* [PATCH v2 1/3] mm/mseal: remove superfluous comments, fix confusion around mm
2026-07-17 17:27 [PATCH v2 0/3] mm/mseal: further cleanups Lorenzo Stoakes (ARM)
@ 2026-07-17 17:27 ` Lorenzo Stoakes (ARM)
2026-07-17 17:27 ` [PATCH v2 2/3] mm/mseal: limit scope of mseal address zero to address zero Lorenzo Stoakes (ARM)
` (2 subsequent siblings)
3 siblings, 0 replies; 5+ messages in thread
From: Lorenzo Stoakes (ARM) @ 2026-07-17 17:27 UTC (permalink / raw)
To: Andrew Morton, Liam R. Howlett, Vlastimil Babka, Jann Horn,
Pedro Falcato, Alexander Viro, Christian Brauner, Jan Kara,
Kees Cook, David Hildenbrand, Mike Rapoport, Suren Baghdasaryan,
Michal Hocko
Cc: ljs, linux-mm, linux-kernel, linux-fsdevel
Remove comment blocks that don't add value and eliminate any confusion
about whether or not we permit mseal()'ing of remote mm's by not passing
through an mm parameter but rather referencing current->mm in each
function.
Also while we're here, avoid an ugly goto by using an else branch, and move
local parameters declarations into reverse xmas tree order.
No functional change intended.
Acked-by: David Hildenbrand (Arm) <david@kernel.org>
Reviewed-by: Pedro Falcato <pfalcato@suse.de>
Signed-off-by: Lorenzo Stoakes (ARM) <ljs@kernel.org>
---
mm/mseal.c | 53 +++++++++++------------------------------------------
1 file changed, 11 insertions(+), 42 deletions(-)
diff --git a/mm/mseal.c b/mm/mseal.c
index 9781647483d1..430a252a6da4 100644
--- a/mm/mseal.c
+++ b/mm/mseal.c
@@ -16,32 +16,11 @@
#include <linux/sched.h>
#include "internal.h"
-/*
- * mseal() disallows an input range which contain unmapped ranges (VMA holes).
- *
- * It disallows unmapped regions from start to end whether they exist at the
- * start, in the middle, or at the end of the range, or any combination thereof.
- *
- * This is because after sealing a range, there's nothing to stop memory mapping
- * of ranges in the remaining gaps later, meaning that the user might then
- * wrongly consider the entirety of the mseal()'d range to be sealed when it
- * in fact isn't.
- */
-
-/*
- * Does the [start, end) range contain any unmapped memory?
- *
- * We ensure that:
- * - start is part of a valid VMA.
- * - end is part of a valid VMA.
- * - no gap (unallocated memory) exists between start and end.
- */
-static bool range_contains_unmapped(struct mm_struct *mm,
- unsigned long start, unsigned long end)
+static bool range_contains_unmapped(unsigned long start, unsigned long end)
{
- struct vm_area_struct *vma;
- unsigned long prev_end = start;
VMA_ITERATOR(vmi, current->mm, start);
+ unsigned long prev_end = start;
+ struct vm_area_struct *vma;
for_each_vma_range(vmi, vma, end) {
if (vma->vm_start > prev_end)
@@ -53,11 +32,10 @@ static bool range_contains_unmapped(struct mm_struct *mm,
return prev_end < end;
}
-static int mseal_apply(struct mm_struct *mm,
- unsigned long start, unsigned long end)
+static int mseal_apply(unsigned long start, unsigned long end)
{
+ VMA_ITERATOR(vmi, current->mm, start);
struct vm_area_struct *vma, *prev;
- VMA_ITERATOR(vmi, mm, start);
/* We know there are no gaps so this will be non-NULL. */
vma = vma_iter_load(&vmi);
@@ -142,10 +120,10 @@ static int mseal_apply(struct mm_struct *mm,
*/
int do_mseal(unsigned long start, size_t len_in, unsigned long flags)
{
- size_t len;
- int ret = 0;
- unsigned long end;
struct mm_struct *mm = current->mm;
+ unsigned long end;
+ int ret = 0;
+ size_t len;
/* Verify flags not set. */
if (flags)
@@ -170,20 +148,11 @@ int do_mseal(unsigned long start, size_t len_in, unsigned long flags)
if (mmap_write_lock_killable(mm))
return -EINTR;
- if (range_contains_unmapped(mm, start, end)) {
+ if (range_contains_unmapped(start, end))
ret = -ENOMEM;
- goto out;
- }
-
- /*
- * Second pass, this should success, unless there are errors
- * from vma_modify_flags, e.g. merge/split error, or process
- * reaching the max supported VMAs, however, those cases shall
- * be rare.
