* [PATCH] lib: utils/irqchip: Add sanity checks to imsic_warm_irqchip_init()
@ 2024-04-09 14:45 Cyan Yang
2024-05-07 12:39 ` Anup Patel
0 siblings, 1 reply; 5+ messages in thread
From: Cyan Yang @ 2024-04-09 14:45 UTC (permalink / raw)
To: opensbi
Add sanity checks for per-HART IMSIC pointer in warm init flow, similar to
what the cold init flow does. This will help prevent any misuse of the
data in scratch space.
Signed-off-by: Cyan Yang <cyan.yang@sifive.com>
---
lib/utils/irqchip/imsic.c | 7 ++++++-
1 file changed, 6 insertions(+), 1 deletion(-)
diff --git a/lib/utils/irqchip/imsic.c b/lib/utils/irqchip/imsic.c
index f2a35c6..e6a3657 100644
--- a/lib/utils/irqchip/imsic.c
+++ b/lib/utils/irqchip/imsic.c
@@ -243,10 +243,15 @@ void imsic_local_irqchip_init(void)
int imsic_warm_irqchip_init(void)
{
+ int rc;
struct imsic_data *imsic = imsic_get_data(current_hartid());
/* Sanity checks */
- if (!imsic || !imsic->targets_mmode)
+ rc = imsic_data_check(imsic);
+ if (rc)
+ return rc;
+
+ if (!imsic->targets_mmode)
return SBI_EINVAL;
/* Disable all interrupts */
--
2.39.3 (Apple Git-146)
^ permalink raw reply related [flat|nested] 5+ messages in thread
* [PATCH] lib: utils/irqchip: Add sanity checks to imsic_warm_irqchip_init()
2024-04-09 14:45 [PATCH] lib: utils/irqchip: Add sanity checks to imsic_warm_irqchip_init() Cyan Yang
@ 2024-05-07 12:39 ` Anup Patel
2024-05-08 6:53 ` Cyan Yang
0 siblings, 1 reply; 5+ messages in thread
From: Anup Patel @ 2024-05-07 12:39 UTC (permalink / raw)
To: opensbi
On Tue, Apr 9, 2024 at 8:25?PM Cyan Yang <cyan.yang@sifive.com> wrote:
>
> Add sanity checks for per-HART IMSIC pointer in warm init flow, similar to
> what the cold init flow does. This will help prevent any misuse of the
> data in scratch space.
Can you elaborate on the kind of misuse which can happen ?
>
> Signed-off-by: Cyan Yang <cyan.yang@sifive.com>
> ---
> lib/utils/irqchip/imsic.c | 7 ++++++-
> 1 file changed, 6 insertions(+), 1 deletion(-)
>
> diff --git a/lib/utils/irqchip/imsic.c b/lib/utils/irqchip/imsic.c
> index f2a35c6..e6a3657 100644
> --- a/lib/utils/irqchip/imsic.c
> +++ b/lib/utils/irqchip/imsic.c
> @@ -243,10 +243,15 @@ void imsic_local_irqchip_init(void)
>
> int imsic_warm_irqchip_init(void)
> {
> + int rc;
> struct imsic_data *imsic = imsic_get_data(current_hartid());
>
> /* Sanity checks */
> - if (!imsic || !imsic->targets_mmode)
> + rc = imsic_data_check(imsic);
> + if (rc)
> + return rc;
> +
> + if (!imsic->targets_mmode)
> return SBI_EINVAL;
>
> /* Disable all interrupts */
> --
> 2.39.3 (Apple Git-146)
>
>
> --
> opensbi mailing list
> opensbi at lists.infradead.org
> http://lists.infradead.org/mailman/listinfo/opensbi
Regards,
Anup
^ permalink raw reply [flat|nested] 5+ messages in thread
* [PATCH] lib: utils/irqchip: Add sanity checks to imsic_warm_irqchip_init()
2024-05-07 12:39 ` Anup Patel
@ 2024-05-08 6:53 ` Cyan Yang
2024-05-14 6:07 ` Anup Patel
0 siblings, 1 reply; 5+ messages in thread
From: Cyan Yang @ 2024-05-08 6:53 UTC (permalink / raw)
To: opensbi
On Tue, May 07, 2024 at 06:09:25PM +0530, Anup Patel wrote:
> On Tue, Apr 9, 2024 at 8:25?PM Cyan Yang <cyan.yang@sifive.com> wrote:
> >
> > Add sanity checks for per-HART IMSIC pointer in warm init flow, similar to
> > what the cold init flow does. This will help prevent any misuse of the
> > data in scratch space.
