Jump to content
ErmaC

Clover problems report & features request

652 posts in this topic

Recommended Posts

Ability to have bigger font size with Text Only mode, and have the menu in the center not top left. On 4K monitors text is very hard to see because the text is too small.

Share this post


Link to post
Share on other sites
Advertisement

A request for ig platform ID in the graphics section for AMD 2200g and 2400g Ryzen Vega chipsets.   Current Clover Graphics screen/Menu only allows Fake ID.   I have managed to spoof the Vega IGPU to KBL CRB and load 3 Intel frame buffers but no accelerators.   I have done this by editing info.plist of KBL kexts.   Having a ig platform option may allow for acceleration for these graphics by spoofing Iris graphics.?

Share this post


Link to post
Share on other sites
On 5/7/2018 at 12:09 AM, Hackmac101 said:

Ability to have bigger font size with Text Only mode, and have the menu in the center not top left. On 4K monitors text is very hard to see because the text is too small.

 

The text is based on the EFI console mode, so the reason it's so small is because that's the size your firmware sets. Usually there's only one or two modes and they don't really change the size of the text, they just draw in differently sized rectangles to allow more characters to be drawn. So changing your resolution to something smaller will make the text bigger, there's no other way to  change the size of the text.

 

On 5/7/2018 at 2:01 AM, Gigamaxx said:

A request for ig platform ID in the graphics section for AMD 2200g and 2400g Ryzen Vega chipsets.   Current Clover Graphics screen/Menu only allows Fake ID.   I have managed to spoof the Vega IGPU to KBL CRB and load 3 Intel frame buffers but no accelerators.   I have done this by editing info.plist of KBL kexts.   Having a ig platform option may allow for acceleration for these graphics by spoofing Iris graphics.?

 

You can inject properties into the _DSM of the device's ACPI with Devices/AddProperties (or you can create your own SSDT that does this and inject it), although not sure that it can do that for those. Also, there is no support for those graphics so you won't be getting it working without another kext to support it. You are confusing the Vega 8 and 11 in those CPUs with the Vega M in the Intels, they are not the same.

Share this post


Link to post
Share on other sites
On 5/8/2018 at 10:47 AM, apianti said:

 

The text is based on the EFI console mode, so the reason it's so small is because that's the size your firmware sets. Usually there's only one or two modes and they don't really change the size of the text, they just draw in differently sized rectangles to allow more characters to be drawn. So changing your resolution to something smaller will make the text bigger, there's no other way to  change the size of the text.

 

 

You can inject properties into the _DSM of the device's ACPI with Devices/AddProperties (or you can create your own SSDT that does this and inject it), although not sure that it can do that for those. Also, there is no support for those graphics so you won't be getting it working without another kext to support it. You are confusing the Vega 8 and 11 in those CPUs with the Vega M in the Intels, they are not the same.

 

I am pretty clear on the Intel vs. AMD Vega chips, not confused .   I am testing both and unfortunately both have only 7mb support so far.  

 

The Intel  i7 8705G has HBM2 and HD 630 but it doesn't have native AMD graphics support.   The closest Vega IDs are 0x69A0, and 0x69AF but neither works to activate the graphics.   These Intel chips have 630 graphics but they are not dsesigned to get acceleration only quick synch so accelerators won't load for these Kaby Lake graphics.

 

The AMD 2200g and 2400g vega chips are using DDR4 memory much like Intel chipsets and cost about 1/5 the price.   However I have been able to edit the info.plist of the Kaby Lake graphics kexts and get KBL CRB listing and Intel frame buffers loaded, but again only 7mb support.   I tested all the AMD kexts edits and Intel as well only the Kaby Lake registers.   Having a ig platform for ATI graphics in Clover would be a nice "trick" tool for testing and may get these chips to achieve acceleration.    ig ID 0x59E30003 may link the accelerators and get these working but I have no option for this.    Could you provide an example on how to do this in Add properties using clover?

 

***Look at the device IDs and Vendor info in the DCPIManager it has both AMD and intel codes.

 

 

Screen Shot 2018-05-10 at 9.27.21 AM.png

Screen Shot 2018-05-11 at 12.07.55 AM.png

Edited by Gigamaxx

Share this post


Link to post
Share on other sites

Fix for 32-bit builds with CLANG38 (stock EDK2, no patches; same as vit's fix for kernel_patcher.c):

 

Spoiler

diff --git a/rEFIt_UEFI/Platform/kext_patcher.c b/rEFIt_UEFI/Platform/kext_patcher.c
index c1b46e82..ddcbf182 100755
--- a/rEFIt_UEFI/Platform/kext_patcher.c
+++ b/rEFIt_UEFI/Platform/kext_patcher.c
@@ -28,7 +28,7 @@ UINTN SearchAndCount(UINT8 *Source, UINT64 SourceSize, UINT8 *Search, UINTN Sear
 {
   UINTN     NumFounds = 0;
   UINT8     *End = Source + SourceSize;
-  
+
   while (Source < End) {
     if (CompareMem(Source, Search, SearchSize) == 0) {
       NumFounds++;
@@ -55,7 +55,7 @@ UINTN SearchAndReplace(UINT8 *Source, UINT64 SourceSize, UINT8 *Search, UINTN Se
   if (!Source || !Search || !Replace || !SearchSize) {
     return 0;
   }
-  
+
   while ((Source < End) && (NoReplacesRestriction || (MaxReplaces > 0))) {
     if (CompareMem(Source, Search, SearchSize) == 0) {
       CopyMem(Source, Replace, SearchSize);
@@ -81,7 +81,7 @@ UINTN SearchAndReplaceTxt(UINT8 *Source, UINT64 SourceSize, UINT8 *Search, UINTN
   if (!Source || !Search || !Replace || !SearchSize) {
     return 0;
   }
-  
+
   while (((Source + SearchSize) <= End) &&
          (NoReplacesRestriction || (MaxReplaces > 0))) { // num replaces
     while (*Source != '\0') {  //comparison
@@ -101,19 +101,19 @@ UINTN SearchAndReplaceTxt(UINT8 *Source, UINT64 SourceSize, UINT8 *Search, UINTN
         Source++;
         Pos++;
       }
-      
+
       if (Pos == SearchEnd) { // pattern found
         Pos = FirstMatch;
         break;
       }
       else
         Pos = NULL;
-      
+
       Source = FirstMatch + 1;
 /*      if (Pos != Search) {
         AsciiPrint("\n");
       } */
-      
+
     }
 
