Re: Alignment of disk-I/O from userland.

From: Don Lewis (truckman_at_FreeBSD.org)
Date: 10/07/03

  • Next message: Sam Leffler: "Re: Alignment of disk-I/O from userland." 
    Date: Mon, 6 Oct 2003 17:48:57 -0700 (PDT)
To: phk@phk.freebsd.dk

    On 7 Oct, Poul-Henning Kamp wrote:

    > But I also just realized a complication I had not thought of earlier,
    > and which may modify our thinking further: This is an issue for
    > all physread()/physwrite() drivers, not just disks.
    >
    > In other words, if I want to write to 1MB blocks to a SCSI tape,
    > and I don't align my in memory buffer sufficiently for the hardware,
    > busdma would have to allocate 1MB of memory (it may _possibly_ be
    > able to do so as disjunct pages rather than consequtively) and copy
    > the entire request over.
    >
    > For disks we can chop the request at sector boundaries or multiple
    > thereoff and deal with it that way, but we don't have that option
    > for scsi_sa or even scsi_pt devices.
    >
    > Currently we impose a 128k upper limit on I/O requests, but we have
    > already more or less agreed that needs to grow into the 4-16MB range
    > soon.

    There might be another reason to other than alignment to copy the data.
    What are the limits on the size of the scatter-gather lists that are
    supported by the DMA engines of commonly available controllers? It is
    unlikely that the user's buffer will be in contiguous memory even if it
    is properly aligned. It might be necessary to copy the the memory into
    some number of buffers, where each buffer is in contiguous physical
    memory, so that reasonable size I/O requests can be supported on
    commonly available controllers. A 4MB request might require 1K
    scatter-gather list entries on a machine with a 4K page size. Ick ...
    _______________________________________________
    freebsd-arch@freebsd.org mailing list
    http://lists.freebsd.org/mailman/listinfo/freebsd-arch
    To unsubscribe, send any mail to "freebsd-arch-unsubscribe@freebsd.org"

  • Next message: Sam Leffler: "Re: Alignment of disk-I/O from userland."

    Relevant Pages

    • Re: Alignment of disk-I/O from userland.
      ... >>the entire request over. ... > unlikely that the user's buffer will be in contiguous memory even if it ... This again is a job for busdma. ...
      (freebsd-arch)
    • Re: [Lit.] Buffer overruns
      ... of physical memory. ... >Overrun in that case was having the bits in a buffer spill over ... >I, an app program, request 100K core from the monitor. ... >>associate the cryptic error message with what they've done wrong. ...
      (sci.crypt)
    • RE: Strange MDL behavior
      ... Since I am in an IRP_MJ_READ request I am expected to write to the ... memory is not mapped to virtual memory yet, so no other thread can write to ... buffer in stead of passing the request on to the disk driver. ... differs = FALSE; ...
      (microsoft.public.win32.programmer.kernel)
    • Re: When to check the return value of malloc
      ... and works on all legitimate size requests, if memory is, in fact, ... request that could exceed the range of an int. ... not too long back - allocation requests exceeding the range of an int ... little less efficiently, with a smaller buffer. ...
      (comp.lang.c)
    • Re: fopen and memory usage
      ... If I remember rightly its nearer 64K nowdays simply because disks are ... memory is cheaper and its a good buffer size. ... Does it use the same buffer space ...
      (Fedora)