From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from kanga.kvack.org (kanga.kvack.org [205.233.56.17]) by smtp.lore.kernel.org (Postfix) with ESMTP id D3DF7C49EA1 for ; Wed, 31 Jul 2024 04:11:59 +0000 (UTC) Received: by kanga.kvack.org (Postfix) id 4DDE16B0085; Wed, 31 Jul 2024 00:11:59 -0400 (EDT) Received: by kanga.kvack.org (Postfix, from userid 40) id 466DA6B0088; Wed, 31 Jul 2024 00:11:59 -0400 (EDT) X-Delivered-To: int-list-linux-mm@kvack.org Received: by kanga.kvack.org (Postfix, from userid 63042) id 2E0E26B0089; Wed, 31 Jul 2024 00:11:59 -0400 (EDT) X-Delivered-To: linux-mm@kvack.org Received: from relay.hostedemail.com (smtprelay0011.hostedemail.com [216.40.44.11]) by kanga.kvack.org (Postfix) with ESMTP id 0E4516B0085 for ; Wed, 31 Jul 2024 00:11:59 -0400 (EDT) Received: from smtpin05.hostedemail.com (a10.router.float.18 [10.200.18.1]) by unirelay07.hostedemail.com (Postfix) with ESMTP id 5720F160188 for ; Wed, 31 Jul 2024 04:11:58 +0000 (UTC) X-FDA: 82398724716.05.C1BC039 Received: from mail-vk1-f182.google.com (mail-vk1-f182.google.com [209.85.221.182]) by imf23.hostedemail.com (Postfix) with ESMTP id 89426140010 for ; Wed, 31 Jul 2024 04:11:56 +0000 (UTC) Authentication-Results: imf23.hostedemail.com; dkim=pass header.d=gmail.com header.s=20230601 header.b=NbFB6wAz; spf=pass (imf23.hostedemail.com: domain of 21cnbao@gmail.com designates 209.85.221.182 as permitted sender) smtp.mailfrom=21cnbao@gmail.com; dmarc=pass (policy=none) header.from=gmail.com ARC-Seal: i=1; s=arc-20220608; d=hostedemail.com; t=1722399088; a=rsa-sha256; cv=none; b=dhilZid7/1p+RmW8AI6ldtAAhNEuHAAn/l1Q2LS+SNmlXwPLgKACL2gtnRaDFcx9wQN5SQ tWisbR4j6hCt9NhRGTCZP1yTlpnYTKYGKWKVo8uZ2A7MoZzZmiGa+e/Nc5nPSIfErJUBPN CkklcrESlDykL2yKTLwL1K/53f1f0cQ= ARC-Authentication-Results: i=1; imf23.hostedemail.com; dkim=pass header.d=gmail.com header.s=20230601 header.b=NbFB6wAz; spf=pass (imf23.hostedemail.com: domain of 21cnbao@gmail.com designates 209.85.221.182 as permitted sender) smtp.mailfrom=21cnbao@gmail.com; dmarc=pass (policy=none) header.from=gmail.com ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=hostedemail.com; s=arc-20220608; t=1722399088; h=from:from:sender: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:dkim-signature; bh=Rg+ne5RbXGzTMhIX6WeWXeIYwsCxgZAOGqT3yzWRW58=; b=HXqGUV2YJtfOWOcnQO7XdD146CNuXGZq+PBSM/r9cDxCfsChcjULtY5c6wgvQB/slY7UmM Kg/7HRiHLJpPqANuTlF/orGpWGrLIb9deK6s7qTRq8VbM8H+qRLDfRHpiEbFegHdq1p4dK zk+XRypBse8eNWrxqdQ626W1JZNdYn8= Received: by mail-vk1-f182.google.com with SMTP id 71dfb90a1353d-4f6c9e8e0f2so1607777e0c.3 for ; Tue, 30 Jul 2024 21:11:56 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20230601; t=1722399115; x=1723003915; darn=kvack.org; h=content-transfer-encoding:cc:to:subject:message-id:date:from :in-reply-to:references:mime-version:from:to:cc:subject:date :message-id:reply-to; bh=Rg+ne5RbXGzTMhIX6WeWXeIYwsCxgZAOGqT3yzWRW58=; b=NbFB6wAzZpbmPwyfYRlwtMP0OVDTnJnVHv5F9G6Jyi5FVQ/wK+wmQ1QhWJzZZKvSdW i4e1J++MGvjpmckfv4hTC1ZL4U+/ozuSknDlwz1zlOPOaYY1+RAJN+lye4UvY81n/zDv Tb8Pv52sD/sllBp/xI05c8k0S595Rls8ZogwqcHnnlcnwtHRqvvBDsZxb07UNeIn62fQ o9eOyQUQHXz+6ZHjmMTK7IaeBChAqu1z85PX2HCL/jNOmwn6fSURC8lkOsjjH1TB0f2A R3cVrpTi7NCtz4AxEdUXbuSxnuxQey6wDNUsLmFHEl7fg2HIpEQKT/BEgJDkXDBrprnH VkLA== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1722399115; x=1723003915; 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=Rg+ne5RbXGzTMhIX6WeWXeIYwsCxgZAOGqT3yzWRW58=; b=VSO/mmuY4cq18Yal69NhDTe3jfQ/DNWniTBSLvFAJW7aKVpiKr4Wh9VCrNDVCzS+yG