linux-kernel - Re: [PATCH] brcmfmac: Avoid keeping power to SDIO card unless WOWL is used

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]

Date:   Fri, 18 Mar 2022 11:47:05 +0100
From:   Yann Gautier <yann.gautier@...s.st.com>
To:     Ulf Hansson <ulf.hansson@...aro.org>,
        Kalle Valo <kvalo@...nel.org>, <linux-wireless@...r.kernel.org>
CC:     Arend van Spriel <aspriel@...il.com>,
        Franky Lin <franky.lin@...adcom.com>,
        Hante Meuleman <hante.meuleman@...adcom.com>,
        Gustavo Padovan <gustavo.padovan@...labora.com>,
        Adrian Ratiu <adrian.ratiu@...labora.com>,
        <brcm80211-dev-list.pdl@...adcom.com>,
        <linux-kernel@...r.kernel.org>,
        Christophe KERELLO - foss <christophe.kerello@...s.st.com>,
        Christophe ROULLIER-SCND-02 <christophe.roullier@...s.st.com>
Subject: Re: [PATCH] brcmfmac: Avoid keeping power to SDIO card unless WOWL is
 used

On 3/17/22 16:28, Ulf Hansson wrote:
> Keeping the power to the SDIO card during system wide suspend, consumes
> energy. Especially on battery driven embedded systems, this can be a
> problem. Therefore, let's change the behaviour into allowing the SDIO card
> to be powered off, unless WOWL is supported and enabled.
> 
> Note that, the downside from this change, is that at system resume the SDIO
> card needs to be re-initialized and the FW must re-programmed. Even it that
> may take some time to complete, it should we worth it, rather than draining
> a battery.
> 
> Signed-off-by: Ulf Hansson <ulf.hansson@...aro.org>
> ---
> 
> Please note that, I have only compile-tested this patch, so I am relying on help
> from Yann and others to run tests on real HW.
> 
> Kind regards
> Ulf Hansson
> 
> ---
>   .../broadcom/brcm80211/brcmfmac/bcmsdh.c      | 33 +++++++++++--------
>   1 file changed, 20 insertions(+), 13 deletions(-)
> 
> diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/bcmsdh.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/bcmsdh.c
> index ac02244a6fdf..351886c9d68e 100644
> --- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/bcmsdh.c
> +++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/bcmsdh.c
> @@ -1119,9 +1119,21 @@ void brcmf_sdio_wowl_config(struct device *dev, bool enabled)
>   {
>   	struct brcmf_bus *bus_if = dev_get_drvdata(dev);
>   	struct brcmf_sdio_dev *sdiodev = bus_if->bus_priv.sdio;
> +	mmc_pm_flag_t pm_caps = sdio_get_host_pm_caps(sdiodev->func1);
>   
> -	brcmf_dbg(SDIO, "Configuring WOWL, enabled=%d\n", enabled);
> -	sdiodev->wowl_enabled = enabled;
> +	/* Power must be preserved to be able to support WOWL. */
> +	if (!(pm_caps & MMC_PM_KEEP_POWER))
> +		goto notsup;
> +
> +	if (sdiodev->settings->bus.sdio.oob_irq_supported ||
> +	    pm_caps & MMC_PM_WAKE_SDIO_IRQ) {
> +		sdiodev->wowl_enabled = enabled;
> +		brcmf_dbg(SDIO, "Configuring WOWL, enabled=%d\n", enabled);
> +		return;
> +	}
> +
> +notsup:
> +	brcmf_dbg(SDIO, "WOWL not supported\n");
>   }
>   
>   #ifdef CONFIG_PM_SLEEP
> @@ -1130,7 +1142,7 @@ static int brcmf_ops_sdio_suspend(struct device *dev)
>   	struct sdio_func *func;
>   	struct brcmf_bus *bus_if;
>   	struct brcmf_sdio_dev *sdiodev;
> -	mmc_pm_flag_t pm_caps, sdio_flags;
> +	mmc_pm_flag_t sdio_flags;
>   	int ret = 0;
>   
>   	func = container_of(dev, struct sdio_func, dev);
> @@ -1142,20 +1154,15 @@ static int brcmf_ops_sdio_suspend(struct device *dev)
>   	bus_if = dev_get_drvdata(dev);
>   	sdiodev = bus_if->bus_priv.sdio;
>   
> -	pm_caps = sdio_get_host_pm_caps(func);
> -
> -	if (pm_caps & MMC_PM_KEEP_POWER) {
> -		/* preserve card power during suspend */
> +	if (sdiodev->wowl_enabled) {
>   		brcmf_sdiod_freezer_on(sdiodev);
>   		brcmf_sdio_wd_timer(sdiodev->bus, 0);
>   
>   		sdio_flags = MMC_PM_KEEP_POWER;
> -		if (sdiodev->wowl_enabled) {
> -			if (sdiodev->settings->bus.sdio.oob_irq_supported)
> -				enable_irq_wake(sdiodev->settings->bus.sdio.oob_irq_nr);
> -			else
> -				sdio_flags |= MMC_PM_WAKE_SDIO_IRQ;
> -		}
> +		if (sdiodev->settings->bus.sdio.oob_irq_supported)
> +			enable_irq_wake(sdiodev->settings->bus.sdio.oob_irq_nr);
> +		else
> +			sdio_flags |= MMC_PM_WAKE_SDIO_IRQ;
>   
>   		if (sdio_set_host_pm_flags(sdiodev->func1, sdio_flags))
>   			brcmf_err("Failed to set pm_flags %x\n", sdio_flags);

Hi Ulf,

You are missing the counter part in brcmf_ops_sdio_resume:

--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/bcmsdh.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/bcmsdh.c
@@ -1190,14 +1190,13 @@ static int brcmf_ops_sdio_resume(struct device *dev)
         if (func->num != 2)
                 return 0;

-       if (!(pm_caps & MMC_PM_KEEP_POWER)) {
+       if (!sdiodev->wowl_enabled) {
                 /* bus was powered off and device removed, probe again */
                 ret = brcmf_sdiod_probe(sdiodev);
                 if (ret)
                         brcmf_err("Failed to probe device on resume\n");
         } else {
-               if (sdiodev->wowl_enabled &&
-                   sdiodev->settings->bus.sdio.oob_irq_supported)
+               if (sdiodev->settings->bus.sdio.oob_irq_supported)
 
disable_irq_wake(sdiodev->settings->bus.sdio.oob_irq_nr);


Else, we get an oops when calling brcmf_sdio_sleep.


Best regards,
Yann