHR6DvbxUXPfMAAM1/VnWtV7dPIm iho2R4F6/+CMLa76Y6+rEWOq7p9hjwTut+u/b8YQXDMwVjvdO4GdUi/tMrYLv0d07QhL FW3g== X-Forwarded-Encrypted: i=1; AJvYcCWlRis8Uk+J4y8LZc7q0CEqnw1tONMs7weAarh2k/GmZEpb5wL6gDZpcaXR1OHwbkpX6m3i9uxjiYA2WUBvFlKALbrQvpI= X-Gm-Message-State: AOJu0Yxwe//jtsCKx2bfD2H5Y7LCnNCBpqoZPa4OrhnU/+pB4p01EpcZ IWg1Gt7jPVCL+vZagrY5kNDxekAxSt8QUiBCGKXzHQXNaL1tEAQCfUmH3x98NkA= X-Google-Smtp-Source: AGHT+IEAziI+9gsftl0DXJJx6vUevNXfKD9RiOu2n+A6mzmWunaKPNCiF/TeS0k0w7vPRaTlX0hQ3w== X-Received: by 2002:a17:90a:ab83:b0:2b1:54e4:e125 with SMTP id n3-20020a17090aab8300b002b154e4e125mr4067812pjq.22.1715092494061; Tue, 07 May 2024 07:34:54 -0700 (PDT) From: Mattias Nissler To: jag.raman@oracle.com, peterx@redhat.com, stefanha@redhat.com, qemu-devel@nongnu.org Cc: Marcel Apfelbaum , =?UTF-8?q?Philippe=20Mathieu-Daud=C3=A9?= , john.levon@nutanix.com, David Hildenbrand , "Michael S. Tsirkin" , Paolo Bonzini , Elena Ufimtseva , Richard Henderson , Mattias Nissler Subject: [PATCH v10 6/7] vfio-user: Message-based DMA support Date: Tue, 7 May 2024 07:34:30 -0700 Message-ID: <20240507143431.464382-7-mnissler@rivosinc.com> X-Mailer: git-send-email 2.43.2 In-Reply-To: <20240507143431.464382-1-mnissler@rivosinc.com> References: <20240507143431.464382-1-mnissler@rivosinc.com> MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Received-SPF: pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) client-ip=209.51.188.17; envelope-from=qemu-devel-bounces+importer2=patchew.org@nongnu.org; helo=lists.gnu.org; Received-SPF: pass client-ip=2607:f8b0:4864:20::42f; envelope-from=mnissler@rivosinc.com; helo=mail-pf1-x42f.google.com X-Spam_score_int: -18 X-Spam_score: -1.9 X-Spam_bar: - X-Spam_report: (-1.9 / 5.0 requ) BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, RCVD_IN_DNSWL_NONE=-0.0001, SPF_HELO_NONE=0.001, SPF_PASS=-0.001 autolearn=ham autolearn_force=no X-Spam_action: no action X-BeenThere: qemu-devel@nongnu.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: qemu-devel-bounces+importer2=patchew.org@nongnu.org Sender: qemu-devel-bounces+importer2=patchew.org@nongnu.org X-ZohoMail-DKIM: pass (identity @rivosinc-com.20230601.gappssmtp.com) X-ZM-MESSAGEID: 1715092650243100006 Content-Type: text/plain; charset="utf-8" Wire up support for DMA for the case where the vfio-user client does not provide mmap()-able file descriptors, but DMA requests must be performed via the VFIO-user protocol. This installs an indirect memory region, which already works for pci_dma_{read,write}, and pci_dma_map works thanks to the existing DMA bounce buffering support. Note that while simple scenarios work with this patch, there's a known race condition in libvfio-user that will mess up the communication channel. See https://github.com/nutanix/libvfio-user/issues/279 for details as well as a proposed fix. Reviewed-by: Jagannathan Raman Signed-off-by: Mattias Nissler --- hw/remote/trace-events | 2 + hw/remote/vfio-user-obj.c | 100 ++++++++++++++++++++++++++++++++------ 2 files changed, 87 insertions(+), 15 deletions(-) diff --git a/hw/remote/trace-events b/hw/remote/trace-events index 0d1b7d56a5..358a68fb34 100644 --- a/hw/remote/trace-events +++ b/hw/remote/trace-events @@ -9,6 +9,8 @@ vfu_cfg_read(uint32_t offset, uint32_t val) "vfu: cfg: 0x%x= -> 0x%x" vfu_cfg_write(uint32_t offset, uint32_t val) "vfu: cfg: 0x%x <- 0x%x" vfu_dma_register(uint64_t gpa, size_t len) "vfu: registering GPA 0x%"PRIx6= 4", %zu bytes" vfu_dma_unregister(uint64_t gpa) "vfu: unregistering GPA 0x%"PRIx64"" +vfu_dma_read(uint64_t gpa, size_t len) "vfu: DMA read 0x%"PRIx64", %zu byt= es" +vfu_dma_write(uint64_t gpa, size_t len) "vfu: DMA write 0x%"PRIx64", %zu b= ytes" vfu_bar_register(int i, uint64_t addr, uint64_t size) "vfu: BAR %d: addr 0= x%"PRIx64" size 0x%"PRIx64"" vfu_bar_rw_enter(const char *op, uint64_t addr) "vfu: %s request for BAR a= ddress 0x%"PRIx64"" vfu_bar_rw_exit(const char *op, uint64_t addr) "vfu: Finished %s of BAR ad= dress 0x%"PRIx64"" diff --git a/hw/remote/vfio-user-obj.c b/hw/remote/vfio-user-obj.c index d9b879e056..a15e291c9a 100644 --- a/hw/remote/vfio-user-obj.c +++ b/hw/remote/vfio-user-obj.c @@ -300,6 +300,63 @@ static ssize_t vfu_object_cfg_access(vfu_ctx_t *vfu_ct= x, char * const buf, return count; } =20 +static MemTxResult vfu_dma_read(void *opaque, hwaddr addr, uint64_t *val, + unsigned size, MemTxAttrs attrs) +{ + MemoryRegion *region =3D opaque; + vfu_ctx_t *vfu_ctx =3D VFU_OBJECT(region->owner)->vfu_ctx; + uint8_t buf[sizeof(uint64_t)]; + + trace_vfu_dma_read(region->addr + addr, size); + + g_autofree dma_sg_t *sg =3D g_malloc0(dma_sg_size()); + vfu_dma_addr_t vfu_addr =3D (vfu_dma_addr_t)(region->addr + addr); + if (vfu_addr_to_sgl(vfu_ctx, vfu_addr, size, sg, 1, PROT_READ) < 0 || + vfu_sgl_read(vfu_ctx, sg, 1, buf) !=3D 0) { + return MEMTX_ERROR; + } + + *val =3D ldn_he_p(buf, size); + + return MEMTX_OK; +} + +static MemTxResult vfu_dma_write(void *opaque, hwaddr addr, uint64_t val, + unsigned size, MemTxAttrs attrs) +{ + MemoryRegion *region =3D opaque; + vfu_ctx_t *vfu_ctx =3D VFU_OBJECT(region->owner)->vfu_ctx; + uint8_t buf[sizeof(uint64_t)]; + + trace_vfu_dma_write(region->addr + addr, size); + + stn_he_p(buf, size, val); + + g_autofree dma_sg_t *sg =3D g_malloc0(dma_sg_size()); + vfu_dma_addr_t vfu_addr =3D (vfu_dma_addr_t)(region->addr + addr); + if (vfu_addr_to_sgl(vfu_ctx, vfu_addr, size, sg, 1, PROT_WRITE) < 0 || + vfu_sgl_write(vfu_ctx, sg, 1, buf) !=3D 0) { + return MEMTX_ERROR; + } + + return MEMTX_OK; +} + +static const MemoryRegionOps vfu_dma_ops =3D { + .read_with_attrs =3D vfu_dma_read, + .write_with_attrs =3D vfu_dma_write, + .endianness =3D DEVICE_HOST_ENDIAN, + .valid =3D { + .min_access_size =3D 1, + .max_access_size =3D 8, + .unaligned =3D true, + }, + .impl =3D { + .min_access_size =3D 1, + .max_access_size =3D 8, + }, +}; + static void dma_register(vfu_ctx_t *vfu_ctx, vfu_dma_info_t *info) { VfuObject *o =3D vfu_get_private(vfu_ctx); @@ -308,17 +365,30 @@ static void dma_register(vfu_ctx_t *vfu_ctx, vfu_dma_= info_t *info) g_autofree char *name =3D NULL; struct iovec *iov =3D &info->iova; =20 - if (!info->vaddr) { - return; - } - name =3D g_strdup_printf("mem-%s-%"PRIx64"", o->device, - (uint64_t)info->vaddr); + (uint64_t)iov->iov_base); =20 subregion =3D g_new0(MemoryRegion, 1); =20 - memory_region_init_ram_ptr(subregion, NULL, name, - iov->iov_len, info->vaddr); + if (info->vaddr) { + memory_region_init_ram_ptr(subregion, OBJECT(o), name, + iov->iov_len, info->vaddr); + } else { + /* + * Note that I/O regions' MemoryRegionOps handle accesses of at mo= st 8 + * bytes at a time, and larger accesses are broken down. However, + * many/most DMA accesses are larger than 8 bytes and VFIO-user can + * handle large DMA accesses just fine, thus this size restriction + * unnecessarily hurts performance, in particular given that each + * access causes a round trip on the VFIO-user socket. + * + * TODO: Investigate how to plumb larger accesses through memory + * regions, possibly by amending MemoryRegionOps or by creating a = new + * memory region type. + */ + memory_region_init_io(subregion, OBJECT(o), &vfu_dma_ops, subregio= n, + name, iov->iov_len); + } =20 dma_as =3D pci_device_iommu_address_space(o->pci_dev); =20 @@ -330,20 +400,20 @@ static void dma_register(vfu_ctx_t *vfu_ctx, vfu_dma_= info_t *info) static void dma_unregister(vfu_ctx_t *vfu_ctx, vfu_dma_info_t *info) { VfuObject *o =3D vfu_get_private(vfu_ctx); + MemoryRegionSection mr_section; AddressSpace *dma_as =3D NULL; - MemoryRegion *mr =3D NULL; - ram_addr_t offset; =20 - mr =3D memory_region_from_host(info->vaddr, &offset); - if (!mr) { + dma_as =3D pci_device_iommu_address_space(o->pci_dev); + + mr_section =3D + memory_region_find(dma_as->root, (hwaddr)info->iova.iov_base, 1); + if (!mr_section.mr) { return; } =20 - dma_as =3D pci_device_iommu_address_space(o->pci_dev); - - memory_region_del_subregion(dma_as->root, mr); + memory_region_del_subregion(dma_as->root, mr_section.mr); =20 - object_unparent((OBJECT(mr))); + object_unparent((OBJECT(mr_section.mr))); =20 trace_vfu_dma_unregister((uint64_t)info->iova.iov_base); } --=20 2.43.2