linux-kernel - Re: [PATCH 2/2] dma: add Qualcomm Technologies HIDMA channel driver

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, 30 Oct 2015 23:50:20 +0100
From:	Arnd Bergmann <arnd@...db.de>
To:	Timur Tabi <timur@...eaurora.org>
Cc:	Sinan Kaya <okaya@...eaurora.org>, dmaengine@...r.kernel.org,
	cov@...eaurora.org, jcm@...hat.com,
	Rob Herring <robh+dt@...nel.org>,
	Pawel Moll <pawel.moll@....com>,
	Mark Rutland <mark.rutland@....com>,
	Ian Campbell <ijc+devicetree@...lion.org.uk>,
	Kumar Gala <galak@...eaurora.org>,
	Vinod Koul <vinod.koul@...el.com>,
	Dan Williams <dan.j.williams@...el.com>,
	devicetree@...r.kernel.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH 2/2] dma: add Qualcomm Technologies HIDMA channel driver

On Friday 30 October 2015 17:36:45 Timur Tabi wrote:
> On 10/30/2015 05:28 PM, Arnd Bergmann wrote:
> >>> > >Why ENODEV? Could you make this handle restarted system calls?
> >> >
> >> >This is the self test code. It gets called from probe. If there is a
> >> >problem with the device or system configuration, I don't want to enable
> >> >this device. I can certainly return a different error code though.
> >> >What's a good code?
> > I see. probe() is not restartable, so it cannot be -ERESTARTSYS.
> >
> > Maybe better use wait_event_timeout and not handle the signals then.
> > It will eventually time out if something goes wrong.
> 
> What about -EPROBE_DEFER?  Isn't that "restartable"?  Granted, it's only 
> supposed to be used if the driver is dependent on another driver to 
> probe, so I'm not sure it applies here.  If the self-test fails, then it 
> is possible that it could succeed later?

No, this is different. The probe function can get called from all sorts
of contexts (sys_init_module, device_create, deferred probing), and not
all of them go back to user space when returning an error, so we cannot
deliver a signal to the calling process this way.

	Arnd
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/