Re: [PATCH V2 1/4] firmware: tegra: reword messaging terminology

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 




Hi Jon,

On 24.1.2019 13.33, Jon Hunter wrote:

On 21/01/2019 12:28, Timo Alho wrote:
As a preparatory change to refactor bpmp driver to support other than
t186/t194 chip generations, reword and slightly refactor some of the
functions to better match with what is actually happening in the
wire-level protocol.

The communication with bpmp is essentially a Remote Procedure Call
consisting of "request" and "response". Either side (BPMP or CPU) can
initiate the communication. The state machine for communication
consists of following steps (from Linux point of view):

Linux initiating the call:
  1) check that channel is free to transmit a request (is_req_channel_free)
  2) copy request message payload to shared location
  3) post the request in channel (post_req)
  4) notify BPMP that channel state has been updated (ring_doorbell)
  5) wait for response (is_resp_ready)
  6) copy response message payload from shared location
  7) acknowledge the response in channel (ack_resp)

BPMP initiating the call:
  1) wait for request (is_req_ready)
  2) copy request message payload from shared location
  3) acknowledge the request in channel (ack_req)
  4) check that channel is free to transmit response (is_resp_channel_free)
  5) copy response message payload to shared location
  6) post the response message to channel (post_resp)
  7) notify BPMP that channel state has been updated (ring_doorbell)

Signed-off-by: Timo Alho <talho@xxxxxxxxxx>
---
  drivers/firmware/tegra/bpmp.c | 106 +++++++++++++++++++++++++++---------------
  1 file changed, 68 insertions(+), 38 deletions(-)

diff --git a/drivers/firmware/tegra/bpmp.c b/drivers/firmware/tegra/bpmp.c
index 689478b..af123de 100644
--- a/drivers/firmware/tegra/bpmp.c
+++ b/drivers/firmware/tegra/bpmp.c
@@ -96,7 +96,7 @@ static bool tegra_bpmp_message_valid(const struct tegra_bpmp_message *msg)
  	       (msg->rx.size == 0 || msg->rx.data);
  }
-static bool tegra_bpmp_master_acked(struct tegra_bpmp_channel *channel)
+static bool tegra_bpmp_is_resp_ready(struct tegra_bpmp_channel *channel)
  {
  	void *frame;
@@ -111,7 +111,12 @@ static bool tegra_bpmp_master_acked(struct tegra_bpmp_channel *channel)
  	return true;
  }
-static int tegra_bpmp_wait_ack(struct tegra_bpmp_channel *channel)
+static bool tegra_bpmp_is_req_ready(struct tegra_bpmp_channel *channel)
+{
+	return tegra_bpmp_is_resp_ready(channel);
+}
+

Any reason not to call this something more generic like
tegra_bpmp_is_message_ready()? I am just wondering if you need to have
this additional function and if it can be avoid. However, not a big
deal, so completely your call.

It is true that it looks bit nonsensical in this patch. However, I made it like this in the anticipation of the follow up patches -- the use of separate req_ready() and resp_ready() becomes obvious once t210 bpmp support is added (where req and resp have different implementation).

So, I'd just leave it as is and I see also Acked this patch already. Thanks.

BR,
Timo



[Index of Archives]     [ARM Kernel]     [Linux ARM]     [Linux ARM MSM]     [Linux USB Devel]     [Video for Linux]     [Linux Audio Users]     [Yosemite News]     [Linux Kernel]     [Linux SCSI]

  Powered by Linux