From: "Liam R. Howlett" <Liam.Howlett@Oracle.com>
To: Lorenzo Stoakes <lstoakes@gmail.com>
Cc: linux-mm@kvack.org, linux-kernel@vger.kernel.org,
Andrew Morton <akpm@linux-foundation.org>,
Alexander Viro <viro@zeniv.linux.org.uk>,
Christian Brauner <brauner@kernel.org>,
Vlastimil Babka <vbabka@suse.cz>,
linux-fsdevel@vger.kernel.org
Subject: Re: [PATCH v2 4/5] mm: abstract merge for new VMAs into vma_merge_new_vma()
Date: Tue, 10 Oct 2023 21:51:40 -0400 [thread overview]
Message-ID: <20231011015140.arngzv47bdyyzfie@revolver> (raw)
In-Reply-To: <8525290591267805ffabf8a31b53f0290a6a4276.1696884493.git.lstoakes@gmail.com>
* Lorenzo Stoakes <lstoakes@gmail.com> [231009 16:53]:
> Only in mmap_region() and copy_vma() do we attempt to merge VMAs which
> occupy entirely new regions of virtual memory.
>
> We can abstract this logic and make the intent of this invocations of it
> completely explicit, rather than invoking vma_merge() with an inscrutable
> wall of parameters.
>
> This also paves the way for a simplification of the core vma_merge()
> implementation, as we seek to make it entirely an implementation detail.
>
> Note that on mmap_region(), VMA fields are initialised to zero, so we can
> simply reference these rather than explicitly specifying NULL.
I don't think that's accurate.. mmap_region() sets the start, end,
offset, flags. It also passes this vma into a driver, so I'm not sure
we can rely on them being anything after that? The whole reason
vma_merge() is attempted in this case is because the driver may have
changed vma->vm_flags on us. Your way may actually be better since the
driver may set something we assume is NULL today.
>
> Reviewed-by: Vlastimil Babka <vbabka@suse.cz>
> Signed-off-by: Lorenzo Stoakes <lstoakes@gmail.com>
> ---
> mm/mmap.c | 27 ++++++++++++++++++++-------
> 1 file changed, 20 insertions(+), 7 deletions(-)
>
> diff --git a/mm/mmap.c b/mm/mmap.c
> index 17c0dcfb1527..33aafd23823b 100644
> --- a/mm/mmap.c
> +++ b/mm/mmap.c
> @@ -2482,6 +2482,22 @@ struct vm_area_struct *vma_modify(struct vma_iterator *vmi,
> return NULL;
> }
>
> +/*
> + * Attempt to merge a newly mapped VMA with those adjacent to it. The caller
> + * must ensure that [start, end) does not overlap any existing VMA.
> + */
> +static struct vm_area_struct *vma_merge_new_vma(struct vma_iterator *vmi,
> + struct vm_area_struct *prev,
> + struct vm_area_struct *vma,
> + unsigned long start,
> + unsigned long end,
> + pgoff_t pgoff)
It's not a coding style, but if you used two tabs here, it may make this
more condensed.
> +{
> + return vma_merge(vmi, vma->vm_mm, prev, start, end, vma->vm_flags,
> + vma->anon_vma, vma->vm_file, pgoff, vma_policy(vma),
> + vma->vm_userfaultfd_ctx, anon_vma_name(vma));
> +}
> +
> /*
> * do_vmi_align_munmap() - munmap the aligned region from @start to @end.
> * @vmi: The vma iterator
> @@ -2837,10 +2853,9 @@ unsigned long mmap_region(struct file *file, unsigned long addr,
> * vma again as we may succeed this time.
> */
> if (unlikely(vm_flags != vma->vm_flags && prev)) {
> - merge = vma_merge(&vmi, mm, prev, vma->vm_start,
> - vma->vm_end, vma->vm_flags, NULL,
> - vma->vm_file, vma->vm_pgoff, NULL,
> - NULL_VM_UFFD_CTX, NULL);
> + merge = vma_merge_new_vma(&vmi, prev, vma,
> + vma->vm_start, vma->vm_end,
> + pgoff);
└ vma->vm_pgoff
> if (merge) {
> /*
> * ->mmap() can change vma->vm_file and fput
> @@ -3382,9 +3397,7 @@ struct vm_area_struct *copy_vma(struct vm_area_struct **vmap,
> if (new_vma && new_vma->vm_start < addr + len)
> return NULL; /* should never get here */
>
> - new_vma = vma_merge(&vmi, mm, prev, addr, addr + len, vma->vm_flags,
> - vma->anon_vma, vma->vm_file, pgoff, vma_policy(vma),
> - vma->vm_userfaultfd_ctx, anon_vma_name(vma));
> + new_vma = vma_merge_new_vma(&vmi, prev, vma, addr, addr + len, pgoff);
> if (new_vma) {
> /*
> * Source vma may have been merged into new_vma
> --
> 2.42.0
>
next prev parent reply other threads:[~2023-10-11 1:52 UTC|newest]
Thread overview: 13+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-10-09 20:53 [PATCH v2 0/5] Abstract vma_merge() and split_vma() Lorenzo Stoakes
2023-10-09 20:53 ` [PATCH v2 1/5] mm: move vma_policy() and anon_vma_name() decls to mm_types.h Lorenzo Stoakes
2023-10-10 6:46 ` Vlastimil Babka
2023-10-09 20:53 ` [PATCH v2 2/5] mm: abstract the vma_merge()/split_vma() pattern for mprotect() et al Lorenzo Stoakes
2023-10-10 7:12 ` Vlastimil Babka
2023-10-10 18:11 ` Lorenzo Stoakes
2023-10-11 2:14 ` Liam R. Howlett
2023-10-11 6:34 ` Lorenzo Stoakes
2023-10-09 20:53 ` [PATCH v2 3/5] mm: make vma_merge() and split_vma() internal Lorenzo Stoakes
2023-10-09 20:53 ` [PATCH v2 4/5] mm: abstract merge for new VMAs into vma_merge_new_vma() Lorenzo Stoakes
2023-10-11 1:51 ` Liam R. Howlett [this message]
2023-10-11 6:48 ` Lorenzo Stoakes
2023-10-09 20:53 ` [PATCH v2 5/5] mm: abstract VMA merge and extend into vma_merge_extend() helper Lorenzo Stoakes
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20231011015140.arngzv47bdyyzfie@revolver \
--to=liam.howlett@oracle.com \
--cc=akpm@linux-foundation.org \
--cc=brauner@kernel.org \
--cc=linux-fsdevel@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=lstoakes@gmail.com \
--cc=vbabka@suse.cz \
--cc=viro@zeniv.linux.org.uk \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).