joevt Posted July 24, 2016 Share Posted July 24, 2016 Hi Can I use clover to remove devices/methods "on the fly" from dsdt? for example, in the attached dsdt (unpatched) I would like to remove (among other) Device (HS04) { Name (_ADR, 0x04) // _ADR: Address } and Scope (_SB.PCI0.XHC.RHUB.HS04) { Method (_UPC, 0, Serialized) // _UPC: USB Port Capabilities { If (!PRTE (_ADR)) { Return (UPCN) /* \_SB_.PCI0.XHC_.RHUB.UPCN */ } Return (UPC3) /* \_SB_.PCI0.XHC_.RHUB.UPC3 */ } Method (_PLD, 0, Serialized) // _PLD: Physical Location of Device { Return (PLD4) /* \_SB_.PCI0.XHC_.RHUB.PLD4 */ } } I think you can disassemble the dsdt to get the bytes for those routines, then use Find and Replace in the config.plist to remove those bytes. http://www.insanelymac.com/forum/topic/312071-please-explain-entrys-in-clover-bootlog/?do=findComment&comment=2239055 https://clover-wiki.zetam.org/Configuration/ACPI#acpi_dsdt_dsdt-patches Make sure the bytes are unique in the DSDT unless you want to remove all occurrences. Link to comment Share on other sites More sharing options...
D-an-W Posted July 25, 2016 Share Posted July 25, 2016 Whats the current best way of "automatically" build the latest Clover version as the app by cvad doesn't seem to want to work at the moment? Link to comment Share on other sites More sharing options...
Fljagd Posted July 25, 2016 Share Posted July 25, 2016 Whats the current best way of "automatically" build the latest Clover version as the app by cvad doesn't seem to want to work at the moment? http://www.insanelymac.com/forum/files/file/589-build-clovercommand/ 2 Link to comment Share on other sites More sharing options...
D-an-W Posted July 25, 2016 Share Posted July 25, 2016 Many thanks Fljagd, will give it a go. 1 Link to comment Share on other sites More sharing options...
Fljagd Posted July 25, 2016 Share Posted July 25, 2016 Many thanks Fljagd, will give it a go. apparently, there are some problems since 2-3 days Link to comment Share on other sites More sharing options...
Micky1979 Posted July 25, 2016 Share Posted July 25, 2016 Yep: unknown scope for symbol _InternalMemSetMem ..something is no longer visible, must be adjusted. Anyway that happen with last revisions of edk2. Link to comment Share on other sites More sharing options...
Slice Posted July 25, 2016 Share Posted July 25, 2016 Yep: unknown scope for symbol _InternalMemSetMem ..something is no longer visible, must be adjusted. Anyway that happen with last revisions of edk2. I am not understanding what is happen with edk2. I can't compile with clang. Similar messages with different symbols. OK with gcc49. FV Space Information DUETEFIMAINFVX64 [35%Full] 2752512 total, 984112 used, 1768400 free - Done - Build end time: 15:28:20, Jul.25 2016 Build total time: 00:04:05 Link to comment Share on other sites More sharing options...
mhaeuser Posted July 25, 2016 Share Posted July 25, 2016 https://github.com/tianocore/edk2/commit/1252a681a67da81bf1304c7ac836f90eab8bb7cb 'protected visibility' sounds related. Link to comment Share on other sites More sharing options...
Micky1979 Posted July 25, 2016 Share Posted July 25, 2016 Looks like the call is inside a statement end not visible outside it. Or someting similar. We should file a bug to edk2 devs. Or look for a new flag in clang.. if exist Link to comment Share on other sites More sharing options...
Slice Posted July 25, 2016 Share Posted July 25, 2016 https://github.com/tianocore/edk2/commit/1252a681a67da81bf1304c7ac836f90eab8bb7cb 'protected visibility' sounds related. Yes, I also narrow down troubleshooting to this stupid commit. Rev 3657 is now compilable. 1 Link to comment Share on other sites More sharing options...
jamiethemorris Posted July 25, 2016 Share Posted July 25, 2016 Alright, I'm having a really weird issue. After updating to El Capitan and also updating my Clover installation, I have to press a key to get into Clover otherwise it just sits at the BIOS splash screen. I'm pretty sure it has something to do with me temporarily removing the drive yesterday. The correct drive is selected in the BIOS, and this only happens with Clover. I'm really confused... Has anyone ever seen this before? Link to comment Share on other sites More sharing options...
D-an-W Posted July 25, 2016 Share Posted July 25, 2016 Clover_v2.3k_r3657 seemed to build ok here 1 Link to comment Share on other sites More sharing options...
lafayen 33 Posted July 25, 2016 Share Posted July 25, 2016 Ok for me thanks 2 Link to comment Share on other sites More sharing options...
