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: <1AE640813FDE7649BE1B193DEA596E880255760D@SHSMSX101.ccr.corp.intel.com>
Date:	Wed, 23 Apr 2014 01:06:56 +0000
From:	"Zheng, Lv" <lv.zheng@...el.com>
To:	"Rafael J. Wysocki" <rjw@...ysocki.net>
CC:	"Wysocki, Rafael J" <rafael.j.wysocki@...el.com>,
	"Brown, Len" <len.brown@...el.com>, Lv Zheng <zetalog@...il.com>,
	"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
	"linux-acpi@...r.kernel.org" <linux-acpi@...r.kernel.org>
Subject: RE: [PATCH 1/4] ACPICA: Add <acpi/platform/aclinuxxf.h> to remove
 mis-ordered inclusion of <acpi/actypes.h> from <acpi/platform/aclinux.h>.

Hi, Rafael

> From: Rafael J. Wysocki [mailto:rjw@...ysocki.net]
> Sent: Tuesday, April 22, 2014 7:16 PM
> 
> On Tuesday, April 22, 2014 02:01:57 AM Zheng, Lv wrote:
> > Hi,
> >
> > <skip>
> >
> > > > +#define ACPI_NATIVE_INTERFACE_HEADER	<acpi/platform/aclinuxxf.h>
> > >
> > > This is not good.
> > >
> > > We don't do things like this in the kernel, because they are confusing and hard
> > > to debug if necessary, so please find a different way to make this work.
> >
> > I use this extra header file to collect:
> 
> I was not talking about the new header, which is basically OK, but about the
> #define above, which is just too confusing to live.
> 
> Please use header file paths directly with #include.

OK.

> 
> > 1. static inline OSL functions
> > 2. divergences of prototypes that haven't been back ported to ACPICA.
> > This file is useful for ACPICA release automation.
> >
> > There are the following concerns that lead to the use of this solution:
> > 1. for this extra header file itself
> >     A. The new header file is OSPM specific, thus it needn't be upstreamed to ACPICA;
> >     B. Since it needn't be upstreamed to ACPICA, ACPICA needn't determine the name of this extra header;
> >     C. It has to be the last file included by <acpi/acpi.h>.
> > 2. for the file that includes this extra header file
> >     A. Currently there is no OSPM specific code in <acpi/acpi.h>.
> > Thus I use a macro so that there is still no OSPM specific code in <acpi/acpi.h> and the name of the extra header can be determined
> by OSPM.
> >
> > If you want another solution, is the following acceptable?
> > 1. In <acpi/platform/aclinux.h>
> >     #define ACPI_INCLUDE_EXTRA_NATIVE_HEADER	1
> > 2. In <acpi/acpi.h> <- this is an ACPICA header file,
> >     #ifdef ACPI_INCLUDE_EXTRA_NATIVE_HEADER
> >     #include <acpi/platform/acextra.h>
> >     #endif
> > Note that in this solution, the name of the extra header file will be determined by ACPICA.
> 
> I think I see what you're trying to do now.  And I see that this ACPI_NATIVE_INTERFACE_HEADER
> thing is already there in the Linus' tree which is not good at all.
> 
> I probably would create an extra ACPICA header, something like <acpios_opt.h>,
> that would be empty for all hosts except for Linux and that would contain the
> stuff you want to put into acextra.h in Linux.
> 
> That would be clean enough I suppose?

Yes, it is clear.
Thanks for the helping.
I'll update this patch and re-send this series.

Best regards
-Lv

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