On Mon, Mar 18, 2019 at 08:20:08PM -0600, George Hilliard wrote: > The buffer descriptor setup loop is correct only if it is setting up at > least one bd struct. Besides, there is an error somewhere if > dma_map_sg() returns 0. So add a paranoid check for this condition. > > Signed-off-by: George Hilliard <thirtythreeforty@xxxxxxxxx> > --- > drivers/staging/mt7621-mmc/sd.c | 7 ++++++- > 1 file changed, 6 insertions(+), 1 deletion(-) > > diff --git a/drivers/staging/mt7621-mmc/sd.c b/drivers/staging/mt7621-mmc/sd.c > index c272aa780719..31c5b44bd29f 100644 > --- a/drivers/staging/mt7621-mmc/sd.c > +++ b/drivers/staging/mt7621-mmc/sd.c > @@ -593,7 +593,12 @@ static void msdc_dma_setup(struct msdc_host *host, struct msdc_dma *dma, > struct bd *bd; > u32 j; > > - BUG_ON(sglen > MAX_BD_NUM); /* not support currently */ > + // Shouldn't happen; we configure the mmc host layer > + // based on MAX_BD_NUM: > + BUG_ON(sglen > MAX_BD_NUM); > + // Correct setup below requires at least one bd > + // (and dma_map_sg should not return 0): > + BUG_ON(sglen == 0); Same here, no new BUG_ON(). To quote one kernel developer, "BUG_ON() means that the programmer is lazy" :) thanks, greg k-h
