From: Georgi Djakov <djakov@kernel.org>
To: Bjorn Andersson <bjorn.andersson@linaro.org>,
Sibi Sankar <quic_sibis@quicinc.com>,
Mathieu Poirier <mathieu.poirier@linaro.org>
Cc: linux-arm-msm@vger.kernel.org, linux-remoteproc@vger.kernel.org,
linux-kernel@vger.kernel.org
Subject: Re: [PATCH] remoteproc: qcom: q6v5: Add interconnect path proxy vote
Date: Mon, 7 Mar 2022 20:35:58 +0200 [thread overview]
Message-ID: <5a9210d9-c726-1ef9-4bf2-716f2ed1fb8b@kernel.org> (raw)
In-Reply-To: <20220225033224.2238425-1-bjorn.andersson@linaro.org>
On 25.02.22 5:32, Bjorn Andersson wrote:
> Many remoteproc instances requires that Linux casts a proxy vote for an
> interconnect path during boot, until they can do it themselves. Add
> support for voting for a single path.
>
> As this is a shared problem between both PAS and MSS drivers, the path
> is acquired and votes casted from the common helper code.
>
> Signed-off-by: Bjorn Andersson <bjorn.andersson@linaro.org>
> ---
>
> Sibi posted recently a patch to add interconnect votes from the modem driver,
> today I needed the same feature for one of the PAS remoteprocs. After
> essentially duplicating Sibi's patch I realized that it doesn't look too bad to
> put this in the common Q6V5 code.
>
> The main difference is that this would be messy if we need to support multiple
> paths, so we probably would have to push it out to the individual drivers at
> that point.
>
> Sibi's patch can be found here.
> https://lore.kernel.org/all/1644813252-12897-3-git-send-email-quic_sibis@quicinc.com/
>
>
> This makes the implementation pick up one path, relevant DT bindings would
> still need to be updated in order be allowed to this in the DeviceTree files.
>
> drivers/remoteproc/qcom_q6v5.c | 21 ++++++++++++++++++++-
> drivers/remoteproc/qcom_q6v5.h | 3 +++
> 2 files changed, 23 insertions(+), 1 deletion(-)
>
> diff --git a/drivers/remoteproc/qcom_q6v5.c b/drivers/remoteproc/qcom_q6v5.c
> index 442a388f8102..5280ec9b5449 100644
> --- a/drivers/remoteproc/qcom_q6v5.c
> +++ b/drivers/remoteproc/qcom_q6v5.c
> @@ -8,6 +8,7 @@
> */
> #include <linux/kernel.h>
> #include <linux/platform_device.h>
> +#include <linux/interconnect.h>
> #include <linux/interrupt.h>
> #include <linux/module.h>
> #include <linux/soc/qcom/qcom_aoss.h>
> @@ -51,9 +52,17 @@ int qcom_q6v5_prepare(struct qcom_q6v5 *q6v5)
> {
> int ret;
>
> + ret = icc_set_bw(q6v5->path, 0, UINT_MAX);
> + if (ret < 0) {
> + dev_err(q6v5->dev, "failed to set bandwidth request\n");
> + return ret;
> + }
> +
> ret = q6v5_load_state_toggle(q6v5, true);
> - if (ret)
> + if (ret) {
> + icc_set_bw(q6v5->path, 0, 0);
> return ret;
> + }
>
> reinit_completion(&q6v5->start_done);
> reinit_completion(&q6v5->stop_done);
> @@ -78,6 +87,9 @@ int qcom_q6v5_unprepare(struct qcom_q6v5 *q6v5)
> disable_irq(q6v5->handover_irq);
> q6v5_load_state_toggle(q6v5, false);
>
> + /* Disable interconnect vote, in case handover never happened */
> + icc_set_bw(q6v5->path, 0, 0);
> +
> return !q6v5->handover_issued;
> }
> EXPORT_SYMBOL_GPL(qcom_q6v5_unprepare);
> @@ -160,6 +172,8 @@ static irqreturn_t q6v5_handover_interrupt(int irq, void *data)
> if (q6v5->handover)
> q6v5->handover(q6v5);
>
> + icc_set_bw(q6v5->path, 0, 0);
> +
> q6v5->handover_issued = true;
>
> return IRQ_HANDLED;
> @@ -332,6 +346,11 @@ int qcom_q6v5_init(struct qcom_q6v5 *q6v5, struct platform_device *pdev,
> return load_state ? -ENOMEM : -EINVAL;
> }
>
> + q6v5->path = devm_of_icc_get(&pdev->dev, NULL);
> + if (IS_ERR(q6v5->path))
> + return dev_err_probe(&pdev->dev, PTR_ERR(q6v5->path),
> + "failed to acquire interconnect path\n");
> +
> return 0;
> }
> EXPORT_SYMBOL_GPL(qcom_q6v5_init);
Probably we should also call icc_put(q6v5->path) in qcom_q6v5_deinit().
Reviewed-by: Georgi Djakov <djakov@kernel.org>
Thanks,
Georgi
next prev parent reply other threads:[~2022-03-07 18:36 UTC|newest]
Thread overview: 4+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-02-25 3:32 [PATCH] remoteproc: qcom: q6v5: Add interconnect path proxy vote Bjorn Andersson
2022-03-07 18:35 ` Georgi Djakov [this message]
2022-03-07 22:42 ` Bjorn Andersson
2022-03-07 23:48 ` Georgi Djakov
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=5a9210d9-c726-1ef9-4bf2-716f2ed1fb8b@kernel.org \
--to=djakov@kernel.org \
--cc=bjorn.andersson@linaro.org \
--cc=linux-arm-msm@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-remoteproc@vger.kernel.org \
--cc=mathieu.poirier@linaro.org \
--cc=quic_sibis@quicinc.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