public inbox for netdev@vger.kernel.org
 help / color / mirror / Atom feed
From: Nathan Chancellor <natechancellor@gmail.com>
To: Ido Schimmel <idosch@idosch.org>
Cc: netdev@vger.kernel.org, davem@davemloft.net,
	jakub.kicinski@netronome.com, jiri@mellanox.com,
	petrm@mellanox.com, mlxsw@mellanox.com,
	Ido Schimmel <idosch@mellanox.com>
Subject: Re: [PATCH net-next 2/2] mlxsw: spectrum_buffers: Calculate the size of the main pool
Date: Tue, 29 Oct 2019 20:31:54 -0700	[thread overview]
Message-ID: <20191030033154.GA43266@ubuntu-m2-xlarge-x86> (raw)
In-Reply-To: <20191023060500.19709-3-idosch@idosch.org>

On Wed, Oct 23, 2019 at 09:05:00AM +0300, Ido Schimmel wrote:
> From: Petr Machata <petrm@mellanox.com>
> 
> Instead of hard-coding the size of the largest pool, calculate it from the
> reported guaranteed shared buffer size and sizes of other pools (currently
> only the CPU port pool).
> 
> Signed-off-by: Petr Machata <petrm@mellanox.com>
> Signed-off-by: Ido Schimmel <idosch@mellanox.com>
> ---
>  .../mellanox/mlxsw/spectrum_buffers.c         | 46 ++++++++++++++-----
>  1 file changed, 34 insertions(+), 12 deletions(-)
> 
> diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_buffers.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_buffers.c
> index 637151682cf2..5fd9a72c8471 100644
> --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_buffers.c
> +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_buffers.c
> @@ -35,6 +35,7 @@ struct mlxsw_sp_sb_cm {
>  };
>  
>  #define MLXSW_SP_SB_INFI -1U
> +#define MLXSW_SP_SB_REST -2U
>  
>  struct mlxsw_sp_sb_pm {
>  	u32 min_buff;
> @@ -421,19 +422,16 @@ static void mlxsw_sp_sb_ports_fini(struct mlxsw_sp *mlxsw_sp)
>  		.freeze_size = _freeze_size,				\
>  	}
>  
> -#define MLXSW_SP1_SB_PR_INGRESS_SIZE	13768608
> -#define MLXSW_SP1_SB_PR_EGRESS_SIZE	13768608
>  #define MLXSW_SP1_SB_PR_CPU_SIZE	(256 * 1000)
>  
>  /* Order according to mlxsw_sp1_sb_pool_dess */
>  static const struct mlxsw_sp_sb_pr mlxsw_sp1_sb_prs[] = {
> -	MLXSW_SP_SB_PR(MLXSW_REG_SBPR_MODE_DYNAMIC,
> -		       MLXSW_SP1_SB_PR_INGRESS_SIZE),
> +	MLXSW_SP_SB_PR(MLXSW_REG_SBPR_MODE_DYNAMIC, MLXSW_SP_SB_REST),
>  	MLXSW_SP_SB_PR(MLXSW_REG_SBPR_MODE_DYNAMIC, 0),
>  	MLXSW_SP_SB_PR(MLXSW_REG_SBPR_MODE_DYNAMIC, 0),
>  	MLXSW_SP_SB_PR(MLXSW_REG_SBPR_MODE_DYNAMIC, 0),
> -	MLXSW_SP_SB_PR_EXT(MLXSW_REG_SBPR_MODE_DYNAMIC,
> -			   MLXSW_SP1_SB_PR_EGRESS_SIZE, true, false),
> +	MLXSW_SP_SB_PR_EXT(MLXSW_REG_SBPR_MODE_DYNAMIC, MLXSW_SP_SB_REST,
> +			   true, false),
>  	MLXSW_SP_SB_PR(MLXSW_REG_SBPR_MODE_DYNAMIC, 0),
>  	MLXSW_SP_SB_PR(MLXSW_REG_SBPR_MODE_DYNAMIC, 0),
>  	MLXSW_SP_SB_PR(MLXSW_REG_SBPR_MODE_DYNAMIC, 0),
> @@ -445,19 +443,16 @@ static const struct mlxsw_sp_sb_pr mlxsw_sp1_sb_prs[] = {
>  			   MLXSW_SP1_SB_PR_CPU_SIZE, true, false),
>  };
>  
> -#define MLXSW_SP2_SB_PR_INGRESS_SIZE	34084800
> -#define MLXSW_SP2_SB_PR_EGRESS_SIZE	34084800
>  #define MLXSW_SP2_SB_PR_CPU_SIZE	(256 * 1000)
>  
>  /* Order according to mlxsw_sp2_sb_pool_dess */
>  static const struct mlxsw_sp_sb_pr mlxsw_sp2_sb_prs[] = {
> -	MLXSW_SP_SB_PR(MLXSW_REG_SBPR_MODE_DYNAMIC,
> -		       MLXSW_SP2_SB_PR_INGRESS_SIZE),
> +	MLXSW_SP_SB_PR(MLXSW_REG_SBPR_MODE_DYNAMIC, MLXSW_SP_SB_REST),
>  	MLXSW_SP_SB_PR(MLXSW_REG_SBPR_MODE_STATIC, 0),
>  	MLXSW_SP_SB_PR(MLXSW_REG_SBPR_MODE_STATIC, 0),
>  	MLXSW_SP_SB_PR(MLXSW_REG_SBPR_MODE_STATIC, 0),
> -	MLXSW_SP_SB_PR_EXT(MLXSW_REG_SBPR_MODE_DYNAMIC,
> -			   MLXSW_SP2_SB_PR_EGRESS_SIZE, true, false),
> +	MLXSW_SP_SB_PR_EXT(MLXSW_REG_SBPR_MODE_DYNAMIC, MLXSW_SP_SB_REST,
> +			   true, false),
>  	MLXSW_SP_SB_PR(MLXSW_REG_SBPR_MODE_STATIC, 0),
>  	MLXSW_SP_SB_PR(MLXSW_REG_SBPR_MODE_STATIC, 0),
>  	MLXSW_SP_SB_PR(MLXSW_REG_SBPR_MODE_STATIC, 0),
> @@ -471,11 +466,33 @@ static const struct mlxsw_sp_sb_pr mlxsw_sp2_sb_prs[] = {
>  
>  static int mlxsw_sp_sb_prs_init(struct mlxsw_sp *mlxsw_sp,
>  				const struct mlxsw_sp_sb_pr *prs,
> +				const struct mlxsw_sp_sb_pool_des *pool_dess,
>  				size_t prs_len)
>  {
> +	/* Round down, unlike mlxsw_sp_bytes_cells(). */
> +	u32 sb_cells = mlxsw_sp->sb->sb_size / mlxsw_sp->sb->cell_size;

