4.20-stable review patch. If anyone has any objections, please let me know.
------------------
From: Uwe Kleine-König u.kleine-koenig@pengutronix.de
commit a55234dabe1f72cf22f9197980751d37e38ba020 upstream.
Commit cbffaf7aa09e ("can: flexcan: Always use last mailbox for TX") introduced a loop letting i run up to (including) ARRAY_SIZE(regs->mb) and in the body accessed regs->mb[i] which is an out-of-bounds array access that then resulted in an access to an reserved register area.
Later this was changed by commit 0517961ccdf1 ("can: flexcan: Add provision for variable payload size") to iterate a bit differently but still runs one iteration too much resulting to call
flexcan_get_mb(priv, priv->mb_count)
which results in a WARN_ON and then a NULL pointer exception. This only affects devices compatible with "fsl,p1010-flexcan", "fsl,imx53-flexcan", "fsl,imx35-flexcan", "fsl,imx25-flexcan", "fsl,imx28-flexcan", so newer i.MX SoCs are not affected.
Fixes: cbffaf7aa09e ("can: flexcan: Always use last mailbox for TX") Signed-off-by: Uwe Kleine-König u.kleine-koenig@pengutronix.de Cc: linux-stable stable@vger.kernel.org # >= 4.20 Signed-off-by: Marc Kleine-Budde mkl@pengutronix.de Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org
--- drivers/net/can/flexcan.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
--- a/drivers/net/can/flexcan.c +++ b/drivers/net/can/flexcan.c @@ -1004,7 +1004,7 @@ static int flexcan_chip_start(struct net } } else { /* clear and invalidate unused mailboxes first */ - for (i = FLEXCAN_TX_MB_RESERVED_OFF_FIFO; i <= ARRAY_SIZE(regs->mb); i++) { + for (i = FLEXCAN_TX_MB_RESERVED_OFF_FIFO; i < ARRAY_SIZE(regs->mb); i++) { priv->write(FLEXCAN_MB_CODE_RX_INACTIVE, ®s->mb[i].can_ctrl); }