linux-kernel - Re: [PATCH] PCI: j721e: In j721e_pcie_suspend_noirq() check reset

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: <20241211091421.4empou7mbm35ynxq@thinkpad>
Date: Wed, 11 Dec 2024 14:44:21 +0530
From: Manivannan Sadhasivam <manivannan.sadhasivam@...aro.org>
To: Thomas Richard <thomas.richard@...tlin.com>
Cc: Bjorn Helgaas <helgaas@...nel.org>, vigneshr@...com, s-vadapalli@...com,
	lpieralisi@...nel.org, kw@...ux.com, robh@...nel.org,
	bhelgaas@...gle.com, theo.lebrun@...tlin.com,
	thomas.petazzoni@...tlin.com, kwilczynski@...nel.org,
	linux-omap@...r.kernel.org, linux-pci@...r.kernel.org,
	linux-arm-kernel@...ts.infradead.org, linux-kernel@...r.kernel.org,
	gregory.clement@...tlin.com, u-kumar1@...com
Subject: Re: [PATCH] PCI: j721e: In j721e_pcie_suspend_noirq() check
 reset_gpio before to use it

On Wed, Dec 11, 2024 at 09:59:30AM +0100, Thomas Richard wrote:
> On 12/10/24 16:42, Bjorn Helgaas wrote:
> > On Mon, Dec 09, 2024 at 12:23:21PM +0100, Thomas Richard wrote:
> >> The reset_gpio is optional, so in j721e_pcie_suspend_noirq() check if it is
> >> not NULL before to use it.
> > 
> > If you have occasion to post a v2, update subject to:
> > 
> >   PCI: j721e: Check reset_gpio for NULL before using it
> > 
> > s/before to use it/before using it/
> > 
> > Did you trip over a NULL pointer dereference here?  Or maybe found via
> > inspection?
> 
> By inspection
> 
> > 
> > It looks like gpiod_set_value_cansleep(desc) *should* be a no-op if
> > desc is NULL, based on this comment [1]:
> > 
> >  * This descriptor validation needs to be inserted verbatim into each
> >  * function taking a descriptor, so we need to use a preprocessor
> >  * macro to avoid endless duplication. If the desc is NULL it is an
> >  * optional GPIO and calls should just bail out.
> > 
> > and the fact that the VALIDATE_DESC_VOID() macro looks like it would
> > return early in that case.
> > 
> > [1] https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/drivers/gpio/gpiolib.c?id=v6.12#n2316
> 
> Oh yes you're right.
> In fact, the if statement in probe() and resume_noirq() is for msleep(),
> not really for gpiod_set_value_cansleep().
> 
> So this patch is useless.
> 

Yes. Almost all of the GPIO APIs accepting desc (except few) use VALIDATE_DESC()
to check for NULL descriptor. So explicit check is not needed.

- Mani

-- 
மணிவண்ணன் சதாசிவம்