netdev.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Simon Horman <horms@kernel.org>
To: Ratheesh Kannoth <rkannoth@marvell.com>
Cc: netdev@vger.kernel.org, linux-kernel@vger.kernel.org,
	sgoutham@marvell.com, davem@davemloft.net, edumazet@google.com,
	kuba@kernel.org, pabeni@redhat.com, bcreeley@amd.com,
	sbhatta@marvell.com, gakula@marvell.com, hkelam@marvell.com,
	sumang@marvell.com
Subject: Re: [PATCH net v3] octeontx2-af: Initialize maps.
Date: Thu, 1 Feb 2024 10:31:19 +0100	[thread overview]
Message-ID: <20240201093119.GB514352@kernel.org> (raw)
In-Reply-To: <20240131024118.254758-1-rkannoth@marvell.com>

On Wed, Jan 31, 2024 at 08:11:18AM +0530, Ratheesh Kannoth wrote:
> kmalloc_array() without __GFP_ZERO flag does not initialize
> memory to zero. This causes issues. Use __GFP_ZERO flag for maps and
> bitmap_zalloc() for bimaps.
> 
> Fixes: dd7842878633 ("octeontx2-af: Add new devlink param to configure maximum usable NIX block LFs")
> Signed-off-by: Ratheesh Kannoth <rkannoth@marvell.com>
> ---
> 
> ChangeLogs:
> v2 -> v3: Used GFP_ZERO for normal map arrays
> v1 -> v2: Used bitmap_zalloc() API.
> v0 -> v1: Removed devm_kcalloc()._
> ---
>  .../ethernet/marvell/octeontx2/af/rvu_npc.c   | 26 ++++++++++---------
>  1 file changed, 14 insertions(+), 12 deletions(-)
> 
> diff --git a/drivers/net/ethernet/marvell/octeontx2/af/rvu_npc.c b/drivers/net/ethernet/marvell/octeontx2/af/rvu_npc.c
> index 167145bdcb75..6a8f0efd96a5 100644
> --- a/drivers/net/ethernet/marvell/octeontx2/af/rvu_npc.c
> +++ b/drivers/net/ethernet/marvell/octeontx2/af/rvu_npc.c
> @@ -1850,8 +1850,8 @@ void npc_mcam_rsrcs_deinit(struct rvu *rvu)
>  {
>  	struct npc_mcam *mcam = &rvu->hw->mcam;
> 
> -	kfree(mcam->bmap);
> -	kfree(mcam->bmap_reverse);
> +	bitmap_free(mcam->bmap);
> +	bitmap_free(mcam->bmap_reverse);
>  	kfree(mcam->entry2pfvf_map);
>  	kfree(mcam->cntr2pfvf_map);
>  	kfree(mcam->entry2cntr_map);
> @@ -1904,13 +1904,11 @@ int npc_mcam_rsrcs_init(struct rvu *rvu, int blkaddr)
>  	mcam->pf_offset = mcam->nixlf_offset + nixlf_count;
> 
>  	/* Allocate bitmaps for managing MCAM entries */
> -	mcam->bmap = kmalloc_array(BITS_TO_LONGS(mcam->bmap_entries),
> -				   sizeof(long), GFP_KERNEL);
> +	mcam->bmap = bitmap_zalloc(mcam->bmap_entries, GFP_KERNEL);
>  	if (!mcam->bmap)
>  		return -ENOMEM;
> 
> -	mcam->bmap_reverse = kmalloc_array(BITS_TO_LONGS(mcam->bmap_entries),
> -					   sizeof(long), GFP_KERNEL);
> +	mcam->bmap_reverse = bitmap_zalloc(mcam->bmap_entries, GFP_KERNEL);
>  	if (!mcam->bmap_reverse)
>  		goto free_bmap;
> 
> @@ -1918,7 +1916,8 @@ int npc_mcam_rsrcs_init(struct rvu *rvu, int blkaddr)
> 
>  	/* Alloc memory for saving entry to RVU PFFUNC allocation mapping */
>  	mcam->entry2pfvf_map = kmalloc_array(mcam->bmap_entries,
> -					     sizeof(u16), GFP_KERNEL);
> +					     sizeof(u16),
> +					     GFP_KERNEL | __GFP_ZERO);

Hi Ratheesh,

The use of bitmap_zalloc()/bitmap_free() looks good to me.
But for the kmalloc_array(..., GFP_KERNEL | __GFP_ZERO) cases
I think kcalloc() is the way to go.

>  	if (!mcam->entry2pfvf_map)
>  		goto free_bmap_reverse;
> 
> @@ -1942,7 +1941,8 @@ int npc_mcam_rsrcs_init(struct rvu *rvu, int blkaddr)
>  		goto free_entry_map;
> 
>  	mcam->cntr2pfvf_map = kmalloc_array(mcam->counters.max,
> -					    sizeof(u16), GFP_KERNEL);
> +					    sizeof(u16),
> +					    GFP_KERNEL | __GFP_ZERO);
>  	if (!mcam->cntr2pfvf_map)
>  		goto free_cntr_bmap;
> 
> @@ -1950,12 +1950,14 @@ int npc_mcam_rsrcs_init(struct rvu *rvu, int blkaddr)
>  	 * counter's reference count.
>  	 */
>  	mcam->entry2cntr_map = kmalloc_array(mcam->bmap_entries,
> -					     sizeof(u16), GFP_KERNEL);
> +					     sizeof(u16),
> +					     GFP_KERNEL | __GFP_ZERO);
>  	if (!mcam->entry2cntr_map)
>  		goto free_cntr_map;
> 
>  	mcam->cntr_refcnt = kmalloc_array(mcam->counters.max,
> -					  sizeof(u16), GFP_KERNEL);
> +					  sizeof(u16),
> +					  GFP_KERNEL | __GFP_ZERO);
>  	if (!mcam->cntr_refcnt)
>  		goto free_entry_cntr_map;
> 
> @@ -1988,9 +1990,9 @@ int npc_mcam_rsrcs_init(struct rvu *rvu, int blkaddr)
>  free_entry_map:
>  	kfree(mcam->entry2pfvf_map);
>  free_bmap_reverse:
> -	kfree(mcam->bmap_reverse);
> +	bitmap_free(mcam->bmap_reverse);
>  free_bmap:
> -	kfree(mcam->bmap);
> +	bitmap_free(mcam->bmap);
> 
>  	return -ENOMEM;
>  }
> --
> 2.25.1
> 

-- 
pw-bot: changes-requested

  parent reply	other threads:[~2024-02-01  9:31 UTC|newest]

Thread overview: 6+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-01-31  2:41 [PATCH net v3] octeontx2-af: Initialize maps Ratheesh Kannoth
2024-01-31 17:30 ` Brett Creeley
2024-02-01  9:31 ` Simon Horman [this message]
2024-02-01  9:35   ` [EXT] " Ratheesh Kannoth
2024-02-01 16:27     ` Jakub Kicinski
2024-02-02  7:32       ` Ratheesh Kannoth

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20240201093119.GB514352@kernel.org \
    --to=horms@kernel.org \
    --cc=bcreeley@amd.com \
    --cc=davem@davemloft.net \
    --cc=edumazet@google.com \
    --cc=gakula@marvell.com \
    --cc=hkelam@marvell.com \
    --cc=kuba@kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=netdev@vger.kernel.org \
    --cc=pabeni@redhat.com \
    --cc=rkannoth@marvell.com \
    --cc=sbhatta@marvell.com \
    --cc=sgoutham@marvell.com \
    --cc=sumang@marvell.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).