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]) (using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits)) (No client certificate requested) by smtp.lore.kernel.org (Postfix) with ESMTPS id A6B40D111A8 for ; Fri, 28 Nov 2025 01:19:08 +0000 (UTC) Received: by kanga.kvack.org (Postfix) id 134A86B000D; Thu, 27 Nov 2025 20:19:08 -0500 (EST) Received: by kanga.kvack.org (Postfix, from userid 40) id 10CB36B000E; Thu, 27 Nov 2025 20:19:08 -0500 (EST) X-Delivered-To: int-list-linux-mm@kvack.org Received: by kanga.kvack.org (Postfix, from userid 63042) id 048876B0023; Thu, 27 Nov 2025 20:19:07 -0500 (EST) X-Delivered-To: linux-mm@kvack.org Received: from relay.hostedemail.com (smtprelay0010.hostedemail.com [216.40.44.10]) by kanga.kvack.org (Postfix) with ESMTP id E71E96B000D for ; Thu, 27 Nov 2025 20:19:07 -0500 (EST) Received: from smtpin19.hostedemail.com (a10.router.float.18 [10.200.18.1]) by unirelay05.hostedemail.com (Postfix) with ESMTP id 798E258677 for ; Fri, 28 Nov 2025 01:19:07 +0000 (UTC) X-FDA: 84158257134.19.FA4BB9B Received: from dggsgout12.his.huawei.com (dggsgout12.his.huawei.com [45.249.212.56]) by imf03.hostedemail.com (Postfix) with ESMTP id 8DC6220007 for ; Fri, 28 Nov 2025 01:19:04 +0000 (UTC) Authentication-Results: imf03.hostedemail.com; spf=pass (imf03.hostedemail.com: domain of wozizhi@huaweicloud.com designates 45.249.212.56 as permitted sender) smtp.mailfrom=wozizhi@huaweicloud.com ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=hostedemail.com; s=arc-20220608; t=1764292745; 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; bh=sHerJEJRO41BGzc3CYRRZvFGZiEOgzzA0n01wZ6r51Y=; b=KlizcAPc5U1n4k70+/qfmTslAto/IiDOToFG9niooD4FRvs5sE6sEqIIQtWiYWBGTlkJHu K192DRkbcHzgOwxCW4Cs7yTr2sWzZ6K2zTUth7pGzUW+LqMuEaADsgcOg/HOaP26i1o8if walqR7Lcbky0hn1iTySx2tuad8HX7Rs= ARC-Authentication-Results: i=1; imf03.hostedemail.com; dkim=none; dmarc=none; spf=pass (imf03.hostedemail.com: domain of wozizhi@huaweicloud.com designates 45.249.212.56 as permitted sender) smtp.mailfrom=wozizhi@huaweicloud.com ARC-Seal: i=1; s=arc-20220608; d=hostedemail.com; t=1764292745; a=rsa-sha256; cv=none; b=vn2zG5mjnXLuLLIz9z4qdZUzlE0wjiDEiSABy1WVcftvmXb/RxUOxndO6tuCdQgY/aakds +8i15Y9XMFbtsE2dRloPXFR/5YRDTvxgCTcF7qSTanpPCbDJ9PfyGwaRwYYdziAk+A4LkK v9Owb2zPZuBp5UUJa9Y3MheF8Pm8UP0= Received: from mail.maildlp.com (unknown [172.19.93.142]) by dggsgout12.his.huawei.com (SkyGuard) with ESMTPS id 4dHb5k5LfbzKHMPm for ; Fri, 28 Nov 2025 09:18:18 +0800 (CST) Received: from mail02.huawei.com (unknown [10.116.40.75]) by mail.maildlp.com (Postfix) with ESMTP id CB66F1A018D for ; Fri, 28 Nov 2025 09:18:59 +0800 (CST) Received: from [10.174.176.88] (unknown [10.174.176.88]) by APP2 (Coremail) with SMTP id Syh0CgCHMXiC+ChptwYJCQ--.54954S3; Fri, 28 Nov 2025 09:18:59 +0800 (CST) Message-ID: <39d99c56-3c2f-46bd-933f-2aef69d169f3@huaweicloud.com> Date: Fri, 28 Nov 2025 09:18:58 +0800 MIME-Version: 1.0 User-Agent: Mozilla Thunderbird Subject: Re: [Bug