     if (!Pos) {
@@ -138,10 +138,10 @@ VOID ExtractKextBundleIdentifier(CHAR8 *Plist)
   CHAR8     *BIStart;
   CHAR8     *BIEnd;
   INTN      DictLevel = 0;
-  
-  
+
+
   gKextBundleIdentifier[0] = '\0';
-  
+
   // start with first <dict>
   Tag = AsciiStrStr(Plist, "<dict>");
   if (Tag == NULL) {
@@ -149,19 +149,19 @@ VOID ExtractKextBundleIdentifier(CHAR8 *Plist)
   }
   Tag += 6;
   DictLevel++;
-  
+
   while (*Tag != '\0') {
-    
+
     if (AsciiStrnCmp(Tag, "<dict>", 6) == 0) {
       // opening dict
       DictLevel++;
       Tag += 6;
-      
+
     } else if (AsciiStrnCmp(Tag, "</dict>", 7) == 0) {
       // closing dict
       DictLevel--;
       Tag += 7;
-      
+
     } else if (DictLevel == 1 && AsciiStrnCmp(Tag, "<key>CFBundleIdentifier</key>", 29) == 0) {
       // BundleIdentifier is next <string>...</string>
       BIStart = AsciiStrStr(Tag + 29, "<string>");
@@ -178,7 +178,7 @@ VOID ExtractKextBundleIdentifier(CHAR8 *Plist)
     } else {
       Tag++;
     }
-    
+
     // advance to next tag
     while (*Tag != '<' && *Tag != '\0') {
       Tag++;
@@ -221,7 +221,7 @@ VOID ATIConnectorsPatchInit(LOADER_ENTRY *Entry)
   //
   // prepar boundle ids
   //
-  
+
   // Lion, SnowLeo 10.6.7 2011 MBP
   AsciiSPrint(ATIKextBundleId[0],
               sizeof(ATIKextBundleId[0]),
@@ -234,9 +234,9 @@ VOID ATIConnectorsPatchInit(LOADER_ENTRY *Entry)
               "com.apple.kext.AMD%sController",
               Entry->KernelAndKextPatches->KPATIConnectorsController
               );
-  
+
   ATIConnectorsPatchInited = TRUE;
-  
+
   //DBG(L"Bundle1: %a\n", ATIKextBundleId[0]);
   //DBG(L"Bundle2: %a\n", ATIKextBundleId[1]);
   //gBS->Stall(10000000);
@@ -247,7 +247,7 @@ VOID ATIConnectorsPatchInit(LOADER_ENTRY *Entry)
 //
 VOID ATIConnectorsPatchRegisterKexts(FSINJECTION_PROTOCOL *FSInject, FSI_STRING_LIST *ForceLoadKexts, LOADER_ENTRY *Entry)
 {
-  
+
   // for future?
   FSInject->AddStringToList(ForceLoadKexts,
                             PoolPrint(L"\\AMD%sController.kext\\Contents\\Info.plist", Entry->KernelAndKextPatches->KPATIConnectorsController)
@@ -259,7 +259,7 @@ VOID ATIConnectorsPatchRegisterKexts(FSINJECTION_PROTOCOL *FSInject, FSI_STRING_
   // SnowLeo
   FSInject->AddStringToList(ForceLoadKexts, L"\\ATIFramebuffer.kext\\Contents\\Info.plist");
   FSInject->AddStringToList(ForceLoadKexts, L"\\AMDFramebuffer.kext\\Contents\\Info.plist");
-  
+
   // dependencies
   FSInject->AddStringToList(ForceLoadKexts, L"\\IOGraphicsFamily.kext\\Info.plist");
   FSInject->AddStringToList(ForceLoadKexts, L"\\ATISupport.kext\\Contents\\Info.plist");
@@ -274,14 +274,14 @@ VOID ATIConnectorsPatchRegisterKexts(FSINJECTION_PROTOCOL *FSInject, FSI_STRING_
 //
 VOID ATIConnectorsPatch(UINT8 *Driver, UINT32 DriverSize, CHAR8 *InfoPlist, UINT32 InfoPlistSize, LOADER_ENTRY *Entry)
 {
-  
+
   UINTN   Num = 0;
-  
+
   DBG_RT(Entry, "\nATIConnectorsPatch: driverAddr = %x, driverSize = %x\nController = %s\n",
          Driver, DriverSize, Entry->KernelAndKextPatches->KPATIConnectorsController);
   ExtractKextBundleIdentifier(InfoPlist);
   DBG_RT(Entry, "Kext: %a\n", gKextBundleIdentifier);
-  
+
   // number of occurences od Data should be 1
   Num = SearchAndCount(Driver, DriverSize, Entry->KernelAndKextPatches->KPATIConnectorsData, Entry->KernelAndKextPatches->KPATIConnectorsDataLen);
   if (Num > 1) {
@@ -290,7 +290,7 @@ VOID ATIConnectorsPatch(UINT8 *Driver, UINT32 DriverSize, CHAR8 *InfoPlist, UINT
     gBS->Stall(5*1000000);
     return;
   }
-  
+
   // patch
   Num = SearchAndReplace(Driver,
                          DriverSize,
@@ -322,9 +322,9 @@ VOID ATIConnectorsPatch(UINT8 *Driver, UINT32 DriverSize, CHAR8 *InfoPlist, UINT
 // Rehabman corrections 2014
 //
 
-UINT8   MovlE2ToEcx[] = { 0xB9, 0xE2, 0x00, 0x00, 0x00 };
-UINT8   MovE2ToCx[]   = { 0x66, 0xB9, 0xE2, 0x00 };
-UINT8   Wrmsr[]       = { 0x0F, 0x30 };
+STATIC UINT8   MovlE2ToEcx[] = { 0xB9, 0xE2, 0x00, 0x00, 0x00 };
+STATIC UINT8   MovE2ToCx[]   = { 0x66, 0xB9, 0xE2, 0x00 };
+STATIC UINT8   Wrmsr[]       = { 0x0F, 0x30 };
 
 VOID AppleIntelCPUPMPatch(UINT8 *Driver, UINT32 DriverSize, CHAR8 *InfoPlist, UINT32 InfoPlistSize, LOADER_ENTRY *Entry)
 {
@@ -397,18 +397,18 @@ VOID AppleIntelCPUPMPatch(UINT8 *Driver, UINT32 DriverSize, CHAR8 *InfoPlist, UI
 // http://www.insanelymac.com/forum/index.php?showtopic=276066
 //
 
-UINT8   LionSearch_X64[]  = { 0x75, 0x30, 0x44, 0x89, 0xf8 };
-UINT8   LionReplace_X64[] = { 0xeb, 0x30, 0x44, 0x89, 0xf8 };
+STATIC UINT8   LionSearch_X64[]  = { 0x75, 0x30, 0x44, 0x89, 0xf8 };
+STATIC UINT8   LionReplace_X64[] = { 0xeb, 0x30, 0x44, 0x89, 0xf8 };
 
-UINT8   LionSearch_i386[]  = { 0x75, 0x3d, 0x8b, 0x75, 0x08 };
-UINT8   LionReplace_i386[] = { 0xeb, 0x3d, 0x8b, 0x75, 0x08 };
+STATIC UINT8   LionSearch_i386[]  = { 0x75, 0x3d, 0x8b, 0x75, 0x08 };
+STATIC UINT8   LionReplace_i386[] = { 0xeb, 0x3d, 0x8b, 0x75, 0x08 };
 
-UINT8   MLSearch[]  = { 0x75, 0x30, 0x89, 0xd8 };
-UINT8   MLReplace[] = { 0xeb, 0x30, 0x89, 0xd8 };
+STATIC UINT8   MLSearch[]  = { 0x75, 0x30, 0x89, 0xd8 };
+STATIC UINT8   MLReplace[] = { 0xeb, 0x30, 0x89, 0xd8 };
 
 //SunKi
-UINT8   MavSearch[]  = { 0x75, 0x2e, 0x0f, 0xb6 };
-UINT8   MavReplace[] = { 0xeb, 0x2e, 0x0f, 0xb6 };
+STATIC UINT8   MavSearch[]  = { 0x75, 0x2e, 0x0f, 0xb6 };
+STATIC UINT8   MavReplace[] = { 0xeb, 0x2e, 0x0f, 0xb6 };
 
 //
 // We can not rely on OSVersion global variable for OS version detection,
@@ -420,19 +420,19 @@ UINT8   MavReplace[] = { 0xeb, 0x2e, 0x0f, 0xb6 };
 
 VOID AppleRTCPatch(UINT8 *Driver, UINT32 DriverSize, CHAR8 *InfoPlist, UINT32 InfoPlistSize, LOADER_ENTRY *Entry)
 {
-  
+
   UINTN   Num = 0;
   UINTN   NumLion_X64 = 0;
   UINTN   NumLion_i386 = 0;
   UINTN   NumML = 0;
   UINTN   NumMav = 0;
-  
+
   DBG_RT(Entry, "\nAppleRTCPatch: driverAddr = %x, driverSize = %x\n", Driver, DriverSize);
   if (Entry->KernelAndKextPatches->KPDebug) {
     ExtractKextBundleIdentifier(InfoPlist);
   }
   DBG_RT(Entry, "Kext: %a\n", gKextBundleIdentifier);
-  
+
   if (is64BitKernel) {
     NumLion_X64 = SearchAndCount(Driver, DriverSize, LionSearch_X64, sizeof(LionSearch_X64));
     NumML  = SearchAndCount(Driver, DriverSize, MLSearch,  sizeof(MLSearch));
@@ -440,7 +440,7 @@ VOID AppleRTCPatch(UINT8 *Driver, UINT32 DriverSize, CHAR8 *InfoPlist, UINT32 In
   } else {
     NumLion_i386 = SearchAndCount(Driver, DriverSize, LionSearch_i386, sizeof(LionSearch_i386));
   }
-  
+
   if (NumLion_X64 + NumLion_i386 + NumML + NumMav > 1) {
     // more then one pattern found - we do not know what to do with it
     // and we'll skipp it
@@ -449,7 +449,7 @@ VOID AppleRTCPatch(UINT8 *Driver, UINT32 DriverSize, CHAR8 *InfoPlist, UINT32 In
     gBS->Stall(5000000);
     return;
   }
-  
+
   if (NumLion_X64 == 1) {
     Num = SearchAndReplace(Driver, DriverSize, LionSearch_X64, sizeof(LionSearch_X64), LionReplace_X64, 1);
     DBG_RT(Entry, "==> Lion X64: %d replaces done.\n", Num);
@@ -508,8 +508,8 @@ VOID CheckForFakeSMC(CHAR8 *InfoPlist, LOADER_ENTRY *Entry)
 //
 // EB9D2D31-2D88-11D3-9A16-0090273F -> EB9D2D35-2D88-11D3-9A16-0090273F
 //
-UINT8   DELL_SMBIOS_GUID_Search[]  = { 0x45, 0x42, 0x39, 0x44, 0x32, 0x44, 0x33, 0x31 };
-UINT8   DELL_SMBIOS_GUID_Replace[] = { 0x45, 0x42, 0x39, 0x44, 0x32, 0x44, 0x33, 0x35 };
+STATIC STATIC UINT8   DELL_SMBIOS_GUID_Search[]  = { 0x45, 0x42, 0x39, 0x44, 0x32, 0x44, 0x33, 0x31 };
+STATIC STATIC UINT8   DELL_SMBIOS_GUID_Replace[] = { 0x45, 0x42, 0x39, 0x44, 0x32, 0x44, 0x33, 0x35 };
 
