From nobody Sun Feb 8 18:44:50 2026 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 343CBC6FD1C for ; Wed, 22 Mar 2023 17:05:25 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S231461AbjCVRFX (ORCPT ); Wed, 22 Mar 2023 13:05:23 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:59216 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S230428AbjCVREl (ORCPT ); Wed, 22 Mar 2023 13:04:41 -0400 Received: from galois.linutronix.de (Galois.linutronix.de [193.142.43.55]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id A96A16546F; Wed, 22 Mar 2023 10:03:54 -0700 (PDT) Date: Wed, 22 Mar 2023 17:03:07 -0000 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=linutronix.de; s=2020; t=1679504588; h=from:from:sender:sender:reply-to: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; bh=sFFEVBWWG4kPKKNpgGxDKlSRPvy9/vlYPq/aKAl24DE=; b=jQTBROYu+lIM2pNlyVxgOBLBukFsaopjBnBl4CiRixZ+cNWm8fgGj/m60gIHRX79CyW4jj lpoBQnaCpr3eDteQBccPc0pexO7vVCggQUyJo/YobSgJXaW+WEkk/ssFBJI7SHEYgRp0ZO zXqVLujoDo/TYRgdKrrg9cwzlkNYeTmVToNc/+X1yQF23aL/XpA+luJNpMKkvu0df/IOAa wpyYTD3/XV/XLPLuThRlrgPqBhTlqyXYH9kCU4ous6RMokJ2oywAZv6wYTyFsyL9NnYg7W 9r/gxrqpK8US2Joc+Aso72dV5B1alM4IqtPPwPoALKUtTshh+EKK+1BK7+Y8qw== DKIM-Signature: v=1; a=ed25519-sha256; c=relaxed/relaxed; d=linutronix.de; s=2020e; t=1679504588; h=from:from:sender:sender:reply-to: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; bh=sFFEVBWWG4kPKKNpgGxDKlSRPvy9/vlYPq/aKAl24DE=; b=JtNBSHvCSXQnap2DsDUxfXRNwsK4xTQC5/FNYTt4vaceC5SFj4T5GvgsuQtsTHYVwnk19+ NTW1M+4YODBhWMAg== From: "tip-bot2 for Kirill A. Shutemov" Sender: tip-bot2@linutronix.de Reply-to: linux-kernel@vger.kernel.org To: linux-tip-commits@vger.kernel.org Subject: [tip: x86/mm] mm: Introduce untagged_addr_remote() Cc: "Kirill A. Shutemov" , Dave Hansen , "Peter Zijlstra (Intel)" , Alexander Potapenko , x86@kernel.org, linux-kernel@vger.kernel.org MIME-Version: 1.0 Message-ID: <167950458787.5837.658737830777605511.tip-bot2@tip-bot2> Robot-ID: Robot-Unsubscribe: Contact to get blacklisted from these emails Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: quoted-printable Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org The following commit has been merged into the x86/mm branch of tip: Commit-ID: 428e106ae1ad4e45d3fd6978a753db475d0d0ec9 Gitweb: https://git.kernel.org/tip/428e106ae1ad4e45d3fd6978a753db475= d0d0ec9 Author: Kirill A. Shutemov AuthorDate: Sun, 12 Mar 2023 14:26:00 +03:00 Committer: Dave Hansen CommitterDate: Thu, 16 Mar 2023 13:08:39 -07:00 mm: Introduce untagged_addr_remote() untagged_addr() removes tags/metadata from the address and brings it to the canonical form. The helper is implemented on arm64 and sparc. Both of them do untagging based on global rules. However, Linear Address Masking (LAM) on x86 introduces per-process settings for untagging. As a result, untagged_addr() is now only suitable for untagging addresses for the current proccess. The new helper untagged_addr_remote() has to be used when the address targets remote process. It requires the mmap lock for target mm to be taken. Signed-off-by: Kirill A. Shutemov Signed-off-by: Dave Hansen Acked-by: Peter Zijlstra (Intel) Tested-by: Alexander Potapenko Link: