All of lore.kernel.org
 help / color / mirror / Atom feed
From: Kalle Valo <kvalo@qca.qualcomm.com>
To: Michal Kazior <michal.kazior@tieto.com>
Cc: linux-wireless@vger.kernel.org, ath10k@lists.infradead.org
Subject: Re: [RFC/RFT 4/7] ath10k: bypass htc for htt tx path
Date: Wed, 19 Feb 2014 16:56:12 +0200	[thread overview]
Message-ID: <87ppmjxib7.fsf@kamboji.qca.qualcomm.com> (raw)
In-Reply-To: <1392629563-31046-5-git-send-email-michal.kazior@tieto.com> (Michal Kazior's message of "Mon, 17 Feb 2014 10:32:40 +0100")

Michal Kazior <michal.kazior@tieto.com> writes:

> Going through full HTC tx path for HTT tx is a
> waste of resources. By skipping it it's possible
> to easily submit scatter-gather to the PCI HIF for
> reduced host CPU load and improved performance.
>
> Signed-off-by: Michal Kazior <michal.kazior@tieto.com>

For a change like this the commit log needs to be more descriptive.
There's no mention about dma pools, nothing about the basic principle
how this works now etc.

> @@ -1067,9 +1067,9 @@ struct ath10k_ce_pipe *ath10k_ce_init(struct ath10k *ar,
>  	 *
>  	 * For the lack of a better place do the check here.
>  	 */
> -	BUILD_BUG_ON(TARGET_NUM_MSDU_DESC >
> +	BUILD_BUG_ON(2*TARGET_NUM_MSDU_DESC >
>  		     (CE_HTT_H2T_MSG_SRC_NENTRIES - 1));
> -	BUILD_BUG_ON(TARGET_10X_NUM_MSDU_DESC >
> +	BUILD_BUG_ON(2*TARGET_10X_NUM_MSDU_DESC >
>  		     (CE_HTT_H2T_MSG_SRC_NENTRIES - 1));
>  
>  	ret = ath10k_pci_wake(ar);
> diff --git a/drivers/net/wireless/ath/ath10k/ce.h b/drivers/net/wireless/ath/ath10k/ce.h
> index 322e929..8eb7f99 100644
> --- a/drivers/net/wireless/ath/ath10k/ce.h
> +++ b/drivers/net/wireless/ath/ath10k/ce.h
> @@ -23,7 +23,7 @@
>  
>  /* Maximum number of Copy Engine's supported */
>  #define CE_COUNT_MAX 8
> -#define CE_HTT_H2T_MSG_SRC_NENTRIES 2048
> +#define CE_HTT_H2T_MSG_SRC_NENTRIES 4096

Also document why you do these changes.

-- 
Kalle Valo

_______________________________________________
ath10k mailing list
ath10k@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/ath10k

WARNING: multiple messages have this Message-ID (diff)
From: Kalle Valo <kvalo@qca.qualcomm.com>
To: Michal Kazior <michal.kazior@tieto.com>
Cc: <ath10k@lists.infradead.org>, <linux-wireless@vger.kernel.org>
Subject: Re: [RFC/RFT 4/7] ath10k: bypass htc for htt tx path
Date: Wed, 19 Feb 2014 16:56:12 +0200	[thread overview]
Message-ID: <87ppmjxib7.fsf@kamboji.qca.qualcomm.com> (raw)
In-Reply-To: <1392629563-31046-5-git-send-email-michal.kazior@tieto.com> (Michal Kazior's message of "Mon, 17 Feb 2014 10:32:40 +0100")

Michal Kazior <michal.kazior@tieto.com> writes:

> Going through full HTC tx path for HTT tx is a
> waste of resources. By skipping it it's possible
> to easily submit scatter-gather to the PCI HIF for
> reduced host CPU load and improved performance.
>
> Signed-off-by: Michal Kazior <michal.kazior@tieto.com>

For a change like this the commit log needs to be more descriptive.
There's no mention about dma pools, nothing about the basic principle
how this works now etc.