report] hash_name() may cross page boundary and trigger sleep in RCU context To: Will Deacon Cc: jack@suse.com, brauner@kernel.org, hch@lst.de, akpm@linux-foundation.org, linux@armlinux.org.uk, linux-fsdevel@vger.kernel.org, linux-kernel@vger.kernel.org, linux-mm@kvack.org, linux-arm-kernel@lists.infradead.org, yangerkun@huawei.com, wangkefeng.wang@huawei.com, pangliyuan1@huawei.com, xieyuanbin1@huawei.com References: <20251126090505.3057219-1-wozizhi@huaweicloud.com> <9ff0d134-2c64-4204-bbac-9fdf0867ac46@huaweicloud.com> From: Zizhi Wo In-Reply-To: <9ff0d134-2c64-4204-bbac-9fdf0867ac46@huaweicloud.com> Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: 8bit X-CM-TRANSID:Syh0CgCHMXiC+ChptwYJCQ--.54954S3 X-Coremail-Antispam: 1UD129KBjvJXoW3JFWkXryxArykAFWDKF1xXwb_yoW3Jr18pr 18Ca4UJFW5Wr1rA3yjqw1DJFy8J3WUJw4UWr1UtF1UZr47Xr1jqr40q3yF934UXr48Xw4U Xr15Jr17Zr1UJFUanT9S1TB71UUUUU7qnTZGkaVYY2UrUUUUjbIjqfuFe4nvWSU5nxnvy2 9KBjDU0xBIdaVrnRJUUUv0b4IE77IF4wAFF20E14v26r4j6ryUM7CY07I20VC2zVCF04k2 6cxKx2IYs7xG6rWj6s0DM7CIcVAFz4kK6r1j6r18M28lY4IEw2IIxxk0rwA2F7IY1VAKz4 vEj48ve4kI8wA2z4x0Y4vE2Ix0cI8IcVAFwI0_tr0E3s1l84ACjcxK6xIIjxv20xvEc7Cj xVAFwI0_Gr1j6F4UJwA2z4x0Y4vEx4A2jsIE14v26rxl6s0DM28EF7xvwVC2z280aVCY1x 0267AKxVW0oVCq3wAS0I0E0xvYzxvE52x082IY62kv0487Mc02F40EFcxC0VAKzVAqx4xG 6I80ewAv7VC0I7IYx2IY67AKxVWUGVWUXwAv7VC2z280aVAFwI0_Jr0_Gr1lOx8S6xCaFV Cjc4AY6r1j6r4UM4x0Y48IcVAKI48JM4IIrI8v6xkF7I0E8cxan2IY04v7MxkF7I0En4kS 14v26r1q6r43MxAIw28IcxkI7VAKI48JMxC20s026xCaFVCjc4AY6r1j6r4UMI8I3I0E5I 8CrVAFwI0_Jr0_Jr4lx2IqxVCjr7xvwVAFwI0_JrI_JrWlx4CE17CEb7AF67AKxVWUtVW8 ZwCIc40Y0x0EwIxGrwCI42IY6xIIjxv20xvE14v26r1j6r1xMIIF0xvE2Ix0cI8IcVCY1x 0267AKxVW8JVWxJwCI42IY6xAIw20EY4v20xvaj40_Jr0_JF4lIxAIcVC2z280aVAFwI0_ Jr0_Gr1lIxAIcVC2z280aVCY1x0267AKxVW8JVW8JrUvcSsGvfC2KfnxnUUI43ZEXa7IUb mii3UUUUU== X-CM-SenderInfo: pzr2x6tkl6x35dzhxuhorxvhhfrp/ X-Rspamd-Queue-Id: 8DC6220007 X-Rspam-User: X-Rspamd-Server: rspam05 X-Stat-Signature: 1g8i5f4r98g4ia4x8ez5h5nc3z1ey4ka X-HE-Tag: 1764292744-852333 X-HE-Meta: U2FsdGVkX1/rr8VPVN3OXmwKYUe0qtPmReu52C0AFszO+TIkOQV2a5wDEWkLlaxaOPOGX/tNLYuU0mxM6nvatrlFZoVPxTAdRMTORfrqHuFBUV7pXumOMKx+dFBVDEw1zP3i3M39R58UOuHiFATUWzgdK2bBxGvEfa1VC4beDO9dYtcKh7HTKh1IhXO+DHN0geGK5a/qTwcu8nHGyeOQ3dbn3N8Txg8RamcC5SWnEXWUt+NZOLT3N9+uDL2fBT5+NTz5W/7xDSZPCzfO7/QWdn6vf77qk/VO1btmK0y4dQd6h7MdaJ+DNxmP9mXkWTY61h5D/AWl0uRhe03OtuctNL14mrzAyD8OR6Dau3TAkKvvOFNqGg08shtC3rvruH+5vKY/5kTWHwKO/SvJfWtOVIlvhEDETwfY8GTrELrvlsmgPB8YOlzFrHWw3UYtnp2NdXKOtq5NwqMkRBMu47aslfKFIDEuSLkjNfUpB2raUZm/nM9lpkjlViIx+n9XoX/eCYXi282fGdPVjzEAOy+30p6juYoRjTzu8wPocCB9grURhj4SQknWdIXG1awUYfpisGqw9dwvokWOj8vOIfBPudt0NRwl1sXjT418OGXlfheWYMotPu4wKifM7+fgT0tAYPHI3E8K4mVTRZJcAd9X82NBzL5ZCn0VeaYAnqnsnXuJbm6HmwReqSTqnYcujDcH2yWMRfvAnrotBGxqx9PfbquWsKbI0T6YyF+A+fiq0unWtwN6F+GNHkbhZfcYcDUrYRULgzQSGksc4qNgam6Iibs2kCKQxZxk5FshjqsydJxmXO1xzUQCpz1PpTebmi5CWRYYFGG+4qLF5Ie4j4rNWph/Gki7VBYC6xsQ+o2q4amXYYxGqoqLOA== 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: 在 2025/11/28 9:17, Zizhi Wo 写道: > > > 在 2025/11/27 