Re: pci_map_single() and pci_unmap_single() pair

From: tcs (dzshen_at_winbond.com.tw)
Date: 05/09/04

  • Next message: P.T. Breuer: "Re: large files: when ubiquitous?" 
    Date: Sun, 09 May 2004 10:58:23 +0800

    hmm, if pci_unmap_single() doesn't do anything useful, then how can we
    release resources required by pci_map_single() back to system?

    thanks

    tcs

    Kris wrote:
    > tcs wrote:
    >
    >> Greetings:
    >> I am maintaing a SCSI device driver and there are some weird behaviors.
    >> The longer the driver is running, the slower it is, I doubt there is
    >> memory
    >> leak problem.
    >>
    >> Here is the question:
    >> What if we use pci_map_single() but in some cases we forget to call
    >> pci_unmap_single(), what's the obviously external behavior?
    >>
    >> thanks
    >>
    >> tcs
    >>
    >>
    >
    > None at all. pci_unmap_single is essentially void.
    >
    > Kris
    >

  • Next message: P.T. Breuer: "Re: large files: when ubiquitous?"

    Relevant Pages

    • Re: pci_map_single() and pci_unmap_single() pair
      ... tcs writes: ... where it frees the IOMMU resources allocated by pci_map_single. ... Basically "Kris" told you that he is only interested in i386; ... later when you need to port your driver to 64bit. ...
      (comp.os.linux.development.system)
    • Re: pci_map_single() and pci_unmap_single() pair
      ... tcs wrote: ... > I am maintaing a SCSI device driver and there are some weird behaviors. ... > The longer the driver is running, the slower it is, I doubt there is memory ...
      (comp.os.linux.development.system)
    • Re: xfree/xorg?
      ... TCS wrote: ... It was simply my experience that the ati drivers didn't work with ... > xorg. ... Hmm, ...
      (comp.os.linux.x)
    Loading