修复睡眠唤醒后初始化失败的问题

[CLAY-BIOS]

No description provided.

[CLAY-BIOS]

Only 0x522710EC was tested.

[Sniki]

This is amazing, i have Lenovo ThinkPads:

• L440
• T440
• T440S
• X240-

And all of them have the same 0x522710EC card reader.
So with this pull request we should finally be able to still use the card reader after wake from sleep, that's amazing.

Will test within a few days and report back with results.

[cholonam]

@CLAY-BIOS Thanks for the PR. I'll wait to see @Sniki test results before merging. I did not have time to look carefully at the PR (my free time is very limited right now), but it seems it's adding a few seconds of wait every time there is a power change? I'm afraid this may make the kext too slow to respond for other chips which do not have this problem.. did you try with shorter times? Maybe this hardcoded value (1000?) could be made a boot parameter defaulting to zero? I am pretty sure for most people zero or some milliseconds will be enough. Thanks again.

[CLAY-BIOS]

@CLAY-BIOS Thanks for the PR. I'll wait to see @Sniki test results before merging. I did not have time to look carefully at the PR (my free time is very limited right now), but it seems it's adding a few seconds of wait every time there is a power change? I'm afraid this may make the kext too slow to respond for other chips which do not have this problem.. did you try with shorter times? Maybe this hardcoded value (1000?) could be made a boot parameter defaulting to zero? I am pretty sure for most people zero or some milliseconds will be enough. Thanks again.

I did some tests and the card reader (0x522710EC) did not work properly when IOSleep() value was below 300.
Some chips with the same problem may require higher values.
I don't think setting IOSleep() to 1000 will have much impact, and even if I set IOSleep() to 5000, the card reader will still work.

[ryahpalma]

Merge, please.

[cholonam]

@Sniki any news on the testing? @ryahpalma could you test these changes? I cannot test since I do not have this chip.

[Sniki]

@cholonam not yet, i am all ready and prepared for testing but i have macOS only on my 13" MacbookPro 2017 which does only have USB-C Ports and i don't have any dongle as i just recently got it.
Waiting for a friend to bring me his Dongle from USB-C to USB3 so i can make the installer on my USB3 stick and place the prepared EFI for install on my Lenovo ThinkPad T440S.

I also got a couple of SD Cards to do some deep testing and play a bit more with that value and find the most optimal one.
I have both old slow SD Cards and newer ones, those should help find the lowest value that are just enough for even the slow cards to load.

So merging is up to you, i should be able to complete the testing by the end of this weekend.

[cholonam]

Sounds good. I think I'll have some time today to upload a pre-release version, in case somebody else can also test it.

[Charlyo]

I would suggest adding a new file (e.g: rts5227.c) with its own methods instead of modifying this one.

[CLAY-BIOS]

I would suggest adding a new file (e.g: rts5227.c) with its own methods instead of modifying this one.

It doesn't seem to make sense to change here.
I have restored it before merging. What I did was add IOSleep() to Sinetek_rtsx::setPowerState.

[cholonam]

I agree with @CLAY-BIOS (I understand those changes are for rts5227, and they probably will affect 5249 and 525A). If you really need those, I think you should add a new file, rts5227.c, based on this one, and use its functions when the chip is detected to actually be rts5227. Otherwise maintenance is unmanageable.

[Charlyo]

repeated

[cholonam]

I just uploaded a new relase. It basically contains @CLAY-BIOS pull request, but I made it boot-configurable, so to get @CLAY-BIOS code, just add rtsx_sleep_wake_delay_ms=1000 as kernel boot parameter. It should be easy to fine-tune the delay by playing with this parameter.

Since the default value is zero, I made it just a normal release and not a pre-release (nothing should break if the parameter is not used).

[Sniki]

@cholonam i installed macOS Big Sur 11.2.1 yesterday on my Lenovo ThinkPad T440S and i can confirm that the Card Reader works after wake from sleep correctly now.

I haven’t played with the values, will do some testing tonight by testing lower values.

If you leave the SD Card Plugged and you put laptop into sleep mode, once it wakes, it gives you the message that the device was incorrectly removed like it does if you remove a USB device without right click > eject.
But after a second it re-mounts correctly automatically.

But i do believe you do get the same message and same behavior even with standard USB ports.

No mimic linux needed, you can add 5227 to working/supported list.

Thanks a lot !

[CLAY-BIOS]

@Sniki Do you have other chips to test? I believe this method may also be applicable to 0x522a. #12
@cholonam This project may be helpful to you, (https://github.com/hlh-restart/rtsx), Thank you for your contribution.

[Sniki]

@CLAY-BIOS unfortunately no, i don't have any other chip besides 0x5227 model.

I just wanted to mention that sleep values like these may be affected by the level of hackintosh optimizations done.
In my testing case i tested it on my X240 and T440S and they both working perfect with just a value of IOSleep(100)

So for Haswell generations ThinkPad(s), i think IOSleep value of (100) works perfect, tested many many times repeatedly and tried going below 100 but whatever went below 100, it didn't work.