include/uapi/linux/usb/ch9.h | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-)
Since commit df8fc4e934c1 ("kbuild: Enable -fstrict-flex-arrays=3"),
UBSAN_BOUNDS no longer pretends 1-element arrays are unbounded. Walking
bmSublinkSpeedAttr will trigger a warning, so make it a proper flexible
array. Add a union to keep the struct size identical for userspace in
case anything was depending on the old size.
False positive warning was:
UBSAN: array-index-out-of-bounds in drivers/usb/host/xhci-hub.c:231:31 index 1 is out of range for type '__le32 [1]'
for this line of code:
ssp_cap->bmSublinkSpeedAttr[offset++] = cpu_to_le32(attr);
Reported-by: Borislav Petkov <bp@alien8.de>
Closes: https://lore.kernel.org/lkml/2023062945-fencing-pebble-0411@gregkh/
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: "Gustavo A. R. Silva" <gustavoars@kernel.org>
Signed-off-by: Kees Cook <keescook@chromium.org>
---
include/uapi/linux/usb/ch9.h | 6 +++++-
1 file changed, 5 insertions(+), 1 deletion(-)
diff --git a/include/uapi/linux/usb/ch9.h b/include/uapi/linux/usb/ch9.h
index b17e3a21b15f..3ff98c7ba7e3 100644
--- a/include/uapi/linux/usb/ch9.h
+++ b/include/uapi/linux/usb/ch9.h
@@ -981,7 +981,11 @@ struct usb_ssp_cap_descriptor {
#define USB_SSP_MIN_RX_LANE_COUNT (0xf << 8)
#define USB_SSP_MIN_TX_LANE_COUNT (0xf << 12)
__le16 wReserved;
- __le32 bmSublinkSpeedAttr[1]; /* list of sublink speed attrib entries */
+ union {
+ __le32 legacy_padding;
+ /* list of sublink speed attrib entries */
+ __DECLARE_FLEX_ARRAY(__le32, bmSublinkSpeedAttr);
+ };
#define USB_SSP_SUBLINK_SPEED_SSID (0xf) /* sublink speed ID */
#define USB_SSP_SUBLINK_SPEED_LSE (0x3 << 4) /* Lanespeed exponent */
#define USB_SSP_SUBLINK_SPEED_LSE_BPS 0
--
2.34.1
On Thu, Jun 29, 2023 at 12:09:00PM -0700, Kees Cook wrote:
> Since commit df8fc4e934c1 ("kbuild: Enable -fstrict-flex-arrays=3"),
> UBSAN_BOUNDS no longer pretends 1-element arrays are unbounded. Walking
> bmSublinkSpeedAttr will trigger a warning, so make it a proper flexible
> array. Add a union to keep the struct size identical for userspace in
> case anything was depending on the old size.
>
> False positive warning was:
>
> UBSAN: array-index-out-of-bounds in drivers/usb/host/xhci-hub.c:231:31 index 1 is out of range for type '__le32 [1]'
>
> for this line of code:
>
> ssp_cap->bmSublinkSpeedAttr[offset++] = cpu_to_le32(attr);
>
> Reported-by: Borislav Petkov <bp@alien8.de>
> Closes: https://lore.kernel.org/lkml/2023062945-fencing-pebble-0411@gregkh/
> Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
> Cc: "Gustavo A. R. Silva" <gustavoars@kernel.org>
> Signed-off-by: Kees Cook <keescook@chromium.org>
> ---
> include/uapi/linux/usb/ch9.h | 6 +++++-
> 1 file changed, 5 insertions(+), 1 deletion(-)
Thanks for the quick response, I'll queue it up after 6.5-rc1 is out.
Boris, can you test if this fixes the issue you see or not?
greg k-h
On Thu, Jun 29, 2023 at 09:17:23PM +0200, Greg Kroah-Hartman wrote:
> Boris, can you test if this fixes the issue you see or not?
Tested-by: Borislav Petkov (AMD) <bp@alien8.de>
--
Regards/Gruss,
Boris.
https://people.kernel.org/tglx/notes-about-netiquette
On Thu, Jun 29, 2023 at 09:17:23PM +0200, Greg Kroah-Hartman wrote:
> On Thu, Jun 29, 2023 at 12:09:00PM -0700, Kees Cook wrote:
> > Since commit df8fc4e934c1 ("kbuild: Enable -fstrict-flex-arrays=3"),
> > UBSAN_BOUNDS no longer pretends 1-element arrays are unbounded. Walking
> > bmSublinkSpeedAttr will trigger a warning, so make it a proper flexible
> > array. Add a union to keep the struct size identical for userspace in
> > case anything was depending on the old size.
> >
> > False positive warning was:
> >
> > UBSAN: array-index-out-of-bounds in drivers/usb/host/xhci-hub.c:231:31 index 1 is out of range for type '__le32 [1]'
> >
> > for this line of code:
> >
> > ssp_cap->bmSublinkSpeedAttr[offset++] = cpu_to_le32(attr);
> >
> > Reported-by: Borislav Petkov <bp@alien8.de>
> > Closes: https://lore.kernel.org/lkml/2023062945-fencing-pebble-0411@gregkh/
> > Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
> > Cc: "Gustavo A. R. Silva" <gustavoars@kernel.org>
> > Signed-off-by: Kees Cook <keescook@chromium.org>
> > ---
> > include/uapi/linux/usb/ch9.h | 6 +++++-
> > 1 file changed, 5 insertions(+), 1 deletion(-)
>
> Thanks for the quick response, I'll queue it up after 6.5-rc1 is out.