20:59, Will Deacon 写道: >> On Wed, Nov 26, 2025 at 05:05:05PM +0800, Zizhi Wo wrote: >>> We're running into the following issue on an ARM32 platform with the >>> linux >>> 5.10 kernel: >>> >>> [] (__dabt_svc) from [] >>> (link_path_walk.part.7+0x108/0x45c) >>> [] (link_path_walk.part.7) from [] >>> (path_openat+0xc4/0x10ec) >>> [] (path_openat) from [] (do_filp_open+0x9c/0x114) >>> [] (do_filp_open) from [] >>> (do_sys_openat2+0x418/0x528) >>> [] (do_sys_openat2) from [] (do_sys_open+0x88/0xe4) >>> [] (do_sys_open) from [] (ret_fast_syscall+0x0/0x58) >>> ... >>> [] (unwind_backtrace) from [] (show_stack+0x20/0x24) >>> [] (show_stack) from [] (dump_stack+0xd8/0xf8) >>> [] (dump_stack) from [] (___might_sleep+0x19c/0x1e4) >>> [] (___might_sleep) from [] >>> (do_page_fault+0x2f8/0x51c) >>> [] (do_page_fault) from [] (do_DataAbort+0x90/0x118) >>> [] (do_DataAbort) from [] (__dabt_svc+0x58/0x80) >>> ... >>> >>> During the execution of hash_name()->load_unaligned_zeropad(), a >>> potential >>> memory access beyond the PAGE boundary may occur. For example, when the >>> filename length is near the PAGE_SIZE boundary. This triggers a page >>> fault, >>> which leads to a call to do_page_fault()->mmap_read_trylock(). If we >>> can't >>> acquire the lock, we have to fall back to the mmap_read_lock() path, >>> which >>> calls might_sleep(). This breaks RCU semantics because path lookup >>> occurs >>> under an RCU read-side critical section. In linux-mainline, arm/arm64 >>> do_page_fault() still has this problem: >>> >>> lock_mm_and_find_vma->get_mmap_lock_carefully->mmap_read_lock_killable. >>> >>> And before commit bfcfaa77bdf0 ("vfs: use 'unsigned long' accesses for >>> dcache name comparison and hashing"), hash_name accessed the name >>> byte by >>> byte. >>> >>> To prevent load_unaligned_zeropad() from accessing beyond the valid >>> memory >>> region, we would need to intercept such cases beforehand? But doing so >>> would require replicating the internal logic of >>> load_unaligned_zeropad(), >>> including handling endianness and constructing the correct value >>> manually. >>> Given that load_unaligned_zeropad() is used in many places across the >>> kernel, we currently haven't found a good solution to address this >>> cleanly. >>> >>> What would be the recommended way to handle this situation? Would >>> appreciate any feedback and guidance from the community. Thanks! >> >> Does it help if you bodge the translation fault handler along the lines >> of the untested diff below? > > Thank you for the solution you provided. However, I seem to have > encountered a bit of a problem. > >> >> Will >> >> --->8 >> >> diff --git a/arch/arm/mm/fault.c b/arch/arm/mm/fault.c >> index bf1577216ffa..b3c81e448798 100644 >> --- a/arch/arm/mm/fault.c >> +++ b/arch/arm/mm/fault.c >> @@ -407,7 +407,7 @@ do_translation_fault(unsigned long addr, unsigned >> int fsr, >>          if (addr < TASK_SIZE) >>                  return do_page_fault(addr, fsr, regs); >> -       