>
> Can you elaborate on the kind of misuse which can happen ?
>
We found a case that if a caller misused this function on a platform without
imsic, the extra check will be needed here since "imsic_get_data()" will not
guarantee to return a null pointer.
It's not a normal case, of course, but since this is not a static function,
having more sanity checks could help prevent the unexpected results.
Regards,
Cyan
> >
> > Signed-off-by: Cyan Yang <cyan.yang@sifive.com>
> > ---
> > lib/utils/irqchip/imsic.c | 7 ++++++-
> > 1 file changed, 6 insertions(+), 1 deletion(-)
> >
> > diff --git a/lib/utils/irqchip/imsic.c b/lib/utils/irqchip/imsic.c
> > index f2a35c6..e6a3657 100644
> > --- a/lib/utils/irqchip/imsic.c
> > +++ b/lib/utils/irqchip/imsic.c
> > @@ -243,10 +243,15 @@ void imsic_local_irqchip_init(void)
> >
> > int imsic_warm_irqchip_init(void)
> > {
> > + int rc;
> > struct imsic_data *imsic = imsic_get_data(current_hartid());
> >
> > /* Sanity checks */
> > - if (!imsic || !imsic->targets_mmode)
> > + rc = imsic_data_check(imsic);
> > + if (rc)
> > + return rc;
> > +
> > + if (!imsic->targets_mmode)
> > return SBI_EINVAL;
> >
> > /* Disable all interrupts */
> > --
> > 2.39.3 (Apple Git-146)
> >
> >
> > --
> > opensbi mailing list
> > opensbi at lists.infradead.org
> > http://lists.infradead.org/mailman/listinfo/opensbi
>
> Regards,
> Anup
^ permalink raw reply [flat|nested] 5+ messages in thread
* [PATCH] lib: utils/irqchip: Add sanity checks to imsic_warm_irqchip_init()
2024-05-08 6:53 ` Cyan Yang
@ 2024-05-14 6:07 ` Anup Patel
2024-05-16 6:54 ` Cyan Yang
0 siblings, 1 reply; 5+ messages in thread
From: Anup Patel @ 2024-05-14 6:07 UTC (permalink / raw)
To: opensbi
On Wed, May 8, 2024 at 12:23?PM Cyan Yang <cyan.yang@sifive.com> wrote:
>
> On Tue, May 07, 2024 at 06:09:25PM +0530, Anup Patel wrote:
> > On Tue, Apr 9, 2024 at 8:25?PM Cyan Yang <cyan.yang@sifive.com> wrote:
> > >
> > > Add sanity checks for per-HART IMSIC pointer in warm init flow, similar to
> > > what the cold init flow does. This will help prevent any misuse of the
> > > data in scratch space.
> >
> > Can you elaborate on the kind of misuse which can happen ?
> >
>
> We found a case that if a caller misused this function on a platform without
> imsic, the extra check will be needed here since "imsic_get_data()" will not
> guarantee to return a null pointer.
>
> It's not a normal case, of course, but since this is not a static function,
> having more sanity checks could help prevent the unexpected results.
