2021-12-20 13:56:31

by Jiasheng Jiang

Subject: [PATCH] sfc: Check null pointer of rx_queue->page_ring

Because of the possible failure of the kcalloc, it should be better to
set rx_queue->page_ptr_mask to 0 when it happens in order to maintain
the consistency.

Fixes: 5a6681e22c14 ("sfc: separate out SFC4000 ("Falcon") support into new sfc-falcon driver")
Signed-off-by: Jiasheng Jiang <[email protected]>
---
drivers/net/ethernet/sfc/rx_common.c | 5 ++++-
1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/sfc/rx_common.c b/drivers/net/ethernet/sfc/rx_common.c
index 68fc7d317693..0983abc0cc5f 100644
--- a/drivers/net/ethernet/sfc/rx_common.c
+++ b/drivers/net/ethernet/sfc/rx_common.c
@@ -150,7 +150,10 @@ static void efx_init_rx_recycle_ring(struct efx_rx_queue *rx_queue)
efx->rx_bufs_per_page);
rx_queue->page_ring = kcalloc(page_ring_size,
sizeof(*rx_queue->page_ring), GFP_KERNEL);
- rx_queue->page_ptr_mask = page_ring_size - 1;
+ if (!rx_queue->page_ring)
+ rx_queue->page_ptr_mask = 0;
+ else
+ rx_queue->page_ptr_mask = page_ring_size - 1;
}

static void efx_fini_rx_recycle_ring(struct efx_rx_queue *rx_queue)
--
2.25.1

2021-12-20 14:16:35

by Greg Kroah-Hartman

Subject: Re: [PATCH] sfc: Check null pointer of rx_queue->page_ring

On Mon, Dec 20, 2021 at 09:56:03PM +0800, Jiasheng Jiang wrote:
> Because of the possible failure of the kcalloc, it should be better to
> set rx_queue->page_ptr_mask to 0 when it happens in order to maintain
> the consistency.
>
> Fixes: 5a6681e22c14 ("sfc: separate out SFC4000 ("Falcon") support into new sfc-falcon driver")
> Signed-off-by: Jiasheng Jiang <[email protected]>
> ---
> drivers/net/ethernet/sfc/rx_common.c | 5 ++++-
> 1 file changed, 4 insertions(+), 1 deletion(-)
>
> diff --git a/drivers/net/ethernet/sfc/rx_common.c b/drivers/net/ethernet/sfc/rx_common.c
> index 68fc7d317693..0983abc0cc5f 100644
> --- a/drivers/net/ethernet/sfc/rx_common.c
> +++ b/drivers/net/ethernet/sfc/rx_common.c
> @@ -150,7 +150,10 @@ static void efx_init_rx_recycle_ring(struct efx_rx_queue *rx_queue)
> efx->rx_bufs_per_page);
> rx_queue->page_ring = kcalloc(page_ring_size,
> sizeof(*rx_queue->page_ring), GFP_KERNEL);
> - rx_queue->page_ptr_mask = page_ring_size - 1;
> + if (!rx_queue->page_ring)
> + rx_queue->page_ptr_mask = 0;
> + else
> + rx_queue->page_ptr_mask = page_ring_size - 1;
> }

Why not return an error?

2021-12-20 14:55:41

by Jiasheng Jiang

Subject: Re: Re: [PATCH] sfc: Check null pointer of rx_queue->page_ring

On Mon, Dec 20, 2021 at 10:16:28PM +0800, Greg KH wrote:
> Why not return an error?

Because I have received the mail from Martin Habets that telling me
it doesn't need to return error code.
Here is the mail.
https://lore.kernel.org/lkml/[email protected]/
On Sun, Dec 19, 2021 at 05:29:48PM +0800, Martin Habets wrote:
> Your predicate is wrong. The code that uses rx_queue->page_ring
> can deal with it being NULL.
> The only thing you might want to do is set rx_queue->page_ptr_mask
> to 0.

Jiasheng

2021-12-22 13:04:04

by Martin Habets

Subject: Re: [PATCH] sfc: Check null pointer of rx_queue->page_ring

On Mon, Dec 20, 2021 at 09:56:03PM +0800, Jiasheng Jiang wrote:
> Because of the possible failure of the kcalloc, it should be better to
> set rx_queue->page_ptr_mask to 0 when it happens in order to maintain
> the consistency.
>
> Fixes: 5a6681e22c14 ("sfc: separate out SFC4000 ("Falcon") support into new sfc-falcon driver")
> Signed-off-by: Jiasheng Jiang <[email protected]>

Acked-by: Martin Habets <[email protected]>

> ---
> drivers/net/ethernet/sfc/rx_common.c | 5 ++++-
> 1 file changed, 4 insertions(+), 1 deletion(-)
>
> diff --git a/drivers/net/ethernet/sfc/rx_common.c b/drivers/net/ethernet/sfc/rx_common.c
> index 68fc7d317693..0983abc0cc5f 100644
> --- a/drivers/net/ethernet/sfc/rx_common.c
> +++ b/drivers/net/ethernet/sfc/rx_common.c
> @@ -150,7 +150,10 @@ static void efx_init_rx_recycle_ring(struct efx_rx_queue *rx_queue)
> efx->rx_bufs_per_page);
> rx_queue->page_ring = kcalloc(page_ring_size,
> sizeof(*rx_queue->page_ring), GFP_KERNEL);
> - rx_queue->page_ptr_mask = page_ring_size - 1;
> + if (!rx_queue->page_ring)
> + rx_queue->page_ptr_mask = 0;
> + else
> + rx_queue->page_ptr_mask = page_ring_size - 1;
> }
>
> static void efx_fini_rx_recycle_ring(struct efx_rx_queue *rx_queue)
> --
> 2.25.1

2021-12-22 20:26:21

by Jakub Kicinski

Subject: Re: [PATCH] sfc: Check null pointer of rx_queue->page_ring

On Wed, 22 Dec 2021 13:03:56 +0000 Martin Habets wrote:
> On Mon, Dec 20, 2021 at 09:56:03PM +0800, Jiasheng Jiang wrote:
> > Because of the possible failure of the kcalloc, it should be better to
> > set rx_queue->page_ptr_mask to 0 when it happens in order to maintain
> > the consistency.
> >
> > Fixes: 5a6681e22c14 ("sfc: separate out SFC4000 ("Falcon") support into new sfc-falcon driver")
> > Signed-off-by: Jiasheng Jiang <[email protected]>
>
> Acked-by: Martin Habets <[email protected]>

Applied, thanks!