linux-kernel - Re: [PATCH 1/1] edac x38: new MC driver module

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, 21 Nov 2008 01:19:41 +0900
From:	Hitoshi Mitake <h.mitake@...il.com>
To:	Russell King <rmk+lkml@....linux.org.uk>
Cc:	Ralf Baechle <ralf@...ux-mips.org>,
	Andrew Morton <akpm@...ux-foundation.org>,
	Doug Thompson <norsk5@...oo.com>, dougthompson@...ssion.com,
	linux-kernel@...r.kernel.org, linux-arch@...r.kernel.org
Subject: Re: [PATCH 1/1] edac x38: new MC driver module

On Tue, 18 Nov 2008 12:32:15 +0000
Russell King <rmk+lkml@....linux.org.uk> wrote:

> On Tue, Nov 18, 2008 at 12:16:20PM +0000, Ralf Baechle wrote:
> > On Sun, Nov 09, 2008 at 11:26:46AM -0800, Andrew Morton wrote:
> > 
> > > Perhaps it would be better to have a CONFIG_ARCH_HAS_READQ and to then
> > > disable these drivers on the architectures which don't provide
> > > readq/writeq support.
> > 
> > And we also need to define the exact semantics.  Questions coming to mind:
> > 
> >   o are implementations performing 2 32-bit accesses acceptable?
> >   o if so, what ordering for the two accesses is acceptable?
> 
> and don't forget to document the semantics.  If we're going to end up
> with CONFIG_ARCH_HAS_READQ which architectures can select, I suggest
> putting it in the help for that symbol.  Why not another random file
> in Documentation/ ?  Because it's a random file in Documentation/
> that'll be overlooked when someone decided to select ARCH_HAS_READQ.
> If it's along side the relevent config option, there is a higher
> chance it will be noticed.
> 

Sorry for my late response...

I knew that implementing architecture-independed readq/writeq is too hard.
To check that implementation is good for every architecture and test that readq/writeq are
difficult works.

So I wrote patch in Andrew's way.
This patch adds ARCH_HAS_READQ to X86_32 and X86_64, adds ARCH_HAS_WRITEQ to X86_64
and adds readq() to X86_32 (writeq is yet).

I want someone to review it. If this patch is good enough, I'll write help document and more patch
adding ARCH_HAS_READQ and ARCH_HAS_WRITEQ to other architectre which has readq/writeq.

description of this patch: Adding config value to x86 architecture to determine existence of readq/writeq


Signed-off-by: Hitoshi Mitake <h.mitake@...il.com>
---
 arch/x86/Kconfig          |    3 +++
 arch/x86/include/asm/io.h |    8 ++++++++
 2 files changed, 11 insertions(+), 0 deletions(-)

diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig
index ac22bb7..8f3c949 100644
--- a/arch/x86/Kconfig
+++ b/arch/x86/Kconfig
@@ -11,9 +11,12 @@ config 64BIT
 
 config X86_32
 	def_bool !64BIT
+	select ARCH_HAS_READQ
 
 config X86_64
 	def_bool 64BIT
+	select ARCH_HAS_READQ
+	select ARCH_HAS_WRITEQ
 
 ### Arch settings
 config X86
diff --git a/arch/x86/include/asm/io.h b/arch/x86/include/asm/io.h
index ac2abc8..2a8fc26 100644
--- a/arch/x86/include/asm/io.h
+++ b/arch/x86/include/asm/io.h
@@ -57,6 +57,14 @@ build_mmio_write(__writeq, "q", unsigned long, "r", )
 /* Let people know we have them */
 #define readq readq
 #define writeq writeq
+
+#else  /* CONFIG_X86_32 */
+
+static inline unsigned long readq(const volatile void __iomem *addr)
+{
+	return readl(addr) | (((u64)readl(addr + 4)) << 32);
+}
+
 #endif
 
 extern int iommu_bio_merge;
-- 
1.5.6.5

--
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/