linux-kernel - Re: [PATCH 03/18] printk: Convert pr_fmt from blank define to KBUILD

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: <8cbdd3e5d08c3faf3e7d10e19e8261f3222ce27c.camel@perches.com>
Date:   Thu, 10 May 2018 19:12:57 -0700
From:   Joe Perches <joe@...ches.com>
To:     Sergey Senozhatsky <sergey.senozhatsky.work@...il.com>
Cc:     Petr Mladek <pmladek@...e.com>,
        Sergey Senozhatsky <sergey.senozhatsky@...il.com>,
        Steven Rostedt <rostedt@...dmis.org>,
        linux-kernel@...r.kernel.org
Subject: Re: [PATCH 03/18] printk: Convert pr_fmt from blank define to
 KBUILD_MODNAME

On Thu, 2018-05-10 at 18:57 -0700, Joe Perches wrote:
> On Fri, 2018-05-11 at 10:51 +0900, Sergey Senozhatsky wrote:
> > On (05/10/18 08:45), Joe Perches wrote:
> > > There are more than 1000 uses of #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
> > > Make that the default define so these uses can be removed later via script.
> > > 
> > > Signed-off-by: Joe Perches <joe@...ches.com>
> > 
> > In general, the idea looks OK to me, so FWIW
> > Acked-by: Sergey Senozhatsky <sergey.senozhatsky@...il.com>
> > 
> > A quick question:
> > 
> > - Would "can be removed later via script" be so trivial?
> >   What about these guys?
> > 
> > drivers/of/address.c:#define pr_fmt(fmt)        "OF: " fmt
> > drivers/of/base.c:#define pr_fmt(fmt)   "OF: " fmt
> > drivers/of/dynamic.c:#define pr_fmt(fmt)        "OF: " fmt
> > drivers/of/fdt.c:#define pr_fmt(fmt)    "OF: fdt: " fmt
> > drivers/of/fdt_address.c:#define pr_fmt(fmt)    "OF: fdt: " fmt
> > drivers/of/irq.c:#define pr_fmt(fmt)    "OF: " fmt
> > drivers/of/of_numa.c:#define pr_fmt(fmt) "OF: NUMA: " fmt
> > drivers/of/of_reserved_mem.c:#define pr_fmt(fmt)        "OF: reserved mem: " fmt
> > drivers/of/overlay.c:#define pr_fmt(fmt)        "OF: overlay: " fmt
> > drivers/of/platform.c:#define pr_fmt(fmt)       "OF: " fmt
> > drivers/of/property.c:#define pr_fmt(fmt)       "OF: " fmt
> > drivers/of/resolver.c:#define pr_fmt(fmt)       "OF: resolver: " fmt
> 
> What about them?
> 
> All those defines above are non-generic as they are
> specified to start with "OF: " and then some optional
> extra bit.
> 
> Why should these defines be updated or modified?

What would be good, and is something that I will
eventually work on, is to make the KBUILD_MODNAME ": "
use a singleton looked up and inserted by the printk
subsystem instead of stored with each format.