HzDjnp2LZm75p4ychn6F1t/BlanqU7mdQwmVQAkcbMQXLJRCoRbrAETmxSIOe1fv5zGS L6sNpQkd7KuaAF4mSZOQ1ffm4V/1IO30RMS265H447V3aoZTKaFkjhnecO97+sRwKBoF n26+1odatvjY2lbgqzHASbddXqiMsHD66qXCw3WfEMlPTEDR3hvgOUcc3UCoCNHc5h5U LmNE1h0rsKNRBpNCnb83sJCoUr3L2ScwW+4K0VJJTC0WZFm7Ijqdyh7DOIferSf6ZGnC y/vw== X-Forwarded-Encrypted: i=1; AJvYcCUrLVYWYKLv3ZEqjTaiJ70wsuC+YxzU3B1QLSq2NA6uCCG7OlSuefgLBfzbR3rjwF/t7vDiCkAXf5B6rciHap5VYco= X-Gm-Message-State: AOJu0YyNTW/8zGxGNob2bDzI90oVXaLf0pRZMB+vRtKrcVPTICp4G7hp s0T362C3Fasf9NHzVE20adrrxDufTWK0St7zzIssbRCZgBDR6lmSeHJwoWg4Qwa0NPYTUt1Jc0f FwXNfYPWsd2ZppdDpuXDr/FIUr+4= X-Google-Smtp-Source: AGHT+IGx8hISC2H0WLvgvyVRjW0c+XgOzVPalC4KfRlnbsMIKUuZoEMS9f9fO16Dt4jspjHCeyZqMPYZxyE0t3aqnAk= X-Received: by 2002:a05:6122:469a:b0:4eb:499a:2453 with SMTP id 71dfb90a1353d-4f6e6904434mr9321710e0c.8.1722399115429; Tue, 30 Jul 2024 21:11:55 -0700 (PDT) MIME-Version: 1.0 References: <20240731000155.109583-1-21cnbao@gmail.com> <20240731000155.109583-2-21cnbao@gmail.com> In-Reply-To: From: Barry Song <21cnbao@gmail.com> Date: Wed, 31 Jul 2024 12:11:41 +0800 Message-ID: Subject: Re: [PATCH RFT v2 1/4] vpda: try to fix the potential crash due to misusing __GFP_NOFAIL To: Jason Wang Cc: akpm@linux-foundation.org, linux-mm@kvack.org, 42.hyeyoo@gmail.com, cl@linux.com, hailong.liu@oppo.com, hch@infradead.org, iamjoonsoo.kim@lge.com, lstoakes@gmail.com, mhocko@suse.com, penberg@kernel.org, rientjes@google.com, roman.gushchin@linux.dev, torvalds@linux-foundation.org, urezki@gmail.com, v-songbaohua@oppo.com, vbabka@suse.cz, virtualization@lists.linux.dev, "Michael S. Tsirkin" , Xuan Zhuo , =?UTF-8?Q?Eugenio_P=C3=A9rez?= , Maxime Coquelin Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable X-Stat-Signature: qutpxwm5stnke7sjgp719kuj141qnmdn X-Rspamd-Queue-Id: 89426140010 X-Rspam-User: X-Rspamd-Server: rspam10 X-HE-Tag: 1722399116-54301 X-HE-Meta: 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 bl8lyUXU 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 X-Bogosity: Ham, tests=bogofilter, spamicity=0.000000, version=1.2.4 Sender: owner-linux-mm@kvack.org Precedence: bulk X-Loop: owner-majordomo@kvack.org List-ID: List-Subscribe: List-Unsubscribe: On Wed, Jul 31, 2024 at 11:58=E2=80=AFAM Jason Wang w= rote: > > On Wed, Jul 31, 2024 at 11:15=E2=80=AFAM Barry Song <21cnbao@gmail.com> w= rote: > > > > On Wed, Jul 31, 2024 at 11:10=E2=80=AFAM Jason Wang wrote: > > > > > > On Wed, Jul 31, 2024 at 8:03=E2=80=AFAM Barry Song <21cnbao@gmail.com= > wrote: > > > > > > > > From: Barry Song > > > > > > > > mm doesn't support non-blockable __GFP_NOFAIL allocation. Because > > > > __GFP_NOFAIL without direct reclamation may just result in a busy > > > > loop within non-sleepable contexts. > > > > > > > > static inline struct page * > > > > __alloc_pages_slowpath(gfp_t gfp_mask, unsigned int order, > > > > struct alloc_contex= t *ac) > > > > { > > > > ... > > > > /* > > > > * Make sure that __GFP_NOFAIL request doesn't leak out and= make sure > > > > * we always retry > > > > */ > > > > if (gfp_mask & __GFP_NOFAIL) { > > > > /* > > > > * All existing users of the __GFP_NOFAIL are block= able, so warn > > > > * of any new users that actually require GFP_NOWAI= T > > > > */ > > > > if (WARN_ON_ONCE_GFP(!can_direct_reclaim, gfp_mask)= ) > > > > goto fail; > > > > ... > > > > } > > > > ... > > > > fail: > > > > warn_alloc(gfp_mask, ac->nodemask, > > > > "page allocation failure: order:%u", order)= ; > > > > got_pg: > > > > return