linux-kernel - Re: [PATCH v14] x86, mce: Add memcpy

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: <20160312171648.GA12167@gmail.com>
Date:	Sat, 12 Mar 2016 18:16:48 +0100
From:	Ingo Molnar <mingo@...nel.org>
To:	Tony Luck <tony.luck@...il.com>
Cc:	Mika Penttilä <mika.penttila@...tfour.com>,
	"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
	Linus Torvalds <torvalds@...ux-foundation.org>,
	Thomas Gleixner <tglx@...utronix.de>,
	Borislav Petkov <bp@...en8.de>,
	Andrew Morton <akpm@...ux-foundation.org>,
	Peter Zijlstra <a.p.zijlstra@...llo.nl>,
	"H. Peter Anvin" <hpa@...or.com>,
	Andy Lutomirski <luto@...capital.net>,
	"Williams, Dan J" <dan.j.williams@...el.com>
Subject: Re: [PATCH v14] x86, mce: Add memcpy_mcsafe()


* Tony Luck <tony.luck@...il.com> wrote:

> On Thu, Mar 10, 2016 at 11:37 AM, Luck, Tony <tony.luck@...el.com> wrote:
> >> But you return 0 == false for success and 1 == true for failure.
> >
> > Aaargh!  -ETOOMUCHSHELLSCRIPTPROGRAMMING
> >
> > -Tony
> 
> Options to fix this:
> 1) Just change the comments in the code.
>      This seems like it would confuse people as I thing most people
>      would expect the "true" return to mean the copy succeeded.
> 2) Reverse the return values.
>      Better that option 1 - but doesn't leave scope to return a count
>      if some future user does want to know where the copy failed.
> 3) Change the return type back from "bool" to "int"
>      0 == success, non-zero == fail (with option to put the non-copied
> byte count in later).
> 4) Something else

Please use the copy_*_user() memory copying API semantics, which are: return 
negative code (-EFAULT) on error, 0 on success.

Don't return 1 and please don't use bools for any memory copy library 
functionality.

Thanks,

	Ingo