 //
 // EB9D2D31-2D88-11D3-9A16-0090273F is the standard SMBIOS Table Type 1 for
@@ -525,19 +525,19 @@ VOID DellSMBIOSPatch(UINT8 *Driver, UINT32 DriverSize, CHAR8 *InfoPlist, UINT32
     // than Yosemite, they are all pure 64bit platforms
     //
     UINTN gPatchCount = 0;
-    
+
     DBG_RT(Entry, "\nDellSMBIOSPatch: driverAddr = %x, driverSize = %x\n", Driver, DriverSize);
     if (Entry->KernelAndKextPatches->KPDebug)
     {
         ExtractKextBundleIdentifier(InfoPlist);
     }
     DBG_RT(Entry, "Kext: %a\n", gKextBundleIdentifier);
-    
+
     //
     // now, let's patch it!
     //
     gPatchCount = SearchAndReplace(Driver, DriverSize, DELL_SMBIOS_GUID_Search, sizeof(DELL_SMBIOS_GUID_Search), DELL_SMBIOS_GUID_Replace, 1);
-    
+
     if (gPatchCount == 1)
     {
         DBG_RT(Entry, "==> AppleSMBIOS: %d replaces done.\n", gPatchCount);
@@ -564,24 +564,24 @@ VOID SNBE_AICPUPatch(UINT8 *Driver, UINT32 DriverSize, CHAR8 *InfoPlist, UINT32
 {
     UINT32 i;
     UINT64 os_ver = AsciiOSVersionToUint64(Entry->OSVersion);
-    
+
     DBG_RT(Entry, "\nSNBE_AICPUPatch: driverAddr = %x, driverSize = %x\n", Driver, DriverSize);
     if (Entry->KernelAndKextPatches->KPDebug) {
         ExtractKextBundleIdentifier(InfoPlist);
     }
-    
+
     DBG_RT(Entry, "Kext: %a\n", gKextBundleIdentifier);
-    
+
     // now let's patch it
     if (os_ver < AsciiOSVersionToUint64("10.9") || os_ver >= AsciiOSVersionToUint64("10.14")) {
         DBG("Unsupported macOS.\nSandyBridge-E requires macOS 10.9 - 10.13.x, aborted\n");
         DBG("SNBE_AICPUPatch() <===FALSE\n");
         return;
     }
-    
+
     if (os_ver < AsciiOSVersionToUint64("10.10")) {
         // 10.9.x
-        UINT8 find[][3] = {
+        STATIC UINT8 find[][3] = {
             { 0x84, 0x2F, 0x01 },
             { 0x3E, 0x75, 0x3A },
             { 0x84, 0x5F, 0x01 },
@@ -590,7 +590,7 @@ VOID SNBE_AICPUPatch(UINT8 *Driver, UINT32 DriverSize, CHAR8 *InfoPlist, UINT32
             { 0xFC, 0x02, 0x74 },
             { 0x01, 0x74, 0x58 }
         };
-        UINT8 repl[][3] = {
+        STATIC UINT8 repl[][3] = {
             { 0x85, 0x2F, 0x01 },
             { 0x3E, 0x90, 0x90 },
             { 0x85, 0x5F, 0x01 },
@@ -599,7 +599,7 @@ VOID SNBE_AICPUPatch(UINT8 *Driver, UINT32 DriverSize, CHAR8 *InfoPlist, UINT32
             { 0xFC, 0x02, 0xEB },
             { 0x01, 0xEB, 0x58 }
         };
-        
+
         for (i = 0; i < 7; i++) {
             if (SearchAndReplace(Driver, DriverSize, find[i], sizeof(find[i]), repl[i], 0)) {
                 DBG("SNBE_AICPUPatch (%d/7) applied\n", i);
@@ -609,12 +609,12 @@ VOID SNBE_AICPUPatch(UINT8 *Driver, UINT32 DriverSize, CHAR8 *InfoPlist, UINT32
         }
     } else if (os_ver < AsciiOSVersionToUint64("10.11")) {
         // 10.10.x
-        UINT8 find[][3] = {
+        STATIC UINT8 find[][3] = {
             { 0x3E, 0x75, 0x39 },
             { 0x74, 0x11, 0xB9 },
             { 0x01, 0x74, 0x56 }
         };
-        UINT8 repl[][3] = {
+        STATIC UINT8 repl[][3] = {
             { 0x3E, 0x90, 0x90 },
             { 0xEB, 0x11, 0xB9 },
             { 0x01, 0xEB, 0x56 }
@@ -626,34 +626,34 @@ VOID SNBE_AICPUPatch(UINT8 *Driver, UINT32 DriverSize, CHAR8 *InfoPlist, UINT32
                 DBG("SNBE_AICPUPatch (%d/7) not apply\n", i);
             }
         }
-        
-        UINT8 find_1[] = { 0xFF, 0x0F, 0x84, 0x2D };
-        UINT8 repl_1[] = { 0xFF, 0x0F, 0x85, 0x2D };
+
+        STATIC UINT8 find_1[] = { 0xFF, 0x0F, 0x84, 0x2D };
+        STATIC UINT8 repl_1[] = { 0xFF, 0x0F, 0x85, 0x2D };
         if (SearchAndReplace(Driver, DriverSize, find_1, sizeof(find_1), repl_1, 0)) {
             DBG("SNBE_AICPUPatch (4/7) applied\n");
         } else {
             DBG("SNBE_AICPUPatch (4/7) not apply\n");
         }
-        
-        
-        UINT8 find_2[] = { 0x01, 0x00, 0x01, 0x0F, 0x84 };
-        UINT8 repl_2[] = { 0x01, 0x00, 0x01, 0x0F, 0x85 };
+
+
+        STATIC UINT8 find_2[] = { 0x01, 0x00, 0x01, 0x0F, 0x84 };
+        STATIC UINT8 repl_2[] = { 0x01, 0x00, 0x01, 0x0F, 0x85 };
         if (SearchAndReplace(Driver, DriverSize, find_2, sizeof(find_2), repl_2, 0)) {
             DBG("SNBE_AICPUPatch (5/7) applied\n");
         } else {
             DBG("SNBE_AICPUPatch (5/7) not apply\n");
         }
-        
-        UINT8 find_3[] = { 0x02, 0x74, 0x0B, 0x41, 0x83, 0xFC, 0x03, 0x75, 0x22, 0xB9, 0x02, 0x06 };
-        UINT8 repl_3[] = { 0x02, 0xEB, 0x0B, 0x41, 0x83, 0xFC, 0x03, 0x75, 0x22, 0xB9, 0x02, 0x06 };
+
+        STATIC UINT8 find_3[] = { 0x02, 0x74, 0x0B, 0x41, 0x83, 0xFC, 0x03, 0x75, 0x22, 0xB9, 0x02, 0x06 };
+        STATIC UINT8 repl_3[] = { 0x02, 0xEB, 0x0B, 0x41, 0x83, 0xFC, 0x03, 0x75, 0x22, 0xB9, 0x02, 0x06 };
         if (SearchAndReplace(Driver, DriverSize, find_3, sizeof(find_3), repl_3, 0)) {
             DBG("SNBE_AICPUPatch (6/7) applied\n");
         } else {
             DBG("SNBE_AICPUPatch (6/7) not apply\n");
         }
-        
-        UINT8 find_4[] = { 0x74, 0x0B, 0x41, 0x83, 0xFC, 0x03, 0x75, 0x11, 0xB9, 0x42, 0x06, 0x00 };
-        UINT8 repl_4[] = { 0xEB, 0x0B, 0x41, 0x83, 0xFC, 0x03, 0x75, 0x11, 0xB9, 0x42, 0x06, 0x00 };
+
+        STATIC UINT8 find_4[] = { 0x74, 0x0B, 0x41, 0x83, 0xFC, 0x03, 0x75, 0x11, 0xB9, 0x42, 0x06, 0x00 };
+        STATIC UINT8 repl_4[] = { 0xEB, 0x0B, 0x41, 0x83, 0xFC, 0x03, 0x75, 0x11, 0xB9, 0x42, 0x06, 0x00 };
         if (SearchAndReplace(Driver, DriverSize, find_4, sizeof(find_4), repl_4, 0)) {
             DBG("SNBE_AICPUPatch (7/7) applied\n");
         } else {
@@ -661,12 +661,12 @@ VOID SNBE_AICPUPatch(UINT8 *Driver, UINT32 DriverSize, CHAR8 *InfoPlist, UINT32
         }
     } else if (os_ver < AsciiOSVersionToUint64("10.12")) {
         // 10.11
-        UINT8 find[][3] = {
+        STATIC UINT8 find[][3] = {
             { 0x3E, 0x75, 0x39 },
             { 0x75, 0x11, 0xB9 },
             { 0x01, 0x74, 0x5F }
         };
-        UINT8 repl[][3] = {
+        STATIC UINT8 repl[][3] = {
             { 0x3E, 0x90, 0x90 },
             { 0xEB, 0x11, 0xB9 },
             { 0x01, 0xEB, 0x5F }
@@ -678,33 +678,33 @@ VOID SNBE_AICPUPatch(UINT8 *Driver, UINT32 DriverSize, CHAR8 *InfoPlist, UINT32
                 DBG("SNBE_AICPUPatch (%d/7) not apply\n", i);
             }
         }
-        
-        UINT8 find_1[] = { 0xFF, 0x0F, 0x84, 0x2D };
-        UINT8 repl_1[] = { 0xFF, 0x0F, 0x85, 0x2D };
+
+        STATIC UINT8 find_1[] = { 0xFF, 0x0F, 0x84, 0x2D };
+        STATIC UINT8 repl_1[] = { 0xFF, 0x0F, 0x85, 0x2D };
         if (SearchAndReplace(Driver, DriverSize, find_1, sizeof(find_1), repl_1, 0)) {
             DBG("SNBE_AICPUPatch (4/7) applied\n");
         } else {
             DBG("SNBE_AICPUPatch (4/7) not apply\n");
         }
-        
-        UINT8 find_2[] = { 0x01, 0x00, 0x01, 0x0F, 0x84 };
-        UINT8 repl_2[] = { 0x01, 0x00, 0x01, 0x0F, 0x85 };
+
+        STATIC UINT8 find_2[] = { 0x01, 0x00, 0x01, 0x0F, 0x84 };
+        STATIC UINT8 repl_2[] = { 0x01, 0x00, 0x01, 0x0F, 0x85 };
         if (SearchAndReplace(Driver, DriverSize, find_2, sizeof(find_2), repl_2, 0)) {
             DBG("SNBE_AICPUPatch (5/7) applied\n");
         } else {
             DBG("SNBE_AICPUPatch (5/7) not apply\n");
         }
-        
-        UINT8 find_3[] = { 0xC9, 0x74, 0x16, 0x0F, 0x32, 0x48, 0x25, 0xFF, 0x0F, 0x00, 0x00, 0x48 };
-        UINT8 repl_3[] = { 0xC9, 0xEB, 0x16, 0x0F, 0x32, 0x48, 0x25, 0xFF, 0x0F, 0x00, 0x00, 0x48 };
+
+        STATIC UINT8 find_3[] = { 0xC9, 0x74, 0x16, 0x0F, 0x32, 0x48, 0x25, 0xFF, 0x0F, 0x00, 0x00, 0x48 };
+        STATIC UINT8 repl_3[] = { 0xC9, 0xEB, 0x16, 0x0F, 0x32, 0x48, 0x25, 0xFF, 0x0F, 0x00, 0x00, 0x48 };
         if (SearchAndReplace(Driver, DriverSize, find_3, sizeof(find_3), repl_3, 0)) {
             DBG("SNBE_AICPUPatch (6/7) applied\n");
         } else {
             DBG("SNBE_AICPUPatch (6/7) not apply\n");
         }
-        
-        UINT8 find_4[] = { 0xC9, 0x74, 0x0C, 0x0F, 0x32, 0x83, 0xE0, 0x1F, 0x42, 0x89, 0x44, 0x3B };
-        UINT8 repl_4[] = { 0xC9, 0xEB, 0x0C, 0x0F, 0x32, 0x83, 0xE0, 0x1F, 0x42, 0x89, 0x44, 0x3B };
+
+        STATIC UINT8 find_4[] = { 0xC9, 0x74, 0x0C, 0x0F, 0x32, 0x83, 0xE0, 0x1F, 0x42, 0x89, 0x44, 0x3B };
+        STATIC UINT8 repl_4[] = { 0xC9, 0xEB, 0x0C, 0x0F, 0x32, 0x83, 0xE0, 0x1F, 0x42, 0x89, 0x44, 0x3B };
         if (SearchAndReplace(Driver, DriverSize, find_4, sizeof(find_4), repl_4, 0)) {
             DBG("SNBE_AICPUPatch (7/7) applied\n");
         } else {
@@ -712,12 +712,12 @@ VOID SNBE_AICPUPatch(UINT8 *Driver, UINT32 DriverSize, CHAR8 *InfoPlist, UINT32
         }
     } else if (os_ver < AsciiOSVersionToUint64("10.13")) {
         // 10.12
-        UINT8 find[][3] = {
+        STATIC UINT8 find[][3] = {
             { 0x01, 0x74, 0x61 },
             { 0x3E, 0x75, 0x38 },
             { 0x75, 0x11, 0xB9 }
         };
-        UINT8 repl[][3] = {
+        STATIC UINT8 repl[][3] = {
             { 0x01, 0xEB, 0x61 },
             { 0x3E, 0x90, 0x90 },
             { 0xEB, 0x11, 0xB9 }
@@ -729,33 +729,33 @@ VOID SNBE_AICPUPatch(UINT8 *Driver, UINT32 DriverSize, CHAR8 *InfoPlist, UINT32
                 DBG("SNBE_AICPUPatch (%d/7) not apply\n", i);
             }
         }
-        
-        UINT8 find_1[] = { 0xFF, 0x0F, 0x84, 0x2D };
-        UINT8 repl_1[] = { 0xFF, 0x0F, 0x85, 0x2D };
+
+        STATIC UINT8 find_1[] = { 0xFF, 0x0F, 0x84, 0x2D };
+        STATIC UINT8 repl_1[] = { 0xFF, 0x0F, 0x85, 0x2D };
         if (SearchAndReplace(Driver, DriverSize, find_1, sizeof(find_1), repl_1, 0)) {
             DBG("SNBE_AICPUPatch (4/7) applied\n");
         } else {
             DBG("SNBE_AICPUPatch (4/7) not apply\n");
         }
 
