linux-mips
[Top] [All Lists]

Re: [PATCH 6/16] define Hit_Invalidate_I to Index_Invalidate_I for loong

To: Ralf Baechle <ralf@linux-mips.org>
Subject: Re: [PATCH 6/16] define Hit_Invalidate_I to Index_Invalidate_I for loongson2
From: Fuxin Zhang <fxzhang@ict.ac.cn>
Date: Wed, 18 Apr 2007 21:51:58 +0800
Cc: tiansm@lemote.com, linux-mips@linux-mips.org, Fuxin Zhang <zhangfx@lemote.com>
In-reply-to: <20070418121139.GF3938@linux-mips.org>
Original-recipient: rfc822;linux-mips@linux-mips.org
References: <11766507651736-git-send-email-tiansm@lemote.com> <11766507661317-git-send-email-tiansm@lemote.com> <11766507661726-git-send-email-tiansm@lemote.com> <11766507662638-git-send-email-tiansm@lemote.com> <11766507661133-git-send-email-tiansm@lemote.com> <11766507661526-git-send-email-tiansm@lemote.com> <11766507662650-git-send-email-tiansm@lemote.com> <20070418121139.GF3938@linux-mips.org>
Sender: linux-mips-bounce@linux-mips.org
User-agent: Icedove 1.5.0.8 (X11/20061116)
Yes, Loongson2 has no Hit_invalidate_I in fact. And although it has 4-way icache, one Index_invalidate_I will invalidate all 4 ways of the same set.

The usermanual and datasheet can be downloaded from:

http://www.lemote.com/upfiles/godson2e-user-manual-V0.6.pdf


Ralf Baechle wrote:
On Sun, Apr 15, 2007 at 11:25:55PM +0800, tiansm@lemote.com wrote:

+#if defined(CONFIG_CPU_LOONGSON2)
+#define Hit_Invalidate_I       0x00

This #ifdef means Index_Invalidate_I and Hit_Invalidate_I will both be
defined as zero, is that really correct?

(This is the point where I would really like to have a CPU manual ...)

+#else
 #define Hit_Invalidate_I       0x10
+#endif

  Ralf





<Prev in Thread] Current Thread [Next in Thread>