linux-kernel - Re: [PATCH v2 3/3] hugetlb: Convert hugetlb_wp() to use struct vm

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-ID: <ZhQvlpOETGxXO_MG@casper.infradead.org>
Date: Mon, 8 Apr 2024 18:55:34 +0100
From: Matthew Wilcox <willy@...radead.org>
To: Vishal Moola <vishal.moola@...il.com>
Cc: Muchun Song <muchun.song@...ux.dev>, linux-kernel@...r.kernel.org,
	akpm@...ux-foundation.org, linux-mm@...ck.org
Subject: Re: [PATCH v2 3/3] hugetlb: Convert hugetlb_wp() to use struct
 vm_fault

On Mon, Apr 08, 2024 at 10:47:12AM -0700, Vishal Moola wrote:
> -static vm_fault_t hugetlb_wp(struct mm_struct *mm, struct vm_area_struct *vma,
> -		       struct folio *pagecache_folio,
> +static vm_fault_t hugetlb_wp(struct folio *pagecache_folio,
>  		       struct vm_fault *vmf)
>  {
> +	struct vm_area_struct *vma = vmf->vma;
> +	struct mm_struct *mm = vma->vm_mm;

I think 'vmf' should be the first parameter, not the second.
Compare:

static vm_fault_t do_page_mkwrite(struct vm_fault *vmf, struct folio *folio)
static inline void wp_page_reuse(struct vm_fault *vmf, struct folio *folio)
static vm_fault_t finish_mkwrite_fault(struct vm_fault *vmf, struct folio *folio)
static vm_fault_t wp_page_shared(struct vm_fault *vmf, struct folio *folio)
vm_fault_t do_set_pmd(struct vm_fault *vmf, struct page *page)
void set_pte_range(struct vm_fault *vmf, struct folio *folio,
                struct page *page, unsigned int nr, unsigned long addr)
int numa_migrate_prep(struct folio *folio, struct vm_fault *vmf,
                      unsigned long addr, int page_nid, int *flags)
static void numa_rebuild_single_mapping(struct vm_fault *vmf, struct vm_area_struct *vma,
                                        unsigned long fault_addr, pte_t *fault_pte,
                                        bool writable)
static void numa_rebuild_large_mapping(struct vm_fault *vmf, struct vm_area_struct *vma,
                                       struct folio *folio, pte_t fault_pte,
                                       bool ignore_writable, bool pte_write_upgrade)
static vm_fault_t wp_huge_pud(struct vm_fault *vmf, pud_t orig_pud)

numa_migrate_prep() is the only one which doesn't have vmf as the first
param.  It's a subtle inconsistency, but one you notice after a while
.. then wish you'd done right the first time, but can't quite bring
yourself to submit a patch to fix ;-)