From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from pegase1.c-s.fr ([93.17.236.30]:2841 "EHLO pegase1.c-s.fr" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1725907AbgCGHFO (ORCPT ); Sat, 7 Mar 2020 02:05:14 -0500 Subject: Re: [PATCH V15] mm/debug: Add tests validating architecture page table helpers References: <61250cdc-f80b-2e50-5168-2ec67ec6f1e6@arm.com> From: Christophe Leroy Message-ID: Date: Sat, 7 Mar 2020 08:05:09 +0100 MIME-Version: 1.0 In-Reply-To: Content-Type: text/plain; charset="utf-8"; format="flowed" Content-Language: fr Content-Transfer-Encoding: 8bit Sender: linux-s390-owner@vger.kernel.org List-ID: To: Anshuman Khandual , Qian Cai Cc: linux-mm@kvack.org, Andrew Morton , Mike Rapoport , Vineet Gupta , Catalin Marinas , Will Deacon , Benjamin Herrenschmidt , Paul Mackerras , Michael Ellerman , Heiko Carstens , Vasily Gorbik , Christian Borntraeger , Thomas Gleixner , Ingo Molnar , Borislav Petkov , "H. Peter Anvin" , "Kirill A . Shutemov" , Paul Walmsley , Palmer Dabbelt , linux-snps-arc@lists.infradead.org, linux-arm-kernel@lists.infradead.org, linuxppc-dev@lists.ozlabs.org, linux-s390@vger.kernel.org, linux-riscv@lists.infradead.org, x86@kernel.org, linux-arch@vger.kernel.org, linux-kernel@vger.kernel.org Le 07/03/2020 à 01:56, Anshuman Khandual a écrit : > > > On 03/07/2020 06:04 AM, Qian Cai wrote: >> >> >>> On Mar 6, 2020, at 7:03 PM, Anshuman Khandual wrote: >>> >>> Hmm, set_pte_at() function is not preferred here for these tests. The idea >>> is to avoid or atleast minimize TLB/cache flushes triggered from these sort >>> of 'static' tests. set_pte_at() is platform provided and could/might trigger >>> these flushes or some other platform specific synchronization stuff. Just >> >> Why is that important for this debugging option? > > Primarily reason is to avoid TLB/cache flush instructions on the system > during these tests that only involve transforming different page table > level entries through helpers. Unless really necessary, why should it > emit any TLB/cache flush instructions ? What's the problem with thoses flushes ? > >> >>> wondering is there specific reason with respect to the soft lock up problem >>> making it necessary to use set_pte_at() rather than a simple WRITE_ONCE() ? >> >> Looks at the s390 version of set_pte_at(), it has this comment, >> vmaddr); >> >> /* >> * Certain architectures need to do special things when PTEs >> * within a page table are directly modified. Thus, the following >> * hook is made available. >> */ >> >> I can only guess that powerpc could be the same here. > > This comment is present in multiple platforms while defining set_pte_at(). > Is not 'barrier()' here alone good enough ? Else what exactly set_pte_at() > does as compared to WRITE_ONCE() that avoids the soft lock up, just trying > to understand. > Argh ! I didn't realise that you were writing directly into the page tables. When it works, that's only by chance I guess. To properly set the page table entries, set_pte_at() has to be used: - On powerpc 8xx, with 16k pages, the page table entry must be copied four times. set_pte_at() does it, WRITE_ONCE() doesn't. - On powerpc book3s/32 (hash MMU), the flag _PAGE_HASHPTE must be preserved among writes. set_pte_at() preserves it, WRITE_ONCE() doesn't. set_pte_at() also does a few other mandatory things, like calling pte_mkpte() So, the WRITE_ONCE() must definitely become a set_pte_at() Christophe