AsusFreak Posted July 25, 2016 Share Posted July 25, 2016 @Micky1979 Can't build here, error messages below. EDIT: Builds fine in ElCap. When running the script in Sierra I see early the script some warnings regarding permissions. Maybe I have something screwed up. ./VfrFormPkg.h:654:47: warning: field 'mEnd' is uninitialized when used here [-Wuninitialized] CIfrOpHeader (EFI_IFR_END_OP, &mEnd->Header) {} ^ ./VfrFormPkg.h:663:64: warning: field 'mDefaultStore' is uninitialized when used here [-Wuninitialized] CIfrOpHeader (EFI_IFR_DEFAULTSTORE_OP, &mDefaultS... ^ ./VfrFormPkg.h:705:49: warning: field 'mForm' is uninitialized when used here [-Wuninitialized] CIfrOpHeader (EFI_IFR_FORM_OP, &mForm->Header) { ^ ./VfrFormPkg.h:737:56: warning: field 'mFormMap' is uninitialized when used here [-Wuninitialized] CIfrOpHeader (EFI_IFR_FORM_MAP_OP, &mFormMap->Header) { ^ ./VfrFormPkg.h:774:56: warning: field 'mVarStore' is uninitialized when used here [-Wuninitialized] CIfrOpHeader (EFI_IFR_VARSTORE_OP, &mVarStore->Header) { ^ ./VfrFormPkg.h:814:63: warning: field 'mVarStoreEfi' is uninitialized when used here [-Wuninitialized] CIfrOpHeader (EFI_IFR_VARSTORE_EFI_OP, &mVarStoreE... ^ ./VfrFormPkg.h:870:78: warning: field 'mVarStoreNameValue' is uninitialized when used here [-Wuninitialized] ...CIfrOpHeader (EFI_IFR_VARSTORE_NAME_VALUE_OP, &mVarStoreNameValue->Heade... ^ ./VfrFormPkg.h:890:51: warning: field 'mImage' is uninitialized when used here [-Wuninitialized] CIfrOpHeader (EFI_IFR_IMAGE_OP, &mImage->Header) { ^ ./VfrFormPkg.h:905:55: warning: field 'mModal' is uninitialized when used here [-Wuninitialized] CIfrOpHeader (EFI_IFR_MODAL_TAG_OP, &mModal->Header) { ^ ./VfrFormPkg.h:916:53: warning: field 'mLocked' is uninitialized when used here [-Wuninitialized] CIfrOpHeader (EFI_IFR_LOCKED_OP, &mLocked->Header) {} ^ ./VfrFormPkg.h:926:49: warning: field 'mRule' is uninitialized when used here [-Wuninitialized] CIfrOpHeader (EFI_IFR_RULE_OP, &mRule->Header) { ^ ./VfrFormPkg.h:948:44: warning: field 'mDefault' is uninitialized when used here [-Wuninitialized] CIfrOpHeader (EFI_IFR_DEFAULT_OP, &mDefault->Header, Size) { ^ ./VfrFormPkg.h:976:44: warning: field 'mDefault' is uninitialized when used here [-Wuninitialized] CIfrOpHeader (EFI_IFR_DEFAULT_OP, &mDefault->Header, sizeof (EFI... ^ ./VfrFormPkg.h:996:50: warning: field 'mValue' is uninitialized when used here [-Wuninitialized] CIfrOpHeader (EFI_IFR_VALUE_OP, &mValue->Header) {} ^ ./VfrFormPkg.h:1006:49: warning: field 'mRead' is uninitialized when used here [-Wuninitialized] CIfrOpHeader (EFI_IFR_READ_OP, &mRead->Header) {} ^ ./VfrFormPkg.h:1016:50: warning: field 'mWrite' is uninitialized when used here [-Wuninitialized] CIfrOpHeader (EFI_IFR_WRITE_OP, &mWrite->Header) {} ^ ./VfrFormPkg.h:1028:38: warning: field 'mGet' is uninitialized when used here [-Wuninitialized] CIfrOpHeader (EFI_IFR_GET_OP, &mGet->Header) { ^ ./VfrFormPkg.h:1048:38: warning: field 'mSet' is uninitialized when used here [-Wuninitialized] CIfrOpHeader (EFI_IFR_SET_OP, &mSet->Header) { ^ ./VfrFormPkg.h:1066:55: warning: field 'mSubtitle' is uninitialized when used here [-Wuninitialized] CIfrOpHeader (EFI_IFR_SUBTITLE_OP, &mSubtitle->Header), ^ ./VfrFormPkg.h:1067:25: warning: field 'mSubtitle' is uninitialized when used here [-Wuninitialized] CIfrStatementHeader (&mSubtitle->Statement) { ^ ./VfrFormPkg.h:1086:48: warning: field 'mText' is uninitialized when used here [-Wuninitialized] CIfrOpHeader (EFI_IFR_TEXT_OP, &mText->Header), ^ ./VfrFormPkg.h:1087:38: warning: field 'mText' is uninitialized when used here [-Wuninitialized] CIfrStatementHeader (&mText->Statement) { ^ ./VfrFormPkg.h:1102:46: warning: field 'mRef' is uninitialized