- */
- ret = mseal_apply(mm, start, end);
+ else
+ ret = mseal_apply(start, end);
-out:
mmap_write_unlock(mm);
return ret;
}
--
2.55.0
^ permalink raw reply [flat|nested] 5+ messages in thread* [PATCH v2 2/3] mm/mseal: limit scope of mseal address zero to address zero
2026-07-17 17:27 [PATCH v2 0/3] mm/mseal: further cleanups Lorenzo Stoakes (ARM)
2026-07-17 17:27 ` [PATCH v2 1/3] mm/mseal: remove superfluous comments, fix confusion around mm Lorenzo Stoakes (ARM)
@ 2026-07-17 17:27 ` Lorenzo Stoakes (ARM)
2026-07-17 17:27 ` [PATCH v2 3/3] mm/mseal: remove further superfluous comments, do_mseal() Lorenzo Stoakes (ARM)
2026-07-17 19:39 ` [PATCH v2 0/3] mm/mseal: further cleanups Andrew Morton
3 siblings, 0 replies; 5+ messages in thread
From: Lorenzo Stoakes (ARM) @ 2026-07-17 17:27 UTC (permalink / raw)
To: Andrew Morton, Liam R. Howlett, Vlastimil Babka, Jann Horn,
Pedro Falcato, Alexander Viro, Christian Brauner, Jan Kara,
Kees Cook, David Hildenbrand, Mike Rapoport, Suren Baghdasaryan,
Michal Hocko
Cc: ljs, linux-mm, linux-kernel, linux-fsdevel
Commit 44f65d900698 ("binfmt_elf: mseal address zero") unconditionally
provided do_mseal() to any internal kernel caller in order to address a
corner case slated for possible removal.
It also incorrectly attempts to mseal without checking to see whether the
mapping even succeeded.
Restrict the scope to the corner case by providing mseal_mmap_page_zero()
which asserts the MMAP_PAGE_ZERO personality.
Avoid unnecessary checks in the start, end range by abstracting the actual
mseal()'ing to mseal_range() and have mseal_mmap_page_zero() call that
instead.
Also only try to seal the VMA if we mapped the VMA. This isn't strictly
necessary as the operation would error out anyway, but it's useless work
and could be problematic if me make future changes to mseal semantics.
Acked-by: David Hildenbrand (Arm) <david@kernel.org>
Signed-off-by: Lorenzo Stoakes (ARM) <ljs@kernel.org>
---
fs/binfmt_elf.c | 7 ++-----
include/linux/mm.h | 8 ++------
mm/mseal.c | 49 +++++++++++++++++++++++++++++++++++--------------
3 files changed, 39 insertions(+), 25 deletions(-)
diff --git a/fs/binfmt_elf.c b/fs/binfmt_elf.c
index 16a56b6b3f6c..e3131a311995 100644
--- a/fs/binfmt_elf.c
+++ b/fs/binfmt_elf.c
@@ -1353,11 +1353,8 @@ static int load_elf_binary(struct linux_binprm *bprm)
emulate the SVr4 behavior. Sigh. */
error = vm_mmap(NULL, 0, PAGE_SIZE, PROT_READ | PROT_EXEC,
MAP_FIXED | MAP_PRIVATE, 0);
-
- retval = do_mseal(0, PAGE_SIZE, 0);
- if (retval)
- pr_warn_ratelimited("pid=%d, couldn't seal address 0, ret=%d.\n",
- task_pid_nr(current), retval);
+ if (!error)
+ mseal_mmap_page_zero();
}
regs = current_pt_regs();
diff --git a/include/linux/mm.h b/include/linux/mm.h
index 550fb92957d1..87feaa5a2b78 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -5291,13 +5291,9 @@ int reserve_mem_find_by_name(const char *name, phys_addr_t *start, phys_addr_t *
int reserve_mem_release_by_name(const char *name);
#ifdef CONFIG_64BIT
-int do_mseal(unsigned long start, size_t len_in, unsigned long flags);
+void mseal_mmap_page_zero(void);
#else
-static inline int do_mseal(unsigned long start, size_t len_in, unsigned long flags)
-{
- /* noop on 32 bit */
- return 0;
