From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mails.dpdk.org (mails.dpdk.org [217.70.189.124]) by inbox.dpdk.org (Postfix) with ESMTP id 748E24298A; Wed, 19 Apr 2023 16:52:27 +0200 (CEST) Received: from mails.dpdk.org (localhost [127.0.0.1]) by mails.dpdk.org (Postfix) with ESMTP id 6306B40A79; Wed, 19 Apr 2023 16:52:27 +0200 (CEST) Received: from us-smtp-delivery-124.mimecast.com (us-smtp-delivery-124.mimecast.com [170.10.129.124]) by mails.dpdk.org (Postfix) with ESMTP id 736524021F for ; Wed, 19 Apr 2023 16:52:25 +0200 (CEST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=redhat.com; s=mimecast20190719; t=1681915945; h=from:from:reply-to:subject:subject:date:date:message-id:message-id: to:to:cc:cc:mime-version:mime-version:content-type:content-type: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references; bh=O98lmjPqLdow2DMpgL5LgEaJW9eHVNZI868KWedCUqI=; b=ETDQzcjZfArrXgUFyglltgr1KYCi7cVwlRDO3s2/l+hzK4fH4ptPgen+W6GqQereRCAP2W Bgs/slWW3MMVrBZl2uJjWsmh/A0Chb/qXnXNl73uUeZU8DPERmOuDeUV0xQtMYUV/fuv6M nKrp3/L1HOhFgKp4jHhD7CenQfDIsu0= Received: from mail-ed1-f69.google.com (mail-ed1-f69.google.com [209.85.208.69]) by relay.mimecast.com with ESMTP with STARTTLS (version=TLSv1.3, cipher=TLS_AES_256_GCM_SHA384) id us-mta-647-wxAMpDxqMOuZaVMfKJcoYw-1; Wed, 19 Apr 2023 10:52:23 -0400 X-MC-Unique: wxAMpDxqMOuZaVMfKJcoYw-1 Received: by mail-ed1-f69.google.com with SMTP id g21-20020a50d0d5000000b005067d6335c3so59023edf.6 for ; Wed, 19 Apr 2023 07:52:23 -0700 (PDT) X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20221208; t=1681915942; x=1684507942; h=content-transfer-encoding:cc:to:subject:message-id:date:from :in-reply-to:references:mime-version:x-gm-message-state:from:to:cc :subject:date:message-id:reply-to; bh=O98lmjPqLdow2DMpgL5LgEaJW9eHVNZI868KWedCUqI=; b=fSua/Mq/X8dUVFlViJssDtfERyMutP05WGRAKVsY+X21rG61yjVXcny0IJgcWQ8z26 2rVjTInLVnBy1CcvyNbdNJ9LNWXzhqwbLGJb8h6szRFCEGj9c60WBf+r56W4ggOM+zQv u8Cz7nzyUs0YzzWr9MwCEHiFbJ2WH4py6Ssd+nuOMBaBM1rZmN6kj6Zco72JD0Rn9R2q JWsC6UjncH38zL4s0GPbPR1JsUe7ZkZIbGmjRCPKFp46RZiV4nFNnXy2KfXi0fW0og4g EG/38K5xg4ZCmZAouBrhGF6/Jy7Ajt7UZOxf8OfGPncg/JxJwe9+TmXlv7NQjoBPoSbC +RmQ== X-Gm-Message-State: AAQBX9eGGdCOwATcxJ8VtJI+T2WPR7qnNabdpurnvULi1TNnVZkOHYii 9m3yibuffmzZgEzmMN893+LcEWXrDWxbG0cqSP2xRjTAP6DKPTgF1S1Rm8aLjP910oREOFwlsJw Ze48IoF+UP8yHSnF9aE4= X-Received: by 2002:a17:906:e290:b0:947:80a4:5cce with SMTP id gg16-20020a170906e29000b0094780a45ccemr13862406ejb.65.1681915942750; Wed, 19 Apr 2023 07:52:22 -0700 (PDT) X-Google-Smtp-Source: AKy350bRkLtgtQL/lkFtMQrbGQyetOCv+8KEl+ewE4fsIZlDG27DMpD3sjcIxz5Y0NkeLnsnDoyIISjzEo+CFfzOnqg= X-Received: by 2002:a17:906:e290:b0:947:80a4:5cce with SMTP id gg16-20020a170906e29000b0094780a45ccemr13862387ejb.65.1681915942473; Wed, 19 Apr 2023 07:52:22 -0700 (PDT) MIME-Version: 1.0 References: <20230331154259.1447831-1-maxime.coquelin@redhat.com> <20230331154259.1447831-6-maxime.coquelin@redhat.com> In-Reply-To: From: Mike Pattrick Date: Wed, 19 Apr 2023 10:52:11 -0400 Message-ID: Subject: Re: [RFC 05/27] vhost: add helper for IOTLB entries shared page check To: Maxime Coquelin Cc: dev@dpdk.org, david.marchand@redhat.com, chenbo.xia@intel.com, fbl@redhat.com, jasowang@redhat.com, cunming.liang@intel.com, xieyongji@bytedance.com, echaudro@redhat.com, eperezma@redhat.com, amorenoz@redhat.com X-Mimecast-Spam-Score: 0 X-Mimecast-Originator: redhat.com Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable X-BeenThere: dev@dpdk.