linux-wireless.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Lorenzo Bianconi <lorenzo@kernel.org>
To: Peter Chiu <chui-hao.chiu@mediatek.com>
Cc: Felix Fietkau <nbd@nbd.name>,
	Lorenzo Bianconi <lorenzo.bianconi@redhat.com>,
	Ryder Lee <ryder.Lee@mediatek.com>,
	Evelyn Tsai <evelyn.tsai@mediatek.com>,
	Shayne Chen <shayne.chen@mediatek.com>,
	linux-wireless@vger.kernel.org,
	linux-mediatek@lists.infradead.org
Subject: Re: [PATCH 2/2] wifi: mt76: mt7915: wed: enable red per-band token drop
Date: Thu, 5 Jan 2023 13:20:53 +0100	[thread overview]
Message-ID: <Y7bApSqBT2LXdBhu@lore-desk> (raw)
In-Reply-To: <20230105114711.31430-2-chui-hao.chiu@mediatek.com>

[-- Attachment #1: Type: text/plain, Size: 4007 bytes --]

> Enable RED to limit the number of token used by each band. If single band
> uses too many tokens, it would hurt the throughput of the other bands.The
> software path can solve this problem by AQL so enable RED for HW path only.
> 
> Reviewed-by: Shayne Chen <shayne.chen@mediatek.com>
> Signed-off-by: Peter Chiu <chui-hao.chiu@mediatek.com>
> ---
>  .../wireless/mediatek/mt76/mt76_connac_mcu.h  |  1 +
>  .../net/wireless/mediatek/mt76/mt7915/mcu.c   | 54 ++++++++++++++++++-
>  .../net/wireless/mediatek/mt76/mt7915/mcu.h   |  1 +
>  3 files changed, 54 insertions(+), 2 deletions(-)
> 
> diff --git a/drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.h b/drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.h
> index f1e942b9a887..855a434ecf54 100644
> --- a/drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.h
> +++ b/drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.h
> @@ -1171,6 +1171,7 @@ enum {
>  	MCU_EXT_CMD_GET_MIB_INFO = 0x5a,
>  	MCU_EXT_CMD_TXDPD_CAL = 0x60,
>  	MCU_EXT_CMD_CAL_CACHE = 0x67,
> +	MCU_EXT_CMD_RED_ENABLE = 0x68,
>  	MCU_EXT_CMD_SET_RADAR_TH = 0x7c,
>  	MCU_EXT_CMD_SET_RDD_PATTERN = 0x7d,
>  	MCU_EXT_CMD_MWDS_SUPPORT = 0x80,
> diff --git a/drivers/net/wireless/mediatek/mt76/mt7915/mcu.c b/drivers/net/wireless/mediatek/mt76/mt7915/mcu.c
> index 37a3c1f082d9..a7bc2043a0db 100644
> --- a/drivers/net/wireless/mediatek/mt76/mt7915/mcu.c
> +++ b/drivers/net/wireless/mediatek/mt76/mt7915/mcu.c
> @@ -2291,6 +2291,57 @@ mt7915_mcu_init_rx_airtime(struct mt7915_dev *dev)
>  				 sizeof(req), true);
>  }
>  
> +static int mt7915_red_set_watermark(struct mt7915_dev *dev)
> +{
> +#define RED_GLOBAL_TOKEN_WATERMARK 2
> +	u16 nbuf = dev->mt76.mmio.wed.wlan.nbuf;

I do not think this patch compile if CONFIG_NET_MEDIATEK_SOC_WED is not enabled

> +	struct {
> +		__le32 args[3];

please remove blank lines.