-}
+static inline void mseal_mmap_page_zero(void) {}
#endif
/*
diff --git a/mm/mseal.c b/mm/mseal.c
index 430a252a6da4..2a516da694c6 100644
--- a/mm/mseal.c
+++ b/mm/mseal.c
@@ -32,7 +32,7 @@ static bool range_contains_unmapped(unsigned long start, unsigned long end)
return prev_end < end;
}
-static int mseal_apply(unsigned long start, unsigned long end)
+static int __mseal_range(unsigned long start, unsigned long end)
{
VMA_ITERATOR(vmi, current->mm, start);
struct vm_area_struct *vma, *prev;
@@ -66,6 +66,38 @@ static int mseal_apply(unsigned long start, unsigned long end)
return 0;
}
+static int mseal_range(unsigned long start, unsigned long end)
+{
+ int err;
+
+ err = mmap_write_lock_killable(current->mm);
+ if (err)
+ return err;
+ if (range_contains_unmapped(start, end))
+ err = -ENOMEM;
+ else
+ err = __mseal_range(start, end);
+ mmap_write_unlock(current->mm);
+ return err;
+}
+
+/**
+ * mseal_mmap_page_zero() - If the MMAP_PAGE_ZERO personality is set, mseal()
+ * the page mapped at address zero.
+ */
+void mseal_mmap_page_zero(void)
+{
+ int err;
+
+ if (WARN_ON_ONCE(!(current->personality & MMAP_PAGE_ZERO)))
+ return;
+
+ err = mseal_range(0, PAGE_SIZE);
+ if (err)
+ pr_warn_ratelimited("pid=%d, couldn't seal address 0, ret=%d.\n",
+ task_pid_nr(current), err);
+}
+
/*
* mseal(2) seals the VM's meta data from
* selected syscalls.
@@ -118,11 +150,9 @@ static int mseal_apply(unsigned long start, unsigned long end)
*
* unseal() is not supported.
*/
-int do_mseal(unsigned long start, size_t len_in, unsigned long flags)
+static int do_mseal(unsigned long start, size_t len_in, unsigned long flags)
{
- struct mm_struct *mm = current->mm;
unsigned long end;
- int ret = 0;
size_t len;
/* Verify flags not set. */
@@ -145,16 +175,7 @@ int do_mseal(unsigned long start, size_t len_in, unsigned long flags)
if (end == start)
return 0;
- if (mmap_write_lock_killable(mm))
- return -EINTR;
-
- if (range_contains_unmapped(start, end))
- ret = -ENOMEM;
- else
- ret = mseal_apply(start, end);
-
- mmap_write_unlock(mm);
- return ret;
+ return mseal_range(start, end);
}
SYSCALL_DEFINE3(mseal, unsigned long, start, size_t, len, unsigned long,
--
2.55.0
^ permalink raw reply [flat|nested] 5+ messages in thread* [PATCH v2 3/3] mm/mseal: remove further superfluous comments, do_mseal()
2026-07-17 17:27 [PATCH v2 0/3] mm/mseal: further cleanups Lorenzo Stoakes (ARM)
2026-07-17 17:27 ` [PATCH v2 1/3] mm/mseal: remove superfluous comments, fix confusion around mm Lorenzo Stoakes (ARM)
2026-07-17 17:27 ` [PATCH v2 2/3] mm/mseal: limit scope of mseal address zero to address zero Lorenzo Stoakes (ARM)
@ 2026-07-17 17:27 ` Lorenzo Stoakes (ARM)
2026-07-17 19:39 ` [PATCH v2 0/3] mm/mseal: further cleanups Andrew Morton
3 siblings, 0 replies; 5+ messages in thread
From: Lorenzo Stoakes (ARM) @ 2026-07-17 17:27 UTC (permalink / raw)
To: Andrew Morton, Liam R. Howlett, Vlastimil Babka, Jann Horn,
Pedro Falcato, Alexander Viro, Christian Brauner, Jan Kara,
Kees Cook, David Hildenbrand, Mike Rapoport, Suren Baghdasaryan,
Michal Hocko
Cc: ljs, linux-mm, linux-kernel, linux-fsdevel
There's no need to abstract do_mseal() any longer so put the system call
implementation in the system call declaration.