if (user_mode(regs)) >> +       if (user_mode(regs) || fsr_fs(fsr) == FSR_FS_INVALID_PAGE) >>                  goto bad_area; > > > > I'm getting an "FSR_FS_INVALID_PAGE undeclared" error during > compilation... > > In which kernel or FSR version was this macro or constant defined Sorry, I didn't see this "#define FSR_FS_INVALID_PAGE". I'll try again right away. Please ignore my previous reply. > >>          index = pgd_index(addr); >> diff --git a/arch/arm/mm/fault.h b/arch/arm/mm/fault.h >> index 9ecc2097a87a..8fb26f85e361 100644 >> --- a/arch/arm/mm/fault.h >> +++ b/arch/arm/mm/fault.h >> @@ -12,6 +12,8 @@ >>   #define FSR_FS3_0              (15) >>   #define FSR_FS5_0              (0x3f) >> +#define FSR_FS_INVALID_PAGE    7 >> + >>   #ifdef CONFIG_ARM_LPAE >>   #define FSR_FS_AEA             17 >> diff --git a/arch/arm/mm/fsr-2level.c b/arch/arm/mm/fsr-2level.c >> index f2be95197265..c7060da345df 100644 >> --- a/arch/arm/mm/fsr-2level.c >> +++ b/arch/arm/mm/fsr-2level.c >> @@ -11,7 +11,7 @@ static struct fsr_info fsr_info[] = { >>          { do_bad,               SIGBUS,  0,             "external >> abort on linefetch"      }, >>          { do_translation_fault, SIGSEGV, SEGV_MAPERR,   "section >> translation fault"        }, >>          { do_bad,               SIGBUS,  0,             "external >> abort on linefetch"      }, >> -       { do_page_fault,        SIGSEGV, SEGV_MAPERR,   "page >> translation fault"           }, >> +       { do_translation_fault, SIGSEGV, SEGV_MAPERR,   "page >> translation fault"           }, >>          { do_bad,               SIGBUS,  0,             "external >> abort on non-linefetch"  }, >>          { do_bad,               SIGSEGV, SEGV_ACCERR,   "section >> domain fault"             }, >>          { do_bad,               SIGBUS,  0,             "external >> abort on non-linefetch"  }, >> diff --git a/arch/arm/mm/fsr-3level.c b/arch/arm/mm/fsr-3level.c >> index d0ae2963656a..19df4af828bd 100644 >> --- a/arch/arm/mm/fsr-3level.c >> +++ b/arch/arm/mm/fsr-3level.c >> @@ -7,7 +7,7 @@ static struct fsr_info fsr_info[] = { >>          { do_bad,               SIGBUS,  0,             "reserved >> translation fault"    }, >>          { do_translation_fault, SIGSEGV, SEGV_MAPERR,   "level 1 >> translation fault"     }, >>          { do_translation_fault, SIGSEGV, SEGV_MAPERR,   "level 2 >> translation fault"     }, >> -       { do_page_fault,        SIGSEGV, SEGV_MAPERR,   "level 3 >> translation fault"     }, >> +       { do_translation_fault, SIGSEGV, SEGV_MAPERR,   "level 3 >> translation fault"     }, >>          { do_bad,               SIGBUS,  0,             "reserved >> access flag fault"    }, >>          { do_bad,               SIGSEGV, SEGV_ACCERR,   "level 1 >> access flag fault"     }, >>          { do_page_fault,        SIGSEGV, SEGV_ACCERR,   "level 2 >> access flag fault"     }, >> >> > > By the way, I tried Al's solution, and this problem didn't reproduce. > > Thanks, > Zizhi Wo