If that's the case then imsic_get_data() should return NULL if
imsic_ptr_offset == 0
Regards,
Anup
>
> Regards,
> Cyan
>
> > >
> > > Signed-off-by: Cyan Yang <cyan.yang@sifive.com>
> > > ---
> > > lib/utils/irqchip/imsic.c | 7 ++++++-
> > > 1 file changed, 6 insertions(+), 1 deletion(-)
> > >
> > > diff --git a/lib/utils/irqchip/imsic.c b/lib/utils/irqchip/imsic.c
> > > index f2a35c6..e6a3657 100644
> > > --- a/lib/utils/irqchip/imsic.c
> > > +++ b/lib/utils/irqchip/imsic.c
> > > @@ -243,10 +243,15 @@ void imsic_local_irqchip_init(void)
> > >
> > > int imsic_warm_irqchip_init(void)
> > > {
> > > + int rc;
> > > struct imsic_data *imsic = imsic_get_data(current_hartid());
> > >
> > > /* Sanity checks */
> > > - if (!imsic || !imsic->targets_mmode)
> > > + rc = imsic_data_check(imsic);
> > > + if (rc)
> > > + return rc;
> > > +
> > > + if (!imsic->targets_mmode)
> > > return SBI_EINVAL;
> > >
> > > /* Disable all interrupts */
> > > --
> > > 2.39.3 (Apple Git-146)
> > >
> > >
> > > --
> > > opensbi mailing list
> > > opensbi at lists.infradead.org
> > > http://lists.infradead.org/mailman/listinfo/opensbi
> >
> > Regards,
> > Anup
^ permalink raw reply [flat|nested] 5+ messages in thread
* [PATCH] lib: utils/irqchip: Add sanity checks to imsic_warm_irqchip_init()
2024-05-14 6:07 ` Anup Patel
@ 2024-05-16 6:54 ` Cyan Yang
0 siblings, 0 replies; 5+ messages in thread
From: Cyan Yang @ 2024-05-16 6:54 UTC (permalink / raw)
To: opensbi
On Tue, May 14, 2024 at 2:08?PM Anup Patel <anup@brainfault.org> wrote:
>
> On Wed, May 8, 2024 at 12:23?PM Cyan Yang <cyan.yang@sifive.com> wrote:
> >
> > On Tue, May 07, 2024 at 06:09:25PM +0530, Anup Patel wrote:
> > > On Tue, Apr 9, 2024 at 8:25?PM Cyan Yang <cyan.yang@sifive.com> wrote:
> > > >
> > > > Add sanity checks for per-HART IMSIC pointer in warm init flow, similar to
> > > > what the cold init flow does. This will help prevent any misuse of the
> > > > data in scratch space.
> > >
> > > Can you elaborate on the kind of misuse which can happen ?
> > >
> >
> > We found a case that if a caller misused this function on a platform without
> > imsic, the extra check will be needed here since "imsic_get_data()" will not
> > guarantee to return a null pointer.
> >
> > It's not a normal case, of course, but since this is not a static function,
> > having more sanity checks could help prevent the unexpected results.
>
> If that's the case then imsic_get_data() should return NULL if
> imsic_ptr_offset == 0
Yes, your suggestion will fix the root cause directly. Also, I found that
imsic_get_target_file() has the same issue. So I will submit another patch
to check imsic_ptr_offset in imsic_get_data() and imsic_file_offset in
imsic_get_target_file() and return NULL if the offset is not set.
Thanks,
Cyan
>
> Regards,
> Anup
>
> >
> > Regards,
> > Cyan
> >
> > > >
> > > > Signed-off-by: Cyan Yang <cyan.yang@sifive.com>
> > > > ---
> > > > lib/utils/irqchip/imsic.c | 7 ++++++-
> > > > 1 file changed, 6 insertions(+), 1 deletion(-)
> > > >
> > > > diff --git a/lib/utils/irqchip/imsic.c b/lib/utils/irqchip/imsic.c
> > > > index f2a35c6..e6a3657 100644
> > > > --- a/lib/utils/irqchip/imsic.c
> > > > +++ b/lib/utils/irqchip/imsic.c
> > > > @@ -243,10 +243,15 @@ void imsic_local_irqchip_init(void)
> > > >
> > > > int imsic_warm_irqchip_init(void)
> > > > {
> > > > + int rc;
> > > > struct imsic_data *imsic = imsic_get_data(current_hartid());
> > > >
> > > > /* Sanity checks */
> > > > - if (!imsic || !imsic->targets_mmode)
> > > > + rc = imsic_data_check(imsic);
> > > > + if (rc)
> > > > + return rc;
> > > > +
> > > > + if (!imsic->targets_mmode)
> > > > return SBI_EINVAL;
> > > >
> > > > /* Disable all interrupts */
> > > > --
> > > > 2.39.3 (Apple Git-146)
> > > >
> > > >
> > > > --
> > > > opensbi mailing list
> > > > opensbi at lists.infradead.org
> > > > http://lists.infradead.org/mailman/listinfo/opensbi
> > >
> > > Regards,
> > > Anup
^ permalink raw reply [flat|nested] 5+ messages in thread
end of thread, other threads:[~2024-05-16 6:54 UTC | newest]
Thread overview: 5+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2024-04-09 14:45 [PATCH] lib: utils/irqchip: Add sanity checks to imsic_warm_irqchip_init() Cyan Yang
2024-05-07 12:39 ` Anup Patel
2024-05-08 6:53 ` Cyan Yang
2024-05-14 6:07 ` Anup Patel
2024-05-16 6:54 ` Cyan Yang
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox