| Age | Commit message (Collapse) | Author | Files | Lines |
|---|
|
Use the pci_enable_x() functions.
The pvscsi controller code will now explicitly set PCI_COMMAND_MEMORY
instead of assuming it has already been enabled.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
|
|
Use the pci_enable_x() functions.
After this change, the megasas driver will no longer enable
PCI_COMMAND_MEMORY accesses, as the megasas driver doesn't actually
map any BARs as memory.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
|
|
Use the pci_enable_x() functions.
The lsi-scsi controller code will now explicitly set PCI_COMMAND_IO
instead of assuming it has already been enabled.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
|
|
Use the pci_enable_x() functions.
The esp-scsi controller code will now explicitly set PCI_COMMAND_IO
instead of assuming it has already been enabled.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
|
|
Use the pci_enable_x() functions.
The ATA controller code will now explicitly set PCI_COMMAND_IO instead
of assuming it has already been enabled.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
|
|
Use the pci_enable_x() functions.
This patch also converts cntl->iobase from a 'u32' to a 'void*' so
that it is clear that the address is a virtual memory address.
After this change, the AHCI driver will no longer enable
PCI_COMMAND_IO io accesses, as the AHCI driver doesn't actually
attempt IO accesses to the device.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
|
|
Add functions to verify and obtain PCI BARs (Base Address Registers).
These new functions check that the requested BAR is of the right type
and appears valid.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
|
|
There are no longer any sleep or yield calls during the usb controller
device scans, so there is no need to run these device scans in a
separate thread.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
|
|
EHCI controller setup needs to occur prior to checking any UHCI or
OHCI ports to ensure a high speed device is not mistakenly configured
on a full speed "companion" controller. However, only the UHCI/OHCI
port scan needs to be delayed, not the full UHCI/OHCI controller init.
This change moves back the ehci controller setup check until port
scan in UHCI/OHCI.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
|
|
Signed-off-by: Cao jin <caoj.fnst@cn.fujitsu.com>
|
|
SMBIOS 2.6+ stores the UUID in a different format, with the first 3
fields in little endian format. This is what modern qemu delivers
and what dmidecode also handles, so let's follow suit too.
Signed-off-by: Cole Robinson <crobinso@redhat.com>
|
|
Simplify the scan_to_keycode[] table by implementing numlock and
capslock checking in the code.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
|
|
The scancode/asciicode pair can be more easily handled as a single
16bit value.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
|
|
The HaveRunPost flag controls whether post or reboot handling is
entered on a reset signal. The flag needs to be set before any other
global variable because an external reboot signal could occur at any
time. (If any global variable is modified prior to setting
HaveRunPost then the code might enter post with global variables in a
dirty state.)
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
|
|
Move the check for soft reboot loops from resume.c to shadow.c and
directly check for the case where the copy of the BIOS in flash
appears to be a memory alias instead. This prevents a hang if an
external reboot request occurs during the BIOS memcpy.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
|
|
If the CBFS header is invalid and points to 0xffffffff it could cause
SeaBIOS to read past the 4GB boundary and cause an exception. Check
the alignment of the header pointer before attempting to access fields
within the header.
Reported-by: "Alex G." <mr.nuke.me@gmail.com>
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
|
|
Since we may detect that HW physical presence is enabled but we do not detect
whether it is actually asserted, we may fail on the TPM menu commands that
require the assertion of physical presence. We therefore cannot set the TPM
into failure mode if we hit this case. Failure should never occur in these
cases if SW physical presence has been asserted.
Signed-off-by: Stefan Berger <stefanb@linux.vnet.ibm.com>
|
|
Remove the usage of PhysicalPresence_CMD_ENABLE from all but
the assert_physical_presence function.
Signed-off-by: Stefan Berger <stefanb@linux.vnet.ibm.com>
|
|
Rework the assertion of physical presence by calling assert_physical_presence
in tpm_setup. This call will assert physical presence if SW assertion is
possible or by checking whether HW physical presence is enabled.