The comment around do_mseal() is strangely formatted, overly long and adds
a lot of superfluous information that the code already provides, so boil it
down to the essentials.
Acked-by: David Hildenbrand (Arm) <david@kernel.org>
Reviewed-by: Pedro Falcato <pfalcato@suse.de>
Signed-off-by: Lorenzo Stoakes (ARM) <ljs@kernel.org>
---
mm/mseal.c | 74 ++++++++++++++------------------------------------------------
1 file changed, 16 insertions(+), 58 deletions(-)
diff --git a/mm/mseal.c b/mm/mseal.c
index 2a516da694c6..7a8ac66dc215 100644
--- a/mm/mseal.c
+++ b/mm/mseal.c
@@ -99,61 +99,25 @@ void mseal_mmap_page_zero(void)
}
/*
- * mseal(2) seals the VM's meta data from
- * selected syscalls.
+ * Seal VMAs in the specified input range to prevent an attacker replacing what
+ * is mapped in the range with something else.
*
- * addr/len: VM address range.
+ * Disallows:
+ * - VMA unmapping, remapping or shrinking.
+ * - Overwriting the VMA with another one via mmap(), mremap() or similar.
+ * - Alteration of properties via mprotect()/pkey_mprotect().
+ * - Destructive madvise() behaviours (like MADV_DONTNEED) on anonymous read-only
+ * ranges.
*
- * The address range by addr/len must meet:
- * start (addr) must be in a valid VMA.
- * end (addr + len) must be in a valid VMA.
- * no gap (unallocated memory) between start and end.
- * start (addr) must be page aligned.
+ * Since unmapped ranges can be mapped at any time, the input range must span
+ * mapped ranges only.
*
- * len: len will be page aligned implicitly.
- *
- * Below VMA operations are blocked after sealing.
- * 1> Unmapping, moving to another location, and shrinking
- * the size, via munmap() and mremap(), can leave an empty
- * space, therefore can be replaced with a VMA with a new
- * set of attributes.
- * 2> Moving or expanding a different vma into the current location,
- * via mremap().
- * 3> Modifying a VMA via mmap(MAP_FIXED).
- * 4> Size expansion, via mremap(), does not appear to pose any
- * specific risks to sealed VMAs. It is included anyway because
- * the use case is unclear. In any case, users can rely on
- * merging to expand a sealed VMA.
- * 5> mprotect and pkey_mprotect.
- * 6> Some destructive madvice() behavior (e.g. MADV_DONTNEED)
- * for anonymous memory, when users don't have write permission to the
- * memory. Those behaviors can alter region contents by discarding pages,
- * effectively a memset(0) for anonymous memory.
- *
- * flags: reserved.
- *
- * return values:
- * zero: success.
- * -EINVAL:
- * invalid input flags.
- * start address is not page aligned.
- * Address range (start + len) overflow.
- * -ENOMEM:
- * addr is not a valid address (not allocated).
- * end (start + len) is not a valid address.
- * a gap (unallocated memory) between start and end.
- * -EPERM:
- * - In 32 bit architecture, sealing is not supported.
- * Note:
- * user can call mseal(2) multiple times, adding a seal on an
- * already sealed memory is a no-action (no error).
- *
- * unseal() is not supported.
+ * The flags parameter is currently reserved.
*/
-static int do_mseal(unsigned long start, size_t len_in, unsigned long flags)
+SYSCALL_DEFINE3(mseal, unsigned long, start, size_t, len, unsigned long, flags)
{
+ size_t len_aligned;
unsigned long end;
- size_t len;
/* Verify flags not set. */
if (flags)
@@ -163,12 +127,12 @@ static int do_mseal(unsigned long start, size_t len_in, unsigned long flags)
if (!PAGE_ALIGNED(start))
return -EINVAL;
- len = PAGE_ALIGN(len_in);
+ len_aligned = PAGE_ALIGN(len);
/* Check to see whether len was rounded up from small -ve to zero. */
- if (len_in && !len)
+ if (len && !len_aligned)
return -EINVAL;
- end = start + len;
+ end = start + len_aligned;
if (end < start)
return -EINVAL;
@@ -177,9 +141,3 @@ static int do_mseal(unsigned long start, size_t len_in, unsigned long flags)
return mseal_range(start, end);
}
-
-SYSCALL_DEFINE3(mseal, unsigned long, start, size_t, len, unsigned long,
- flags)
-{
- return do_mseal(start, len, flags);
-}
--
2.55.0
^ permalink raw reply [flat|nested] 5+ messages in thread* Re: [PATCH v2 0/3] mm/mseal: further cleanups
2026-07-17 17:27 [PATCH v2 0/3] mm/mseal: further cleanups Lorenzo Stoakes (ARM)
` (2 preceding siblings ...)
