[PATCH v18] arm64: mm: Handle Granule Protection Faults (GPFs)

[PATCH v18] arm64: mm: Handle Granule Protection Faults (GPFs)

[Suzuki K Poulose]

From: Steven Price <steven.price@arm.com>

If the host attempts to access granules that have been delegated for use
in a realm these accesses will be caught and will trigger a Granule
Protection Fault (GPF).

A fault during a page walk signals a bug in the kernel and is handled by
oopsing the kernel. A non-page walk fault could be caused by user space
having access to a page which has been delegated to the kernel and will
trigger a SIGBUS to allow debugging why user space is trying to access a
delegated page.

There is work in progress to unmap the guest_memfd backed private pages from the
linear map. Until we get that support, we could get spurious GPFs from within
the kernel, e.g., load_unaligned_zeropad(). So, try to fix them up for now.

Reviewed-by: Suzuki K Poulose <suzuki.poulose@arm.com>
Reviewed-by: Gavin Shan <gshan@redhat.com>
Reviewed-by: Catalin Marinas <catalin.marinas@arm.com>
Signed-off-by: Steven Price <steven.price@arm.com>
Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
---
Changes since v17:
 * Pass untagged address to die_kernel_fault() - Sashiko
 * Explicitly check !user_mode() for fixups - Catalin
 * Switch to BUS_OBJERR for si_code from SI_KERNEL - Catalin
 * Clarify the commit description about the upcoming work on
   unmapping guest_memfd backed pages from linear map
Changes since v16:
 * Update the commit description to indicate why we try to fixup GPFs
Changes since v10:
 * Don't call arm64_notify_die() in do_gpf() but simply return 1.
Changes since v2:
 * Include missing "Granule Protection Fault at level -1"
---
 arch/arm64/mm/fault.c | 30 ++++++++++++++++++++++++------
 1 file changed, 24 insertions(+), 6 deletions(-)

diff --git a/arch/arm64/mm/fault.c b/arch/arm64/mm/fault.c
index 75c3e463df2ef..dc3a87902a60c 100644
--- a/arch/arm64/mm/fault.c
+++ b/arch/arm64/mm/fault.c
@@ -914,6 +914,24 @@ static int do_tag_check_fault(unsigned long far, unsigned long esr,
 	return 0;
 }
 
+static int do_gpf_ptw(unsigned long far, unsigned long esr, struct pt_regs *regs)
+{
+	const struct fault_info *inf = esr_to_fault_info(esr);
+	unsigned long addr = untagged_addr(far);
+
+	die_kernel_fault(inf->name, addr, esr, regs);
+	return 0;
+}
+
+static int do_gpf(unsigned long far, unsigned long esr, struct pt_regs *regs)
+{
+	if (!user_mode(regs) && !is_el1_instruction_abort(esr) &&
+	    fixup_exception(regs, esr))
+		return 0;
+
+	return 1;
+}
+
 static const struct fault_info fault_info[] = {
 	{ do_bad,		SIGKILL, SI_KERNEL,	"ttbr address size fault"	},
 	{ do_bad,		SIGKILL, SI_KERNEL,	"level 1 address size fault"	},
@@ -950,12 +968,12 @@ static const struct fault_info fault_info[] = {
 	{ do_bad,		SIGKILL, SI_KERNEL,	"unknown 32"			},
 	{ do_alignment_fault,	SIGBUS,  BUS_ADRALN,	"alignment fault"		},
 	{ do_bad,		SIGKILL, SI_KERNEL,	"unknown 34"			},
-	{ do_bad,		SIGKILL, SI_KERNEL,	"unknown 35"			},
-	{ do_bad,		SIGKILL, SI_KERNEL,	"unknown 36"			},
-	{ do_bad,		SIGKILL, SI_KERNEL,	"unknown 37"			},
-	{ do_bad,		SIGKILL, SI_KERNEL,	"unknown 38"			},
-	{ do_bad,		SIGKILL, SI_KERNEL,	"unknown 39"			},
-	{ do_bad,		SIGKILL, SI_KERNEL,	"unknown 40"			},
+	{ do_gpf_ptw,		SIGKILL, SI_KERNEL,	"level -1 granule protection fault (translation table walk)" },
+	{ do_gpf_ptw,		SIGKILL, SI_KERNEL,	"level 0 granule protection fault (translation table walk)" },
+	{ do_gpf_ptw,		SIGKILL, SI_KERNEL,	"level 1 granule protection fault (translation table walk)" },
+	{ do_gpf_ptw,		SIGKILL, SI_KERNEL,	"level 2 granule protection fault (translation table walk)" },
+	{ do_gpf_ptw,		SIGKILL, SI_KERNEL,	"level 3 granule protection fault (translation table walk)" },
+	{ do_gpf,		SIGBUS,  BUS_OBJERR,	"granule protection fault" },
 	{ do_bad,		SIGKILL, SI_KERNEL,	"level -1 address size fault"	},
 	{ do_bad,		SIGKILL, SI_KERNEL,	"unknown 42"			},
 	{ do_translation_fault,	SIGSEGV, SEGV_MAPERR,	"level -1 translation fault"	},
-- 
2.43.0

Re: [PATCH v18] arm64: mm: Handle Granule Protection Faults (GPFs)

[Catalin Marinas]

