linux-kernel - Re: [PATCH v2 01/10] interconnect: qcom: rpm: Don't set QoS params before non-zero bw is requested

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <1039a507-c4cd-e92f-dc29-1e2169ce5078@linaro.org>
Date:   Tue, 10 Jan 2023 14:00:47 +0000
From:   Bryan O'Donoghue <bryan.odonoghue@...aro.org>
To:     Konrad Dybcio <konrad.dybcio@...aro.org>,
        linux-arm-msm@...r.kernel.org, andersson@...nel.org,
        agross@...nel.org, krzysztof.kozlowski@...aro.org
Cc:     marijn.suijten@...ainline.org, Georgi Djakov <djakov@...nel.org>,
        linux-pm@...r.kernel.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH v2 01/10] interconnect: qcom: rpm: Don't set QoS params
 before non-zero bw is requested

On 10/01/2023 13:21, Konrad Dybcio wrote:
> Until now, the icc-rpm driver unconditionally set QoS params, even on
> empty requests. This is superfluous and the downstream counterpart does
> not do it. Follow it by doing the same.
> 
> Signed-off-by: Konrad Dybcio <konrad.dybcio@...aro.org>
> ---
>   drivers/interconnect/qcom/icc-rpm.c | 6 ++++++
>   1 file changed, 6 insertions(+)
> 
> diff --git a/drivers/interconnect/qcom/icc-rpm.c b/drivers/interconnect/qcom/icc-rpm.c
> index df3196f72536..361dcbf3386f 100644
> --- a/drivers/interconnect/qcom/icc-rpm.c
> +++ b/drivers/interconnect/qcom/icc-rpm.c
> @@ -191,6 +191,12 @@ static int qcom_icc_qos_set(struct icc_node *node, u64 sum_bw)
>   	struct qcom_icc_provider *qp = to_qcom_provider(node->provider);
>   	struct qcom_icc_node *qn = node->data;
>   
> +	/* Defer setting QoS until the first non-zero bandwidth request. */
> +	if (!(node->avg_bw || node->peak_bw)) {
> +		dev_dbg(node->provider->dev, "NOT Setting QoS for %s\n", qn->name);
> +		return 0;
> +	}
> +
>   	dev_dbg(node->provider->dev, "Setting QoS for %s\n", qn->name);
>   
>   	switch (qp->type) {

I still think you should include the original logic on the else, for the 
minimum case of silicon that predates the 5.4 kernel release.

/* Clear bandwidth registers */
set_qos_bw_regs(base, mas_index, 0, 0, 0, 0, 0);

Either that or get the relevant silicon engineers at qcom to say the 
host side port write is redundant.

---
bod