-        UINT8 find_2[] = { 0x01, 0x00, 0x01, 0x0F, 0x84 };
-        UINT8 repl_2[] = { 0x01, 0x00, 0x01, 0x0F, 0x85 };
+        STATIC UINT8 find_2[] = { 0x01, 0x00, 0x01, 0x0F, 0x84 };
+        STATIC UINT8 repl_2[] = { 0x01, 0x00, 0x01, 0x0F, 0x85 };
         if (SearchAndReplace(Driver, DriverSize, find_2, sizeof(find_2), repl_2, 0)) {
             DBG("SNBE_AICPUPatch (5/7) applied\n");
         } else {
             DBG("SNBE_AICPUPatch (5/7) not apply\n");
         }
-        
-        UINT8 find_3[] = { 0xC9, 0x74, 0x15, 0x0F, 0x32, 0x25, 0xFF, 0x0F, 0x00, 0x00, 0x48 };
-        UINT8 repl_3[] = { 0xC9, 0xEB, 0x15, 0x0F, 0x32, 0x25, 0xFF, 0x0F, 0x00, 0x00, 0x48 };
+
+        STATIC UINT8 find_3[] = { 0xC9, 0x74, 0x15, 0x0F, 0x32, 0x25, 0xFF, 0x0F, 0x00, 0x00, 0x48 };
+        STATIC UINT8 repl_3[] = { 0xC9, 0xEB, 0x15, 0x0F, 0x32, 0x25, 0xFF, 0x0F, 0x00, 0x00, 0x48 };
         if (SearchAndReplace(Driver, DriverSize, find_3, sizeof(find_3), repl_3, 0)) {
             DBG("SNBE_AICPUPatch (6/7) applied\n");
         } else {
             DBG("SNBE_AICPUPatch (6/7) not apply\n");
         }
 
