On Fri, Apr 06, 2018 at 01:52:47PM +0200, Sergio Paracuellos wrote:
> On Fri, Apr 6, 2018 at 1:07 PM, Dan Carpenter <dan.carpenter@xxxxxxxxxx> wrote:
> > On Fri, Apr 06, 2018 at 12:38:16PM +0200, Sergio Paracuellos wrote:
> >> This commit removes KS_WLAN_MEM_FLAG definition which is
> >> hiding GFP_ATOMIC along the code. It is better to use directly
> >> GFP_ATOMIC in kmalloc's because this it is easier to read.
> >>
> >> Signed-off-by: Sergio Paracuellos <sergio.paracuellos@xxxxxxxxx>
> >> ---
> >> drivers/staging/ks7010/ks_hostif.c | 6 ++----
> >> 1 file changed, 2 insertions(+), 4 deletions(-)
> >>
> >> diff --git a/drivers/staging/ks7010/ks_hostif.c b/drivers/staging/ks7010/ks_hostif.c
> >> index e336d25..208c91e 100644
> >> --- a/drivers/staging/ks7010/ks_hostif.c
> >> +++ b/drivers/staging/ks7010/ks_hostif.c
> >> @@ -38,8 +38,6 @@ static inline unsigned int cnt_smeqbody(struct ks_wlan_private *priv)
> >> return (sme_cnt + SME_EVENT_BUFF_SIZE) % SME_EVENT_BUFF_SIZE;
> >> }
> >>
> >> -#define KS_WLAN_MEM_FLAG (GFP_ATOMIC)
> >> -
> >> static inline u8 get_byte(struct ks_wlan_private *priv)
> >> {
> >> u8 data;
> >> @@ -1044,7 +1042,7 @@ static void *hostif_generic_request(size_t size, int event)
> >> {
> >> struct hostif_hdr *p;
> >>
> >> - p = kzalloc(hif_align_size(size), KS_WLAN_MEM_FLAG);
> >> + p = kzalloc(hif_align_size(size), GFP_ATOMIC);
> >
> > If there were a ton of these, then I probably would be OK with a
> > straight sed like this. But there are only two places. Probably
> > neither allocation needs to be atomic. Could you do some research to
> > see if we can just use GFP_KERNEL here?
>
> I think is correct as it is because it is in a tasklet context so we
> cannot sleep.
>
I see that the original author thought some of this could be called in
the IRQ handler as well. Nevermind then. This patch is fine.
regards,
dan carpenter
_______________________________________________
devel mailing list
devel@xxxxxxxxxxxxxxxxxxxxxx
http://driverdev.linuxdriverproject.org/mailman/listinfo/driverdev-devel