> +
> +		u8 cmd;
> +		u8 version;
> +		u8 __rsv1[4];
> +		u16 len;

I guess it should be __le16 here

> +
> +		__le16 high_mark;
> +		__le16 low_mark;
> +		u8 __rsv2[12];
> +	} __packed req = {
> +		.args[0] = cpu_to_le32(MCU_WA_PARAM_RED_SETTING),
> +		.cmd = RED_GLOBAL_TOKEN_WATERMARK,
> +		.len = cpu_to_le16(sizeof(req) - sizeof(req.args)),
> +
> +		.high_mark = cpu_to_le16(nbuf - 256),
> +		.low_mark = cpu_to_le16(nbuf - 256 - 1536),
> +	};
> +
> +	return mt76_mcu_send_msg(&dev->mt76, MCU_WA_PARAM_CMD(SET), &req,
> +				 sizeof(req), false);
> +}
> +
> +int mt7915_mcu_set_red(struct mt7915_dev *dev, bool enabled)
> +{
> +#define RED_DISABLE		0
> +#define RED_BY_WA_ENABLE	2
> +	int ret;
> +	u32 red_type = enabled ? RED_BY_WA_ENABLE : RED_DISABLE;
> +	__le32 req = cpu_to_le32(red_type);
> +
> +	if (enabled) {
> +		ret = mt7915_red_set_watermark(dev);
> +		if (ret < 0)
> +			return ret;
> +	}
> +
> +	ret = mt76_mcu_send_msg(&dev->mt76, MCU_EXT_CMD(RED_ENABLE), &req,
> +				sizeof(req), false);
> +	if (ret < 0)
> +		return ret;
> +
> +	return mt7915_mcu_wa_cmd(dev, MCU_WA_PARAM_CMD(SET),
> +				 MCU_WA_PARAM_RED, enabled, 0);
> +}
> +
>  int mt7915_mcu_init_firmware(struct mt7915_dev *dev)
>  {
>  	int ret;
> @@ -2339,8 +2390,7 @@ int mt7915_mcu_init_firmware(struct mt7915_dev *dev)
>  	if (ret)
>  		return ret;
>  
> -	return mt7915_mcu_wa_cmd(dev, MCU_WA_PARAM_CMD(SET),
> -				 MCU_WA_PARAM_RED, 0, 0);
> +	return mt7915_mcu_set_red(dev, mtk_wed_device_active(&dev->mt76.mmio.wed));
>  }
>  
>  int mt7915_mcu_init(struct mt7915_dev *dev)
> diff --git a/drivers/net/wireless/mediatek/mt76/mt7915/mcu.h b/drivers/net/wireless/mediatek/mt76/mt7915/mcu.h
> index 29b5434bfdb8..b9ea297f382c 100644
> --- a/drivers/net/wireless/mediatek/mt76/mt7915/mcu.h
> +++ b/drivers/net/wireless/mediatek/mt76/mt7915/mcu.h
> @@ -278,6 +278,7 @@ enum {
>  	MCU_WA_PARAM_PDMA_RX = 0x04,
>  	MCU_WA_PARAM_CPU_UTIL = 0x0b,
>  	MCU_WA_PARAM_RED = 0x0e,
> +	MCU_WA_PARAM_RED_SETTING = 0x40,
>  };
>  
>  enum mcu_mmps_mode {
> -- 
> 2.18.0
> 

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 228 bytes --]

  reply	other threads:[~2023-01-05 12:21 UTC|newest]

Thread overview: 4+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-01-05 11:47 [PATCH 1/2] wifi: mt76: mt7915: set sku initial value to zero Peter Chiu
2023-01-05 11:47 ` [PATCH 2/2] wifi: mt76: mt7915: wed: enable red per-band token drop Peter Chiu
2023-01-05 12:20   ` Lorenzo Bianconi [this message]
2023-01-05 13:54   ` kernel test robot

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=Y7bApSqBT2LXdBhu@lore-desk \
    --to=lorenzo@kernel.org \
    --cc=chui-hao.chiu@mediatek.com \
    --cc=evelyn.tsai@mediatek.com \
    --cc=linux-mediatek@lists.infradead.org \
    --cc=linux-wireless@vger.kernel.org \
    --cc=lorenzo.bianconi@redhat.com \
    --cc=nbd@nbd.name \
    --cc=ryder.Lee@mediatek.com \
    --cc=shayne.chen@mediatek.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).