page; > > > > } > > > > > > > > Let's move the memory allocation out of the atomic context and use > > > > the normal sleepable context to get pages. > > > > > > > > [RFT]: This has only been compile-tested; I'd prefer if the VDPA ma= intainers > > > > handles it. > > > > > > > > Cc: "Michael S. Tsirkin" > > > > Cc: Jason Wang > > > > Cc: Xuan Zhuo > > > > Cc: "Eugenio P=C3=A9rez" > > > > Cc: Maxime Coquelin > > > > Signed-off-by: Barry Song > > > > --- > > > > drivers/vdpa/vdpa_user/iova_domain.c | 31 +++++++++++++++++++++++-= ---- > > > > drivers/vdpa/vdpa_user/iova_domain.h | 5 ++++- > > > > drivers/vdpa/vdpa_user/vduse_dev.c | 4 +++- > > > > 3 files changed, 33 insertions(+), 7 deletions(-) > > > > > > > > diff --git a/drivers/vdpa/vdpa_user/iova_domain.c b/drivers/vdpa/vd= pa_user/iova_domain.c > > > > index 791d38d6284c..9318f059a8b5 100644 > > > > --- a/drivers/vdpa/vdpa_user/iova_domain.c > > > > +++ b/drivers/vdpa/vdpa_user/iova_domain.c > > > > @@ -283,7 +283,23 @@ int vduse_domain_add_user_bounce_pages(struct = vduse_iova_domain *domain, > > > > return ret; > > > > } > > > > > > > > -void vduse_domain_remove_user_bounce_pages(struct vduse_iova_domai= n *domain) > > > > +struct page **vduse_domain_alloc_pages_to_remove_bounce(struct vdu= se_iova_domain *domain) > > > > +{ > > > > + struct page **pages; > > > > + unsigned long count, i; > > > > + > > > > + if (!domain->user_bounce_pages) > > > > + return NULL; > > > > + > > > > + count =3D domain->bounce_size >> PAGE_SHIFT; > > > > + pages =3D kmalloc_array(count, sizeof(*pages), GFP_KERNEL |= __GFP_NOFAIL); > > > > + for (i =3D 0; i < count; i++) > > > > + pages[i] =3D alloc_page(GFP_KERNEL | __GFP_NOFAIL); > > > > + > > > > + return pages; > > > > +} > > > > + > > > > +void vduse_domain_remove_user_bounce_pages(struct vduse_iova_domai= n *domain, struct page **pages) > > > > { > > > > struct vduse_bounce_map *map; > > > > unsigned long i, count; > > > > @@ -294,15 +310,16 @@ void vduse_domain_remove_user_bounce_pages(st= ruct vduse_iova_domain *domain) > > > > > > > > count =3D domain->bounce_size >> PAGE_SHIFT; > > > > for (i =3D 0; i < count; i++) { > > > > - struct page *page =3D NULL; > > > > + struct page *page =3D pages[i]; > > > > > > > > map =3D &domain->bounce_maps[i]; > > > > - if (WARN_ON(!map->bounce_page)) > > > > + if (WARN_ON(!map->bounce_page)) { > > > > + put_page(page); > > > > continue; > > > > + } > > > > > > > > /* Copy user page to kernel page if it's in use */ > > > > if (map->orig_phys !=3D INVALID_PHYS_ADDR) { > > > > - page =3D alloc_page(GFP_ATOMIC | __GFP_NOFA= IL); > > > > memcpy_from_page(page_address(page), > > > > map->bounce_page, 0, PAGE_= SIZE); > > > > } > > > > @@ -310,6 +327,7 @@ void vduse_domain_remove_user_bounce_pages(stru= ct vduse_iova_domain *domain) > > > > map->bounce_page =3D page; > > > > } > > > > domain->user_bounce_pages =3D false; > > > > + kfree(pages); > > > > out: > > > > write_unlock(&domain->bounce_lock); > > > > } > > > > @@ -543,10 +561,13 @@ static int vduse_domain_mmap(struct file *fil= e, struct vm_area_struct *vma) > > > > static int vduse_domain_release(struct inode *inode, struct file *= file) > > > > { > > > > struct vduse_iova_domain *domain =3D file->private_data; > > > > + struct page **pages; > > > > + > > > > + pages =3D vduse_domain_alloc_pages_to_remove_bounce(domain)= ; > > > > > > > > spin_lock(&domain->iotlb_lock); > > > > vduse_iotlb_del_range(domain, 