Re: [PATCH] Staging: rtl8723bs: core: Replace rtw_malloc with kmalloc

From: Bhanusree Mahesh
Date: Fri Apr 19 2019 - 15:00:06 EST


On Fri, 19 Apr 2019 at 16:52, Greg Kroah-Hartman
<gregkh@xxxxxxxxxxxxxxxxxxx> wrote:
>
> On Fri, Apr 19, 2019 at 03:52:43PM +0530, Bhanusree Pola wrote:
> > Replace rtw_malloc with kmalloc to make code OS independent
> > use kmalloc second argument as GFP_ATOMIC as these are called by functions
> > that holds lock.
> >
> > Signed-off-by: Bhanusree Pola <bhanusreemahesh@xxxxxxxxx>
> > ---
> > drivers/staging/rtl8723bs/core/rtw_ap.c | 4 ++--
> > 1 file changed, 2 insertions(+), 2 deletions(-)
> >
> > diff --git a/drivers/staging/rtl8723bs/core/rtw_ap.c b/drivers/staging/rtl8723bs/core/rtw_ap.c
> > index 18fabf5..6a6683c 100644
> > --- a/drivers/staging/rtl8723bs/core/rtw_ap.c
> > +++ b/drivers/staging/rtl8723bs/core/rtw_ap.c
> > @@ -128,7 +128,7 @@ static void update_BCNTIM(struct adapter *padapter)
> > }
> >
> > if (remainder_ielen > 0) {
> > - pbackup_remainder_ie = rtw_malloc(remainder_ielen);
> > + pbackup_remainder_ie = kmalloc(remainder_ielen,GFP_ATOMIC);
>
> Always run checkpatch.pl on your patches so you do nto get grumpy
> maintainers telling you to run checkpatch.pl on your code :)

>
> Why not fix up all of the callers of this function?

There are many callers of this function. Should I send the whole thing
as of patch series?

>And are you sure
> that GFP_ATOMIC is the correct thing to do here?

yes, because it is called by the function which holds the lock.
correct me if I'm wrong.

>
> thanks,
>
> greg k-h