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: DPDK patches and discussions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: dev-bounces@dpdk.org On Wed, Apr 19, 2023 at 5:35=E2=80=AFAM Maxime Coquelin wrote: > > Hi Mike, > > On 4/17/23 21:39, Mike Pattrick wrote: > > Hi Maxime, > > > > On Fri, Mar 31, 2023 at 11:43=E2=80=AFAM Maxime Coquelin > > wrote: > >> > >> This patch introduces a helper to check whether two IOTLB > >> entries share a page. > >> > >> Signed-off-by: Maxime Coquelin > >> --- > >> lib/vhost/iotlb.c | 25 ++++++++++++++++++++----- > >> 1 file changed, 20 insertions(+), 5 deletions(-) > >> > >> diff --git a/lib/vhost/iotlb.c b/lib/vhost/iotlb.c > >> index e8f1cb661e..d919f74704 100644 > >> --- a/lib/vhost/iotlb.c > >> +++ b/lib/vhost/iotlb.c > >> @@ -23,6 +23,23 @@ struct vhost_iotlb_entry { > >> > >> #define IOTLB_CACHE_SIZE 2048 > >> > >> +static bool > >> +vhost_user_iotlb_share_page(struct vhost_iotlb_entry *a, struct vhost= _iotlb_entry *b, > >> + uint64_t align) > >> +{ > >> + uint64_t a_end, b_start; > >> + > >> + if (a =3D=3D NULL || b =3D=3D NULL) > >> + return false; > >> + > >> + /* Assumes entry a lower than entry b */ > >> + RTE_ASSERT(a->uaddr < b->uaddr); > >> + a_end =3D RTE_ALIGN_CEIL(a->uaddr + a->size, align); > >> + b_start =3D RTE_ALIGN_FLOOR(b->uaddr, align); > >> + > >> + return a_end > b_start; > > > > This may be a very minor detail, but it might improve readability if > > the a_end calculation used addr + size - 1 and the return conditional > > used >=3D. > > That's actually how I initially implemented it, but discussing with > David we agreed it would be better that way for consistency with > vhost_user_iotlb_clear_dump(). I can get behind consistency. Acked-by: Mike Pattrick > > Regards, > Maxime > > > > > Cheers, > > M > > > > > >> +} > >> + > >> static void > >> vhost_user_iotlb_set_dump(struct virtio_net *dev, struct vhost_iotlb= _entry *node) > >> { > >> @@ -37,16 +54,14 @@ static void > >> vhost_user_iotlb_clear_dump(struct virtio_net *dev, struct vhost_iot= lb_entry *node, > >> struct vhost_iotlb_entry *prev, struct vhost_iotlb_en= try *next) > >> { > >> - uint64_t align, mask; > >> + uint64_t align; > >> > >> align =3D hua_to_alignment(dev->mem, (void *)(uintptr_t)node-= >uaddr); > >> - mask =3D ~(align - 1); > >> > >> /* Don't disable coredump if the previous node is in the same= page */ > >> - if (prev =3D=3D NULL || (node->uaddr & mask) !=3D ((prev->uadd= r + prev->size - 1) & mask)) { > >> + if (!vhost_user_iotlb_share_page(prev, node, align)) { > >> /* Don't disable coredump if the next node is in the = same page */ > >> - if (next =3D=3D NULL || > >> - ((node->uaddr + node->size - 1) & mask= ) !=3D (next->uaddr & mask)) > >> + if (!vhost_user_iotlb_share_page(node, next, align)) > >> mem_set_dump((void *)(uintptr_t)node->uaddr, = node->size, false, align); > >> } > >> } > >> -- > >> 2.39.2 > >> > > >