> @@ -1067,9 +1067,9 @@ struct ath10k_ce_pipe *ath10k_ce_init(struct ath10k *ar,
>  	 *
>  	 * For the lack of a better place do the check here.
>  	 */
> -	BUILD_BUG_ON(TARGET_NUM_MSDU_DESC >
> +	BUILD_BUG_ON(2*TARGET_NUM_MSDU_DESC >
>  		     (CE_HTT_H2T_MSG_SRC_NENTRIES - 1));
> -	BUILD_BUG_ON(TARGET_10X_NUM_MSDU_DESC >
> +	BUILD_BUG_ON(2*TARGET_10X_NUM_MSDU_DESC >
>  		     (CE_HTT_H2T_MSG_SRC_NENTRIES - 1));
>  
>  	ret = ath10k_pci_wake(ar);
> diff --git a/drivers/net/wireless/ath/ath10k/ce.h b/drivers/net/wireless/ath/ath10k/ce.h
> index 322e929..8eb7f99 100644
> --- a/drivers/net/wireless/ath/ath10k/ce.h
> +++ b/drivers/net/wireless/ath/ath10k/ce.h
> @@ -23,7 +23,7 @@
>  
>  /* Maximum number of Copy Engine's supported */
>  #define CE_COUNT_MAX 8
> -#define CE_HTT_H2T_MSG_SRC_NENTRIES 2048
> +#define CE_HTT_H2T_MSG_SRC_NENTRIES 4096

Also document why you do these changes.

-- 
Kalle Valo

  reply	other threads:[~2014-02-19 14:56 UTC|newest]