The TPM menu will only be shown if physical presence is asserted or HW
physical presence is enabled after this call.
Signed-off-by: Stefan Berger <stefanb@linux.vnet.ibm.com>
|
|
Extend assert_physical_presence with checks for hardware physical presence
support. If hardware physical presence is enabled, we return 0.
Signed-off-by: Stefan Berger <stefanb@linux.vnet.ibm.com>
|
|
Move assert_physical_presence and dependencies in front of tpm_startup
so that the next patches can assert physical presence after TPM_ORD_Startup
ran.
Signed-off-by: Stefan Berger <stefanb@linux.vnet.ibm.com>
|
|
Copy the digest into the response of a HashLogExtendEvent API call.
Signed-off-by: Stefan Berger <stefanb@linux.vnet.ibm.com>
|
|
The usbdev->slotid field is xhci specific and on xhci it is always
reachable from usbdev->defpipe->slotid.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
|
|
Don't use the return codes from the 16bit BIOS spec in the internal
menu functions. Only the 16bit BIOS interface code should need to
handle the details of that spec. For functions that need to return
the TIS command status, return those codes directly instead of via a
pointer parameter.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
|
|
Don't use the return codes from the 16bit BIOS spec in the internal
tpmhw functions. Only the 16bit BIOS interface code should need to
handle the details of that spec.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
|
|
Don't use the return codes from the 16bit BIOS spec in the internal
tpm_log_event() and tpm_log_extend_event() functions. Only the 16bit
BIOS interface code should need to handle the details of that spec.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
|
|
Don't use the return codes from the 16bit BIOS spec in the internal
function build_and_send_cmd(). Instead, return the TIS command status
code of the command or -1 if there was a command transmission failure.
This eliminates the need for a returnCode pointer parameter.
Also, implement debugging dprintf() in build_and_send_cmd() instead of
in every caller. This replaces the command name with the integer
command id, but it does make the debugging more consistent.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
|
|
The 16bit BIOS interface shouldn't be able to shutdown the TPM. Move
the check for tpm_is_working() and tpm_set_failure() to the only
caller of tpm_log_extend_event() that may shutdown the TPM.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
|
|
The low-level measurement functions already handle error conditions,
there is no need to check for the errors in the high level measurement
functions.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
|
|
The callers of the measurements don't care what happens, so no need to
return a status.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
|
|
There are no longer any callers that use the response buffer.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
|
|
Introduce helper function to call the TPM_ORD_GetCapability command.
Update all get capability callers to use this helper.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
|
|
The TPM ACPI tables are only scanned once at startup and the code can
rely on that. Merge find_tcpa_table() into find_tcpa_by_rsdp(), merge
get_lasa_base_ptr() into reset_acpi_log(), and merge tpm_acpi_init()
into tpm_setup().
The tpm_state structure is now only used for TCPA tracking.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
|
|
The 16bit BIOS interface should only shutdown on request from that
interface - errors from the tcp or acpi log setup should not shutdown
the interface. (Errors from those functions will cause the TPM to be
in a "not working" state which will cause all the 16bit interface
functions to fail.)
Centralize the checking for the interface shutdown condition in
tpm_interrupt_handler32().
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
|
|
Move the hardware interface functions (tpmhw_probe, tpmhw_is_present,
tpmhw_transmit, and tpmhw_set_timeouts) to tpm_drivers.c code, and
only export those functions. This simplifies the hardware interface.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
|
|
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
|
|
Perform the hardware probe once during setup instead of checking if
the probe has been completed on each measurement event.
Don't probe for hardware during BIOS interface detection. Just check
if the previous probe found the hardware.
Unify has_working_tpm() with similar tpm_is_working().
