Add WIFI support for RDA target UNO_91H #9501

caixue1102 · 2019-01-25T02:13:18Z

Description

Add support wifi for RDA target UNO_91H.
Tested with GCC_ARM, IAR and ARMCC.
Test report here:
RDA_UNO_91H_test_report.zip

Pull request type

[ ] Fix
[ ] Refactor
[x] Target update
[ ] Functionality change
[ ] Docs update
[ ] Test update
[ ] Breaking change

Reviewers

ciarmcom · 2019-01-25T04:00:30Z

@caixue1102, thank you for your changes.
@ARMmbed/mbed-os-maintainers @ARMmbed/mbed-os-test @ARMmbed/mbed-os-ipcore please review.

0xc0170

Looks fine, just that I would run astyle for implementation files of Mbed OS functionality (like emac port , or wifi interface - RdaWiFiInterface).

+1 for providing licenses for libs

SeppoTakalo

Driver looks OK, but I'm requesting changes for couple of things.

License header copy&pasted from other files and not updated to reflect this work. Now headers don't reflect the original author of these files, but instead say Copyright Arm
Small code style changes requested.
Usage of mbed_mac_address() without providing implementation for it.
Copied LwIP headers without clear explanation of why
tab used as indentation marker in targets.json file.

SeppoTakalo · 2019-01-25T09:49:34Z

features/netsocket/emac-drivers/TARGET_RDA_EMAC/RdaWiFiInterface.cpp

+            }
+        }
+    } else
+        find = true;


Our style guide requires one line if-else clauses to have brackets as well.

SeppoTakalo · 2019-01-25T09:50:19Z

features/netsocket/emac-drivers/TARGET_RDA_EMAC/RdaWiFiInterface.cpp

+        find = true;
+
+    if (find == false) {
+        printf("can not find the ap.\r\n");


Use mbed_trace or other loggers.
I don't think using printf() inside driver is acceptable, as you cannot disable the output from build time.

SeppoTakalo · 2019-01-25T09:52:14Z

features/netsocket/emac-drivers/TARGET_RDA_EMAC/RdaWiFiInterface.cpp

+          _blocking);
+
+    if (ret == NSAPI_ERROR_DHCP_FAILURE)
+        ret = NSAPI_ERROR_CONNECTION_TIMEOUT;


Why do you translate DHCP_FAILURE to CONNECTION_TIMEOUT?

SeppoTakalo · 2019-01-25T09:53:57Z

features/netsocket/emac-drivers/TARGET_RDA_EMAC/RdaWiFiInterface.cpp

+
+
+nsapi_error_t RDAWiFiInterface::connect()
+{


I don't see the point why this cannot call RDAWiFiInterface::connect(const char *ssid, const char *paswd, nsapi_security_t security, uint8_t channel)

SeppoTakalo · 2019-01-25T09:57:21Z

features/netsocket/emac-drivers/TARGET_RDA_EMAC/RdaWiFiInterface.cpp

+    if(sta_state < 2)
+        return NSAPI_ERROR_NO_CONNECTION;
+
+    init();


Why do you call init() here?

I would have assumed you to call power_down() or equivalent.

SeppoTakalo · 2019-01-25T10:01:35Z

features/netsocket/emac-drivers/TARGET_RDA_EMAC/RdaWiFiInterface.h

@@ -0,0 +1,135 @@
+/* LWIP implementation of NetworkInterfaceAPI
+ * Copyright (c) 2019 ARM Limited


Same here, as previous.. You should not copy&paste the license section.

SeppoTakalo · 2019-01-25T10:04:48Z

.../netsocket/emac-drivers/TARGET_RDA_EMAC/lwip-wifi/arch/TARGET_RDA/TARGET_UNO_91H/inc/dhcps.h

+ */
+
+#ifndef __DHCPS_H__
+#define __DHCPS_H__


Why this file is copied from LwIP?

SeppoTakalo · 2019-01-25T10:05:10Z

...et/emac-drivers/TARGET_RDA_EMAC/lwip-wifi/arch/TARGET_RDA/TARGET_UNO_91H/inc/lwipopts_conf.h

+
+
+#ifndef LWIPOPTS_CONF_H
+#define LWIPOPTS_CONF_H


Why this file is copied from LwIP?

SeppoTakalo · 2019-01-25T10:11:22Z

features/netsocket/emac-drivers/TARGET_RDA_EMAC/rda5981x_emac.cpp

+
+bool RDA5981x_EMAC::get_hwaddr(uint8_t *addr) const
+{
+    mbed_mac_address((char *)addr);


I don't see this function used anywhere else in the driver, are you sure that it actually provides the MAC address that this hardware uses?

If you don't provide implementation of mbed_mac_address() it gives you pseudo MAC address that is generated in build time.

SeppoTakalo · 2019-01-25T10:13:10Z

targets/targets.json

-            "ANALOGIN",
-            "FLASH",
-            "TRNG"
+	    "USTICKER",


Tabulator used as indentation here.. Use spaces as the rest of the file.