This causes a link time error on arm32. It can be simply reproduced with
the following configs + multi_v7_defconfig:

CONFIG_MLXSW_CORE=y
CONFIG_MLXSW_PCI=y
CONFIG_NET_SWITCHDEV=y
CONFIG_VLAN_8021Q=y
CONFIG_MLXSW_SPECTRUM=y

arm-linux-gnueabi-ld: drivers/net/ethernet/mellanox/mlxsw/spectrum_buffers.o: in function `mlxsw_sp_buffers_init':
spectrum_buffers.c:(.text+0x1c88): undefined reference to `__aeabi_uldivmod'

It can be solved by something like this but I am not sure if that is
proper or not since div_u64 returns a u64, which would implicitly get
converted to u32. I can submit it as a formal patch if needed but I
figured I would reach out first in case you want to go in a different
direction.

diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_buffers.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_buffers.c
index 33a978af80d6..968f0902e4fe 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_buffers.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_buffers.c
@@ -470,7 +470,7 @@ static int mlxsw_sp_sb_prs_init(struct mlxsw_sp *mlxsw_sp,
 				size_t prs_len)
 {
 	/* Round down, unlike mlxsw_sp_bytes_cells(). */
-	u32 sb_cells = mlxsw_sp->sb->sb_size / mlxsw_sp->sb->cell_size;
+	u32 sb_cells = div_u64(mlxsw_sp->sb->sb_size, mlxsw_sp->sb->cell_size);
 	u32 rest_cells[2] = {sb_cells, sb_cells};
 	int i;
 	int err;

  reply	other threads:[~2019-10-30  3:32 UTC|newest]

Thread overview: 6+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-10-23  6:04 [PATCH net-next 0/2] mlxsw: Update main pool computation and pool size limits Ido Schimmel
2019-10-23  6:04 ` [PATCH net-next 1/2] mlxsw: spectrum: Use guaranteed buffer size as pool size limit Ido Schimmel
2019-10-23  6:05 ` [PATCH net-next 2/2] mlxsw: spectrum_buffers: Calculate the size of the main pool Ido Schimmel
2019-10-30  3:31   ` Nathan Chancellor [this message]
2019-10-30  7:57     ` Ido Schimmel
2019-10-24  4:31 ` [PATCH net-next 0/2] mlxsw: Update main pool computation and pool size limits David Miller

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=20191030033154.GA43266@ubuntu-m2-xlarge-x86 \
    --to=natechancellor@gmail.com \
    --cc=davem@davemloft.net \
    --cc=idosch@idosch.org \
    --cc=idosch@mellanox.com \
    --cc=jakub.kicinski@netronome.com \
    --cc=jiri@mellanox.com \
    --cc=mlxsw@mellanox.com \
    --cc=netdev@vger.kernel.org \
    --cc=petrm@mellanox.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