| Message ID | 20220513044143.1045728-7-tzungbi@kernel.org (mailing list archive) |
|---|---|
| State | Accepted |
| Commit | ddec8e9e90cea8e8430b04a01adce7fb196d95c6 |
| Headers | show |
| Series | platform/chrome: get rid of BUG_ON() | expand |
On Thu, May 12, 2022 at 9:42 PM Tzung-Bi Shih <tzungbi@kernel.org> wrote: > > In the context, the following conditions are always false: > > - `todo` < 0 > Suppose that EC_SPI_FRAME_START is found at the last byte of transfer. > > In the case, `ptr` == `end` - 1. As a result, `todo` must be 0. > > - `todo` > `ec_dev->din_size` > Suppose that there is no preamble bytes. EC_SPI_FRAME_START is found at > the first byte of transfer. > > In the case, `end` == `ptr` + EC_MSG_PREAMBLE_COUNT. > As a result, `todo` == EC_MSG_PREAMBLE_COUNT - 1. > However, it already checked `ec_dev->din_size` < EC_MSG_PREAMBLE_COUNT at > the beginning of function. > > Drop the unneeded BUG_ON(). > > Signed-off-by: Tzung-Bi Shih <tzungbi@kernel.org> Reviewed-by: Guenter Roeck <groeck@chromium.org> > --- > Changes from v1: > - Separate from the original 6th patch. > - Drop the BUG_ON() instead of returning -EINVAL. > > drivers/platform/chrome/cros_ec_spi.c | 2 -- > 1 file changed, 2 deletions(-) > > diff --git a/drivers/platform/chrome/cros_ec_spi.c b/drivers/platform/chrome/cros_ec_spi.c > index 589f18e9537d..5264615f46af 100644 > --- a/drivers/platform/chrome/cros_ec_spi.c > +++ b/drivers/platform/chrome/cros_ec_spi.c > @@ -237,7 +237,6 @@ static int cros_ec_spi_receive_packet(struct cros_ec_device *ec_dev, > * start of our buffer > */ > todo = end - ++ptr; > - BUG_ON(todo < 0 || todo > ec_dev->din_size); > todo = min(todo, need_len); > memmove(ec_dev->din, ptr, todo); > ptr = ec_dev->din + todo; > @@ -345,7 +344,6 @@ static int cros_ec_spi_receive_response(struct cros_ec_device *ec_dev, > * start of our buffer > */ > todo = end - ++ptr; > - BUG_ON(todo < 0 || todo > ec_dev->din_size); > todo = min(todo, need_len); > memmove(ec_dev->din, ptr, todo); > ptr = ec_dev->din + todo; > -- > 2.36.0.550.gb090851708-goog >
diff --git a/drivers/platform/chrome/cros_ec_spi.c b/drivers/platform/chrome/cros_ec_spi.c index 589f18e9537d..5264615f46af 100644 --- a/drivers/platform/chrome/cros_ec_spi.c +++ b/drivers/platform/chrome/cros_ec_spi.c @@ -237,7 +237,6 @@ static int cros_ec_spi_receive_packet(struct cros_ec_device *ec_dev, * start of our buffer */ todo = end - ++ptr; - BUG_ON(todo < 0 || todo > ec_dev->din_size); todo = min(todo, need_len); memmove(ec_dev->din, ptr, todo); ptr = ec_dev->din + todo; @@ -345,7 +344,6 @@ static int cros_ec_spi_receive_response(struct cros_ec_device *ec_dev, * start of our buffer */ todo = end - ++ptr; - BUG_ON(todo < 0 || todo > ec_dev->din_size); todo = min(todo, need_len); memmove(ec_dev->din, ptr, todo); ptr = ec_dev->din + todo;
In the context, the following conditions are always false: - `todo` < 0 Suppose that EC_SPI_FRAME_START is found at the last byte of transfer. In the case, `ptr` == `end` - 1. As a result, `todo` must be 0. - `todo` > `ec_dev->din_size` Suppose that there is no preamble bytes. EC_SPI_FRAME_START is found at the first byte of transfer. In the case, `end` == `ptr` + EC_MSG_PREAMBLE_COUNT. As a result, `todo` == EC_MSG_PREAMBLE_COUNT - 1. However, it already checked `ec_dev->din_size` < EC_MSG_PREAMBLE_COUNT at the beginning of function. Drop the unneeded BUG_ON(). Signed-off-by: Tzung-Bi Shih <tzungbi@kernel.org> --- Changes from v1: - Separate from the original 6th patch. - Drop the BUG_ON() instead of returning -EINVAL. drivers/platform/chrome/cros_ec_spi.c | 2 -- 1 file changed, 2 deletions(-)