On Tue, 2014-05-20 at 22:30 -0700, Jordan Justen wrote:
> On Tue, May 20, 2014 at 7:46 PM, Fan, Jeff <jeff....@intel.com> wrote:
> > Mike,
> >
> > I have one question, how do you find this issue?  Do you encounter
> > one real issue on your platform or you find it by code review?
We experienced a long boot time. This patch was in series that tried to
fix this.
And from the timeouts perspective it increases possible delay.
But in case of virtual machine these conditions must be very fast,
because of required flags can be set immediately after some command
accepted. In case of real floppy device there could be a real timeouts
that really required for access to physical device. So using seconds
here would be right.

> 
> I will say that I've seen timeouts on floppy often account for a large
> part of the OVMF boot time. I haven't figured out the correct way to
> fix it though.
I can point to another place where unconditional exists:
look at Recalibrate() function:

    for (Index = 0; Index < sizeof (FDD_COMMAND_PACKET2); Index++) {
      if (EFI_ERROR (DataOutByte (FdcDev, CommandPointer++))) {
        return EFI_DEVICE_ERROR;
      }
    }
    //
    // Experience value
    //
    MicroSecondDelay (250000);
    //
    // need modify according to 1.44M or 2.88M
    //
    if (EFI_ERROR (SenseIntStatus (FdcDev, &StatusRegister0,
&PresentCylinderNumber))) {
      return EFI_DEVICE_ERROR;
    }

We need to optimize it.

> It is fairly unfortunate considering floppies are almost never used with OVMF.

What about unattended installations?

> But, Mike's change seems to increase the timeout, so I guess that
> won't help OVMF. :)

Will try to improve  that :)


------------------------------------------------------------------------------
"Accelerate Dev Cycles with Automated Cross-Browser Testing - For FREE
Instantly run your Selenium tests across 300+ browser/OS combos.
Get unparalleled scalability from the best Selenium testing platform available
Simple to use. Nothing to install. Get started now for free."
http://p.sf.net/sfu/SauceLabs
_______________________________________________
edk2-devel mailing list
edk2-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/edk2-devel

Reply via email to