From: "Liam R. Howlett" <Liam.Howlett@oracle.com>
To: Pedro Falcato <pedro.falcato@gmail.com>
Cc: Andrew Morton <akpm@linux-foundation.org>,
Vlastimil Babka <vbabka@suse.cz>,
Lorenzo Stoakes <lorenzo.stoakes@oracle.com>,
linux-mm@kvack.org, linux-kernel@vger.kernel.org,
oliver.sang@intel.com, torvalds@linux-foundation.org,
jeffxu@google.com, Michael Ellerman <mpe@ellerman.id.au>
Subject: Re: [PATCH v2 5/6] mseal: Replace can_modify_mm_madv with a vma variant
Date: Fri, 9 Aug 2024 12:27:03 -0400 [thread overview]
Message-ID: <skp4dvg6h4otefmknyrg3jabqpwh4ixdi6zq645ij4wc3phmep@nzi6piu4gkk7> (raw)
In-Reply-To: <20240807211309.2729719-6-pedro.falcato@gmail.com>
* Pedro Falcato <pedro.falcato@gmail.com> [240807 17:13]:
> Replace can_modify_mm_madv() with a single vma variant, and associated
> checks in madvise.
>
> While we're at it, also invert the order of checks in:
> if (unlikely(is_ro_anon(vma) && !can_modify_vma(vma))
>
> Checking if we can modify the vma itself (through vm_flags) is
> certainly cheaper than is_ro_anon() due to arch_vma_access_permitted()
> looking at e.g pkeys registers (with extra branches) in some
> architectures.
Doesn't this also allow for partial madvise success? If you pass a
range across vmas, then it will fail once it encounters an mseal'ed vma.
This change should probably be reflected in the change log.
>
> Signed-off-by: Pedro Falcato <pedro.falcato@gmail.com>
> ---
> mm/internal.h | 6 ++----
> mm/madvise.c | 13 +++----------
> mm/mseal.c | 17 ++++-------------
> 3 files changed, 9 insertions(+), 27 deletions(-)
>
> diff --git a/mm/internal.h b/mm/internal.h
> index 90f50f3c4cf..3f9a5c17626 100644
> --- a/mm/internal.h
> +++ b/mm/internal.h
> @@ -1373,8 +1373,7 @@ static inline bool can_modify_vma(struct vm_area_struct *vma)
>
> bool can_modify_mm(struct mm_struct *mm, unsigned long start,
> unsigned long end);
> -bool can_modify_mm_madv(struct mm_struct *mm, unsigned long start,
> - unsigned long end, int behavior);
> +bool can_modify_vma_madv(struct vm_area_struct *vma, int behavior);
> #else
> static inline int can_do_mseal(unsigned long flags)
> {
> @@ -1387,8 +1386,7 @@ static inline bool can_modify_mm(struct mm_struct *mm, unsigned long start,
> return true;
> }
>
> -static inline bool can_modify_mm_madv(struct mm_struct *mm, unsigned long start,
> - unsigned long end, int behavior)
> +static inline bool can_modify_vma_madv(struct vm_area_struct *vma, int behavior)
> {
> return true;
> }
> diff --git a/mm/madvise.c b/mm/madvise.c
> index 89089d84f8d..4e64770be16 100644
> --- a/mm/madvise.c
> +++ b/mm/madvise.c
> @@ -1031,6 +1031,9 @@ static int madvise_vma_behavior(struct vm_area_struct *vma,
> struct anon_vma_name *anon_name;
> unsigned long new_flags = vma->vm_flags;
>
> + if (unlikely(!can_modify_vma_madv(vma, behavior)))
> + return -EPERM;
> +
> switch (behavior) {
> case MADV_REMOVE:
> return madvise_remove(vma, prev, start, end);
> @@ -1448,15 +1451,6 @@ int do_madvise(struct mm_struct *mm, unsigned long start, size_t len_in, int beh
> start = untagged_addr_remote(mm, start);
> end = start + len;
>
> - /*
> - * Check if the address range is sealed for do_madvise().
> - * can_modify_mm_madv assumes we have acquired the lock on MM.
> - */
> - if (unlikely(!can_modify_mm_madv(mm, start, end, behavior))) {
> - error = -EPERM;
> - goto out;
> - }
> -
Funny, this check stopped populate madvise operations. The new code
does not, which is probably better and fine.
