On 22.10.2018 12.12, Thierry Reding wrote:
On Thu, Oct 18, 2018 at 08:55:40PM +0300, Timo Alho wrote:
Add a helper function to check that firmware is supporting a given MRQ
command.
Signed-off-by: Timo Alho <talho@xxxxxxxxxx>
---
drivers/firmware/tegra/bpmp.c | 28 ++++++++++++++++++++++++++++
include/soc/tegra/bpmp.h | 7 +++++++
2 files changed, 35 insertions(+)
diff --git a/drivers/firmware/tegra/bpmp.c b/drivers/firmware/tegra/bpmp.c
index 41448ba..79859ab 100644
--- a/drivers/firmware/tegra/bpmp.c
+++ b/drivers/firmware/tegra/bpmp.c
@@ -470,6 +470,34 @@ void tegra_bpmp_free_mrq(struct tegra_bpmp *bpmp, unsigned int mrq, void *data)
}
EXPORT_SYMBOL_GPL(tegra_bpmp_free_mrq);
+bool tegra_bpmp_mrq_is_supported(struct tegra_bpmp *bpmp, unsigned int mrq)
+{
+ struct mrq_query_abi_request req = { .mrq = cpu_to_le32(mrq) };
+ struct mrq_query_abi_response resp;
+ struct tegra_bpmp_message msg = {
+ .mrq = MRQ_QUERY_ABI,
+ .tx = {
+ .data = &req,
+ .size = sizeof(req),
+ },
+ .rx = {
+ .data = &resp,
+ .size = sizeof(resp),
+ },
+ };
+ int ret;
+
+ ret = tegra_bpmp_transfer(bpmp, &msg);
+ if (ret != 0 || msg.rx.ret != 0) {
Can somebody remind me of the difference between these two? It looks
like we don't have a description of msg.rx.ret anywhere. It comes from
tegra_bpmp_channel.ib->code, which in turn is set directly by BPMP as > part of a transfer.
This is correct. 'ret' is nonzero if the "wire-level" communication
between CPU and BPMP fails. In turn, 'msg.rx.ret' is the first word of
the response payload from bpmp. A nonzero value indicates that the
request made to BPMP was somehow ill-formatted. Possible reasons could
be e.g.:
* Use of non-existing MRQ code
* Use of non-existing resource identifier (clk, reset, power-domain, ...)
What are the values that it can take?
Error codes are listed in very end of bpmp-abi.h
I think we could use some kerneldoc for struct tegra_bpmp_mb_data in
include/soc/tegra/bpmp.h.
+ /* something went wrong; assume not supported */
+ dev_warn(bpmp->dev, "tegra_bpmp_transfer failed\n");
This should probably contain an error code of some sort to help narrow
down why it fails. Also, please use a more human readable error message
for consistency within the driver. Something along these lines perhaps?
dev_warn(bpmp->dev, "transfer failed: %d\n", ret);
Or perhaps more specifically:
dev_warn(bpmp->dev, "ABI query MRQ failed: %d\n", ret);
Sure, I'll make the change.