2026-07-17 17:27 ` [PATCH v2 3/3] mm/mseal: remove further superfluous comments, do_mseal() Lorenzo Stoakes (ARM)
@ 2026-07-17 19:39 ` Andrew Morton
3 siblings, 0 replies; 5+ messages in thread
From: Andrew Morton @ 2026-07-17 19:39 UTC (permalink / raw)
To: Lorenzo Stoakes (ARM)
Cc: Liam R. Howlett, Vlastimil Babka, Jann Horn, Pedro Falcato,
Alexander Viro, Christian Brauner, Jan Kara, Kees Cook,
David Hildenbrand, Mike Rapoport, Suren Baghdasaryan,
Michal Hocko, linux-mm, linux-kernel, linux-fsdevel
On Fri, 17 Jul 2026 18:27:08 +0100 "Lorenzo Stoakes (ARM)" <ljs@kernel.org> wrote:
>
> The mseal implementation is still rather confusing, so tighten things up a
> little.
>
> The only user of do_mseal() outside of the system call is the MMAP_PAGE_ZERO
> process personality - retain better control over how mseal is utilised by
> providing mseal_mmap_page_zero() for this instead.
>
> The comments are overly long and confusion, so cut them down so they're a lot
> clearer.
>
> Remove confusing mm_struct params (mseal can not be used on remote mm's) and
> wrap the actual system call logic into the system call declaration.
Thanks, I've updated mm-new to this version.
> v2:
> * Added tags (thanks everyone!)
> * Abstracted mm as per David.
> * Renamed [__]mseal() to [__]mseal_range() as per Pedro.
> * Moved to reverse xmas tree declarations.
> * Tweaked commit message for 2/3.
Here's how v2 altered mm.git:
mm/mseal.c | 16 ++++++++--------
1 file changed, 8 insertions(+), 8 deletions(-)
--- a/mm/mseal.c~b
+++ a/mm/mseal.c
@@ -18,9 +18,9 @@
static bool range_contains_unmapped(unsigned long start, unsigned long end)
{
- struct vm_area_struct *vma;
- unsigned long prev_end = start;
VMA_ITERATOR(vmi, current->mm, start);
+ unsigned long prev_end = start;
+ struct vm_area_struct *vma;
for_each_vma_range(vmi, vma, end) {
if (vma->vm_start > prev_end)
@@ -32,10 +32,10 @@ static bool range_contains_unmapped(unsi
return prev_end < end;
}
-static int __mseal(unsigned long start, unsigned long end)
+static int __mseal_range(unsigned long start, unsigned long end)
{
- struct vm_area_struct *vma, *prev;
VMA_ITERATOR(vmi, current->mm, start);
+ struct vm_area_struct *vma, *prev;
/* We know there are no gaps so this will be non-NULL. */
vma = vma_iter_load(&vmi);
@@ -66,7 +66,7 @@ static int __mseal(unsigned long start,
return 0;
}
-static int mseal(unsigned long start, unsigned long end)
+static int mseal_range(unsigned long start, unsigned long end)
{
int err;
@@ -76,7 +76,7 @@ static int mseal(unsigned long start, un
if (range_contains_unmapped(start, end))
err = -ENOMEM;
else
- err = __mseal(start, end);
+ err = __mseal_range(start, end);
mmap_write_unlock(current->mm);
return err;
}
@@ -92,7 +92,7 @@ void mseal_mmap_page_zero(void)
if (WARN_ON_ONCE(!(current->personality & MMAP_PAGE_ZERO)))
return;
- err = mseal(0, PAGE_SIZE);
+ err = mseal_range(0, PAGE_SIZE);
if (err)
pr_warn_ratelimited("pid=%d, couldn't seal address 0, ret=%d.\n",
task_pid_nr(current), err);
@@ -139,5 +139,5 @@ SYSCALL_DEFINE3(mseal, unsigned long, st
if (end == start)
return 0;
- return mseal(start, end);
+ return mseal_range(start, end);
}
_
^ permalink raw reply [flat|nested] 5+ messages in thread