> blk_start_plug(&plug);
> switch (behavior) {
> case MADV_POPULATE_READ:
> @@ -1470,7 +1464,6 @@ int do_madvise(struct mm_struct *mm, unsigned long start, size_t len_in, int beh
> }
> blk_finish_plug(&plug);
>
> -out:
> if (write)
> mmap_write_unlock(mm);
> else
> diff --git a/mm/mseal.c b/mm/mseal.c
> index 4591ae8d29c..6559242dd05 100644
> --- a/mm/mseal.c
> +++ b/mm/mseal.c
> @@ -67,24 +67,15 @@ bool can_modify_mm(struct mm_struct *mm, unsigned long start, unsigned long end)
> }
>
> /*
> - * Check if the vmas of a memory range are allowed to be modified by madvise.
> - * the memory ranger can have a gap (unallocated memory).
> - * return true, if it is allowed.
> + * Check if a vma is allowed to be modified by madvise.
> */
> -bool can_modify_mm_madv(struct mm_struct *mm, unsigned long start, unsigned long end,
> - int behavior)
> +bool can_modify_vma_madv(struct vm_area_struct *vma, int behavior)
> {
> - struct vm_area_struct *vma;
> -
> - VMA_ITERATOR(vmi, mm, start);
> -
> if (!is_madv_discard(behavior))
> return true;
>
> - /* going through each vma to check. */
> - for_each_vma_range(vmi, vma, end)
> - if (unlikely(is_ro_anon(vma) && !can_modify_vma(vma)))
> - return false;
> + if (unlikely(!can_modify_vma(vma) && is_ro_anon(vma)))
> + return false;
>
> /* Allow by default. */
> return true;
> --
> 2.46.0
>
next prev parent reply other threads:[~2024-08-09 16:27 UTC|newest]
Thread overview: 26+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <20240807211309.2729719-1-pedro.falcato@gmail.com>
2024-08-08 23:12 ` [PATCH v2 0/6] mm: Optimize mseal checks Andrew Morton
2024-08-09 0:34 ` Pedro Falcato
2024-08-09 1:02 ` Jeff Xu
2024-08-09 19:34 ` Liam R. Howlett
2024-08-15 22:10 ` Jeff Xu
2024-08-16 1:58 ` Pedro Falcato
2024-08-16 17:07 ` Jeff Xu
2024-08-16 18:09 ` Pedro Falcato
2024-08-16 18:20 ` Jeff Xu
2024-08-16 18:26 ` Pedro Falcato
2024-08-16 18:42 ` Jeff Xu
2024-08-16 17:30 ` Jeff Xu
[not found] ` <20240807211309.2729719-2-pedro.falcato@gmail.com>
2024-08-09 9:57 ` [PATCH v2 1/6] mm: Move can_modify_vma to mm/internal.h Lorenzo Stoakes
[not found] ` <20240807211309.2729719-5-pedro.falcato@gmail.com>
2024-08-09 16:05 ` [PATCH v2 4/6] mm/mremap: Replace can_modify_mm with can_modify_vma Liam R. Howlett
2024-08-09 18:45 ` Pedro Falcato
2024-08-12 15:22 ` Liam R. Howlett
[not found] ` <20240807211309.2729719-3-pedro.falcato@gmail.com>
2024-08-09 16:15 ` [PATCH v2 2/6] mm/munmap: " Liam R. Howlett
2024-08-09 16:48 ` Liam R. Howlett
2024-08-09 18:53 ` Pedro Falcato
2024-08-09 19:24 ` Liam R. Howlett
2024-08-12 14:30 ` Jeff Xu
2024-08-12 16:57 ` Liam R. Howlett
2024-08-12 17:38 ` Jeff Xu
2024-08-12 19:25 ` Liam R. Howlett
[not found] ` <20240807211309.2729719-6-pedro.falcato@gmail.com>
2024-08-09 16:27 ` Liam R. Howlett [this message]
2024-08-15 21:11 ` [PATCH v2 0/6] mm: Optimize mseal checks Jeff Xu
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=skp4dvg6h4otefmknyrg3jabqpwh4ixdi6zq645ij4wc3phmep@nzi6piu4gkk7 \
--to=liam.howlett@oracle.com \
--cc=akpm@linux-foundation.org \
--cc=jeffxu@google.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=lorenzo.stoakes@oracle.com \
--cc=mpe@ellerman.id.au \
--cc=oliver.sang@intel.com \
--cc=pedro.falcato@gmail.com \
--cc=torvalds@linux-foundation.org \
--cc=vbabka@suse.cz \
/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