SeppoTakalo · 2019-01-25T10:18:08Z

Also, I would like to see test results as instructed in https://os.mbed.com/docs/mbed-os/v5.11/porting/wifi-port.html#testing

So provide log file from mbed test command running all network socket tests as well as WiFi tests.

See the Socket test plan what kind of mbed_app.json is required to run it properly.

caixue1102 · 2019-01-28T09:40:16Z

thank you for the suggestions。

coding style changes are made according to your suggestions.
return values in connect function have been modified by determining the cause of the connection errors.
you are right, we finally called RDAWiFiInterface::connect(const char *ssid, const char *paswd, nsapi_security_t security, uint8_t channel) in RDAWiFiInterface::connect();
init() is useless in disconnect, so it has been removed.
Copyright has been modified in all the files.
LwIP headers are useless and they were removed.
mbed_mac_address is used in wifi library function, and final used by the hardware.

caixue1102 · 2019-01-30T07:18:36Z

Also, I would like to see test results as instructed in https://os.mbed.com/docs/mbed-os/v5.11/porting/wifi-port.html#testing

So provide log file from mbed test command running all network socket tests as well as WiFi tests.

See the Socket test plan what kind of mbed_app.json is required to run it properly.

The test report pasted in the description is runned on the newest TEST case （netsocket and network）with 'mbed test' command , is there anything wrong with my operation? could you please give me some pointers?

caixue1102 · 2019-01-30T07:24:11Z

@SeppoTakalo please help to review the new changes, thank you very much~

SeppoTakalo · 2019-01-30T09:44:35Z

@caixue1102 Sorry, I did not spot the attached log files.

Those are fine and look OK.

SeppoTakalo · 2019-01-30T09:50:12Z

features/netsocket/emac-drivers/TARGET_RDA_EMAC/RdaWiFiInterface.cpp

+{
+#if 1
+    return connect(_ssid, _pass, _security, _channel);
+#else


In the future, you might want to remove the dead code from the driver.
When left there, it soon gets outdated, and might be misleading for the next developer who tries to understand this driver.

you are right, and I have remove the dead code just now.

VeijoPesonen · 2019-01-30T11:46:16Z

features/netsocket/emac-drivers/TARGET_RDA_EMAC/rda5981x_emac.cpp

+        }
+
+    }
+    else


Please see https://os.mbed.com/docs/mbed-os/v5.11/contributing/style.html

0xc0170

2 requests, the rest looks fine

SPDX identifier addition
fixing coding style in the implementation files (emac/wifiinterface)

0xc0170 · 2019-01-30T12:42:19Z

features/netsocket/emac-drivers/TARGET_RDA_EMAC/rda5981x_emac.h

@@ -0,0 +1,160 @@
+/* Copyright (c) 2019 Unisoc Communications Inc.
+ *


can you add SPDX identifier to new files please?

* SPDX-License-Identifier: Apache-2.0

caixue1102 · 2019-01-31T03:12:11Z

@0xc0170 @VeijoPesonen Thank you, now coding style is fixed and SPDX identifier is added.

0xc0170 · 2019-01-31T14:03:15Z

CI started

caixue1102 · 2019-02-11T11:17:21Z

@0xc0170 thank you, this has been modified

0xc0170 · 2019-02-11T11:31:18Z

CI restarted

mbed-ci · 2019-02-11T15:16:03Z

Test run: FAILED

Summary: 6 of 8 test jobs failed
Build number : 5
Build artifacts

Failed test jobs:

jenkins-ci/mbed-os-ci_mbed2-build-ARM
jenkins-ci/mbed-os-ci_mbed2-build-IAR
jenkins-ci/mbed-os-ci_build-ARM
jenkins-ci/mbed-os-ci_mbed2-build-GCC_ARM
jenkins-ci/mbed-os-ci_build-IAR
jenkins-ci/mbed-os-ci_build-GCC_ARM

0xc0170 · 2019-02-11T15:20:54Z

Please review build artifacts, there are failures:

[DEBUG] Output: "./cmsis/TARGET_CORTEX_M/mbed_fault_handler.c", line 23: Error: #5: cannot open source input file "device.h": No such file or directory

caixue1102 · 2019-02-12T03:14:49Z

@0xc0170 Thank you，I have modified target.json and now it can be compiled successfully using -m RDA5981X, I didn't find the failed reason before, because I used -m UNO_91H for all tests in the past.and now using -m RDA5981X can not get a emac.bin.

cmonr · 2019-02-12T04:19:26Z

CI started

mbed-ci · 2019-02-12T06:35:25Z

Test run: SUCCESS

Summary: 12 of 12 test jobs passed
Build number : 6
Build artifacts

0xc0170 · 2019-02-12T12:48:01Z

targets/targets.json

-            "ANALOGIN",
-            "FLASH",
-            "TRNG"
+        "USTICKER",


I've noticed this should be aligned as it was previously (4 spaces more to the right) ?