-        UINT8 find_4[] = { 0xC9, 0x74, 0x0C, 0x0F, 0x32, 0x83, 0xE0, 0x1F, 0x42, 0x89, 0x44, 0x3B };
-        UINT8 repl_4[] = { 0xC9, 0xEB, 0x0C, 0x0F, 0x32, 0x83, 0xE0, 0x1F, 0x42, 0x89, 0x44, 0x3B };
+        STATIC UINT8 find_4[] = { 0xC9, 0x74, 0x0C, 0x0F, 0x32, 0x83, 0xE0, 0x1F, 0x42, 0x89, 0x44, 0x3B };
+        STATIC UINT8 repl_4[] = { 0xC9, 0xEB, 0x0C, 0x0F, 0x32, 0x83, 0xE0, 0x1F, 0x42, 0x89, 0x44, 0x3B };
         if (SearchAndReplace(Driver, DriverSize, find_4, sizeof(find_4), repl_4, 0)) {
             DBG("SNBE_AICPUPatch (7/7) applied\n");
         } else {
@@ -763,12 +763,12 @@ VOID SNBE_AICPUPatch(UINT8 *Driver, UINT32 DriverSize, CHAR8 *InfoPlist, UINT32
         }
     } else if (os_ver < AsciiOSVersionToUint64("10.14")) {
         // 10.13
-        UINT8 find[][3] = {
+        STATIC UINT8 find[][3] = {
             { 0x01, 0x74, 0x61 },
             { 0x3E, 0x75, 0x38 },
             { 0x75, 0x11, 0xB9 }
         };
-        UINT8 repl[][3] = {
+        STATIC UINT8 repl[][3] = {
             { 0x01, 0xEB, 0x61 },
             { 0x3E, 0x90, 0x90 },
             { 0xEB, 0x11, 0xB9 }
@@ -780,40 +780,40 @@ VOID SNBE_AICPUPatch(UINT8 *Driver, UINT32 DriverSize, CHAR8 *InfoPlist, UINT32
                 DBG("SNBE_AICPUPatch (%d/7) not apply\n", i);
             }
         }
-        
-        UINT8 find_1[] = { 0xFF, 0x0F, 0x84, 0xD3 };
-        UINT8 repl_1[] = { 0xFF, 0x0F, 0x85, 0xD3 };
+
+        STATIC UINT8 find_1[] = { 0xFF, 0x0F, 0x84, 0xD3 };
+        STATIC UINT8 repl_1[] = { 0xFF, 0x0F, 0x85, 0xD3 };
         if (SearchAndReplace(Driver, DriverSize, find_1, sizeof(find_1), repl_1, 0)) {
             DBG("SNBE_AICPUPatch (4/7) applied\n");
         } else {
             DBG("SNBE_AICPUPatch (4/7) not apply\n");
         }
-        
-        UINT8 find_2[] = { 0x01, 0x00, 0x01, 0x0F, 0x84 };
-        UINT8 repl_2[] = { 0x01, 0x00, 0x01, 0x0F, 0x85 };
+
+        STATIC UINT8 find_2[] = { 0x01, 0x00, 0x01, 0x0F, 0x84 };
+        STATIC UINT8 repl_2[] = { 0x01, 0x00, 0x01, 0x0F, 0x85 };
         if (SearchAndReplace(Driver, DriverSize, find_2, sizeof(find_2), repl_2, 0)) {
             DBG("SNBE_AICPUPatch (5/7) applied\n");
         } else {
             DBG("SNBE_AICPUPatch (5/7) not apply\n");
         }
-        
-        UINT8 find_3[] = { 0xC9, 0x74, 0x14, 0x0F, 0x32, 0x25, 0xFF, 0x0F, 0x00, 0x00, 0x6B };
-        UINT8 repl_3[] = { 0xC9, 0xEB, 0x14, 0x0F, 0x32, 0x25, 0xFF, 0x0F, 0x00, 0x00, 0x6B};
+
+        STATIC UINT8 find_3[] = { 0xC9, 0x74, 0x14, 0x0F, 0x32, 0x25, 0xFF, 0x0F, 0x00, 0x00, 0x6B };
+        STATIC UINT8 repl_3[] = { 0xC9, 0xEB, 0x14, 0x0F, 0x32, 0x25, 0xFF, 0x0F, 0x00, 0x00, 0x6B};
         if (SearchAndReplace(Driver, DriverSize, find_3, sizeof(find_3), repl_3, 0)) {
             DBG("SNBE_AICPUPatch (6/7) applied\n");
         } else {
             DBG("SNBE_AICPUPatch (6/7) not apply\n");
         }
-        
-        UINT8 find_4[] = { 0xC9, 0x74, 0x0C, 0x0F, 0x32, 0x83, 0xE0, 0x1F, 0x42, 0x89, 0x44, 0x3B };
-        UINT8 repl_4[] = { 0xC9, 0xEB, 0x0C, 0x0F, 0x32, 0x83, 0xE0, 0x1F, 0x42, 0x89, 0x44, 0x3B };
+
+        STATIC UINT8 find_4[] = { 0xC9, 0x74, 0x0C, 0x0F, 0x32, 0x83, 0xE0, 0x1F, 0x42, 0x89, 0x44, 0x3B };
+        STATIC UINT8 repl_4[] = { 0xC9, 0xEB, 0x0C, 0x0F, 0x32, 0x83, 0xE0, 0x1F, 0x42, 0x89, 0x44, 0x3B };
         if (SearchAndReplace(Driver, DriverSize, find_4, sizeof(find_4), repl_4, 0)) {
             DBG("SNBE_AICPUPatch (7/7) applied\n");
         } else {
             DBG("SNBE_AICPUPatch (7/7) not apply\n");
         }
     }
-    
+
     if (Entry->KernelAndKextPatches->KPDebug) {
         gBS->Stall(5000000);
     }
@@ -828,31 +828,31 @@ VOID SNBE_AICPUPatch(UINT8 *Driver, UINT32 DriverSize, CHAR8 *InfoPlist, UINT32
 // find: 0x48, 0x81, 0xFB, 0x00, 0x00, 0x00, 0x40
 // repl: 0x48, 0x81, 0xFB, 0x00, 0x00, 0x00, 0x80
 //
-UINT8 BroadwellE_IOPCI_Find[] = { 0x48, 0x81, 0xFB, 0x00, 0x00, 0x00, 0x40 };
-UINT8 BroadwellE_IOPCI_Repl[] = { 0x48, 0x81, 0xFB, 0x00, 0x00, 0x00, 0x80 };
+STATIC UINT8 BroadwellE_IOPCI_Find[] = { 0x48, 0x81, 0xFB, 0x00, 0x00, 0x00, 0x40 };
+STATIC UINT8 BroadwellE_IOPCI_Repl[] = { 0x48, 0x81, 0xFB, 0x00, 0x00, 0x00, 0x80 };
 
 VOID BDWE_IOPCIPatch(UINT8 *Driver, UINT32 DriverSize, CHAR8 *InfoPlist, UINT32 InfoPlistSize, LOADER_ENTRY *Entry)
 {
     UINTN count = 0;
-    
+
     DBG_RT(Entry, "\nBDWE_IOPCIPatch: driverAddr = %x, driverSize = %x\n", Driver, DriverSize);
     if (Entry->KernelAndKextPatches->KPDebug) {
         ExtractKextBundleIdentifier(InfoPlist);
     }
-    
+
     DBG_RT(Entry, "Kext: %a\n", gKextBundleIdentifier);
     //
     // now, let's patch it!
     //
     count = SearchAndReplace(Driver, DriverSize, BroadwellE_IOPCI_Find, sizeof(BroadwellE_IOPCI_Find), BroadwellE_IOPCI_Repl, 0);
-    
+
     if (count) {
         DBG_RT(Entry, "==> IOPCIFamily: %d replaces done.\n", count);
     }
     else {
         DBG_RT(Entry, "==> Patterns not found - patching NOT done.\n");
     }
-    
+
     if (Entry->KernelAndKextPatches->KPDebug) {
         gBS->Stall(5000000);
     }
@@ -878,7 +878,7 @@ VOID AnyKextPatch(UINT8 *Driver, UINT32 DriverSize, CHAR8 *InfoPlist, UINT32 Inf
 {
   UINTN   Num = 0;
   INTN    Ind;
-  
+
   DBG_RT(Entry, "\nAnyKextPatch %d: driverAddr = %x, driverSize = %x\nAnyKext = %a\n",
          N, Driver, DriverSize, Entry->KernelAndKextPatches->KextPatches[N].Label);
 
@@ -914,7 +914,7 @@ VOID AnyKextPatch(UINT8 *Driver, UINT32 DriverSize, CHAR8 *InfoPlist, UINT32 Inf
       DBG_RT(Entry, "%c", Entry->KernelAndKextPatches->KextPatches[N].Patch[Ind]);
     }
     DBG_RT(Entry, "' \n");
-    
+
     Num = SearchAndReplaceTxt((UINT8*)InfoPlist,
                            InfoPlistSize,
                            Entry->KernelAndKextPatches->KextPatches[N].Data,
@@ -922,7 +922,7 @@ VOID AnyKextPatch(UINT8 *Driver, UINT32 DriverSize, CHAR8 *InfoPlist, UINT32 Inf
                            Entry->KernelAndKextPatches->KextPatches[N].Patch,
                            -1);
   }
-  
+
   if (Entry->KernelAndKextPatches->KPDebug) {
     if (Num > 0) {
       DBG_RT(Entry, "==> patched %d times!\n", Num);
@@ -940,17 +940,17 @@ VOID AnyKextPatch(UINT8 *Driver, UINT32 DriverSize, CHAR8 *InfoPlist, UINT32 Inf
 VOID KextPatcherRegisterKexts(FSINJECTION_PROTOCOL *FSInject, FSI_STRING_LIST *ForceLoadKexts, LOADER_ENTRY *Entry)
 {
   INTN i;
-  
+
   if (Entry->KernelAndKextPatches->KPATIConnectorsController != NULL) {
     ATIConnectorsPatchRegisterKexts(FSInject, ForceLoadKexts, Entry);
   }
-  
+
   for (i = 0; i < Entry->KernelAndKextPatches->NrKexts; i++) {
     FSInject->AddStringToList(ForceLoadKexts,
                               PoolPrint(L"\\%a.kext\\Contents\\Info.plist",
                                         Entry->KernelAndKextPatches->KextPatches[i].Name) );
   }
-  
+
 }
 
