2014-04-18 07:26:14

by Alexey Khoroshilov

Subject: [PATCH] [SCSI] bfa: allocate memory with GFP_ATOMIC in spinlock context

bfa_fcb_pbc_vport_create() is called only from bfa_fcs_pbc_vport_init(),
that is called only from bfad_drv_start() with bfad_lock spinlock held.
So the patch replaces GFP_KERNEL with GFP_ATOMIC to avoid
sleeping in atomic spinlock context.

Found by Linux Driver Verification project (linuxtesting.org).

Signed-off-by: Alexey Khoroshilov <[email protected]>
---
drivers/scsi/bfa/bfad.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/scsi/bfa/bfad.c b/drivers/scsi/bfa/bfad.c
index cc0fbcdc5192..8f46d1b72fb8 100644
--- a/drivers/scsi/bfa/bfad.c
+++ b/drivers/scsi/bfa/bfad.c
@@ -652,7 +652,7 @@ bfad_vport_create(struct bfad_s *bfad, u16 vf_id,
unsigned long flags;
struct completion fcomp;

- vport = kzalloc(sizeof(struct bfad_vport_s), GFP_KERNEL);
+ vport = kzalloc(sizeof(struct bfad_vport_s), GFP_ATOMIC);
if (!vport) {
rc = BFA_STATUS_ENOMEM;
goto ext;
--
1.8.3.2

2014-04-18 07:50:20

by Alexey Khoroshilov

Subject: Re: [PATCH] [SCSI] bfa: allocate memory with GFP_ATOMIC in spinlock context

Please ignore this patch, it will resend a correct version of it.

On 18.04.2014 00:25, Alexey Khoroshilov wrote:
> bfa_fcb_pbc_vport_create() is called only from bfa_fcs_pbc_vport_init(),
> that is called only from bfad_drv_start() with bfad_lock spinlock held.
> So the patch replaces GFP_KERNEL with GFP_ATOMIC to avoid
> sleeping in atomic spinlock context.
>
> Found by Linux Driver Verification project (linuxtesting.org).
>
> Signed-off-by: Alexey Khoroshilov <[email protected]>
> ---
> drivers/scsi/bfa/bfad.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/drivers/scsi/bfa/bfad.c b/drivers/scsi/bfa/bfad.c
> index cc0fbcdc5192..8f46d1b72fb8 100644
> --- a/drivers/scsi/bfa/bfad.c
> +++ b/drivers/scsi/bfa/bfad.c
> @@ -652,7 +652,7 @@ bfad_vport_create(struct bfad_s *bfad, u16 vf_id,
> unsigned long flags;
> struct completion fcomp;
>
> - vport = kzalloc(sizeof(struct bfad_vport_s), GFP_KERNEL);
> + vport = kzalloc(sizeof(struct bfad_vport_s), GFP_ATOMIC);
> if (!vport) {
> rc = BFA_STATUS_ENOMEM;
> goto ext;