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 bombadil.infradead.org (bombadil.infradead.org [198.137.202.133]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.lore.kernel.org (Postfix) with ESMTPS id A4463C433F5 for ; Thu, 2 Dec 2021 18:20:24 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; q=dns/txt; c=relaxed/relaxed; d=lists.infradead.org; s=bombadil.20210309; h=Sender: Content-Transfer-Encoding:Content-Type:List-Subscribe:List-Help:List-Post: List-Archive:List-Unsubscribe:List-Id:In-Reply-To:MIME-Version:References: Message-ID:Subject:Cc:To:From:Date:Reply-To:Content-ID:Content-Description: Resent-Date:Resent-From:Resent-Sender:Resent-To:Resent-Cc:Resent-Message-ID: List-Owner; bh=kKTZWDaUCh2gIRMIPVAXk3Lntr3KcWsJQc99jIB33K0=; b=tUDxhxfbAUCF5T rMX5MGhUXAQ9G5xnhMSOllHdt9stkBCMDPMKRvW3Kk68jwT9npx9Eo2vJpYHo+r7ztY0zwSGe7hQt SY4JM9fnUPuVMPrdpchLvAHXq7venUi4lNwD4kQ3dvu4+tasb5gxscEQpAIGKN8BQYXHFUeVnMvpT MDfzImhV52nvTOWEdWp/oa9RVlNRp9LvscxQTBOszzn2weFD1vFh3Ktut/JOZf1BPmphDTtqD70hX 8u1tZeWU1JlKNIPNwq7xo+/yEi2l6NuOXridaVtXLJPdB1EnhDi7+iOPrqHaXcxwfEMVUt+d0XuM9 aCTPeb4LLcetc0qSJBeA==; Received: from localhost ([::1] helo=bombadil.infradead.org) by bombadil.infradead.org with esmtp (Exim 4.94.2 #2 (Red Hat Linux)) id 1msqgr-00DGns-FO; Thu, 02 Dec 2021 18:20:17 +0000 Received: from foss.arm.com ([217.140.110.172]) by bombadil.infradead.org with esmtp (Exim 4.94.2 #2 (Red Hat Linux)) id 1msqgd-00DGmN-S7; Thu, 02 Dec 2021 18:20:05 +0000 Received: from usa-sjc-imap-foss1.foss.arm.com (unknown [10.121.207.14]) by usa-sjc-mx-foss1.foss.arm.com (Postfix) with ESMTP id A0AF31435; Thu, 2 Dec 2021 10:20:01 -0800 (PST) Received: from arm.com (arrakis.cambridge.arm.com [10.1.196.175]) by usa-sjc-imap-foss1.foss.arm.com (Postfix) with ESMTPSA id 353A23F766; Thu, 2 Dec 2021 10:20:00 -0800 (PST) Date: Thu, 2 Dec 2021 18:19:58 +0000 From: Catalin Marinas To: Kuan-Ying Lee Cc: Andrew Morton , Matthias Brugger , chinwen.chang@mediatek.com, nicholas.tang@mediatek.com, james.hsu@mediatek.com, yee.lee@mediatek.com, linux-mm@kvack.org, linux-kernel@vger.kernel.org, linux-arm-kernel@lists.infradead.org, linux-mediatek@lists.infradead.org Subject: Re: [PATCH] kmemleak: fix kmemleak false positive report with HW tag-based kasan enable Message-ID: References: <20211118054426.4123-1-Kuan-Ying.Lee@mediatek.com> MIME-Version: 1.0 Content-Disposition: inline In-Reply-To: <20211118054426.4123-1-Kuan-Ying.Lee@mediatek.com> X-CRM114-Version: 20100106-BlameMichelson ( TRE 0.8.0 (BSD) ) MR-646709E3 X-CRM114-CacheID: sfid-20211202_102004_053011_856DE48E X-CRM114-Status: GOOD ( 15.07 ) X-BeenThere: linux-mediatek@lists.infradead.org X-Mailman-Version: 2.1.34 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: "Linux-mediatek" Errors-To: linux-mediatek-bounces+linux-mediatek=archiver.kernel.org@lists.infradead.org On Thu, Nov 18, 2021 at 01:44:19PM +0800, Kuan-Ying Lee wrote: > With HW tag-based kasan enable, We will get the warning > when we free object whose address starts with 0xFF. > > It is because kmemleak rbtree stores tagged object and > this freeing object's tag does not match with rbtree object. > > In the example below, kmemleak rbtree stores the tagged object in > the kmalloc(), and kfree() gets the pointer with 0xFF tag. > > Call sequence: > ptr = kmalloc(size, GFP_KERNEL); > page = virt_to_page(ptr); > kfree(page_address(page)); > ptr = kmalloc(size, GFP_KERNEL); > > Call sequence like that may cause the warning as following: > 1) Freeing unknown object: > In kfree(), we will get free unknown object warning in kmemleak_free(). > Because object(0xFx) in kmemleak rbtree and pointer(0xFF) in kfree() have > different tag. > > 2) Overlap existing: > When we allocate that object with the same hw-tag again, we will > find the overlap in the kmemleak rbtree and kmemleak thread will > be killed. > > [ 116.685312] kmemleak: Freeing unknown object at 0xffff000003f88000 > [ 116.686422] CPU: 5 PID: 177 Comm: cat Not tainted 5.16.0-rc1-dirty #21 > [ 116.687067] Hardware name: linux,dummy-virt (DT) > [ 116.687496] Call trace: > [ 