 //
@@ -960,7 +960,7 @@ VOID KextPatcherRegisterKexts(FSINJECTION_PROTOCOL *FSInject, FSI_STRING_LIST *F
 VOID PatchKext(UINT8 *Driver, UINT32 DriverSize, CHAR8 *InfoPlist, UINT32 InfoPlistSize, LOADER_ENTRY *Entry)
 {
   INT32 i;
-  
+
   if (Entry->KernelAndKextPatches->KPATIConnectorsController != NULL) {
     //
     // ATIConnectors
@@ -977,9 +977,9 @@ VOID PatchKext(UINT8 *Driver, UINT32 DriverSize, CHAR8 *InfoPlist, UINT32 InfoPl
       return;
     }
   }
-  
+
   ExtractKextBundleIdentifier(InfoPlist);
-  
+
   if (Entry->KernelAndKextPatches->KPAppleIntelCPUPM &&
       (AsciiStrStr(InfoPlist,
                    "<string>com.apple.driver.AppleIntelCPUPowerManagement</string>") != NULL)) {
@@ -1034,7 +1034,7 @@ VOID PatchKext(UINT8 *Driver, UINT32 DriverSize, CHAR8 *InfoPlist, UINT32 InfoPl
       }
     }
 //  }
-  
+
   //
   // Check for FakeSMC (InjectKexts if no FakeSMC)
   //
@@ -1071,43 +1071,43 @@ UINT64 GetPlistHexValue(CHAR8 *Plist, CHAR8 *Key, CHAR8 *WholePlist)
   UINTN     IDLen;
   CHAR8     Buffer[48];
   //static INTN   DbgCount = 0;
-  
+
   // search for Key
   Value = AsciiStrStr(Plist, Key);
   if (Value == NULL) {
     //DBG(L"\nNo key: %a\n", Key);
     return 0;
   }
-  
+
   // search for <integer
   IntTag = AsciiStrStr(Value, "<integer");
   if (IntTag == NULL) {
     DBG(L"\nNo integer\n");
     return 0;
   }
-  
+
   // find <integer end
   Value = AsciiStrStr(IntTag, ">");
   if (Value == NULL) {
     DBG(L"\nNo <integer end\n");
     return 0;
   }
-  
+
   if (Value[-1] != '/') {
-    
+
     // normal case: value is here
     NumValue = AsciiStrHexToUint64(Value + 1);
     return NumValue;
-    
+
   }
-  
+
   // it might be a reference: IDREF="173"/>
   Value = AsciiStrStr(IntTag, "<integer IDREF=\"");
   if (Value != IntTag) {
     DBG(L"\nNo <integer IDREF=\"\n");
     return 0;
   }
-  
+
   // compose <integer ID="xxx" in the Buffer
   IDStart = AsciiStrStr(IntTag, "\"") + 1;
   IDEnd = AsciiStrStr(IDStart, "\"");
@@ -1130,14 +1130,14 @@ UINT64 GetPlistHexValue(CHAR8 *Plist, CHAR8 *Key, CHAR8 *WholePlist)
    DBG(L"Searching: '%a'\n", Buffer);
    }
    */
-  
+
   // and search whole plist for ID
   IntTag = AsciiStrStr(WholePlist, Buffer);
   if (IntTag == NULL) {
     DBG(L"\nNo %a\n", Buffer);
     return 0;
   }
-  
+
   // got it. find closing >
   /*
    if (DbgCount < 3) {
@@ -1154,10 +1154,10 @@ UINT64 GetPlistHexValue(CHAR8 *Plist, CHAR8 *Key, CHAR8 *WholePlist)
     DBG(L"\nInvalid <integer IDREF end\n");
     return 0;
   }
-  
+
   // we should have value now
   NumValue = AsciiStrHexToUint64(Value + 1);
-  
+
   /*
    if (DbgCount < 3) {
    AsciiStrnCpy(Buffer, IntTag, sizeof(Buffer) - 1);
@@ -1166,7 +1166,7 @@ UINT64 GetPlistHexValue(CHAR8 *Plist, CHAR8 *Key, CHAR8 *WholePlist)
    }
    DbgCount++;
    */
-  
+
   return NumValue;
 }
 
@@ -1208,10 +1208,10 @@ VOID PatchPrelinkedKexts(LOADER_ENTRY *Entry)
   //INTN      DbgCount = 0;
   UINT32    KextAddr;
   UINT32    KextSize;
-  
-  
+
+
   WholePlist = (CHAR8*)(UINTN)PrelinkInfoAddr;
-  
+
   //
   // Detect FakeSMC and if present then
   // disable kext injection InjectKexts().
@@ -1222,12 +1222,12 @@ VOID PatchPrelinkedKexts(LOADER_ENTRY *Entry)
   // But searching through the whole prelink info
   // works and that's the reason why it is here.
   //
-  
+
   //Slice
   // I see no reason to disable kext injection if FakeSMC found in cache
   //since rev4240 we have manual kext inject disable
   CheckForFakeSMC(WholePlist, Entry);
-  
+
   DictPtr = WholePlist;
   while ((DictPtr = AsciiStrStr(DictPtr, "dict>")) != NULL) {
     if (DictPtr[-1] == '<') {
@@ -1238,16 +1238,16 @@ VOID PatchPrelinkedKexts(LOADER_ENTRY *Entry)
         InfoPlistStart = DictPtr - 1;
       }
     } else if (DictPtr[-2] == '<' && DictPtr[-1] == '/') {
-      
+
       // closing dict
       if (DictLevel == 2 && InfoPlistStart != NULL) {
         // kext end
         InfoPlistEnd = DictPtr + 5 /* "dict>" */;
-        
+
         // terminate Info.plist with 0
         SavedValue = *InfoPlistEnd;
         *InfoPlistEnd = '\0';
-        
+
         // get kext address from _PrelinkExecutableSourceAddr
         // truncate to 32 bit to get physical addr
         KextAddr = (UINT32)GetPlistHexValue(InfoPlistStart, kPrelinkExecutableSourceKey, WholePlist);
@@ -1256,9 +1256,9 @@ VOID PatchPrelinkedKexts(LOADER_ENTRY *Entry)
         KextAddr += KernelSlide;
         // and adjust for AptioFixDrv's KernelRelocBase
         KextAddr += (UINT32)KernelRelocBase;
-        
+
         KextSize = (UINT32)GetPlistHexValue(InfoPlistStart, kPrelinkExecutableSizeKey, WholePlist);
-        
+
         /*if (DbgCount < 3
          || DbgCount == 100 || DbgCount == 101 || DbgCount == 102
          ) {
@@ -1267,7 +1267,7 @@ VOID PatchPrelinkedKexts(LOADER_ENTRY *Entry)
          gBS->Stall(20000000);
          }
          */
-        
+
         // patch it
         PatchKext(
                   (UINT8*)(UINTN)KextAddr,
@@ -1276,7 +1276,7 @@ VOID PatchPrelinkedKexts(LOADER_ENTRY *Entry)
                   (UINT32)(InfoPlistEnd - InfoPlistStart),
                   Entry
                   );
-        
+
         // return saved char
         *InfoPlistEnd = SavedValue;
         //DbgCount++;
@@ -1302,16 +1302,16 @@ VOID PatchLoadedKexts(LOADER_ENTRY *Entry)
 	struct OpaqueDTPropertyIterator OPropIter;
 	DTPropertyIterator	PropIter = &OPropIter;
 	//UINTN               DbgCount = 0;
-  
-  
+
+
   DBG(L"\nPatchLoadedKexts ... dtRoot = %p\n", dtRoot);
-  
+
   if (!dtRoot) {
     return;
   }
-  
+
   DTInit(dtRoot);
-  
+
   if (DTLookupEntry(NULL,"/chosen/memory-map", &MMEntry) == kSuccess) {
     if (DTCreatePropertyIteratorNoAlloc(MMEntry, PropIter) == kSuccess) {
       while (DTIterateProperties(PropIter, &PropName) == kSuccess) {
@@ -1320,15 +1320,15 @@ VOID PatchLoadedKexts(LOADER_ENTRY *Entry)
           // PropEntry _DeviceTreeBuffer is the value of Driver-XXXXXX property
           PropEntry = (_DeviceTreeBuffer*)(((UINT8*)PropIter->currentProperty) + sizeof(DeviceTreeNodeProperty));
           //if (DbgCount < 3) DBG(L"%a: paddr = %x, length = %x\n", PropName, PropEntry->paddr, PropEntry->length);
-          
+
           // PropEntry->paddr points to _BooterKextFileInfo
           KextFileInfo = (_BooterKextFileInfo *)(UINTN)PropEntry->paddr;
-          
+
           // Info.plist should be terminated with 0, but will also do it just in case
           InfoPlist = (CHAR8*)(UINTN)KextFileInfo->infoDictPhysAddr;
           SavedValue = InfoPlist[KextFileInfo->infoDictLength];
           InfoPlist[KextFileInfo->infoDictLength] = '\0';
-          
+
           PatchKext(
                     (UINT8*)(UINTN)KextFileInfo->executablePhysAddr,
                     KextFileInfo->executableLength,

 

 

Edited by TheRacerMaster

Share this post


Link to post
Share on other sites

Clover 4482 has broken kext injection of my ApplePS2SmartTouchPad.kext so i have no touchpad. External mouse works fine but trackpad is no longer detected. Was working fine with 4480 not sure about 4481 as i don't have it.

