linux-arm-msm.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Bhaumik Bhatt <bbhatt@codeaurora.org>
To: Loic Poulain <loic.poulain@linaro.org>
Cc: manivannan.sadhasivam@linaro.org, hemantk@codeaurora.org,
	linux-arm-msm@vger.kernel.org
Subject: Re: [PATCH 3/8] mhi: pci_generic: Enable burst mode for hardware channels
Date: Fri, 13 Nov 2020 09:06:02 -0800	[thread overview]
Message-ID: <09b9acb6d808268f7eaea5409b90203b@codeaurora.org> (raw)
In-Reply-To: <1605279602-18749-4-git-send-email-loic.poulain@linaro.org>

Hi Loic,
On 2020-11-13 06:59, Loic Poulain wrote:
> Hardware channels have a feature called burst mode that allows to
> queue transfer ring element(s) (TRE) to a channel without ringing
> the device doorbell. In that mode, the device is polling the channel
> context for new elements. This reduces the frequency of host initiated
> doorbells and increase throughput.
> 
> Create a new dedicated macro for hardware channels with burst enabled.
> 
> Signed-off-by: Loic Poulain <loic.poulain@linaro.org>
> ---
>  drivers/bus/mhi/pci_generic.c | 34 ++++++++++++++++++++++++++++++++--
>  1 file changed, 32 insertions(+), 2 deletions(-)
> 
> diff --git a/drivers/bus/mhi/pci_generic.c 
> b/drivers/bus/mhi/pci_generic.c
> index 09c6b26..0c07cf5 100644
> --- a/drivers/bus/mhi/pci_generic.c
> +++ b/drivers/bus/mhi/pci_generic.c
> @@ -78,6 +78,36 @@ struct mhi_pci_dev_info {
>  		.offload_channel = false,	\
>  	}
> 
> +#define MHI_CHANNEL_CONFIG_HW_UL(ch_num, ch_name, el_count, ev_ring) \
> +	{						\
> +		.num = ch_num,				\
> +		.name = ch_name,			\
> +		.num_elements = el_count,		\
> +		.event_ring = ev_ring,			\
> +		.dir = DMA_TO_DEVICE,			\
> +		.ee_mask = BIT(MHI_EE_AMSS),		\
> +		.pollcfg = 0,				\
> +		.doorbell = MHI_DB_BRST_ENABLE,	\
> +		.lpm_notify = false,			\
> +		.offload_channel = false,		\
> +		.doorbell_mode_switch = false,		\
doorbell_mode_switch should be true for channel 100 (UL). Any reason why 
the boolean
fields are not yet configurable from the parameters to this macro? That 
will be
needed going forward so it should be changed now.

doorbell also may not be "enabled" by default. Basically, let's have 
something that
can be easily configured for different kinds of HW channels.
> +	}						\
> +
> +#define MHI_CHANNEL_CONFIG_HW_DL(ch_num, ch_name, el_count, ev_ring) \
> +	{						\
> +		.num = ch_num,				\
> +		.name = ch_name,			\
> +		.num_elements = el_count,		\
> +		.event_ring = ev_ring,			\
> +		.dir = DMA_FROM_DEVICE,			\
> +		.ee_mask = BIT(MHI_EE_AMSS),		\
> +		.pollcfg = 0,				\
> +		.doorbell = MHI_DB_BRST_ENABLE,	\
> +		.lpm_notify = false,			\
> +		.offload_channel = false,		\
> +		.doorbell_mode_switch = false,		\
> +	}
> +
>  #define MHI_EVENT_CONFIG_DATA(ev_ring)		\
>  	{					\
>  		.num_elements = 128,		\
> @@ -112,8 +142,8 @@ static const struct mhi_channel_config
> modem_qcom_v1_mhi_channels[] = {
>  	MHI_CHANNEL_CONFIG_DL(15, "QMI", 4, 0),
>  	MHI_CHANNEL_CONFIG_UL(20, "IPCR", 8, 0),
>  	MHI_CHANNEL_CONFIG_DL(21, "IPCR", 8, 0),
> -	MHI_CHANNEL_CONFIG_UL(100, "IP_HW0", 128, 1),
> -	MHI_CHANNEL_CONFIG_DL(101, "IP_HW0", 128, 2),
> +	MHI_CHANNEL_CONFIG_HW_UL(100, "IP_HW0", 128, 1),
> +	MHI_CHANNEL_CONFIG_HW_DL(101, "IP_HW0", 128, 2),
>  };
> 
>  static const struct mhi_event_config modem_qcom_v1_mhi_events[] = {
Thanks,
Bhaumik
-- 
The Qualcomm Innovation Center, Inc. is a member of the Code Aurora 
Forum,
a Linux Foundation Collaborative Project

  reply	other threads:[~2020-11-13 17:06 UTC|newest]

Thread overview: 27+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-11-13 14:59 [PATCH 0/8] mhi: pci_generic: Misc improvements Loic Poulain
2020-11-13 14:59 ` [PATCH 1/8] mhi: pci-generic: Increase number of hardware events Loic Poulain
2020-11-13 16:56   ` Bhaumik Bhatt
2020-11-18  3:54   ` Hemant Kumar
2020-11-18  8:26     ` Loic Poulain
2020-11-19  1:50       ` Hemant Kumar
2020-11-19  9:42         ` Loic Poulain
2020-11-13 14:59 ` [PATCH 2/8] mhi: pci-generic: Perform hard reset on remove Loic Poulain
2020-11-19  1:46   ` Hemant Kumar
2020-11-19  9:21     ` Loic Poulain
2020-11-25 17:41       ` Jeffrey Hugo
2020-11-27 16:21         ` Loic Poulain
2020-12-02  1:54           ` Bhaumik Bhatt
2020-11-13 14:59 ` [PATCH 3/8] mhi: pci_generic: Enable burst mode for hardware channels Loic Poulain
2020-11-13 17:06   ` Bhaumik Bhatt [this message]
2020-11-13 14:59 ` [PATCH 4/8] mhi: pci_generic: Add support for reset Loic Poulain
2020-11-13 17:12   ` Bhaumik Bhatt
2020-11-13 17:42     ` Loic Poulain
2020-11-13 17:44       ` Bhaumik Bhatt
2020-11-13 14:59 ` [PATCH 5/8] mhi: pci_generic: Add suspend/resume/recovery procedure Loic Poulain
2020-11-19  2:21   ` Hemant Kumar
2020-11-19  9:54     ` Loic Poulain
2020-11-13 15:00 ` [PATCH 6/8] mhi: pci_generic: Add PCI error handlers Loic Poulain
2020-11-13 15:00 ` [PATCH 7/8] mhi: pci_generic: Add health-check Loic Poulain
2020-11-13 15:00 ` [PATCH 8/8] mhi: pci_generic: Increase controller timeout value Loic Poulain
2020-11-13 17:08   ` Bhaumik Bhatt
2020-11-25 20:23   ` Hemant Kumar

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=09b9acb6d808268f7eaea5409b90203b@codeaurora.org \
    --to=bbhatt@codeaurora.org \
    --cc=hemantk@codeaurora.org \
    --cc=linux-arm-msm@vger.kernel.org \
    --cc=loic.poulain@linaro.org \
    --cc=manivannan.sadhasivam@linaro.org \
    /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).