port tlucks upgrade delay into rehabman code.

[ESProcessing]

Please review these changes. Added additional delay for:
Broadcom BCM94352Z NGFF M.2 WiFi WLAN Bluetooth 4.0 802.11ac up to 867 Mbps card 22*30mm
on Lenovo T460.

[RehabMan]

Seems the only change effected here is an IOSleep(0) call at line 165.
Because mUpgradeDelay is not initialized until later.

Is that the intention?

[ESProcessing]

That is not the intention. In Tlucks code there are two calls to IOSleep(200). The positions should be in my code, i.e. t470-mbp:BrcmPatchRAM sjk$ fgrep -A 2 -B 2 tjl * BrcmPatchRAM.cpp-#endif BrcmPatchRAM.cpp- BrcmPatchRAM.cpp:// tjl (sjk) port forward. BrcmPatchRAM.cpp- IOSleep(mUpgradeDelay); BrcmPatchRAM.cpp- -- -- BrcmPatchRAM.cpp- DeviceState previousState = kUnknown; BrcmPatchRAM.cpp-#endif BrcmPatchRAM.cpp: // tjl (sjk) port forward. BrcmPatchRAM.cpp- IOSleep(mUpgradeDelay); BrcmPatchRAM.cpp- IOLockLock(mCompletionLock); Cheers, Scott.

…

On Feb 27, 2019, at 10:43 PM, RehabMan ***@***.***> wrote: Seems the only change effected here is an IOSleep(0) call at line 165. Because mUpgradeDelay is not initialized until later. Is that the intention? — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub <#63 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/Aeaub4nGHYCBGrxV9l87DA1kS3ewy_Wwks5vR1B8gaJpZM4bV4BW>.

[ESProcessing]

Just looked at and tested the code. It is not the first IOSleep at line 164 of BrcmPatchRam.cpp, that I just commented out in my fork. It is the one at line 1221 that gives me stability across sleep/wake cycles.

[ESProcessing]

Please check my latest fork. I moved the new variable initialization up, and have the two IOSleep call in.

…

On Feb 27, 2019, at 10:43 PM, RehabMan ***@***.***> wrote: Seems the only change effected here is an IOSleep(0) call at line 165. Because mUpgradeDelay is not initialized until later. Is that the intention? — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub <#63 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/Aeaub4nGHYCBGrxV9l87DA1kS3ewy_Wwks5vR1B8gaJpZM4bV4BW>.

[ESProcessing]

I have been running these changes for 3 weeks without a single drop of BT over many sleep/wake cycles.

bpr_probedelay=200 bpr_upgradedelay=250 bpr_initialdelay=400 bpr_postresetdelay=400

The bpr_upgradedelay=250 is the new delay I ported from TLUCK.