Thread overview: 80+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2014-02-17  9:32 [RFC/RFT 0/7] ath10k: performance improvements Michal Kazior
2014-02-17  9:32 ` Michal Kazior
2014-02-17  9:32 ` [RFC/RFT 1/7] ath10k: remove DMA mapping wrappers Michal Kazior
2014-02-17  9:32   ` Michal Kazior
2014-02-19 12:37   ` Kalle Valo
2014-02-19 12:37     ` Kalle Valo
2014-02-17  9:32 ` [RFC/RFT 2/7] ath10k: remove is_aborted from skb_cb Michal Kazior
2014-02-17  9:32   ` Michal Kazior
2014-02-17  9:32 ` [RFC/RFT 3/7] ath10k: replace send_head() with tx_sg() Michal Kazior
2014-02-17  9:32   ` Michal Kazior
2014-02-19 12:48   ` Kalle Valo
2014-02-19 12:48     ` Kalle Valo
2014-02-19 13:25     ` Michal Kazior
2014-02-19 13:25       ` Michal Kazior
2014-02-19 14:18       ` Kalle Valo
2014-02-19 14:18         ` Kalle Valo
2014-02-20  6:43         ` Michal Kazior
2014-02-20  6:43           ` Michal Kazior
2014-02-24 11:46           ` Kalle Valo
2014-02-24 11:46             ` Kalle Valo
2014-02-17  9:32 ` [RFC/RFT 4/7] ath10k: bypass htc for htt tx path Michal Kazior
2014-02-17  9:32   ` Michal Kazior
2014-02-19 14:56   ` Kalle Valo [this message]
2014-02-19 14:56     ` Kalle Valo
2014-02-17  9:32 ` [RFC/RFT 5/7] ath10k: batch htt tx/rx completions Michal Kazior
2014-02-17  9:32   ` Michal Kazior
2014-02-19 15:10   ` Kalle Valo
2014-02-19 15:10     ` Kalle Valo
2014-02-20 11:23     ` Michal Kazior
2014-02-20 11:23       ` Michal Kazior
2014-02-24 11:49       ` Kalle Valo
2014-02-24 11:49         ` Kalle Valo
2014-02-17  9:32 ` [RFC/RFT 6/7] ath10k: remove pci completion list Michal Kazior
2014-02-17  9:32   ` Michal Kazior
2014-02-17  9:32 ` [RFC/RFT 7/7] ath10k: minimize coherent dma accesses Michal Kazior
2014-02-17  9:32   ` Michal Kazior
2014-02-17 15:01 ` [RFC/RFT 0/7] ath10k: performance improvements Kalle Valo
2014-02-17 15:01   ` Kalle Valo
2014-02-19 15:16 ` Kalle Valo
2014-02-19 15:16   ` Kalle Valo
2014-02-26 11:34 ` [PATCH v2 0/8] " Michal Kazior
2014-02-26 11:34   ` Michal Kazior
2014-02-26 11:34   ` [PATCH v2 1/8] ath10k: remove DMA mapping wrappers Michal Kazior
2014-02-26 11:34     ` Michal Kazior
2014-02-26 11:34   ` [PATCH v2 2/8] ath10k: remove is_aborted from skb_cb Michal Kazior
2014-02-26 11:34     ` Michal Kazior
2014-02-26 12:09   ` [PATCH v2 0/8] ath10k: performance improvements Michal Kazior
2014-02-26 12:09     ` Michal Kazior
2014-02-27  7:19 ` [PATCH v3 " Michal Kazior
2014-02-27  7:19   ` Michal Kazior
2014-02-27  7:19   ` [PATCH v3 1/8] ath10k: remove DMA mapping wrappers Michal Kazior
2014-02-27  7:19     ` Michal Kazior
2014-02-27  7:19   ` [PATCH v3 2/8] ath10k: remove is_aborted from skb_cb Michal Kazior
2014-02-27  7:19     ` Michal Kazior
2014-02-27  7:19   ` [PATCH v3 3/8] ath10k: replace send_head() with tx_sg() Michal Kazior
2014-02-27  7:19     ` Michal Kazior
2014-02-27  7:19   ` [PATCH v3 4/8] ath10k: bypass htc for htt tx path Michal Kazior
2014-02-27  7:19     ` Michal Kazior
2014-02-28  9:06     ` Kalle Valo
2014-02-28  9:06       ` Kalle Valo
2014-02-28  9:15       ` Michal Kazior
2014-02-28  9:15         ` Michal Kazior
2014-02-28  9:28         ` Kalle Valo
2014-02-28  9:28           ` Kalle Valo
2014-02-28  9:54           ` Michal Kazior
2014-02-28  9:54             ` Michal Kazior
2014-02-27  7:19   ` [PATCH v3 5/8] ath10k: batch htt tx/rx completions Michal Kazior
2014-02-27  7:19     ` Michal Kazior
2014-02-28  9:00     ` Kalle Valo
2014-02-28  9:00       ` Kalle Valo
2014-02-28  9:07       ` Michal Kazior
2014-02-28  9:07         ` Michal Kazior
2014-02-27  7:19   ` [PATCH v3 6/8] ath10k: reduce htt tx/rx spinlock overhead Michal Kazior
2014-02-27  7:19     ` Michal Kazior
2014-02-27  7:19   ` [PATCH v3 7/8] ath10k: remove pci completion list Michal Kazior
2014-02-27  7:19     ` Michal Kazior
2014-02-27  7:19   ` [PATCH v3 8/8] ath10k: minimize coherent dma accesses Michal Kazior
2014-02-27  7:19     ` Michal Kazior
2014-02-28 10:13   ` [PATCH v3 0/8] ath10k: performance improvements Kalle Valo
2014-02-28 10:13     ` Kalle Valo

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=87ppmjxib7.fsf@kamboji.qca.qualcomm.com \
    --to=kvalo@qca.qualcomm.com \
    --cc=ath10k@lists.infradead.org \
    --cc=linux-wireless@vger.kernel.org \
    --cc=michal.kazior@tieto.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.