src/arch/x86/interrupts.cc (Diff revision 1)

I'm not sure what the semantics are supposed to be, but isn't the "clockEdge" function in the clocked object what is more or less expressed in the if/else statement?

Show all issues

1. Nilay Vaish March 12, 2013, 10:32 a.m. (March 12, 2013, 10:32 a.m.)

   While you are right about that the fact that clockEdge() can be used,
   the check that I am adding would still be required. This is because
   newCount can be 0, in which case you would end up scheduling an event
   at the same tick, while the intention of the Linux kernel is to actually
   shutdown the apic timer.

2. Andreas Hansson March 12, 2013, 6:18 p.m. (March 12, 2013, 6:18 p.m.)

   Makes sense

src/dev/x86/i8042.cc (Diff revision 1)

Is this something that happens at boot with never kernels? How difficult would it be to implement it "properly"?

Show all issues

1. Nilay Vaish March 12, 2013, 10:32 a.m. (March 12, 2013, 10:32 a.m.)

   You really think we need a keyboard or a mouse in a simulator!
   I am not implementing them anyway.

2. Andreas Hansson March 12, 2013, 6:18 p.m. (March 12, 2013, 6:18 p.m.)

   I'm mostly curious why the behaviour changed.

3. Nilay Vaish March 19, 2013, 10:34 a.m. (March 19, 2013, 10:34 a.m.)

   The following lines were added to the file drivers/input/serio/i8042-x86ia64io.h
   
   
         /*
          * A20 was already enabled during early kernel init. But some buggy
          * BIOSes (in MSI Laptops) require A20 to be enabled using 8042 to
          * resume from S3. So we do it here and hope that nothing breaks.
          */
         i8042_command(&a20_on, 0x10d1);
         i8042_command(NULL, 0x00ff);    /* Null command for SMM firmware */

Ship It!