From: Stephen Gallimore <[email protected]> Clocks implementing the get_parent() op may return an invalid parent index if the hardware is in an undefined state when the clock is created. However the calls of get_parent() in clk.c do not check that the returned index is in range before using it to dereference the clock's parents[] and parent_names[] arrays.
This patch adds checks against the number of clock parents to prevent an incorrect access to the clock state. This does not otherwise change the use of get_parent() and will leave a clock, with an undefined parent, orphaned until a valid parent is set through the clock API. Signed-off-by: Stephen Gallimore <[email protected]> --- Notes: There are two clocks, clk-mux and the OMAP2 dpll, that currently return -EINVAL, which as the return type of get_parent() is a u8 will result in an index value of 234. Mike is aware of this and was already thinking about changing the get_parent() prototype and usage. That should not remove the need for the tests added by this patch, but may require them to be modified later depending on exactly what Mike decides to do. drivers/clk/clk.c | 7 +++++-- 1 files changed, 5 insertions(+), 2 deletions(-) diff --git a/drivers/clk/clk.c b/drivers/clk/clk.c index 54a191c..b5032d0 100644 --- a/drivers/clk/clk.c +++ b/drivers/clk/clk.c @@ -1309,6 +1309,8 @@ static struct clk *__clk_init_parent(struct clk *clk) */ index = clk->ops->get_parent(clk->hw); + if (index >= clk->num_parents) + goto out; if (!clk->parents) clk->parents = @@ -1630,8 +1632,9 @@ int __clk_init(struct device *dev, struct clk *clk) hlist_for_each_entry_safe(orphan, tmp2, &clk_orphan_list, child_node) { if (orphan->ops->get_parent) { i = orphan->ops->get_parent(orphan->hw); - if (!strcmp(clk->name, orphan->parent_names[i])) - __clk_reparent(orphan, clk); + if (i < orphan->num_parents) + if (!strcmp(clk->name, orphan->parent_names[i])) + __clk_reparent(orphan, clk); continue; } -- 1.7.6.5 -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [email protected] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