Edited by SavageAUS

Share this post


Link to post
Share on other sites

32-bit builds appear to be broken with GCC 8:

Clover/rEFIt_UEFI/Platform/AcpiPatcher.c:222:22: error: cast from pointer to integer of different size [-Werror=pointer-to-int-cast]
       gBS->FreePages((EFI_PHYSICAL_ADDRESS)XsdtEntryFromIndex(Index), XsdtReplaceSizes[Index]);
                      ^
#define XsdtEntryFromIndex(index) (EFI_ACPI_DESCRIPTION_HEADER*)(UINTN)ReadUnaligned64(XsdtEntryPtrFromIndex(index))

EFI_PHYSICAL_ADDRESS is 64-bit, so this is casting from a 32-bit pointer to a 64-bit integer.

 

Also, r4482's plist.c changes (specifically, the "end of cache" freeing) appear to cause memory corruption. Removing this fixes kext injection issues (plist.c memory corruption causes some kext versions to display <null string> in bdmesg).

Edited by TheRacerMaster

Share this post


Link to post
Share on other sites
1 hour ago, TheRacerMaster said:

32-bit builds appear to be broken with GCC 8:


Clover/rEFIt_UEFI/Platform/AcpiPatcher.c:222:22: error: cast from pointer to integer of different size [-Werror=pointer-to-int-cast]
       gBS->FreePages((EFI_PHYSICAL_ADDRESS)XsdtEntryFromIndex(Index), XsdtReplaceSizes[Index]);
                      ^

#define XsdtEntryFromIndex(index) (EFI_ACPI_DESCRIPTION_HEADER*)(UINTN)ReadUnaligned64(XsdtEntryPtrFromIndex(index))

EFI_PHYSICAL_ADDRESS is 64-bit, so this is casting from a 32-bit pointer to a 64-bit integer.

 

Also, r4482's plist.c changes (specifically, the "end of cache" freeing) appear to cause memory corruption. Removing this fixes kext injection issues (plist.c memory corruption causes some kext versions to display <null string> in bdmesg).

Thanks,

STATIC arrays accepted in 4484,

plist.c fixed in 4483.

EFI_PHYSICAL_ADDRESS will be fixed.  EDITED: 4485

Share this post


Link to post
Share on other sites

Hi,

it appears that 

        <key>FakeID</key>
        <dict>
            <key>LAN</key>
            <string>0x435411ab</string>
        </dict>

is broken since rev 4497

OK with 4495.

Is it a bug in new injection method ?

 

Share this post


Link to post
Share on other sites

Zenith432, why did you change architecture 32+64 to 64 in clover utilities? This breaks 32-bit compilation I still use. 10.14 has nothing to do to fat binaries, please revert the change back.

Share this post


Link to post
Share on other sites
9 hours ago, vit9696 said:

Zenith432, why did you change architecture 32+64 to 64 in clover utilities? This breaks 32-bit compilation I still use. 10.14 has nothing to do to fat binaries, please revert the change back.

Explained here

Which compilation did it break?

 

FWIW: I know why 32-bit build of clover-genconfig defines MDE_CPU_X64.  It's because in project.pbxproj, HEADER_SEARCH_PATHS is always MdePkg/Include/X64, so ProcessorBind.h is taken from there and code always uses 64 EDK2 defs.  In particular, the lines

Quote
#if defined(MDE_CPU_IA32)
printf("32 bit generator\n");
#elif defined(MDE_CPU_X64)
printf("64 bit generator\n");

in clover-genconfig.c are a joke.

 

PS: So it seems even 32-bit build of clover-genconfig wants to parse binary format of SETTINGS_DATA used by 64-bit Clover.  Is this right?  Are we looking for version to parse SETTINGS_DATA for 32-bit Clover?  Should 32-bit support be fully discontinued?  I need to know what the intention was to decide on resolution.

 

PPS: Including X64 version of ProcessorBind.h is 32-bit build of clover-genconfig cannot work anyway.  The struct SETTINGS_DATA has pointers in it, which are a different size, so 32-bit build will not end up with correct structure for 64-bit Clover.

 

@Slice: did you bother to test new code in r4501?  I see in main() in clover-genconfig.c call to GetOFVariable("device-properties", ...) which uses global gPlatform before it's been initialized.

Edited by Zenith432

Share this post


Link to post
Share on other sites

@Zenith432, the obvious issue here is that you enforced clover-genconfig being 64-bit, and thus it does not run on a 32-bit only CPU (and yes, that is my case, which I still rarely use for testing some really old legacy projects). So it is very clear how and why this is broken :).

 

Sure, nobody did request you to support a 32-bit only binary (which I agree is inadequate), yet a fat binary should successfully be able to generate a 32-bit config on a 32-bit only system, and a 64-bit config on a 64-bit system. For hybrid ones there have always been an arch util (man arch).

 

To fix the actual header paths bug you need to change the Xcode project header inclusion paths. If you click on + in Xcode project prefs it would let you define different header paths depending on the architecture. This should have been done right previously, yet some crazy attempts to make it "64-bit only" must have borked it.

 

Thanks~

Share this post


Link to post
Share on other sites

The 32-bit build of clover-genconfig didn't work before.  It was built, but it didn't do anything. It used arch i386 and X64/ProcessorBind.h which gives a different sizeof(SETTINGS_DATA) than compiling with arch x86_64.  In clover-genconfig.c there is a sanity check for sizeof(SETTINGS_DATA) being the same as the size of the data being parsed and it exited with an error message.

The reason for removal of 32-bit build was that after reversion of MdePke/Include/Base.h, it needs MDE_CPU_IA32 for arch i386 or it generates a compile error.  Right now, both xcodeproj and Makefile for clover-genconfig do not result in any errors, but do not build 32-bitnary either.

I'll only have access to macos on Monday, but when I do I'll change xcodeproj to use IA32/ProcessorBind.h for arch i386, return 32-bit into fat binary.  It is the only workable solution, because hybrid option does not yield a usable program.

Share this post


Link to post
Share on other sites
On 5/26/2018 at 8:19 AM, Zenith432 said:

 

@Slice: did you bother to test new code in r4501?  I see in main() in clover-genconfig.c call to GetOFVariable("device-properties", ...) which uses global gPlatform before it's been initialized.

4501 is not working version. I commit it to have an ability get my draft from other place.

I want to seriously advance clover-genconfig.

Share this post


Link to post
Share on other sites
On 5/26/2018 at 2:28 AM, Mr MagOO said:

Hi,

it appears that 

        <key>FakeID</key>
        <dict>
            <key>LAN</key>
            <string>0x435411ab</string>
        </dict>

is broken since rev 4497

OK with 4495.

Is it a bug in new injection method ?

 

@Slice : is there a chance the FakeID injection is broken or do I need to change syntax or something to make it work post rev 4496 ?

Share this post


Link to post
Share on other sites
6 hours ago, Mr MagOO said:

@Slice : is there a chance the FakeID injection is broken or do I need to change syntax or something to make it work post rev 4496 ?

Yes, you have to add

Quote

    <key>Devices</key>
    <dict>

        <key>LANInject</key>
        <true/>

 

Sorry for the back incompatibility.

I have to disable automatic injection to not mix with new method of Properties injection.

Share this post


Link to post
Share on other sites

@Slice

The latest config-sample.plist you committed has embedded nulls in it.  For example, after the string "Apple WiFi card".  There are others.

Also, the file was previously sorted.  It makes it hard to merge changes when each commit is randomly sorted from previous one.

Share this post


Link to post
Share on other sites
15 hours ago, Slice said:

Yes, you have to add

Sorry for the back incompatibility.

I have to disable automatic injection to not mix with new method of Properties injection.

