Redirect int10 calls to serial console output.
Parse serial input and queue key events.
The serial console can work both as primary display
and in parallel to another vga display (splitmode).
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
The C code only uses _cfuncX_ prefixes for parameters to the call32(),
stack_hop_back(), and call32_params() functions. It's simpler to use
macro wrappers around those functions which provide the required
prefix.
This also changes the parameter order of stack_hop() and
stack_hop_back() to use the more natural (func, params) ordering.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Some Chromebooks (with Baytrail CPUs) apparently do not support
routing of legacy interrupts. This patch adds minimal support for
running SeaBIOS in such an environment. Even with this patch, it is
known that old operating systems and even some recent bootloaders will
not function without real hardware interrupts.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Add a build time config option to remove support for RTC timer
interrupts along with the associated bios calls requiring that
support.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Some recent hardware has trouble with routing PS2 port interrupts
while the interrupt controller is in legacy routing mode. This patch
adds a config mechanism (via "etc/ps2-poll-only") to force the PS2
code into a polling only mode so that interrupts are not required.
It is not recommended to use this polling mode on hardware that does
properly support PS2 irqs, because some very old (DOS-era) programs
depend on the BIOS PS2 irq behavior.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
This patch implements the TCG BIOS interrupt handler 1ah. It is for
example used by trusted grub.
This patch adds an implementation of SHA1 (following NIST specs., IETF RFC 3147
and Wikipedia) for speeding up measurements of code. Trusted Grub for example
makes use of this interface and measures (calculates SHA1) of the Linux kernel
and initrd. Those files can be rather large and hunting their bytes through
the TIS interface as part of the int handler commands invoked by trusted grub
does take quite some time due to the many vmexits the interface is creating
(one per byte).
There is also a threshold for the size of data to hash (100k) below which
the TPM is used and above the internal faster SHA1 algorithm is used.
This patch for example enables trusted grub to interact with the TPM
and take additional measurements.
Signed-off-by: Stefan Berger <stefanb@linux.vnet.ibm.com>
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Group the Real Time Clock code into hw/rtc.[ch].
Also, use rtc_read/write/mask function naming (instead of
inb/outb_cmos) to be more consistent with other register accessors.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Move the hardware setup to the hw/timer.c code. This eliminates the
need for a separate hw/pit.h file with definitions.
Also, move the IRQ counting code (which is dependent on the BDA) from
hw/timer.c to clock.c.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Rename the check_tsc() function to timer_check(). The CPU TSC is
often not the basis of the timer, so use a more appropriate name.
Convert all callers that were using u64 for the timers to use u32.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
The TICKS_PER_DAY setting is a bios standard and needs to be 1573040
for compatibility. However, there are actually ~1573042.24 ticks in a
day. So, only use TICKS_PER_DAY when working with the BDA
timer_counter - not when calculating any internal times.
The PIT hz is actually 143181800 / 12 (~1193181.667). This can be
accurately encoded as PMTIMER hz / 3. Because the PIT hz is usually
multiplied and divided by other numbers, we can use the PMTIMER hz and
defer the division by 3 to improve accuracy.
When doing division for delay time calculations, always round up the
division so the delay is never less than the requested time.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Preface PIC functions with a pic_ to provide a more consistent
naming.
Convert the irqmask code to a more consistent
pic_irqmask_read/write/mask form.
Move code from pic.h to pic.c.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
The PCI-BIOS entry point can be called in 16bit protected mode, so
separate its entry code from the legacy 0x1a code.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Convert all users of the alternative variable exports to VARFSEG.
There isn't a significant distinction between the existing types of
exports, so it's simpler to just use one type going forward.
The new VARFSEG declaration is only emitting when in 32bit mode, so
update and move some variables as needed.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Avoid reading/writing to cmos at runtime to get the QEMU century
information. Instead, read it at startup and cache the info.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
The POST phase has to invoke many initialization functions, and these
functions can have complex inter-dependencies. Try to categorize the
functions into 4 classes:
preinit - functions called very early in POST where function ordering
is very important and the code has limited access to other
interfaces.
init - functions that initialize internal interfaces and standard
external interfaces. This code is generally not dependent on
particular hardware and typically does not communicate directly
with any hardware devices.
setup - functions which access hardware or are dependent on particular
hardware or platform devices.
prepboot - functions that finalize internal interfaces and that
prepare for the boot phase.