116.687792] dump_backtrace+0x0/0x1ac > [ 116.688255] show_stack+0x1c/0x30 > [ 116.688663] dump_stack_lvl+0x68/0x84 > [ 116.689096] dump_stack+0x1c/0x38 > [ 116.689499] kmemleak_free+0x6c/0x70 > [ 116.689919] slab_free_freelist_hook+0x104/0x200 > [ 116.690420] kmem_cache_free+0xa8/0x3d4 > [ 116.690845] test_version_show+0x270/0x3a0 > [ 116.691344] module_attr_show+0x28/0x40 > [ 116.691789] sysfs_kf_seq_show+0xb0/0x130 > [ 116.692245] kernfs_seq_show+0x30/0x40 > [ 116.692678] seq_read_iter+0x1bc/0x4b0 > [ 116.692678] seq_read_iter+0x1bc/0x4b0 > [ 116.693114] kernfs_fop_read_iter+0x144/0x1c0 > [ 116.693586] generic_file_splice_read+0xd0/0x184 > [ 116.694078] do_splice_to+0x90/0xe0 > [ 116.694498] splice_direct_to_actor+0xb8/0x250 > [ 116.694975] do_splice_direct+0x88/0xd4 > [ 116.695409] do_sendfile+0x2b0/0x344 > [ 116.695829] __arm64_sys_sendfile64+0x164/0x16c > [ 116.696306] invoke_syscall+0x48/0x114 > [ 116.696735] el0_svc_common.constprop.0+0x44/0xec > [ 116.697263] do_el0_svc+0x74/0x90 > [ 116.697665] el0_svc+0x20/0x80 > [ 116.698261] el0t_64_sync_handler+0x1a8/0x1b0 > [ 116.698695] el0t_64_sync+0x1ac/0x1b0 > ... > [ 117.520301] kmemleak: Cannot insert 0xf2ff000003f88000 into the object search tree (overlaps existing) > [ 117.521118] CPU: 5 PID: 178 Comm: cat Not tainted 5.16.0-rc1-dirty #21 > [ 117.521827] Hardware name: linux,dummy-virt (DT) > [ 117.522287] Call trace: > [ 117.522586] dump_backtrace+0x0/0x1ac > [ 117.523053] show_stack+0x1c/0x30 > [ 117.523578] dump_stack_lvl+0x68/0x84 > [ 117.524039] dump_stack+0x1c/0x38 > [ 117.524472] create_object.isra.0+0x2d8/0x2fc > [ 117.524975] kmemleak_alloc+0x34/0x40 > [ 117.525416] kmem_cache_alloc+0x23c/0x2f0 > [ 117.525914] test_version_show+0x1fc/0x3a0 > [ 117.526379] module_attr_show+0x28/0x40 > [ 117.526827] sysfs_kf_seq_show+0xb0/0x130 > [ 117.527363] kernfs_seq_show+0x30/0x40 > [ 117.527848] seq_read_iter+0x1bc/0x4b0 > [ 117.528320] kernfs_fop_read_iter+0x144/0x1c0 > [ 117.528809] generic_file_splice_read+0xd0/0x184 > [ 117.529316] do_splice_to+0x90/0xe0 > [ 117.529734] splice_direct_to_actor+0xb8/0x250 > [ 117.530227] do_splice_direct+0x88/0xd4 > [ 117.530686] do_sendfile+0x2b0/0x344 > [ 117.531154] __arm64_sys_sendfile64+0x164/0x16c > [ 117.531673] invoke_syscall+0x48/0x114 > [ 117.532111] el0_svc_common.constprop.0+0x44/0xec > [ 117.532621] do_el0_svc+0x74/0x90 > [ 117.533048] el0_svc+0x20/0x80 > [ 117.533461] el0t_64_sync_handler+0x1a8/0x1b0 > [ 117.533950] el0t_64_sync+0x1ac/0x1b0 > [ 117.534625] kmemleak: Kernel memory leak detector disabled > [ 117.535201] kmemleak: Object 0xf2ff000003f88000 (size 128): > [ 117.535761] kmemleak: comm "cat", pid 177, jiffies 4294921177 > [ 117.536339] kmemleak: min_count = 1 > [ 117.536718] kmemleak: count = 0 > [ 117.537068] kmemleak: flags = 0x1 > [ 117.537429] kmemleak: checksum = 0 > [ 117.537806] kmemleak: backtrace: > [ 117.538211] kmem_cache_alloc+0x23c/0x2f0 > [ 117.538924] test_version_show+0x1fc/0x3a0 > [ 117.539393] module_attr_show+0x28/0x40 > [ 117.539844] sysfs_kf_seq_show+0xb0/0x130 > [ 117.540304] kernfs_seq_show+0x30/0x40 > [ 117.540750] seq_read_iter+0x1bc/0x4b0 > [ 117.541206] kernfs_fop_read_iter+0x144/0x1c0 > [ 117.541687] generic_file_splice_read+0xd0/0x184 > [ 117.542182] do_splice_to+0x90/0xe0 > [ 117.542611] splice_direct_to_actor+0xb8/0x250 > [ 117.543097] do_splice_direct+0x88/0xd4 > [ 117.543544] do_sendfile+0x2b0/0x344 > [ 117.543983] __arm64_sys_sendfile64+0x164/0x16c > [ 117.544471] invoke_syscall+0x48/0x114 > [ 117.544917] el0_svc_common.constprop.0+0x44/0xec > [ 117.545416] do_el0_svc+0x74/0x90 > [ 117.554100] kmemleak: Automatic memory scanning thread ended > > Signed-off-by: Kuan-Ying Lee I was wondering whether we should just give up the tag when storing the object->pointer and avoid any later untagging when searching the rb tree. But, if we want to keep that for debugging, fine by me as well. Reviewed-by: Catalin Marinas -- Catalin _______________________________________________ Linux-mediatek mailing list Linux-mediatek@lists.infradead.org http://lists.infradead.org/mailman/listinfo/linux-mediatek