 

Well, something is really happening with LANInjection=true ...

but I don't know what because I lost my video with a black screen boot :)

I've got this in Devices and Graphics :

	<key>Devices</key>
	<dict>
		<key>Audio</key>
		<dict>
			<key>Inject</key>
			<string>no</string>
		</dict>
        <key>LANInjection</key>
		<true/>
		<key>FakeID</key>
		<dict>
			<key>LAN</key>
			<string>0x435411ab</string>
		</dict>
		<key>USB</key>
		<dict>
			<key>AddClockID</key>
			<true/>
			<key>FixOwnership</key>
			<true/>
			<key>HighCurrent</key>
			<true/>
			<key>Inject</key>
			<true/>
		</dict>
	</dict>

	<key>Graphics</key>
	<dict>
		<key>Inject</key>
		<dict>
			<key>ATI</key>
			<false/>
			<key>Intel</key>
			<false/>
			<key>NVidia</key>
			<true/>
		</dict>
		<key>NvidiaGeneric</key>
		<true/>
		<key>VideoPorts</key>
		<integer>3</integer>
	</dict>

Is there any change for video injection ?

edit : {censored}, my fault, all is OK now with rev 4507.

Thank you Slice!

 

Edited by Mr MagOO

Share this post


Link to post
Share on other sites
3 hours ago, Zenith432 said:

@Slice

The latest config-sample.plist you committed has embedded nulls in it.  For example, after the string "Apple WiFi card".  There are others.

Also, the file was previously sorted.  It makes it hard to merge changes when each commit is randomly sorted from previous one.

It is PlistEditPro mischief. I corrected this in 4507. I will no more use this strange application.

Share this post


Link to post
Share on other sites
5 hours ago, Zenith432 said:

@Slice

The latest config-sample.plist you committed has embedded nulls in it.  For example, after the string "Apple WiFi card".  There are others.

 

It looks like nulls are known problem

Quote

  CopyMem(configBuffer, buffer, bufferSize);
  for (i=0; i<bufferSize; i++) {
    if (configBuffer == 0) {
      configBuffer = 0x20;  //replace random zero bytes to spaces
    }
  }

 

But I am not sure if this is good solution. Anyway we live with this workaround some years.

Share this post


Link to post
Share on other sites
13 hours ago, Zenith432 said:

@Slice

The latest config-sample.plist you committed has embedded nulls in it.  For example, after the string "Apple WiFi card".  There are others.

 

clover-genconfig corrected to not produce nulls. rev 4508.

 

Sorry about 32 bits, Xcode 9.3.1 don't like it.

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

  • Recently Browsing   0 members

    No registered users viewing this page.

  • Similar Content

    • By Reel
      All work except camera
      You need SwitchResX that the screen does not freeze
         


    • By RougeOne
      Budget Workstation MacOS 10.14.1 build: z370 AORUS Gaming 5 - i5-8600K - RX 580 - UHD 630
      flawless iGPU/dGPU integration, fast FCPX editing/rendering, great overall performance for ~$1200 / ~£1000
       
      Update: updated WhateverGreen and Lilu, then updated OS to 10.14.3 and everything is fine! Even iMessage, which previously I hadn't been able to make work, is now working - possibly the update flushed the NVRAM?
       
      I'm gonna come straight out and admit it - I used newb tools which are considered taboo on InsanelyMac, and so this isn't so much an installation guide as it is a green flag to anyone considering the same hardware. My first and previous Hackintosh project was just over ten years ago and this site was massively helpful, so this is my (very small) way of giving back.
       
      EFI DIRECTORY
      http://s000.tinyupload.com/index.php?file_id=97606131925137056003
       
      HARDWARE
      CPU: Intel i5-8600K 3.6GHz hexacore (contains UHD 630 iGPU)
      CPU Cooler: be quiet! Pure Rock Slim (no paste needed, and fan facing front)
      Mobo: GIGABYTE z370 AORUS Gaming 5 (WIFI)
      RAM: 1x 8GB CORSAIR Vengeance LPX C16 2666MHz (in first slot)
      GFX Card: Sapphire RX 580 NITRO+ SE
      SATA: used ports 0, 1, 3, 5 (HDD problems when connected to port 4)
      HDD: generic
      DVD Drive: generic
      PSU: EVGA 500W Semi Modular (but you should get 650W!)
      Case: CoolerMaster CM590 III
      Fans: 2x stock case fans (one sucky on front, one blowy on back), 3x be quiet! BL046 120mm fans (two blowy on top, one sucky on front). Airflow near-perfectly balanced.
       
      ADVICE
      With the Clover configuration, you want to pay close attention to all things relating to graphics. There are some specific settings for ACPI, Boot, Devices and Graphics which should be copied exactly in order to get the iGPU and dGPU to work in perfect harmony. For instance, make sure you use ig-platform-id 59120003 rather than 59120000 if you want FCPX to be usable. With the right settings, FCPX will not only run, but run flawlessly.
       
      I was not able to figure out how to make NVRAM/PRAM behave itself, and I got errors when I tried to issue relevant commands. As a result, I could not at first get iMessage to work properly. It worked once on my first installation attempt but not from subsequent attempts. This Clover configuration therefore has no consideration whatsoever toward iMessage reliability - however, iMessage ended up working somehow, anyway - I noticed it after I'd updated to 10.14.3 but this may just have been coincidence.
       
      Onboard WiFi does not work. There is no driver for it. I'm using wired ethernet to a bridge router as it was cheaper than getting a compatible PCI card.
       
      Audio is working well after I installed the z370 audio patch, then the ALC1220 codecs, then AppleALC.
       
      With kexts for generic 3rd party SATA and generic 3rd party eSATA, my HDDs and DVD drive work perfectly well.
       
      Sleep/Wake functions run perfectly. iCloud functions run perfectly. Shutdown leaves the mobo LEDs on, annoyingly.
       
      My USB 3 ports seem to suffer from low power output, and I don't know why.
       
      Overall, this system appears to run about as well under MacOS 10.14 Mojave as it does under Windows 7. I would solidly recommend this build to anyone looking to build a computer for workstation purposes, on a budget.
    • By JhonneR
      Good morning, everyone, I'm sorry if I was wrong in choosing the topic but I need help.
      I have found it extremely difficult to mount a Hackintosh in my acer.
      Setup;
      I5 2430m 2.4 6GB RAM
      Mobo P5we0 - Intel intel Graphics 3000
      At the moment I have not found out what my audio codecs are.
      but I'll leave the link on the motherboard.
      however, my real problem is several.
      I have already used the itakos image, I have already used several images that could be installed normally and that after installing the image I do not have internet via cable or wireless.
      And another, I already looked in other forums and I know that I will be called a donkey in the logical question but I have not been able to install the [url=&quot;http://www.insanelymac.com/forum/topic/279450-why-insanelymac-does-not-support-tonymacx86/&quot;]#####[/url] or clover to start from the Hard Disk, so I do not know what to do.At the moment I still do not know if there has already been someone who has already installed the High Sierra or just the Sierra in this notebook model and obtained permanent success.
      If anyone can help me, I'll be grateful.
      Installations without success, but with the system already installed in ACER;
      SIERRA
      HIGH SIERRA
      MOJAVE
      All with the same symptom. Starting only by the bootable USB stick and without internet connection via cable and wireless.
      Appreciate ! Acer 5750 Official Page
    • By Slice
      Since rev 4844 Vector Themes are supported and there are ready-to-use Clovy by Clovy, cesium by Slice and BGM_SVG by Blackosx.
      You may see it's structure to create own theme
      -------------------------------------------------------------------------------------------------------------------------------------------------------
       
       
      Now I want to add vector graphics support in Clover. See rev 4560 and later.
      It is not working yet but designers may begin to create Vector Themes.
      It supposed to consist of SVG elements and has design size. It will be rendered to any screen size scaled from design size.
       
      What application in macOS can create SVG graphics?
      Inkscape is not working in macOS 10.11+. Pity.
      LibreOffice Draw works with SVG but buggy.
      Boxy SVG cost 10$ but looks good enough. It creates the best in simplicity files and have more then enough features.
      Illustrator is good but expensive.
       
      How to improve SVG file?
      Clover has restricted support for SVG. It is your job to make compatible file and as small as possible to speedup rendering.
      Some helps:
      Help:Inkscape – From invalid to valid SVG Inkscape files
      From invalid to valid SVG Adobe Illustrator files
      From invalid to valid SVG files of other editors: BKchem, ChemDraw and CorelDRAW
      Help:Illustrator – Assistance with creating and saving SVG images in Adobe Illustrator that will pass W3C validation
      User:Quibik/Cleaning up SVG files manually
      Later I will write own instructions specific to Clover abilities.
       
      How to create SVG fonts?
      You can google to find ready-to-use SVG fonts.  I found some problems with too beaty fonts: slow rendering and overflow crash. Be careful.
      You can get ttf or otf fonts and convert them into svg by using online WEB services. Not a problem to google.
      But then I want to find a way to simplify the font to reduce a size and speedup rendering.
      You can create own font by FontForge It is opensource and available for Windows, Mac and GNU+Linux. It creates otf font which you can convert to svg font.
       
      Pictures from Badruzeus
      https://www.insanelymac.com/forum/applications/core/interface/file/attachment.php?id=301597
×