This patch attempts to normalize the suffixes - functions that used
_init(), _setup(), _finalize(), or similar that did not follow the
above pattern were renamed. Other than function name changes, there
should be no code impact to this patch.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
This patch makes seabios use the acpi pmtimer instead of tsc for
timekeeping. The pmtimer has a fixed frequency and doesn't need
calibration, thus it doesn't suffer from calibration errors due to a
loaded host machine.
[ v4: mask port ioport read ]
[ v2: add CONFIG_PMTIMER ]
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
The winxp boot loader does something curious - it sets an int 0x1c
handler, records the stack location, and then spins in place with irqs
enabled. The 0x1c handler alters the memory just past the stack
pointer so that when the timer irq returns the code jumps to a new
location and stop spinning. The winxp code relies on the fact that a
hw irq will always place 6 bytes at a specific location and that it
can alter those bytes.
The ecdc655a patch does a full backup/restore of the register state.
Unfortunately, the restore overwrites the changes made by the winxp
0x1c handler.
This patch reverts much of ecdc655a. Hardware irqs are still handled
on the extra stack, but only the essential register state is backed up
and restored.
Also, stack_hop_back is changed to only use %sp when changing states -
this enables the entry code to store just %esp instead of both %esp
and %sp.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Jump into the extra stack for all hardware irq handlers. This reduces
the overall stack requirements of SeaBIOS.
Replace all users of call16_simpint with call16_int. Only the
hardware irq handlers used the old call, and they need to use the new
call to ensure the extra stack is properly re-entrant.
Also, pass in a 'struct bregs' to the hardware irq handlers now. It
was not done previously to save stack space. Now that the extra stack
is used, that is no longer an issue.
Note that should an old OS invoke a hardware irq in 16bit protected
mode, then this patch could break that OS. However, the chances of
this causing a regression seem small as several existing hardware irq
handlers already do not work in 16bit protected mode.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
It appears real machines sometimes have a flaky RTC, so avoid using
the RTC irq during boot. Instead, use a delay based on the standard
timer irq.
This also optimizes CONFIG_THREAD_OPTIONROMS users as it is no longer
necessary to use preemption - the wait_irq() call handles task
switching natively.
Increase the time waiting for rtc from 3ms to 15ms - only 3ms is
needed on real hardware, but scheduling delays on qemu can make this
longer. Extending the time prevents annoying debugging messages.
Some old programs will spin on a clock/keyboard call with irqs
disabled. They assume the BIOS will enable irqs and allow key events
and clock events to occur.
So, enable irqs in those functions that a caller might "spin" on.
The existing code always checks for USB "interrupt in" events every
millisecond. Although that's okay, it consumes extra bandwidth. This
change interrupt checks to be scheduled according to their requested
interval time.
Don't use "fail" in the debug output - as this confuses users.
When reporting on an invalid parameter - use the word "invalid".
When reporting on an unimplemented call - state it is unimplemented.
Add separate debug levels for unimplemented vs invalid calls.
Also, increase the debug level of several entry points.
When experimental support for parallelizing option roms and hardware
init (default disabled) is selected, add support for checking on
hardware init progress from the RTC irq handler.
Enable ability for RTC to be turned on for additional users.
Allow regular option roms (not just vga option roms) to run in
parallel with hardware init.
Don't use stack in transition32 / transition16 until new mode is
entered.
Also, cleanup leaking of data handlers in usb code.
Also, decrease frequency of iomemcpy checks (every 2K instead of 1K).
Add new function yield() which will permit irqs to trigger.
The yield() call enables irqs to occur in 32bit mode.
Add [num]sleep calls that yield instead of just spinning.
Rename existing int 1586 usleep call to biosusleep.
Convert many calls to mdelay to msleep.
This adds preliminary support for USB controllers and keyboards.
Add support for UHCI controllers.
Add support for "HID" USB keyboards.
Also, fix bug in hexdump() - len need not be power of 4.
Add wait_irq() - it's more efficient than looping with cpu_relax().
Also, move kbd irq enables down - only kbd_command needs it.
Also, make some minor code layout improvements to kbd.c.
Gerd Hoffmann