linux-mips
[Top] [All Lists]

Re: flush_kernel_vmap_range() invalidate_kernel_vmap_range() API not exi

To: Ralf Baechle <ralf@linux-mips.org>
Subject: Re: flush_kernel_vmap_range() invalidate_kernel_vmap_range() API not exists for MIPS
From: naveen yadav <yad.naveen@gmail.com>
Date: Fri, 24 Jun 2011 10:36:13 +0530
Cc: Sergei Shtylyov <sshtylyov@mvista.com>, Christoph Hellwig <hch@lst.de>, linux-mips@linux-mips.org
Dkim-signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:mime-version:in-reply-to:references:date :message-id:subject:from:to:cc:content-type :content-transfer-encoding; bh=jY3ZJjoiT4fwPASUxbTRA0U5FSzf0nk94ojcnvteRlQ=; b=QkOayfOUB6PbLrVgYIkx8Fs1CPmTxbyIJwrQjCIZFwBH3CnS2cO5equhYnJxGM4kVI hvliiQq2z9t3XZvz6mqdi8CGqIMmmMiFULHJmforcxcs/gjzaf5karVv3RfZV4vgJP57 surXVWODmetVn02NbC3F9PjbW77m8j2F7R/6Q=
Domainkey-signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc:content-type:content-transfer-encoding; b=Nm6IvKpWk15rIcPD5wX0E9YQrXv4E0YZhWjsrtgL72Xhif+IaT5tXC5nqDHFGJ7P5v dRZGab1pd8HPyQoxsG0NTtZlLrdygzsiGinQS5JWS4FyCRBAkfIUjA4C4A9K1YGFCzG2 CXdr962RX4751VK+Q1txYfJzI1rtDS0r87KOs=
In-reply-to: <20110620095625.GA3227@linux-mips.org>
References: <AANLkTimkh2QLvupu+62NGrKfqRb_gC7KLCAKkEoS9N9N@mail.gmail.com> <20110325172709.GC8483@linux-mips.org> <BANLkTimo6BEgDnTh+sPVR+MELyxiwJoFGw@mail.gmail.com> <20110616180250.GA13025@lst.de> <20110617152028.GA14107@linux-mips.org> <4DFCA2DD.9060707@mvista.com> <20110620095625.GA3227@linux-mips.org>
Sender: linux-mips-bounce@linux-mips.org
Dear Ralf and Sergei,

If we have L2 cache, then we need to invalidate them also ?

our system  have L2 cache(write back).

Regards


On Mon, Jun 20, 2011 at 3:26 PM, Ralf Baechle <ralf@linux-mips.org> wrote:
> On Sat, Jun 18, 2011 at 05:06:37PM +0400, Sergei Shtylyov wrote:
>
>> >+static inline void flush_kernel_vmap_range(void *vaddr, int size)
>> >+{
>> >+    if (cpu_has_dc_aliases)
>> >+            __flush_kernel_vmap_range((unsigned long) vaddr, size);
>> >+}
>> >+
>> >+static inline void invalidate_kernel_vmap_range(void *vaddr, int size)
>> >+{
>> >+    if (cpu_has_dc_aliases)
>> >+            __flush_kernel_vmap_range((unsigned long) vaddr, size);
>>
>>    Not __invalidate_kernel_vmap_range()?
>
> No, for the moment both just do a writeback + invalidate.  I may change
> that for something more efficient later once I understand the consequences
> of this.
>
>  Ralf
>

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