when used here [-Wuninitialized] CIfrOpHeader (EFI_IFR_REF_OP, &mRef->Header), ^ ./VfrFormPkg.h:1103:36: warning: field 'mRef' is uninitialized when used here [-Wuninitialized] CIfrQuestionHeader (&mRef->Question) { ^ ./VfrFormPkg.h:1118:47: warning: field 'mRef2' is uninitialized when used here [-Wuninitialized] CIfrOpHeader (EFI_IFR_REF_OP, &mRef2->Header, sizeof (EFI... ^ ./VfrFormPkg.h:1119:37: warning: field 'mRef2' is uninitialized when used here [-Wuninitialized] CIfrQuestionHeader (&mRef2->Question) { ^ ./VfrFormPkg.h:1139:47: warning: field 'mRef3' is uninitialized when used here [-Wuninitialized] CIfrOpHeader (EFI_IFR_REF_OP, &mRef3->Header, sizeof (EFI... ^ ./VfrFormPkg.h:1140:37: warning: field 'mRef3' is uninitialized when used here [-Wuninitialized] CIfrQuestionHeader (&mRef3->Question) { ^ ./VfrFormPkg.h:1165:47: warning: field 'mRef4' is uninitialized when used here [-Wuninitialized] CIfrOpHeader (EFI_IFR_REF_OP, &mRef4->Header, sizeof(EFI_... ^ ./VfrFormPkg.h:1166:37: warning: field 'mRef4' is uninitialized when used here [-Wuninitialized] CIfrQuestionHeader (&mRef4->Question) { ^ ./VfrFormPkg.h:1196:46: warning: field 'mRef5' is uninitialized when used here [-Wuninitialized] CIfrOpHeader (EFI_IFR_REF_OP, &mRef5->Header, sizeof (EFI_... ^ ./VfrFormPkg.h:1197:36: warning: field 'mRef5' is uninitialized when used here [-Wuninitialized] CIfrQuestionHeader (&mRef5->Question) { ^ ./VfrFormPkg.h:1207:64: warning: field 'mResetButton' is uninitialized when used here [-Wuninitialized] CIfrOpHeader (EFI_IFR_RESET_BUTTON_OP, &mResetBut... ^ ./VfrFormPkg.h:1208:25: warning: field 'mResetButton' is uninitialized when used here [-Wuninitialized] CIfrStatementHeader (&mResetButton->Statement) { ^ ./VfrFormPkg.h:1223:58: warning: field 'mCheckBox' is uninitialized when used here [-Wuninitialized] CIfrOpHeader (EFI_IFR_CHECKBOX_OP, &mCheckBox->Header), ^ ./VfrFormPkg.h:1224:43: warning: field 'mCheckBox' is uninitialized when used here [-Wuninitialized] CIfrQuestionHeader (&mCheckBox->Question) { ^ ./VfrFormPkg.h:1263:52: warning: field 'mAction' is uninitialized when used here [-Wuninitialized] CIfrOpHeader (EFI_IFR_ACTION_OP, &mAction->Header), ^ ./VfrFormPkg.h:1264:39: warning: field 'mAction' is uninitialized when used here [-Wuninitialized] CIfrQuestionHeader (&mAction->Question) { ^ ./VfrFormPkg.h:1279:48: warning: field 'mDate' is uninitialized when used here [-Wuninitialized] CIfrOpHeader (EFI_IFR_DATE_OP, &mDate->Header), ^ ./VfrFormPkg.h:1280:37: warning: field 'mDate' is uninitialized when used here [-Wuninitialized] CIfrQuestionHeader (&mDate->Question) { ^ ./VfrFormPkg.h:1322:55: warning: field 'mNumeric' is uninitialized when used here [-Wuninitialized] CIfrOpHeader (EFI_IFR_NUMERIC_OP, &mNumeric->Header), ^ ./VfrFormPkg.h:1323:41: warning: field 'mNumeric' is uninitialized when used here [-Wuninitialized] CIfrQuestionHeader (&mNumeric->Question), ^ ./VfrFormPkg.h:1324:41: warning: field 'mNumeric' is uninitialized when used here [-Wuninitialized] CIfrMinMaxStepData (&mNumeric->data, TRUE) { ^ ./VfrFormPkg.h:1383:52: warning: field 'mOneOf' is uninitialized when used here [-Wuninitialized] CIfrOpHeader (EFI_IFR_ONE_OF_OP, &mOneOf->Header), ^ ./VfrFormPkg.h:1384:39: warning: field 'mOneOf' is uninitialized when used here [-Wuninitialized] CIfrQuestionHeader (&mOneOf->Question), ^ ./VfrFormPkg.h:1385:39: warning: field 'mOneOf' is uninitialized when used here [-Wuninitialized] CIfrMinMaxStepData (&mOneOf->data) { ^ ./VfrFormPkg.h:1440:52: warning: field 'mString' is uninitialized when used here [-Wuninitialized] CIfrOpHeader (EFI_IFR_STRING_OP, &mString->Header), ^ Link to comment Share on other sites More sharing options...
