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 vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id 2BA4DC64ED9 for ; Mon, 20 Feb 2023 08:29:21 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S231128AbjBTI3T (ORCPT ); Mon, 20 Feb 2023 03:29:19 -0500 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:42566 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S231129AbjBTI3Q (ORCPT ); Mon, 20 Feb 2023 03:29:16 -0500 Received: from mx1.emlix.com (mx1.emlix.com [136.243.223.33]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 7DAEC12F1E; Mon, 20 Feb 2023 00:29:11 -0800 (PST) Received: from mailer.emlix.com (unknown [81.20.119.6]) (using TLSv1.2 with cipher ADH-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by mx1.emlix.com (Postfix) with ESMTPS id 167505F997; Mon, 20 Feb 2023 09:29:09 +0100 (CET) From: Rolf Eike Beer To: Matthew Wilcox , linux-arch@vger.kernel.org, Alexandre Ghiti Cc: Yin Fengwei , linux-mm@kvack.org, linux-alpha@vger.kernel.org, linux-csky@vger.kernel.org, linux-m68k@lists.linux-m68k.org, linux-mips@vger.kernel.org, Dinh Nguyen , linux-parisc@vger.kernel.org, linux-sh@vger.kernel.org, linux-arm-kernel@lists.infradead.org, loongarch@lists.linux.dev, openrisc@lists.librecores.org, linuxppc-dev@lists.ozlabs.org, linux-riscv@lists.infradead.org, sparclinux@vger.kernel.org, linux-xtensa@linux-xtensa.org Subject: Re: API for setting multiple PTEs at once Date: Mon, 20 Feb 2023 09:29:58 +0100 Message-ID: <5649318.DvuYhMxLoT@devpool47> Organization: emlix GmbH In-Reply-To: <0bf59207-838b-2a0b-a95e-925a6bbf1913@ghiti.fr> References: <0bf59207-838b-2a0b-a95e-925a6bbf1913@ghiti.fr> MIME-Version: 1.0 Content-Type: multipart/signed; boundary="nextPart5907765.lOV4Wx5bFT"; micalg="pgp-sha256"; protocol="application/pgp-signature" Precedence: bulk List-ID: X-Mailing-List: linux-mips@vger.kernel.org --nextPart5907765.lOV4Wx5bFT Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset="UTF-8"; protected-headers="v1" From: Rolf Eike Beer Subject: Re: API for setting multiple PTEs at once Date: Mon, 20 Feb 2023 09:29:58 +0100 Message-ID: <5649318.DvuYhMxLoT@devpool47> Organization: emlix GmbH In-Reply-To: <0bf59207-838b-2a0b-a95e-925a6bbf1913@ghiti.fr> MIME-Version: 1.0 On Dienstag, 14. Februar 2023 10:55:43 CET Alexandre Ghiti wrote: > Hi Matthew, >=20 > On 2/7/23 21:27, Matthew Wilcox wrote: > > On Thu, Feb 02, 2023 at 09:14:23PM +0000, Matthew Wilcox wrote: > >> For those of you not subscribed, linux-mm is currently discussing > >> how best to handle page faults on large folios. I simply made it work > >> when adding large folio support. Now Yin Fengwei is working on > >> making it fast. > >=20 > > OK, here's an actual implementation: > >=20 > > https://lore.kernel.org/linux-mm/20230207194937.122543-3-willy@infradea= d.o > > rg/ > >=20 > > It survives a run of xfstests. If your architecture doesn't store its > > PFNs at PAGE_SHIFT, you're going to want to implement your own set_ptes= (), > > or you'll see entirely the wrong pages mapped into userspace. You may > > also wish to implement set_ptes() if it can be done more efficiently > > than __pte(pteval(pte) + PAGE_SIZE). > >=20 > > Architectures that implement things like flush_icache_page() and > > update_mmu_cache() may want to propose batched versions of those. > > That's alpha, csky, m68k, mips, nios2, parisc, sh, > > arm, loongarch, openrisc, powerpc, riscv, sparc and xtensa. > > Maintainers BCC'd, mailing lists CC'd. > >=20 > > I'm happy to collect implementations and submit them as part of a v6. >=20 > Please find below the riscv implementation for set_ptes: >=20 > diff --git a/arch/riscv/include/asm/pgtable.h > b/arch/riscv/include/asm/pgtable.h > index ebee56d47003..10bf812776a4 100644 > --- a/arch/riscv/include/asm/pgtable.h > +++ b/arch/riscv/include/asm/pgtable.h > @@ -463,6 +463,20 @@ static inline void set_pte_at(struct mm_struct *mm, > __set_pte_at(mm, addr, ptep, pteval); > } >=20 > +#define set_ptes set_ptes > +static inline void set_ptes(struct mm_struct *mm, unsigned long addr, > + pte_t *ptep, pte_t pte, unsigned int nr) > +{ > + for (;;) { > + set_pte_at(mm, addr, ptep, pte); > + if (--nr =3D=3D 0) > + break; > + ptep++; > + addr +=3D PAGE_SIZE; > + pte =3D __pte(pte_val(pte) + (1 << _PAGE_PFN_SHIFT)); > + } > +} Given that this is the same code as the original version (surprise!), what= =20 about doing something like this in the generic code instead: #ifndef PTE_PAGE_STEP #define PTE_PAGE_STEP PAGE_SIZE #endif [=E2=80=A6] > + pte =3D __pte(pte_val(pte) + PTE_PAGE_STEP); The name of the define is an obvious candidate for bike-shedding, feel free= to=20 name it as you want. Or if that isn't sound enough maybe introduce something like: static inline pte_t set_ptes_next_pte(pte_t pte) { return __pte(pte_val(pte) + (1 << _PAGE_PFN_SHIFT)); } Greetings, Eike =2D-=20 Rolf Eike Beer, emlix GmbH, http://www.emlix.com =46on +49 551 30664-0, Fax +49 551 30664-11 Gothaer Platz 3, 37083 G=C3=B6ttingen, Germany Sitz der Gesellschaft: G=C3=B6ttingen, Amtsgericht G=C3=B6ttingen HR B 3160 Gesch=C3=A4ftsf=C3=BChrung: Heike Jordan, Dr. Uwe Kracke =E2=80=93 Ust-IdNr= =2E: DE 205 198 055 emlix - smart embedded open source --nextPart5907765.lOV4Wx5bFT Content-Type: application/pgp-signature; name="signature.asc" Content-Description: This is a digitally signed message part. Content-Transfer-Encoding: 7Bit -----BEGIN PGP SIGNATURE----- iLMEAAEIAB0WIQQ/Uctzh31xzAxFCLur5FH7Xu2t/AUCY/MvhgAKCRCr5FH7Xu2t /E96A/9/8q/Nw4RXmkex96nOqUtHteq72XctyEiwy7GrvS5dHVEeF79ebvwfpWIc Q6IMXFvld2oH9gZNXFrKV13KlDICP8qZscK88++MqxHdNVgMw/o6sU5yAdaiaKEh /bf9Rzx9VJTkDekQ7nP4YDLaAybLYSJ3fXX9PXr+j3HfhBsJxg== =hoWZ -----END PGP SIGNATURE----- --nextPart5907765.lOV4Wx5bFT--