On Sun, Sep 13, 2026 at 08:04:58AM +0100, Suzuki K Poulose wrote:
> From: Steven Price <steven.price@arm.com>
> 
> If the host attempts to access granules that have been delegated for use
> in a realm these accesses will be caught and will trigger a Granule
> Protection Fault (GPF).
> 
> A fault during a page walk signals a bug in the kernel and is handled by
> oopsing the kernel. A non-page walk fault could be caused by user space
> having access to a page which has been delegated to the kernel and will
> trigger a SIGBUS to allow debugging why user space is trying to access a
> delegated page.
> 
> There is work in progress to unmap the guest_memfd backed private pages from the
> linear map. Until we get that support, we could get spurious GPFs from within
> the kernel, e.g., load_unaligned_zeropad(). So, try to fix them up for now.
> 
> Reviewed-by: Suzuki K Poulose <suzuki.poulose@arm.com>
> Reviewed-by: Gavin Shan <gshan@redhat.com>
> Reviewed-by: Catalin Marinas <catalin.marinas@arm.com>
> Signed-off-by: Steven Price <steven.price@arm.com>
> Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
> ---
> Changes since v17:
>  * Pass untagged address to die_kernel_fault() - Sashiko
>  * Explicitly check !user_mode() for fixups - Catalin
>  * Switch to BUS_OBJERR for si_code from SI_KERNEL - Catalin
>  * Clarify the commit description about the upcoming work on
>    unmapping guest_memfd backed pages from linear map
> Changes since v16:
>  * Update the commit description to indicate why we try to fixup GPFs
> Changes since v10:
>  * Don't call arm64_notify_die() in do_gpf() but simply return 1.
> Changes since v2:
>  * Include missing "Granule Protection Fault at level -1"
> ---
>  arch/arm64/mm/fault.c | 30 ++++++++++++++++++++++++------
>  1 file changed, 24 insertions(+), 6 deletions(-)
> 
> diff --git a/arch/arm64/mm/fault.c b/arch/arm64/mm/fault.c
> index 75c3e463df2ef..dc3a87902a60c 100644
> --- a/arch/arm64/mm/fault.c
> +++ b/arch/arm64/mm/fault.c
> @@ -914,6 +914,24 @@ static int do_tag_check_fault(unsigned long far, unsigned long esr,
>  	return 0;
>  }
>  
> +static int do_gpf_ptw(unsigned long far, unsigned long esr, struct pt_regs *regs)
> +{
> +	const struct fault_info *inf = esr_to_fault_info(esr);
> +	unsigned long addr = untagged_addr(far);
> +
> +	die_kernel_fault(inf->name, addr, esr, regs);
> +	return 0;
> +}
> +
> +static int do_gpf(unsigned long far, unsigned long esr, struct pt_regs *regs)
> +{
> +	if (!user_mode(regs) && !is_el1_instruction_abort(esr) &&
> +	    fixup_exception(regs, esr))
> +		return 0;
> +
> +	return 1;
> +}

We discussed briefly offline. With the latest patches around, would we
ever end up with private memory mapped in the VMM and hence the GPF? If
not, I would still keep this handling but add a
WARN_ON_ONCE(user_mode(regs)).

However, can we end up delegating a non-guest_memfd memslot page as
protected?

I played a bit with codex and it reckons it's possible if a guest_memfd
memslot is deleted after its IPA range has been initialised with
RIPAS=RAM. Removing the memslot unmaps and undelegates any data pages
but leaves the RMM state as RAM. The VMM can then install an ordinary
memslot over the same GPA range.

A subsequent private-IPA S2 fault sees the non-guest_memfd slot, takes
user_mem_abort(), GUPs the user page and passes it to
realm_map_protected(). The userspace mapping remains present, so a later
EL0 access can generate a GPF.

What's worse, I think it can even trick the kernel into doing a memcpy()
access (via GUP). Hmm, does such ordinary slot page even remain pinned?
There are other kernel parts that could access it.

I don't think it changes this patch but if the above is possible, we
should definitely get it tightened on the other series (and here we can
add the warning).

-- 
Catalin

Re: [PATCH v18] arm64: mm: Handle Granule Protection Faults (GPFs)

[Catalin Marinas]

On Wed, Sep 16, 2026 at 05:39:31PM +0100, Catalin Marinas wrote:
> On Sun, Sep 13, 2026 at 08:04:58AM +0100, Suzuki K Poulose wrote:
> > +static int do_gpf(unsigned long far, unsigned long esr, struct pt_regs *regs)
> > +{
> > +	if (!user_mode(regs) && !is_el1_instruction_abort(esr) &&
> > +	    fixup_exception(regs, esr))
> > +		return 0;
> > +
> > +	return 1;
> > +}
> 
> We discussed briefly offline. With the latest patches around, would we
> ever end up with private memory mapped in the VMM and hence the GPF? If
> not, I would still keep this handling but add a
> WARN_ON_ONCE(user_mode(regs)).

Actually, even better:

	WARN_ON_ONCE(is_ttbr0_addr(untagged_addr(far)));

to capture uaccess. The user_mode() test you have above can stay the
same.

-- 
Catalin

Re: [PATCH v18] arm64: mm: Handle Granule Protection Faults (GPFs)

[Catalin Marinas]

On Wed, Sep 16, 2026 at 05:39:24PM +0100, Catalin Marinas wrote:
> However, can we end up delegating a non-guest_memfd memslot page as
> protected?
> 
> I played a bit with codex and it reckons it's possible if a guest_memfd
> memslot is deleted after its IPA range has been initialised with
> RIPAS=RAM. Removing the memslot unmaps and undelegates any data pages
> but leaves the RMM state as RAM. The VMM can then install an ordinary
> memslot over the same GPA range.

After some more digging, I think RIPAS becomes DESTROYED after
undelegation but it doesn't change much. A subsequent guest access to
the private IPA still exits to the host which will attempt to delegate
it again even if it's no longer guest_memfd. RIPAS remains DESTROYED but
GPT is now REALM (and potentially a valid user mapping).

-- 
Catalin