linux-kernel - Re: [Patch] [mm] More driver core fixes for -mm

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:	Tue, 25 Jul 2006 01:08:52 -0700
From:	Andrew Morton <akpm@...l.org>
To:	Cornelia Huck <cornelia.huck@...ibm.com>
Cc:	linux-kernel@...r.kernel.org, greg@...ah.com
Subject: Re: [Patch] [mm] More driver core fixes for -mm

On Fri, 21 Jul 2006 15:20:00 +0200
Cornelia Huck <cornelia.huck@...ibm.com> wrote:

> 
> I've looked some more into the __must_check stuff in the driver core,
> and tried to fix some functions (especially device_add() is a bit of a
> beast; I split off helper functions).

OK.

> Question: What is considered "good style" concerning symlinks? I would
> think I should remove symlinks I created, but most places don't seem to
> do this...

Removing symlinks seems like a good idea.  Leaving them around might cause
a subsequent driver load to fail due to EEXIST (assuming that the caller
checks error codes, as if).

I assume you're referring to error paths here?

> -- 
> Cornelia Huck
> Linux for zSeries Developer
> Tel.: +49-7031-16-4837, Mail: cornelia.huck@...ibm.com
> 
> From: Cornelia Huck <cornelia.huck@...ibm.com>
> 
> Fix missing checks of return codes for driver model functions called in
> the driver core.
> 
> Also fix bus_attach_device(), which didn't take into account that
> device_attach() may return 0 or 1 on success.
> 

Yes, this was nasty (oopses).

> @@ -401,13 +401,33 @@ int bus_attach_device(struct device * de
>  
>  	if (bus) {
>  		ret = device_attach(dev);
> -		if (ret == 0)
> +		if (ret >= 0)
>  			klist_add_tail(&dev->knode_bus, &bus->klist_devices);
>  	}

But I made bus_attach_device() convert the positive return value to zero. 
See
ftp://ftp.kernel.org/pub/linux/kernel/people/akpm/patches/2.6/2.6.18-rc1/2.6.18-rc1-mm2/hot-fixes/drivers-base-check-errors-fix.patch.

Is there a reason to propagate this irritating "1" back out of
bus_attach_device() as well?
-
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/