REFPLTB-3230 REFPLTB-3233 REFPLTB-3234: Potential memory leak at PSM_Get_Record_Value2 call #75

pganna856 · 2025-01-09T07:20:12Z

Reason for change: Added code to release the memory.
Test Procedure: 1) Added the change suggested in the ticket to generate the leak in the api which is calling during the service start.
2) Valgrind command is added in service file as given below,
    ExecStart=/bin/sh -c 'valgrind --tool=memcheck --leak-check=yes --show-reachable=yes --num-callers=20 --track-fds=yes /usr/bin/OneWifi -subsys eRT.'
3) Increased the file size in journald.conf file
4) Stopped the service using systemctl command, it will be automatically try to restart using rpiwifiinitialized.path
5) Executed "journalctl -u onewifi" immediately with the stop command, redirected to some text file in the /tmp folder
   Console command: "systemctl stop onewifi.service;journalctl -u onewifi > /tmp/leak_report.txt"
Risks: None

Signed-off-by: Prabhudas Gannavarapu <[email protected]>

Impacted Platforms: All rdkb platforms Reason for change: Used monitor variable instead of core variable in whix to fetch the vap enable status. Test Procedure: 1. Load the device with above mentioned build. 2. Change the log interval to 15mins using below, dmcli eRT setv Device.DeviceInfo.X_RDKCENTRAL-COM_WIFI_TELEMETRY.LogInterval int 300. 3.Check the mesh backhaul status for 13th and 14th vap using, dmcli eRT getv Device.WiFi.SSID.13.Status dmcli eRT getv Device.WiFi.SSID.14.Status 4. Change the enable value of those backhaul vaps in Wifi_VIF_Config table by, /usr/opensync/tools/ovsh U Wifi_VIF_Config -w if_name==wl1.7 enabled:=true /usr/opensync/tools/ovsh U Wifi_VIF_Config -w if_name==wl0.7 enabled:=true 5. Check the status for mesh backhaul vaps again by following step 3. 6. Finally check for the WIFI_VAP_PERCENT_UP marker value contains correct value for the mesh backhaul vaps 13 and 14 as per the changed enable value in /rdklogs/logs/wifihealth.txt file. 7. If the marker doesn't contain the enable value as per the change , then issue is reproduced. Risks: Low Priority: P2 Change-Id: Id4a3c589b691b9177ac24fdbd14d19decadcc941 Signed-off-by:[email protected]

RDKB-58273: VAP UP PERCENT reports incorrect value for offline vaps.

* RDKB-57995: [Onewifi] OneWifi memory Optimization Reason for change: OneWifi memory Optimization and few DML API Abstraction. Below APIs are Abstracted: getDeviceMac sendWebpaMsg push_data_to_ssp_queue getPartnerId Get_PSM_Record_Status Set_PSM_Record_Status Test Procedure: 1) Load OneWifi build. 2) Test Basic sanity. 3) Test Levl and csi feature. Risks: Low Priority: P1 Change-Id: I10400d5b204d4402a6cc2caa90f60a8f2c6671da Signed-off-by: [email protected] * RDKB-57995: [Onewifi] OneWifi memory Optimization Reason for change: OneWifi memory Optimization and few DML API Abstraction. Below APIs are Abstracted: getDeviceMac sendWebpaMsg push_data_to_ssp_queue getPartnerId Get_PSM_Record_Status Set_PSM_Record_Status Test Procedure: 1) Load OneWifi build. 2) Test Basic sanity. 3) Test Levl and csi feature. Risks: Low Priority: P1 Change-Id: I10400d5b204d4402a6cc2caa90f60a8f2c6671da Signed-off-by: [email protected] --------- Signed-off-by: [email protected]

…ntral#71) Impacted Platforms: OneWifi platforms Reason for change: Implement Rbus functions for TCM as part of Epic Test Procedure: Load the build and test the changes as per ACs Risks: Low Signed-off-by:[email protected]

RDKB-58301 Remove remaining pending monitor depend Impacted Platforms: All OneWifi Platform Reason for change: WHiX app framework should not have any dependancy with monitor. Test Procedure: WHiX app should collect stats and work as expected. Priority: P1

Address compilation issues seen in comcast jenkins build

Address few more compilation issue seen on jenkins build.

When the feature flag FEATURE_SUPPORT_ECOPOWERDOWN is enabled, compiler error is seen in webconfig_hal_single_radio_apply function. This commit addresses that.

