Hi!
James Tabor a écrit :
What happen to
the com stuff! 19200 baud! WTH!
Boot up normal, until after the serial driver loads.
(ldr/loader.c:309) Could not open module file:
\SystemRoot\system32\drivers\mpu401.sys
DriverBase for \SystemRoot\system32\drivers\null.sys:
9d62e000
DriverBase for \SystemRoot\system32\drivers\serial.sys: 9d646000
(io/device.c:171) IoAttachDeviceToDeviceStackSafe() badly
implemented!.ààààààààààààààààààààààààààààààààààààààààààààààààà
àààààààààààààààààààààààààààààààààààà
That's strange.
Indeed, serial.sys driver modifies serial port configuration. But it
sets it to 19200 bauds, 8 data bits, no parity, 1 stop bit, like in
boot/freeldr/freeldr/comm/rs232.c
Which debug environment are you using? qemu, vmware, real hardware? real
serial cable or output to some named pipe? which tool are you using to
get debug output?
IoAttachDeviceToDeviceStackSafe() warning is not very important. It
should be used instead of IoAttachDeviceToDeviceStack() to prevent a
race condition, but this race condition actually exists in its code...
Hervé
PS1: old serial driver did absolutely nothing so that's normal that
reverting to it fixes the problem
PS2: actually, expect conflicts because serial ports are controlled *at
least* in:
- boot/freeldr/freeldr/comm/rs232.c
- drivers/dd/serial/
- drivers/input/sermouse/
- hal/halx86/generic/kdbg.c