0xc0170 · 2019-02-12T12:48:44Z

@caixue1102 Please send a new PR fixing the alignment or amend it here, let us know (will start CI)

caixue1102 · 2019-02-12T15:12:47Z

@0xc0170 thank you and it's modified.

cmonr · 2019-02-12T15:22:18Z

CI started

mbed-ci · 2019-02-12T17:55:15Z

Test run: FAILED

Summary: 3 of 8 test jobs failed
Build number : 7
Build artifacts

Failed test jobs:

jenkins-ci/mbed-os-ci_build-ARM
jenkins-ci/mbed-os-ci_build-GCC_ARM
jenkins-ci/mbed-os-ci_build-IAR

alekla01 · 2019-02-12T18:27:36Z

Restarted CI after #9677

mbed-ci · 2019-02-12T19:50:56Z

Test run: SUCCESS

Summary: 12 of 12 test jobs passed
Build number : 8
Build artifacts

ciarmcom requested review from a team January 25, 2019 04:00

ciarmcom added the needs: review label Jan 25, 2019

OPpuolitaival approved these changes Jan 25, 2019

View reviewed changes

0xc0170 reviewed Jan 25, 2019

View reviewed changes

SeppoTakalo suggested changes Jan 25, 2019

View reviewed changes

caixue1102 force-pushed the mbed-os-rda branch 2 times, most recently from 3c82a00 to c0cad4e Compare January 28, 2019 09:30

SeppoTakalo approved these changes Jan 30, 2019

View reviewed changes

SeppoTakalo reviewed Jan 30, 2019

View reviewed changes

caixue1102 force-pushed the mbed-os-rda branch from c0cad4e to a55c96c Compare January 30, 2019 10:11

SeppoTakalo approved these changes Jan 30, 2019

View reviewed changes

VeijoPesonen reviewed Jan 30, 2019

View reviewed changes

VeijoPesonen approved these changes Jan 30, 2019

View reviewed changes

0xc0170 requested changes Jan 30, 2019

View reviewed changes

cmonr added needs: work release-version: 5.11.4 and removed needs: review labels Jan 30, 2019

caixue1102 force-pushed the mbed-os-rda branch from a55c96c to 3c6079e Compare January 31, 2019 03:04

cmonr added needs: CI and removed needs: work labels Jan 31, 2019

caixue1102 force-pushed the mbed-os-rda branch from c8c8d92 to ef6d7e1 Compare February 11, 2019 11:15

caixue1102 force-pushed the mbed-os-rda branch from ef6d7e1 to 88888db Compare February 12, 2019 03:12

cmonr added needs: CI and removed needs: work labels Feb 12, 2019

0xc0170 added ready for merge and removed needs: CI labels Feb 12, 2019

0xc0170 reviewed Feb 12, 2019

View reviewed changes

0xc0170 added needs: work and removed ready for merge labels Feb 12, 2019

Add WIFI support for RDA target UNO_91H

d92e33d

caixue1102 force-pushed the mbed-os-rda branch from 88888db to d92e33d Compare February 12, 2019 15:09

0xc0170 added needs: CI and removed needs: work labels Feb 12, 2019

cmonr added ready for merge and removed needs: CI labels Feb 12, 2019

cmonr merged commit 3e6cd4e into ARMmbed:master Feb 13, 2019

cmonr removed the ready for merge label Feb 13, 2019

		@@ -0,0 +1,135 @@
		/* LWIP implementation of NetworkInterfaceAPI
		* Copyright (c) 2019 ARM Limited

+                      }
+                  }
+                  else

		@@ -0,0 +1,160 @@
		/* Copyright (c) 2019 Unisoc Communications Inc.
		*



		nsapi_error_t RDAWiFiInterface::connect()
		{



		#ifndef LWIPOPTS_CONF_H
		#define LWIPOPTS_CONF_H

Add WIFI support for RDA target UNO_91H #9501

Add WIFI support for RDA target UNO_91H #9501

Conversation

caixue1102 commented Jan 25, 2019

Description

Pull request type

Reviewers

ciarmcom commented Jan 25, 2019

0xc0170 left a comment

Choose a reason for hiding this comment

SeppoTakalo left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

SeppoTakalo commented Jan 25, 2019

caixue1102 commented Jan 28, 2019

caixue1102 commented Jan 30, 2019

caixue1102 commented Jan 30, 2019

SeppoTakalo commented Jan 30, 2019

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

0xc0170 left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

caixue1102 commented Jan 31, 2019

0xc0170 commented Jan 31, 2019

caixue1102 commented Feb 11, 2019

0xc0170 commented Feb 11, 2019

mbed-ci commented Feb 11, 2019

0xc0170 commented Feb 11, 2019

caixue1102 commented Feb 12, 2019

cmonr commented Feb 12, 2019

mbed-ci commented Feb 12, 2019

Choose a reason for hiding this comment

0xc0170 commented Feb 12, 2019

caixue1102 commented Feb 12, 2019

cmonr commented Feb 12, 2019

mbed-ci commented Feb 12, 2019

alekla01 commented Feb 12, 2019

mbed-ci commented Feb 12, 2019