Micky1979 Posted July 25, 2016 Share Posted July 25, 2016 (edited) Is not the script.. is edk2 with clang 800+ (also is in Beta ...). BTW EFI_IFR_END *mEnd; should be I guess EFI_IFR_END *mEnd = NULL; ... and so on for others in VfrFormPkg.h. Cannot exclude other similar stuff will fails (they can't are warnings) else where (I'm in bed, wont test now ). EDIT for fdisk440 can't say but should be something related linking to a shared/static libgcc apparently edk2 devs does not care about clang with their last(s) commit Edited July 26, 2016 by Micky1979 Link to comment Share on other sites More sharing options...
AsusFreak Posted July 26, 2016 Share Posted July 26, 2016 Thanks for the reply, I'll just follow the thread... 1 Link to comment Share on other sites More sharing options...
chris1111 Posted July 26, 2016 Share Posted July 26, 2016 Everything's fine here Xcode 7.3.1 OS X 10.11.6 Thanks Slice Link to comment Share on other sites More sharing options...
arcade33 Posted July 26, 2016 Share Posted July 26, 2016 on EL CAPITAN 10.11.6 ALL IS OK !!!!! XCODE 7.3.1 THANKS SO MUCH SLICE Link to comment Share on other sites More sharing options...
lafayen 33 Posted July 26, 2016 Share Posted July 26, 2016 Update 3657 to 3659 ok MacOs Sierra Thanks 1 Link to comment Share on other sites More sharing options...
Micky1979 Posted July 26, 2016 Share Posted July 26, 2016 Guys, thanks for using my script ... but consider to move to my topic for that, this thread has nothing to do with it. Code (what matter) is done by Slice & co, so may you intrested in posting how your PC works booting with Clover here , the script is only an easy way to automate what devs do everyday manually, and don't make the build better than other, no magician Link to comment Share on other sites More sharing options...
Slice Posted July 26, 2016 Share Posted July 26, 2016 Yup, not a big deal to take a good working script and compile good working sources to obtain a result. I just doubt in intermediate revisions sometimes with not finished logics. 2 Link to comment Share on other sites More sharing options...
Micky1979 Posted July 26, 2016 Share Posted July 26, 2016 Not sure what you say is good for me EDIT BTW agreed ("results"/"revisions" now I see your edit ). For that users should look at commits done (Clover and edk2) before update, not always there are "hotfix" or thoroughly tested code. Also you wait before posting a new package on source forge (stable release? I hope this is clear to users ... mmhm ..maybe). 1 Link to comment Share on other sites More sharing options...
cecekpawon Posted July 27, 2016 Share Posted July 27, 2016 I see edk2 has flexible "CONF_PATH" right now (Windows) ... Link to comment Share on other sites More sharing options...
mhaeuser Posted July 27, 2016 Share Posted July 27, 2016 Unfortunately not... CONF_PATH as a variable is 'flexible' now, though it is not used everywhere. For example, build uses WorkspaceDir + 'Conf', so, that would needed to be fixed as well. Link to comment Share on other sites More sharing options...
cecekpawon Posted July 27, 2016 Share Posted July 27, 2016 Uhm @DF tomorrow will do test to set that variable outside workspace before calling edksetup.. Previously I use build "-conf" options to set a custom conf path with ebuild, Windows untested https://sourceforge.net/p/cloverefiboot/code/HEAD/tree/ebuild.sh#l392 Link to comment Share on other sites More sharing options...
Recommended Posts