I'm going to send this before -rc1, since we've had another report[1] that
was fixed by this. Given the verification there and Boris's testing, I
think this is good to land. I'll toss it in -next now and send it to
Linus on Friday after making sure there are no more surprises.
-Kees
[1] https://lore.kernel.org/lkml/DA3FEB08-DF39-406B-89CC-9076CFCF597A@kernel.org/
--
Kees Cook
On Wed, Jul 05, 2023 at 02:11:03PM -0700, Kees Cook wrote:
> On Thu, Jun 29, 2023 at 09:17:23PM +0200, Greg Kroah-Hartman wrote:
> > On Thu, Jun 29, 2023 at 12:09:00PM -0700, Kees Cook wrote:
> > > Since commit df8fc4e934c1 ("kbuild: Enable -fstrict-flex-arrays=3"),
> > > UBSAN_BOUNDS no longer pretends 1-element arrays are unbounded. Walking
> > > bmSublinkSpeedAttr will trigger a warning, so make it a proper flexible
> > > array. Add a union to keep the struct size identical for userspace in
> > > case anything was depending on the old size.
> > >
> > > False positive warning was:
> > >
> > > UBSAN: array-index-out-of-bounds in drivers/usb/host/xhci-hub.c:231:31 index 1 is out of range for type '__le32 [1]'
> > >
> > > for this line of code:
> > >
> > > ssp_cap->bmSublinkSpeedAttr[offset++] = cpu_to_le32(attr);
> > >
> > > Reported-by: Borislav Petkov <bp@alien8.de>
> > > Closes: https://lore.kernel.org/lkml/2023062945-fencing-pebble-0411@gregkh/
> > > Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
> > > Cc: "Gustavo A. R. Silva" <gustavoars@kernel.org>
> > > Signed-off-by: Kees Cook <keescook@chromium.org>
> > > ---
> > > include/uapi/linux/usb/ch9.h | 6 +++++-
> > > 1 file changed, 5 insertions(+), 1 deletion(-)
> >
> > Thanks for the quick response, I'll queue it up after 6.5-rc1 is out.
>
> I'm going to send this before -rc1, since we've had another report[1] that
> was fixed by this. Given the verification there and Boris's testing, I
> think this is good to land. I'll toss it in -next now and send it to
> Linus on Friday after making sure there are no more surprises.
>
> -Kees
>
> [1] https://lore.kernel.org/lkml/DA3FEB08-DF39-406B-89CC-9076CFCF597A@kernel.org/
No problem:
Reviewed-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
On 6/29/23 13:09, Kees Cook wrote:
> Since commit df8fc4e934c1 ("kbuild: Enable -fstrict-flex-arrays=3"),
> UBSAN_BOUNDS no longer pretends 1-element arrays are unbounded. Walking
> bmSublinkSpeedAttr will trigger a warning, so make it a proper flexible
> array. Add a union to keep the struct size identical for userspace in
> case anything was depending on the old size.
>
> False positive warning was:
>
> UBSAN: array-index-out-of-bounds in drivers/usb/host/xhci-hub.c:231:31 index 1 is out of range for type '__le32 [1]'
>
> for this line of code:
>
> ssp_cap->bmSublinkSpeedAttr[offset++] = cpu_to_le32(attr);
>
> Reported-by: Borislav Petkov <bp@alien8.de>
> Closes: https://lore.kernel.org/lkml/2023062945-fencing-pebble-0411@gregkh/
> Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
> Cc: "Gustavo A. R. Silva" <gustavoars@kernel.org>
> Signed-off-by: Kees Cook <keescook@chromium.org>
Reviewed-by: Gustavo A. R. Silva <gustavoars@kernel.org>
Thanks!
--
Gustavo
> ---
> include/uapi/linux/usb/ch9.h | 6 +++++-
> 1 file changed, 5 insertions(+), 1 deletion(-)
>
> diff --git a/include/uapi/linux/usb/ch9.h b/include/uapi/linux/usb/ch9.h
> index b17e3a21b15f..3ff98c7ba7e3 100644
> --- a/include/uapi/linux/usb/ch9.h
> +++ b/include/uapi/linux/usb/ch9.h
> @@ -981,7 +981,11 @@ struct usb_ssp_cap_descriptor {
> #define USB_SSP_MIN_RX_LANE_COUNT (0xf << 8)
> #define USB_SSP_MIN_TX_LANE_COUNT (0xf << 12)
> __le16 wReserved;
> - __le32 bmSublinkSpeedAttr[1]; /* list of sublink speed attrib entries */
> + union {
> + __le32 legacy_padding;
> + /* list of sublink speed attrib entries */
> + __DECLARE_FLEX_ARRAY(__le32, bmSublinkSpeedAttr);
> + };
> #define USB_SSP_SUBLINK_SPEED_SSID (0xf) /* sublink speed ID */
> #define USB_SSP_SUBLINK_SPEED_LSE (0x3 << 4) /* Lanespeed exponent */
> #define USB_SSP_SUBLINK_SPEED_LSE_BPS 0
© 2016 - 2026 Red Hat, Inc.