[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <alpine.LFD.2.01.0905311659350.3435@localhost.localdomain>
Date: Sun, 31 May 2009 17:02:03 -0700 (PDT)
From: Linus Torvalds <torvalds@...ux-foundation.org>
To: Andrew Morton <akpm@...ux-foundation.org>
cc: Alexey Dobriyan <adobriyan@...il.com>,
Matt Helsley <matthltc@...ibm.com>, xemul@...allels.com,
containers@...ts.linux-foundation.org,
linux-kernel@...r.kernel.org, dave@...ux.vnet.ibm.com,
mingo@...e.hu
Subject: Re: [PATCH 14/38] Remove struct mm_struct::exe_file et al
On Sun, 31 May 2009, Andrew Morton wrote:
>
> It's arbitrary! If we were to gain more performance benefit by
> aggregating processes under, say, the fs_struct then that's the way the
> kernel would have been implemented.
What?
That's like saying that "if quarks were the size of soccer-balls, we'd all
be pink".
It makes no sense.
When you claim that it's arbitrary, you're totally ignoring reality, and
CPU design.
The fact is, TLB's are very fundamental to task switching. filesystems are
not. It's that simple. IOW, it's not at all arbitrary, it's a direct
result of how the world works. The "world" for the kernel is the reality
of CPU's, and in that reality, TLB switching is a huge factor.
There is nothing arbitrary about it at all.
Linus
--
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/
Powered by blists - more mailing lists