Skylake HD530 - Sleeps but won't wake

[goodwin_c]

no HD540 is different with HD530, apple choose HD540, it has not wake issue

You are wrong. Apple chose 540 for low end MBP. But on high end they are using switchable graphics with HD530 and AMD Polaris.

 

[zebrum]

You are wrong. Apple chose 540 for low end MBP. But on high end they are using switchable graphics with HD530 and AMD Polaris.

yep unbelievably the CPU they chose for the high end pro only has the HD 530, those CPU came out Sept 2015 but Intel delayed the Iris Pro quad core mobile CPUs til Jan 2016 and increased the price and dropped the fastest by 100Mhz. I'm guessing Apple waited but then said "screw you Intel!" and went for the earlier model! Intel likely didn't know that Apple had a low-powered AMD partnership up their sleeves.

 

[goodwin_c]

Here is DSDT patch for working HD530 under Sierra. Have fun!

# inject "hda-gfx"="onboard-1" into IGPU (GFX0 originally)

into device label GFX0 set_label begin IGPU end;
into_all all code_regex GFX0 replaceall_matched begin IGPU end;
into method label _DSM parent_label IGPU remove_entry;
into device label IGPU insert begin
Method (_DSM, 4, NotSerialized)\n
{\n
    If (LEqual (Arg2, Zero)) { Return (Buffer() { 0x03 } ) }\n
    Store (Package()\n
        {\n
            "model", Buffer() { "Intel HD Graphics 530" },\n
            "hda-gfx", Buffer() { "onboard-1" },\n
            "AAPL,ig-platform-id", Buffer() { 0x00, 0x00, 0x1b, 0x19 },\n
            "AAPL,Gfx324", Buffer() { 0x00, 0x00, 0x00, 0x01 },\n
            "AAPL,GfxYTile", Buffer() { 0x00, 0x00, 0x00, 0x01 },\n
            "AAPL00,PanelPowerUp", Buffer() { 0x00, 0x00, 0x00, 0x30 },\n
            "AAPL00,PanelPowerOn", Buffer() { 0x00, 0x00, 0x01, 0x19 },\n
            "AAPL00,PanelPowerOff", Buffer() { 0x00, 0x00, 0x00, 0x11 },\n
            "AAPL00,PanelPowerDown", Buffer() { 0x00, 0x00, 0x00, 0x3C },\n
            "AAPL00,PanelCycleDelay", Buffer() { 0x00, 0x00, 0x00, 0xFA },\n
            "device-colors", Buffer() { 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x9D, 0x9D, 0xA0, 0x00, 0x00, 0x02, 0x00  },\n
        }, Local0)\n
    DTGP (Arg0, Arg1, Arg2, Arg3, RefOf (Local0))\n
    Return (Local0)\n
}\n
end;

 

[napoleonn]

You are wrong. Apple chose 540 for low end MBP. But on high end they are using switchable graphics with HD530 and AMD Polaris.

I am saying that there is no macbook pro with just internal HD530, so this wake issue will not be fixed.

Here is DSDT patch for working HD530 under Sierra. Have fun!

# inject "hda-gfx"="onboard-1" into IGPU (GFX0 originally)

into device label GFX0 set_label begin IGPU end;
into_all all code_regex GFX0 replaceall_matched begin IGPU end;
into method label _DSM parent_label IGPU remove_entry;
into device label IGPU insert begin
Method (_DSM, 4, NotSerialized)\n
{\n
    If (LEqual (Arg2, Zero)) { Return (Buffer() { 0x03 } ) }\n
    Store (Package()\n
        {\n
            "model", Buffer() { "Intel HD Graphics 530" },\n
            "hda-gfx", Buffer() { "onboard-1" },\n
            "AAPL,ig-platform-id", Buffer() { 0x00, 0x00, 0x1b, 0x19 },\n
            "AAPL,Gfx324", Buffer() { 0x00, 0x00, 0x00, 0x01 },\n
            "AAPL,GfxYTile", Buffer() { 0x00, 0x00, 0x00, 0x01 },\n
            "AAPL00,PanelPowerUp", Buffer() { 0x00, 0x00, 0x00, 0x30 },\n
            "AAPL00,PanelPowerOn", Buffer() { 0x00, 0x00, 0x01, 0x19 },\n
            "AAPL00,PanelPowerOff", Buffer() { 0x00, 0x00, 0x00, 0x11 },\n
            "AAPL00,PanelPowerDown", Buffer() { 0x00, 0x00, 0x00, 0x3C },\n
            "AAPL00,PanelCycleDelay", Buffer() { 0x00, 0x00, 0x00, 0xFA },\n
            "device-colors", Buffer() { 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x9D, 0x9D, 0xA0, 0x00, 0x00, 0x02, 0x00  },\n
        }, Local0)\n
    DTGP (Arg0, Arg1, Arg2, Arg3, RefOf (Local0))\n
    Return (Local0)\n
}\n
end;

I think this patch data is from https://pikeralpha.wordpress.com/2016/10/30/aapl-properties-for-skylake-graphics/
but it's like the buffer data should be reverse direction for Gfx324, GfxYTile, PanelXXXXXX

 

[goodwin_c]

I am saying that there is no macbook pro with just internal HD530, so this wake issue will not be fixed.


I think this patch data is from https://pikeralpha.wordpress.com/2016/10/30/aapl-properties-for-skylake-graphics/
but it's like the buffer data should be reverse direction for Gfx324, GfxYTile, PanelXXXXXX

Yes it is based on data from Piker's blog. And yes, in buffer data should be in reverse order, as you can see in my patch. And it is already proven that it is working just fine.

 

[bombardier10]

Installed patch as SSDT and it is working very well ) Motherboard Gigabyte Z170 HD3P and MacOS Sierra .
But no wake after sleep at all .

 

[bombardier10]

In config.plist NO Inject Intel and ig-platform-id 0x19120000.

 

[zebrum]

Seems like Intel fixed it in the Windows driver on June 10, 2016, version 15.40.25.4463.
https://downloadmirror.intel.com/26078/eng/ReleaseNotes_15.40.4463.pdf

But also possibly sooner because a beta user stated:
Where did you get the 4331 driver? That includes important fixes that appeared (IIRC) in the 4326 beta driver on the Intel website (includes the fix for wake from sleep hang).
http://en.community.dell.com/owners-club/alienware/f/3746/t/19667727?pi21932=2

So as soon as Apple release the build of Sierra for the 2016 MacBook 15" with Touch Bar it should hopefully contain the newer driver and then this should be resolved. Someone will hopefully copy the driver off for us so can get it before the next version of Sierra is distributed. Hopefully this fix did make it into the Mac version though!

 

[masterscarhead1]

Anyone know if the patch for the menu bar glitch would also fix the sleep wake issue?

 

[beelzebozo]

Anyone know if the patch for the menu bar glitch would also fix the sleep wake issue?

It doesn't.