linux-kernel - Re: [PATCH v2] usb: dwc2: Fix build in periphal-only mode

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] [day] [month] [year] [list]

Message-ID: <d577970e-7b86-1b1b-b31a-e432f1dc1c64@raspberrypi.com>
Date:   Fri, 28 May 2021 11:04:08 +0100
From:   Phil Elwell <phil@...pberrypi.com>
To:     Greg Kroah-Hartman <gregkh@...uxfoundation.org>
Cc:     Minas Harutyunyan <hminas@...opsys.com>,
        Sasha Levin <sashal@...nel.org>,
        Artur Petrosyan <Arthur.Petrosyan@...opsys.com>,
        linux-usb@...r.kernel.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH v2] usb: dwc2: Fix build in periphal-only mode

On 28/05/2021 10:51, Greg Kroah-Hartman wrote:

> On Fri, May 28, 2021 at 10:37:48AM +0100, Phil Elwell wrote:
>> Hi Greg,
>>
>> On 28/05/2021 10:23, Greg Kroah-Hartman wrote:
>>> On Fri, May 28, 2021 at 10:13:50AM +0100, Phil Elwell wrote:
>>>> The bus_suspended member of struct dwc2_hsotg is only present in builds
>>>> that support host-mode.
>>>>
>>>> Fixes: 24d209dba5a3 ("usb: dwc2: Fix hibernation between host and device modes.")
>>>> Signed-off-by: Phil Elwell <phil@...pberrypi.com>
>>>> ---
>>>>    drivers/usb/dwc2/core_intr.c | 4 ++++
>>>>    1 file changed, 4 insertions(+)
>>>>
>>>> v2: Correct commit hash used in the Fixes line.
>>>>
>>>> diff --git a/drivers/usb/dwc2/core_intr.c b/drivers/usb/dwc2/core_intr.c
>>>> index a5ab03808da6..03d0c034cf57 100644
>>>> --- a/drivers/usb/dwc2/core_intr.c
>>>> +++ b/drivers/usb/dwc2/core_intr.c
>>>> @@ -725,7 +725,11 @@ static inline void dwc_handle_gpwrdn_disc_det(struct dwc2_hsotg *hsotg,
>>>>    	dwc2_writel(hsotg, gpwrdn_tmp, GPWRDN);
>>>>    	hsotg->hibernated = 0;
>>>> +
>>>> +#if IS_ENABLED(CONFIG_USB_DWC2_HOST) ||	\
>>>> +	IS_ENABLED(CONFIG_USB_DWC2_DUAL_ROLE)
>>>>    	hsotg->bus_suspended = 0;
>>>> +#endif
>>>>    	if (gpwrdn & GPWRDN_IDSTS) {
>>>>    		hsotg->op_state = OTG_STATE_B_PERIPHERAL;
>>>> -- 
>>>> 2.25.1
>>>>
>>> I do not understand, the field in the structure is present for all, why
>>> is this crazy #if needed here?
>>>
>>> I see that the commit you reference here did add the new line to set
>>> bus_suspended, which seemed to be the point here.  Why will the #if
>>> values matter here?
>> Sorry to waste your brain cycles on this. There is a problem, but it only
>> exists in branches where the blamed commit (24d209dba5a3) has been
>> back-ported as a Fix, because it depends on commit 012466fc8ccc which isn't
>> a Fix and therefore
>> hasn't been back-ported. Sadly 012466fc8ccc doesn't back-port cleanly on its
>> own - either more cherry-picks or a temporary patch like mine will be
>> needed.
> So should we revert this commit from the stable releases where it showed
> up?  Which ones specifically?
>
> If so, please let me and stable@...r.kernel.org know and we can take
> care of it there.

Reverting back-ports of 24d209dba5a3 would be sufficient, although you are then 
left with the problem that 24d209dba5a3 was intended to address.

I'll email the stable list.

Phil