linux-kernel - Re: [PATCH v4] usb: typec: intel_pmc_mux: Expose IOM port status to debugfs

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: <2023052326-outbreak-pliable-2d21@gregkh>
Date:   Tue, 23 May 2023 08:46:17 +0100
From:   Greg KH <gregkh@...uxfoundation.org>
To:     Rajat Khandelwal <rajat.khandelwal@...ux.intel.com>
Cc:     heikki.krogerus@...ux.intel.com, linux-usb@...r.kernel.org,
        linux-kernel@...r.kernel.org
Subject: Re: [PATCH v4] usb: typec: intel_pmc_mux: Expose IOM port status to
 debugfs

On Tue, May 23, 2023 at 01:13:37PM +0530, Rajat Khandelwal wrote:
> Hi,
> 
> On 5/22/2023 9:56 PM, Greg KH wrote:
> > On Mon, May 22, 2023 at 03:24:13PM +0530, Rajat Khandelwal wrote:
> > > Hi,
> > > 
> > > On 5/22/2023 1:07 PM, Greg KH wrote:
> > > > On Mon, May 22, 2023 at 12:33:43PM +0530, Rajat Khandelwal wrote:
> > > > > IOM status has a crucial role during debugging to check the
> > > > > current state of the type-C port.
> > > > > There are ways to fetch the status, but all those require the
> > > > > IOM port status offset, which could change with platform.
> > > > > 
> > > > > Make a debugfs directory for intel_pmc_mux and expose the status
> > > > > under it per port basis.
> > > > > 
> > > > > Signed-off-by: Rajat Khandelwal <rajat.khandelwal@...ux.intel.com>
> > > > > Reviewed-by: Heikki Krogerus <heikki.krogerus@...ux.intel.com>
> > > > > ---
> > > > > 
> > > > > v4:
> > > > > 1. Maintain a root directory for PMC module and incorporate devices
> > > > > under it
> > > > > 2. Add the debugfs module under '/sys/kernel/debug/usb'
> > > > > 3. Use the platform device 'pmc->dev' to assign the device's name
> > > > > 
> > > > > v3: Allocate the debugfs directory name for the platform device with
> > > > > its ACPI dev name included
> > > > > 
> > > > > v2:
> > > > > 1. Remove static declaration of the debugfs root for 'intel_pmc_mux'
> > > > > 2. Remove explicitly defined one-liner functions
> > > > > 
> > > > >    drivers/usb/typec/mux/intel_pmc_mux.c | 54 ++++++++++++++++++++++++++-
> > > > >    1 file changed, 53 insertions(+), 1 deletion(-)
> > > > > 
> > > > > diff --git a/drivers/usb/typec/mux/intel_pmc_mux.c b/drivers/usb/typec/mux/intel_pmc_mux.c
> > > > > index 34e4188a40ff..f400094c76f2 100644
> > > > > --- a/drivers/usb/typec/mux/intel_pmc_mux.c
> > > > > +++ b/drivers/usb/typec/mux/intel_pmc_mux.c
> > > > > @@ -15,6 +15,8 @@
> > > > >    #include <linux/usb/typec_mux.h>
> > > > >    #include <linux/usb/typec_dp.h>
> > > > >    #include <linux/usb/typec_tbt.h>
> > > > > +#include <linux/debugfs.h>
> > > > > +#include <linux/usb.h>
> > > > >    #include <asm/intel_scu_ipc.h>
> > > > > @@ -143,8 +145,14 @@ struct pmc_usb {
> > > > >    	struct acpi_device *iom_adev;
> > > > >    	void __iomem *iom_base;
> > > > >    	u32 iom_port_status_offset;
> > > > > +
> > > > > +#ifdef CONFIG_DEBUG_FS
> > > > > +	struct dentry *dentry;
> > > > > +#endif
> > > > No need for the #ifdef anymore, right?  In fact, I think it will break
> > > > the build if you have it this way and CONFIG_DEBUG_FS is not enabled,
> > > > right?
> > > I guess you're right. Maybe it'd have been fine if the rest of the
> > > debugfs stuff was also enclosed within the conditional macros.
> > Which is not needed and not good kernel coding style, so it's right that
> > they are not there.
> 
> Sure.
> 
> > 
> > > Anyways, removing it seems appropriate now.
> > > 
> > > Also, is it OK to send you v5 on the public list directly?
> > Where else would you send it?
> 
> I was under the impression that you'd again require the review
> internally under Intel before releasing out to public list.

As you all know, that is always required.  Ideally it would have caught
build breakages like this the first time around, so maybe it isn't
working as well as it should be :(

greg k-h