| Message ID | 20190115194625.GA1074@kadam (mailing list archive) |
|---|---|
| State | Accepted, archived |
| Headers | show |
| Series | [v2] clk: ti: Fix error handling in ti_clk_parse_divider_data() | expand |
Quoting Dan Carpenter (2019-01-15 11:46:25) > The ti_clk_parse_divider_data() function is only called from > _get_div_table_from_setup(). That function doesn't look at the return > value but instead looks at the "*table" pointer. In this case, if the > kcalloc() fails then *table is NULL (which means success). It should > instead be an error pointer. > > The ti_clk_parse_divider_data() function has two callers. One checks > for errors and the other doesn't. I have fixed it so now both handle > errors. > > Fixes: 4f6be5655dc9 ("clk: ti: divider: add driver internal API for parsing divider data") > Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com> > --- Applied to clk-fixes I'm going to add Tero's ack to this because it isn't really that different.
On 24/01/2019 21:24, Stephen Boyd wrote: > Quoting Dan Carpenter (2019-01-15 11:46:25) >> The ti_clk_parse_divider_data() function is only called from >> _get_div_table_from_setup(). That function doesn't look at the return >> value but instead looks at the "*table" pointer. In this case, if the >> kcalloc() fails then *table is NULL (which means success). It should >> instead be an error pointer. >> >> The ti_clk_parse_divider_data() function has two callers. One checks >> for errors and the other doesn't. I have fixed it so now both handle >> errors. >> >> Fixes: 4f6be5655dc9 ("clk: ti: divider: add driver internal API for parsing divider data") >> Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com> >> --- > > Applied to clk-fixes > > I'm going to add Tero's ack to this because it isn't really that > different. > Yeah, thats ok. Thanks Stephen. -Tero -- Texas Instruments Finland Oy, Porkkalankatu 22, 00180 Helsinki. Y-tunnus/Business ID: 0615521-4. Kotipaikka/Domicile: Helsinki
diff --git a/drivers/clk/ti/divider.c b/drivers/clk/ti/divider.c index 8d77090ad94a..0241450f3eb3 100644 --- a/drivers/clk/ti/divider.c +++ b/drivers/clk/ti/divider.c @@ -403,8 +403,10 @@ int ti_clk_parse_divider_data(int *div_table, int num_dividers, int max_div, num_dividers = i; tmp = kcalloc(valid_div + 1, sizeof(*tmp), GFP_KERNEL); - if (!tmp) + if (!tmp) { + *table = ERR_PTR(-ENOMEM); return -ENOMEM; + } valid_div = 0; *width = 0; @@ -439,6 +441,7 @@ struct clk_hw *ti_clk_build_component_div(struct ti_clk_divider *setup) { struct clk_omap_divider *div; struct clk_omap_reg *reg; + int ret; if (!setup) return NULL; @@ -458,6 +461,12 @@ struct clk_hw *ti_clk_build_component_div(struct ti_clk_divider *setup) div->flags |= CLK_DIVIDER_POWER_OF_TWO; div->table = _get_div_table_from_setup(setup, &div->width); + if (IS_ERR(div->table)) { + ret = PTR_ERR(div->table); + kfree(div); + return ERR_PTR(ret); + } + div->shift = setup->bit_shift; div->latch = -EINVAL;
The ti_clk_parse_divider_data() function is only called from _get_div_table_from_setup(). That function doesn't look at the return value but instead looks at the "*table" pointer. In this case, if the kcalloc() fails then *table is NULL (which means success). It should instead be an error pointer. The ti_clk_parse_divider_data() function has two callers. One checks for errors and the other doesn't. I have fixed it so now both handle errors. Fixes: 4f6be5655dc9 ("clk: ti: divider: add driver internal API for parsing divider data") Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com> --- v2: I somehow didn't compile the v1 of this patch so it had a couple stupid bugs. I'm very sorry. drivers/clk/ti/divider.c | 11 ++++++++++- 1 file changed, 10 insertions(+), 1 deletion(-)