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
|
|
Avoid macros and use regular struct definitions for the request and
response headers. This simplifies the header and reduces the need for
casts in the code.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
|
|
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
|
|
Fixes the following build error:
Building ld scripts
Traceback (most recent call last):
File "./scripts/buildversion.py", line 134, in <module>
main()
File "./scripts/buildversion.py", line 114, in main
cleanbuild, toolstr = tool_versions(options.tools)
File "./scripts/buildversion.py", line 90, in tool_versions
vers[isbinutils] = "mixed"
NameError: global name 'vers' is not defined
Makefile:160: recipe for target 'out/romlayout16.lds' failed
Signed-off-by: Roger Pau Monné <roger.pau@citrix.com>
|
|
The sector count is a 64bit number that is often reported as a 32bit
number (due to limitations in dprintf). Consistently use "%u"
reporting to avoid confusing negative numbers.
Reported-by: Tobias Diedrich <tobiasdiedrich@gmail.com>
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
|
|
The scsi_is_ready() function may be called from a thread, and it is
not valid to call printf() from a thread. Convert printf() to
dprintf() to avoid this possibility.
This does mean that cdrom detection (from cdrom_boot() ) may not give
notification of slow cdrom drives to a user. However, the extra
medium detection time is unlikely to be large anyway.
Reported-by: Tobias Diedrich <tobiasdiedrich@gmail.com>
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
|
|
The NMI could occur when already on the extra stack, which would
corrupt it. Always use the current stack on an NMI to avoid this.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
|
|
If the set_address command fails, attempt to free up the slot resource
associated with the attempt.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
|
|
Report the port state on a port status change notification. Clear the
change bits so further change notifications are delivered.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
|
|
Some USB3 controllers (and/or devices) need additional time after the
device is detected to place the port in an enabled state. Wait for
the controller to report enabled before proceeding. This fixes
several reports of devices that showed a "stall" error (cc 4) during
set address.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
|
|
PCIe downstream ports (Root Ports and switches Downstream Ports) appear
to firmware as PCI-PCI bridges and a 4K IO space is allocated for them
even if there is no device behind them requesting IO space,
all that for hotplug purpose.
However, PCIe devices can work without IO, so there is no need
to allocate IO space for hotplug.
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Marcel Apfelbaum <marcel@redhat.com>
|
|
This patch adds an new menu entry to the main menu. This menu item enables
the user to enter a TPM control menu which allows control of those aspects
of the TPM's state that can only be controlled while in the firmware
and while physical presence can be asserted.
If the machine has a TPM, the boot menu will look as follows, with
the new menu item accessible by pressing the 't' key.
Select boot device:
1. ata0-1: QEMU HARDDISK ATA-7 Hard-Disk (6144 MiBytes)
2. Legacy option rom
3. iPXE (PCI 00:03.0)
t. TPM Menu
Upon pressing t the TPM submenu will be shown:
The Trusted Platform Module (TPM) is a hardware device in this machine.
It can help verify the integrity of system software.
The current state of the TPM is:
Enabled and active
Ownership has not been taken
A user can take ownership of the TPM
Available options are:
d. Disable the TPM
v. Deactivate the TPM
p. Prevent installation of an owner
If no change is desired or if this menu was reached by mistake, press ESC to
reboot the machine.
The TPM menu only shows those options that are currently accessible considering
the state of the TPM.
The patch adds several functions for sending those messages to the TPM
required for supporting those menu items.
Signed-off-by: Stefan Berger <stefanb@linux.vnet.ibm.com>
|
|
Check the length parameter that indicates the length of the array
for whether it has a minimum value of 0x200.
Signed-off-by: Stefan Berger <stefanb@linux.vnet.ibm.com>
|
|
Remove the check for a working TPM from the TPM interrupt handler. This then
allows the individual API calls to return information even if the TPM was not
working correctly. Some API calls will still run into the check.
Signed-off-by: Stefan Berger <stefanb@linux.vnet.ibm.com>
|
