Remove the 1 second timeout applied to hw_protection_shutdown after an
EC panic. On some platforms this 1 second timeout is insufficient to
allow the filesystem to fully sync. Independently the EC will force a
full system reset after a short period. So this backup timeout is
unnecessary.

Signed-off-by: Rob Barnes <robbarnes@google.com>
Reviewed-by: Guenter Roeck <groeck@chromium.org>
Link: https://lore.kernel.org/r/20230802175847.1.Ie9fc53b6a1f4c6661c5376286a50e0cf51b3e961@changeid
Signed-off-by: Tzung-Bi Shih <tzungbi@kernel.org>

Update cros_ec_lpc_pm_ops to call cros_ec_lpc_prepare() during PM
.prepare() and cros_ec_lpc_complete() during .complete(). This moves the
host command that the AP sends and allows the EC to log entry/exit of
AP's suspend/resume more accurately.

Reviewed-by: Raul E Rangel <rrangel@chromium.org>
Signed-off-by: Tim Van Patten <timvp@chromium.org>
Signed-off-by: Tzung-Bi Shih <tzungbi@kernel.org>
Link: https://lore.kernel.org/r/20230515142552.1.I17cae37888be3a8683911991602f18e482e7a621@changeid

Create a uevent when an EC panic is detected. This will allow udev rules
to trigger when a panic occurs. For example, a udev rule could be added to
capture an EC coredump. This approach avoids the need to stuff all the
processing into the driver.

Signed-off-by: Rob Barnes <robbarnes@google.com>
Reviewed-by: Prashant Malani <pmalani@chromium.org>
Signed-off-by: Tzung-Bi Shih <tzungbi@kernel.org>
Link: https://lore.kernel.org/r/20230509232624.3120347-1-robbarnes@google.com

Clang static analysis reports this problem
drivers/platform/chrome/cros_ec_lpc.c:379:13: warning: The left operand
  of '!=' is a garbage value [core.UndefinedBinaryOperatorResult]
    if (buf[0] != 'E' || buf[1] != 'C') {
         ~~~~~~ ^

The check depends on the side effect of the read.  When the read fails
or is short, a buf containing garbage could be mistaken as correct.

Signed-off-by: Tom Rix <trix@redhat.com>
Signed-off-by: Tzung-Bi Shih <tzungbi@kernel.org>
Link: https://lore.kernel.org/r/20230110193611.3573777-1-trix@redhat.com

When an EC panic is reported, attempt an orderly shutdown.
Force a shutdown after a brief timeout if the orderly shutdown
fails for any reason.

Using the common hw_protection_shutdown utility function since
an EC panic has the potential to cause hw damage.

This is all best effort. EC should also force a hard reset after a
short timeout.

Signed-off-by: Rob Barnes <robbarnes@google.com>
Reviewed-by: Prashant Malani <pmalani@chromium.org>
Signed-off-by: Tzung-Bi Shih <tzungbi@kernel.org>
Link: https://lore.kernel.org/r/20230104011524.369764-3-robbarnes@google.com

Add handler for CrOS EC panic events. When a panic is reported,
immediately poll for EC log.

This should result in the log leading to the EC panic being
preserved.

ACPI_NOTIFY_CROS_EC_PANIC is defined in coreboot at
https://review.coreboot.org/plugins/gitiles/coreboot/+/refs/heads/master/src/ec/google/chromeec/acpi/ec.asl

Signed-off-by: Rob Barnes <robbarnes@google.com>
Reviewed-by: Prashant Malani <pmalani@chromium.org>
Signed-off-by: Tzung-Bi Shih <tzungbi@kernel.org>
Link: https://lore.kernel.org/r/20230104011524.369764-2-robbarnes@google.com

This reverts commit bd88b965ae8c ("platform/chrome: cros_ec_lpc: Mark
PROBE_PREFER_ASYNCHRONOUS"), and then some.

It has been reported that there are issues with 'cros-ec-keyb' devices
that are children of this. As noted in the initial patch for its ACPI
support (commit ba0f32141bc5 ("Input: cros_ec_keyb - handle x86
detachable/convertible Chromebooks")), it's possible to probe an ACPI
child device before its parent is probed -- hence the need for
EPROBE_DEFER. Unfortunately, poking your parent's dev_get_drvdata()
isn't safe with asynchronous probe, as there's no locking, and the
ordering is all wrong anyway (drvdata is set before the device is
*really* ready).

Because this parent/child relationship has known issues, let's go the
other direction and force synchronous probe, until we resolve the
issues.

Possible solutions involve adding device links, so we ensure the child
doesn't probe before the parent is done; or perhaps some other larger
refactoring (auxiliary bus?). But that might take a little more effort
and review, as there are many other potential sub-devices of
cros_ec_lpc that could need patching.

Note that we don't have the same problem for non-ACPI cros-ec hosts,
like cros-ec-spi (commit 015e4b05c377 ("platform/chrome: cros_ec_spi:
Set PROBE_PREFER_ASYNCHRONOUS")), because its sub-devices aren't created
until cros_ec_register(), or they don't exist at all (e.g., FPMCU uses).

Fixes: bd88b965ae8c ("platform/chrome: cros_ec_lpc: Mark PROBE_PREFER_ASYNCHRONOUS")
Signed-off-by: Brian Norris <briannorris@chromium.org>
Reviewed-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
Reviewed-by: Guenter Roeck <groeck@chromium.org>
Signed-off-by: Tzung-Bi Shih <tzungbi@kernel.org>
Link: https://lore.kernel.org/r/20221111231302.3458191-1-briannorris@chromium.org

This takes on the order of 60ms to probe on some systems, so let it
probe asynchronously. It shouldn't have any dependencies that aren't
handled cleanly.

Signed-off-by: Brian Norris <briannorris@chromium.org>
Signed-off-by: Tzung-Bi Shih <tzungbi@kernel.org>
Link: https://lore.kernel.org/r/20221101152132.v2.2.Ib1036816e77aba71ebc16b71f7615c55d054689c@changeid

Disregarding the weird global state hiding in this cros_ec_lpc_mec_*()
stuff, it belongs in device probe. We shouldn't assume we can access
hardware resources when the device isn't attached to the driver.

Signed-off-by: Brian Norris <briannorris@chromium.org>
Signed-off-by: Tzung-Bi Shih <tzungbi@kernel.org>
Link: https://lore.kernel.org/r/20221101152132.v2.1.I0728421299079b104710c202d5d7095b2674fd8c@changeid

It's pointless (and invalid) to destroy a statically allocated mutex in
cros_ec_lpc_mec_destroy().

Let's remove it.

Signed-off-by: Tzung-Bi Shih <tzungbi@kernel.org>
Reviewed-by: Guenter Roeck <groeck@chromium.org>
Reviewed-by: Brian Norris <briannorris@chromium.org>
Link: https://lore.kernel.org/r/20221031050657.3899359-1-tzungbi@kernel.org