https://lore.kernel.org/all/20230312112612.31869-6-kirill.shutemov%40= linux.intel.com --- arch/sparc/include/asm/uaccess_64.h | 2 ++ drivers/vfio/vfio_iommu_type1.c | 2 +- fs/proc/task_mmu.c | 9 +++++++-- include/linux/mm.h | 11 ----------- include/linux/uaccess.h | 22 ++++++++++++++++++++++ mm/gup.c | 4 ++-- mm/madvise.c | 5 +++-- mm/migrate.c | 11 ++++++----- 8 files changed, 43 insertions(+), 23 deletions(-) diff --git a/arch/sparc/include/asm/uaccess_64.h b/arch/sparc/include/asm/u= access_64.h index 94266a5..b825a5d 100644 --- a/arch/sparc/include/asm/uaccess_64.h +++ b/arch/sparc/include/asm/uaccess_64.h @@ -8,8 +8,10 @@ =20 #include #include +#include #include #include +#include =20 #include #include diff --git a/drivers/vfio/vfio_iommu_type1.c b/drivers/vfio/vfio_iommu_type= 1.c index 493c31d..3d4dd94 100644 --- a/drivers/vfio/vfio_iommu_type1.c +++ b/drivers/vfio/vfio_iommu_type1.c @@ -580,7 +580,7 @@ static int vaddr_get_pfns(struct mm_struct *mm, unsigne= d long vaddr, goto done; } =20 - vaddr =3D untagged_addr(vaddr); + vaddr =3D untagged_addr_remote(mm, vaddr); =20 retry: vma =3D vma_lookup(mm, vaddr); diff --git a/fs/proc/task_mmu.c b/fs/proc/task_mmu.c index 6a96e17..29fd6b1 100644 --- a/fs/proc/task_mmu.c +++ b/fs/proc/task_mmu.c @@ -1689,8 +1689,13 @@ static ssize_t pagemap_read(struct file *file, char = __user *buf, =20 /* watch out for wraparound */ start_vaddr =3D end_vaddr; - if (svpfn <=3D (ULONG_MAX >> PAGE_SHIFT)) - start_vaddr =3D untagged_addr(svpfn << PAGE_SHIFT); + if (svpfn <=3D (ULONG_MAX >> PAGE_SHIFT)) { + ret =3D mmap_read_lock_killable(mm); + if (ret) + goto out_free; + start_vaddr =3D untagged_addr_remote(mm, svpfn << PAGE_SHIFT); + mmap_read_unlock(mm); + } =20 /* Ensure the address is inside the task */ if (start_vaddr > mm->task_size) diff --git a/include/linux/mm.h b/include/linux/mm.h index 1f79667..289ae4c 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -96,17 +96,6 @@ extern int mmap_rnd_compat_bits __read_mostly; #include #include =20 -/* - * Architectures that support memory tagging (assigning tags to memory reg= ions, - * embedding these tags into addresses that point to these memory regions,= and - * checking that the memory and the pointer tags match on memory accesses) - * redefine this macro to strip tags from pointers. - * It's defined as noop for architectures that don't support memory taggin= g. - */ -#ifndef untagged_addr -#define untagged_addr(addr) (addr) -#endif - #ifndef __pa_symbol #define __pa_symbol(x) __pa(RELOC_HIDE((unsigned long)(x), 0)) #endif diff --git a/include/linux/uaccess.h b/include/linux/uaccess.h index ab97281..3064314 100644 --- a/include/linux/uaccess.h +++ b/include/linux/uaccess.h @@ -11,6 +11,28 @@ #include =20 /* + * Architectures that support memory tagging (assigning tags to memory reg= ions, + * embedding these tags into addresses that point to these memory regions,= and + * checking that the memory and the pointer tags match on memory accesses) + * redefine this macro to strip tags from pointers. + * + * Passing down mm_struct allows to define untagging rules on per-process + * basis. + * + * It's defined as noop for architectures that don't support memory taggin= g. + */ +#ifndef untagged_addr +#define untagged_addr(addr) (addr) +#endif + +#ifndef untagged_addr_remote +#define untagged_addr_remote(mm, addr) ({ \ + mmap_assert_locked(mm); \ + untagged_addr(addr); \ +}) +#endif + +/* * Architectures should provide two primitives (raw_copy_{to,from}_user()) * and get rid of their private instances of copy_{to,from}_user() and * __copy_{to,from}_user{,_inatomic}(). diff --git a/mm/gup.c b/mm/gup.c index eab18ba..5ee8b68 100644 --- a/mm/gup.c +++ b/mm/gup.c @@ -1085,7 +1085,7 @@ static long __get_user_pages(struct mm_struct *mm, if (!nr_pages) return 0; =20 - start =3D untagged_addr(start); + start =3D untagged_addr_remote(mm, start); =20 VM_BUG_ON(!!pages !=3D !!(gup_flags & (FOLL_GET | FOLL_PIN))); =20 @@ -1259,7 +1259,7 @@ int fixup_user_fault(struct mm_struct *mm, struct vm_area_struct *vma; vm_fault_t ret; =20 - address =3D untagged_addr(address); + address =3D untagged_addr_remote(mm, address); =20 if (unlocked) fault_flags |=3D FAULT_FLAG_ALLOW_RETRY | FAULT_FLAG_KILLABLE; diff --git a/mm/madvise.c b/mm/madvise.c index 340125d..d4b67f3 100644 --- a/mm/madvise.c +++ b/mm/madvise.c @@ -1402,8 +1402,6 @@ int do_madvise(struct mm_struct *mm, unsigned long st= art, size_t len_in, int beh size_t len; struct blk_plug plug; =20 - start =3D untagged_addr(start); - if (!madvise_behavior_valid(behavior)) return -EINVAL; =20 @@ -1435,6 +1433,9 @@ int do_madvise(struct mm_struct *mm, unsigned long st= art, size_t len_in, int beh mmap_read_lock(mm); } =20 + start =3D untagged_addr_remote(mm, start); + end =3D start + len; + blk_start_plug(&plug); error =3D madvise_walk_vmas(mm, start, end, behavior, madvise_vma_behavior); diff --git a/mm/migrate.c b/mm/migrate.c index 98f1c11..8cd11bc 100644 --- a/mm/migrate.c +++ b/mm/migrate.c @@ -2097,15 +2097,18 @@ static int do_move_pages_to_node(struct mm_struct *= mm, * target node * 1 - when it has been queued */ -static int add_page_for_migration(struct mm_struct *mm, unsigned long addr, +static int add_page_for_migration(struct mm_struct *mm, const void __user = *p, int node, struct list_head *pagelist, bool migrate_all) { struct vm_area_struct *vma; + unsigned long addr; struct page *page; int err; bool isolated; =20 mmap_read_lock(mm); + addr =3D (unsigned long)untagged_addr_remote(mm, p); + err =3D -EFAULT; vma =3D vma_lookup(mm, addr); if (!vma || !vma_migratable(vma)) @@ -2211,7 +2214,6 @@ static int do_pages_move(struct mm_struct *mm, nodema= sk_t task_nodes, =20 for (i =3D start =3D 0; i < nr_pages; i++) { const void __user *p; - unsigned long addr; int node; =20 err =3D -EFAULT; @@ -2219,7 +2221,6 @@ static int do_pages_move(struct mm_struct *mm, nodema= sk_t task_nodes, goto out_flush; if (get_user(node, nodes + i)) goto out_flush; - addr =3D (unsigned long)untagged_addr(p); =20 err =3D -ENODEV; if (node < 0 || node >=3D MAX_NUMNODES) @@ -2247,8 +2248,8 @@ static int do_pages_move(struct mm_struct *mm, nodema= sk_t task_nodes, * Errors in the page lookup or isolation are not fatal and we simply * report them via status */ - err =3D add_page_for_migration(mm, addr, current_node, - &pagelist, flags & MPOL_MF_MOVE_ALL); + err =3D add_page_for_migration(mm, p, current_node, &pagelist, + flags & MPOL_MF_MOVE_ALL); =20 if (err > 0) { /* The page is successfully queued for migration */