[Telink]: Fix entropy source failed crash on boot

[rikorsev]

Problem

Fix crash on startup for Telink platform with following log:

*** Booting Zephyr OS build zephyr-v2.5.0-7511-g083f3065b44a  ***
I: Notifier: StateChanged (0x00000001) [Ip6+]
I: Init CHIP stack
E: 75 [CR]mbedTLS error: CTR_DRBG - The entropy source failed
E: 80 [DL]Entropy initialization failed: Error CHIP:0x000000AC
E: PlatformMgr().InitChipStack() failed
E: Exited with code ac
I: 8 Sectors of 256 bytes
I: alloc wra: 0, e8
I: data wra: 0, 0
I: Notifier: StateChanged (0x00038200) [NetData PanId NetName ExtPanId]

Investigation shows that the issue occurs due to changes made by this #9420 PR

Did previous solution works for nRF platform?

Any way, according to previous solution, srand() were called twice.
First time in ${chip_root}/src/platform/Zephyr/PlatformManagerImpl.cpp in _InitChipStack
Second time in ${chip_root}/src/platform/Entropy.cpp in InitEntropy

Change overview

• Remove srand call from ${chip_root}/src/platform/Zephyr/PlatformManagerImpl.cpp
• Move Zephyr related entropy driver calls to app_entropy_source function
• Add chip::Crypto::add_entropy_source call in _InitChipStack method

Testing

Manually. Steps:

1. Build lighting example
2. Flash it on Telink device
3. Check if it does not crash on boot
4. Pair device over chip-tool
5. Send chip-tool onoff on 1 command
6. Check if it works correctly

[github-actions]

Size increase report for "esp32-example-build" from e945d16

File	Section	File	VM
chip-ipv6only-app.elf	.flash.text	172	172
chip-temperature-measurement-app.elf	.flash.text	32	32
chip-shell.elf	.flash.text	43	48

Full report output

BLOAT REPORT

Files found only in the build output:
    report.csv

Comparing ./master_artifact/chip-ipv6only-app.elf and ./pull_artifact/chip-ipv6only-app.elf:

sections,vmsize,filesize
[Unmapped],0,3924
.flash.text,172,172

Comparing ./master_artifact/chip-temperature-measurement-app.elf and ./pull_artifact/chip-temperature-measurement-app.elf:

sections,vmsize,filesize
.flash.text,32,32
[Unmapped],0,-32

Comparing ./master_artifact/chip-lock-app.elf and ./pull_artifact/chip-lock-app.elf:

sections,vmsize,filesize

Comparing ./master_artifact/chip-pigweed-app.elf and ./pull_artifact/chip-pigweed-app.elf:

sections,vmsize,filesize

Comparing ./master_artifact/chip-persistent-storage.elf and ./pull_artifact/chip-persistent-storage.elf:

sections,vmsize,filesize

Comparing ./master_artifact/chip-bridge-app.elf and ./pull_artifact/chip-bridge-app.elf:

sections,vmsize,filesize

Comparing ./master_artifact/chip-shell.elf and ./pull_artifact/chip-shell.elf:

sections,vmsize,filesize
[Unmapped],0,4053
.flash.text,48,43

Comparing ./master_artifact/chip-all-clusters-app.elf and ./pull_artifact/chip-all-clusters-app.elf:

sections,vmsize,filesize

[github-actions]

Size increase report for "nrfconnect-example-build" from e945d16

File	Section	File	VM
chip-lock.elf	text	104	104
chip-lock.elf	device_handles	-8	-8
chip-shell.elf	text	104	104
chip-shell.elf	device_handles	-8	-8

Full report output

BLOAT REPORT

Files found only in the build output:
    report.csv

Comparing ./master_artifact/chip-lock.elf and ./pull_artifact/chip-lock.elf:

sections,vmsize,filesize
.debug_info,0,864
.debug_loc,0,217
.debug_line,0,125
text,104,104
.strtab,0,103
.symtab,0,96
.debug_abbrev,0,59
.debug_frame,0,28
.debug_str,0,19
.debug_ranges,0,16
.debug_aranges,0,8
.shstrtab,0,1
device_handles,-8,-8

Comparing ./master_artifact/chip-shell.elf and ./pull_artifact/chip-shell.elf:

sections,vmsize,filesize
.debug_info,0,864
.debug_loc,0,185
.debug_line,0,125
text,104,104
.strtab,0,103
.symtab,0,96
.debug_abbrev,0,59
.debug_frame,0,28
.debug_str,0,19
.debug_ranges,0,16
.debug_aranges,0,8
.shstrtab,0,1
device_handles,-8,-8

[kghost]

Is it possible to automactically verify the issue by CI ? In case we may have regression problems like this one.