Fix compile error seen in webconfig_hal_single_radio_apply

…Get_Record_Value2 call Reason for change: Added code to release the memory. Test Procedure: 1) Added the change suggested in the ticket to generate the leak in the api which is calling during the service start. 2) Valgrind command is added in service file as given below, ExecStart=/bin/sh -c 'valgrind --tool=memcheck --leak-check=yes --show-reachable=yes --num-callers=20 --track-fds=yes /usr/bin/OneWifi -subsys eRT.' 3) Increased the file size in journald.conf file 4) Stopped the service using systemctl command, it will be automatically try to restart using rpiwifiinitialized.path 5) Executed "journalctl -u onewifi" immediately with the stop command, redirected to some text file in the /tmp folder Console command: "systemctl stop onewifi.service;journalctl -u onewifi > /tmp/leak_report.txt" Risks: None Signed-off-by: Prabhudas Gannavarapu <[email protected]>

amarnathhullur · 2025-01-09T17:29:47Z

source/ccsp/ccsp_api.c

-            return strValue;
+            strncpy(strVal, strValue, (strlen(strValue) + 1));
+            ((CCSP_MESSAGE_BUS_INFO *)bus_handle)->freefunc(strValue);
+            return strVal;


There is a return of locally allocated static string. This will go out of context when the function returns and the caller might be accessing invalid memory.

amarnathhullur · 2025-01-09T17:30:04Z

source/dml/tr_181/ml/cosa_wifi_internal.c

-            return strValue;
+            strncpy(strVal, strValue, (strlen(strValue) + 1));
+            ((CCSP_MESSAGE_BUS_INFO *)bus_handle)->freefunc(strValue);
+            return strVal;


Same comment as above

address the comments

CLAassistant · 2025-01-28T16:25:34Z

Thank you for your submission! We really appreciate it. Like many open source projects, we ask that you all sign our Contributor License Agreement before we can accept your contribution.
7 out of 8 committers have signed the CLA.

✅ Aniket0606
✅ vivianecordeiro-sky
✅ amarnathhullur
✅ HarshavardhanP1
✅ pganna856
✅ narendradandu
✅ KarthikeyanNanjundan
❌ Sanjay Venkatesan

Sanjay Venkatesan seems not to be a GitHub user. You need a GitHub account to be able to sign the CLA. If you have already a GitHub account, please add the email address used for this commit to your account.
_{You have signed the CLA already but the status is still pending? Let us recheck it.}

narendradandu · 2025-02-06T14:29:52Z

close this, there is already #76 for this.

Sanjay Venkatesan and others added 13 commits January 2, 2025 15:17

Merge pull request rdkcentral#67 from SanjayVenkatesan190/develop

01accff

RDKB-58273: VAP UP PERCENT reports incorrect value for offline vaps.

Typo correction of define CONFIG_HW_CAPABILITIES (rdkcentral#69)

b41a103

Address compilation issues seen in comcast jenkins build

dd679cc

Merge pull request rdkcentral#72 from amarnathhullur/develop

cde56bc

Address compilation issues seen in comcast jenkins build

Address few more compilation issue seen on jenkins build.

fced3ab

Merge pull request rdkcentral#73 from amarnathhullur/develop

bd60e6d

Address few more compilation issue seen on jenkins build.

Fix compile error seen in webconfig_hal_single_radio_apply

8721f17

When the feature flag FEATURE_SUPPORT_ECOPOWERDOWN is enabled, compiler error is seen in webconfig_hal_single_radio_apply function. This commit addresses that.

Merge pull request rdkcentral#74 from amarnathhullur/develop

525b110

Fix compile error seen in webconfig_hal_single_radio_apply

amarnathhullur reviewed Jan 9, 2025

View reviewed changes

narendradandu previously approved these changes Jan 9, 2025

View reviewed changes

VikasBagora force-pushed the develop branch from 525b110 to 48ca2ac Compare January 11, 2025 02:42

pganna856 closed this Feb 6, 2025

pganna856 deleted the mem_leak_3230_33_34 branch February 6, 2025 18:41

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

REFPLTB-3230 REFPLTB-3233 REFPLTB-3234: Potential memory leak at PSM_Get_Record_Value2 call #75

REFPLTB-3230 REFPLTB-3233 REFPLTB-3234: Potential memory leak at PSM_Get_Record_Value2 call #75

pganna856 commented Jan 9, 2025

amarnathhullur Jan 9, 2025

amarnathhullur Jan 9, 2025

CLAassistant commented Jan 28, 2025 •

edited

Loading

narendradandu commented Feb 6, 2025

REFPLTB-3230 REFPLTB-3233 REFPLTB-3234: Potential memory leak at PSM_Get_Record_Value2 call #75

REFPLTB-3230 REFPLTB-3233 REFPLTB-3234: Potential memory leak at PSM_Get_Record_Value2 call #75

Conversation

pganna856 commented Jan 9, 2025

amarnathhullur Jan 9, 2025

Choose a reason for hiding this comment

amarnathhullur Jan 9, 2025

Choose a reason for hiding this comment

CLAassistant commented Jan 28, 2025 • edited Loading

narendradandu commented Feb 6, 2025

CLAassistant commented Jan 28, 2025 •

edited

Loading