From: Chris Lew <quic_clew@quicinc.com>
To: Joe Hattori <joe@pf.is.s.u-tokyo.ac.jp>, <andersson@kernel.org>,
<mathieu.poirier@linaro.org>, <dmitry.baryshkov@linaro.org>
Cc: <linux-remoteproc@vger.kernel.org>
Subject: Re: [PATCH] remoteproc: qcom: pas: Remove subdevs on the error path of adsp_probe()
Date: Wed, 6 Nov 2024 18:07:56 -0800 [thread overview]
Message-ID: <2eb54159-7e92-42a7-bbb7-1b4848f49a80@quicinc.com> (raw)
In-Reply-To: <20241105092055.255127-1-joe@pf.is.s.u-tokyo.ac.jp>
Hi Joe,
On 11/5/2024 1:20 AM, Joe Hattori wrote:
> Current implementation of adsp_probe() does not remove the subdevs of
> struct qcom_adsp *adsp on the error path. Fix this bug by calling
> qcom_remove_{ssr,sysmon,pdm,smd,glink}_subdev(), qcom_q6v5_deinit(), and
> adsp_unassign_memory_region() appropriately.
>
> Fixes: 5b9f51b200dc ("remoteproc: qcom: enable in-kernel PD mapper")
The ssr, sysmon, glink subdevs and qcom_q6v5_init() are the calls that
leak resources. I think this issue existed before the in-kernel PD
mapper patches so this Fixes tag isn't accurate.
Would this apply to the other remoteproc files like qcom_q6v5_adsp.c?
> Signed-off-by: Joe Hattori <joe@pf.is.s.u-tokyo.ac.jp>
> ---
> drivers/remoteproc/qcom_q6v5_pas.c | 20 +++++++++++++++-----
> 1 file changed, 15 insertions(+), 5 deletions(-)
>
> diff --git a/drivers/remoteproc/qcom_q6v5_pas.c b/drivers/remoteproc/qcom_q6v5_pas.c
> index ef82835e98a4..27b23a000c7a 100644
> --- a/drivers/remoteproc/qcom_q6v5_pas.c
> +++ b/drivers/remoteproc/qcom_q6v5_pas.c
> @@ -759,16 +759,16 @@ static int adsp_probe(struct platform_device *pdev)
>
> ret = adsp_init_clock(adsp);
> if (ret)
> - goto free_rproc;
> + goto unassign_mem;
>
> ret = adsp_init_regulator(adsp);
> if (ret)
> - goto free_rproc;
> + goto unassign_mem;
>
> ret = adsp_pds_attach(&pdev->dev, adsp->proxy_pds,
> desc->proxy_pd_names);
> if (ret < 0)
> - goto free_rproc;
> + goto unassign_mem;
> adsp->proxy_pd_count = ret;
>
> ret = qcom_q6v5_init(&adsp->q6v5, pdev, rproc, desc->crash_reason_smem, desc->load_state,
> @@ -784,18 +784,28 @@ static int adsp_probe(struct platform_device *pdev)
> desc->ssctl_id);
> if (IS_ERR(adsp->sysmon)) {
> ret = PTR_ERR(adsp->sysmon);
> - goto detach_proxy_pds;
> + goto deinit_remove_pdm_smd_glink;
> }
>
> qcom_add_ssr_subdev(rproc, &adsp->ssr_subdev, desc->ssr_name);
> ret = rproc_add(rproc);
> if (ret)
> - goto detach_proxy_pds;
> + goto remove_ssr_sysmon;
>
> return 0;
>
> +remove_ssr_sysmon:
> + qcom_remove_ssr_subdev(rproc, &adsp->ssr_subdev);
> + qcom_remove_sysmon_subdev(adsp->sysmon);
> +deinit_remove_pdm_smd_glink: > + qcom_remove_pdm_subdev(rproc, &adsp->pdm_subdev);
> + qcom_remove_smd_subdev(rproc, &adsp->smd_subdev);
> + qcom_remove_glink_subdev(rproc, &adsp->glink_subdev);
> + qcom_q6v5_deinit(&adsp->q6v5); > detach_proxy_pds:
> adsp_pds_detach(adsp, adsp->proxy_pds, adsp->proxy_pd_count);
> +unassign_mem:
> + adsp_unassign_memory_region(adsp);
> free_rproc:
> device_init_wakeup(adsp->dev, false);
>
prev parent reply other threads:[~2024-11-07 2:08 UTC|newest]
Thread overview: 2+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-11-05 9:20 [PATCH] remoteproc: qcom: pas: Remove subdevs on the error path of adsp_probe() Joe Hattori
2024-11-07 2:07 ` Chris Lew [this message]
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=2eb54159-7e92-42a7-bbb7-1b4848f49a80@quicinc.com \
--to=quic_clew@quicinc.com \
--cc=andersson@kernel.org \
--cc=dmitry.baryshkov@linaro.org \
--cc=joe@pf.is.s.u-tokyo.ac.jp \
--cc=linux-remoteproc@vger.kernel.org \
--cc=mathieu.poirier@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 read-only IMAP folder(s) and NNTP newsgroup(s).