0, ULLONG_MAX); > > > > - vduse_domain_remove_user_bounce_pages(domain); > > > > + vduse_domain_remove_user_bounce_pages(domain, pages); > > > > vduse_domain_free_kernel_bounce_pages(domain); > > > > spin_unlock(&domain->iotlb_lock); > > > > put_iova_domain(&domain->stream_iovad); > > > > diff --git a/drivers/vdpa/vdpa_user/iova_domain.h b/drivers/vdpa/vd= pa_user/iova_domain.h > > > > index f92f22a7267d..17efa5555b3f 100644 > > > > --- a/drivers/vdpa/vdpa_user/iova_domain.h > > > > +++ b/drivers/vdpa/vdpa_user/iova_domain.h > > > > @@ -74,7 +74,10 @@ void vduse_domain_reset_bounce_map(struct vduse_= iova_domain *domain); > > > > int vduse_domain_add_user_bounce_pages(struct vduse_iova_domain *d= omain, > > > > struct page **pages, int cou= nt); > > > > > > > > -void vduse_domain_remove_user_bounce_pages(struct vduse_iova_domai= n *domain); > > > > +void vduse_domain_remove_user_bounce_pages(struct vduse_iova_domai= n *domain, > > > > + struct page **pages); > > > > + > > > > +struct page **vduse_domain_alloc_pages_to_remove_bounce(struct vdu= se_iova_domain *domain); > > > > > > > > void vduse_domain_destroy(struct vduse_iova_domain *domain); > > > > > > > > diff --git a/drivers/vdpa/vdpa_user/vduse_dev.c b/drivers/vdpa/vdpa= _user/vduse_dev.c > > > > index 7ae99691efdf..5d8d5810df57 100644 > > > > --- a/drivers/vdpa/vdpa_user/vduse_dev.c > > > > +++ b/drivers/vdpa/vdpa_user/vduse_dev.c > > > > @@ -1030,6 +1030,7 @@ static int vduse_dev_queue_irq_work(struct vd= use_dev *dev, > > > > static int vduse_dev_dereg_umem(struct vduse_dev *dev, > > > > u64 iova, u64 size) > > > > { > > > > + struct page **pages; > > > > int ret; > > > > > > > > mutex_lock(&dev->mem_lock); > > > > @@ -1044,7 +1045,8 @@ static int vduse_dev_dereg_umem(struct vduse_= dev *dev, > > > > if (dev->umem->iova !=3D iova || size !=3D dev->domain->bou= nce_size) > > > > goto unlock; > > > > > > > > - vduse_domain_remove_user_bounce_pages(dev->domain); > > > > + pages =3D vduse_domain_alloc_pages_to_remove_bounce(dev->do= main); > > > > + vduse_domain_remove_user_bounce_pages(dev->domain, pages); > > > > unpin_user_pages_dirty_lock(dev->umem->pages, > > > > dev->umem->npages, true); > > > > atomic64_sub(dev->umem->npages, &dev->umem->mm->pinned_vm); > > > > > > We miss a kfree(pages); here? > > no. > > i've moved it into vduse_domain_remove_user_bounce_pages. > > Ok, but it seems tricky e.g allocated by the caller but freed in > callee. And I think I missed some important issues in the previous > review: The check of user_bounce_pages must be done under the > bounce_lock, otherwise it might race with umem_reg. > > So in the case of release(), we know the device is gone, so there's no > need to allocate pages that will be released soon. So we can pass NULL > as a hint and just assign bounce_page to NULL in > vduse_domain_remove_user_bounce_pages(). > > And in the case of vduse_dev_dereg_umem(), we need to allocate the > pages without checking user_bounce_pages. So in > vduse_domain_remove_user_bounce_pages() if we can free the allocated > pages as well as the pages in the following check > > if (!domain->user_bounce_pages) > goto out; > > What do you think? I am not a vdpa guy, but changing the current logic is another patch. >From mm perspective, I can only address the __GFP_NOFAIL issue. I actually prefer you guys handle it directly:-) I'd rather report a BUG instead. TBH, I know nothing about vpda. > > Thanks > > > > > > > > > Thanks > > > > > > > -- > > > > 2.34.1 > > > > > > > Thanks Barry