Re: [PATCH 2/2] powerpc/spufs: stop using access_ok

2020-04-29 Thread Christophe Leroy
Le 30/04/2020 à 07:39, Christoph Hellwig a écrit : On Thu, Apr 30, 2020 at 08:39:00AM +0800, Jeremy Kerr wrote: Hi Christophe, Just use the proper non __-prefixed get/put_user variants where that is not done yet. But it means you are doing the access_ok() check everytime, which is what is

Re: [PATCH 2/2] powerpc/spufs: stop using access_ok

2020-04-29 Thread Christoph Hellwig
On Thu, Apr 30, 2020 at 08:39:00AM +0800, Jeremy Kerr wrote: > Hi Christophe, > > > > Just use the proper non __-prefixed get/put_user variants where > > > that is not done yet. > > > > But it means you are doing the access_ok() check everytime, which is > > what is to be avoided by doing the ac

Re: [PATCH 2/2] powerpc/spufs: stop using access_ok

2020-04-29 Thread Jeremy Kerr
Hi Christophe, > > Just use the proper non __-prefixed get/put_user variants where > > that is not done yet. > > But it means you are doing the access_ok() check everytime, which is > what is to be avoided by doing the access_ok() once then using the > __-prefixed variant. 5 out of 8 of these

Re: [PATCH 2/2] powerpc/spufs: stop using access_ok

2020-04-29 Thread Christophe Leroy
Le 29/04/2020 à 09:03, Jeremy Kerr a écrit : From: Christoph Hellwig Just use the proper non __-prefixed get/put_user variants where that is not done yet. But it means you are doing the access_ok() check everytime, which is what is to be avoided by doing the access_ok() once then using th

[PATCH 2/2] powerpc/spufs: stop using access_ok

2020-04-29 Thread Jeremy Kerr
From: Christoph Hellwig Just use the proper non __-prefixed get/put_user variants where that is not done yet. Signed-off-by: Christoph Hellwig Signed-off-by: Jeremy Kerr --- arch/powerpc/platforms/cell/spufs/file.c | 42 +--- 1 file changed, 8 insertions(+), 34 deletions(-