xfs
[Top] [All Lists]

Re: TAKE - fix dbench on lowmem systems

To: Steve Lord <lord@xxxxxxx>
Subject: Re: TAKE - fix dbench on lowmem systems
From: "Amit D. Chaudhary" <amitc@xxxxxxxxxxx>
Date: Tue, 11 Dec 2001 16:23:10 -0800
Cc: linux-xfs list <linux-xfs@xxxxxxxxxxx>
References: <200112112047.fBBKlRw21471@xxxxxxxxxxxxxxxxxxxx>
Sender: owner-linux-xfs@xxxxxxxxxxx
User-agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:0.9.5) Gecko/20011012
Steve,

What is the typical memory on such a machine? Does it affect specific kernels?

Thanks
Amit

Steve Lord wrote:

This cures a dbench hang on low memory machines, it quite probably
fixes a number of other vm related interactions with XFS.

Date:  Tue Dec 11 12:47:18 PST 2001
Workarea:  jen.americas.sgi.com:/src/lord/xfs-linux.2.4

The following file(s) were checked into:
  bonnie.engr.sgi.com:/isms/slinx/2.4.x-xfs


Modid:  2.4.x-xfs:slinx:108247a
linux/fs/buffer.c - 1.97
        - Export the end_buffer_io_sync function for pagebuf and add BH_Delay
back into the busy bits
linux/fs/pagebuf/page_buf_io.c - 1.103
        - Change convert_page to refile the buffer it is about to write and to
          unlock the page before returning. We now use end_buffer_io_sync as the
          I/O completion function.







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