Slice Posted January 7, 2018 Share Posted January 7, 2018 @apianti Thank you Works here HP ProBook 6570b WithOut EmuV First of all thanks to vit9696. Hi there, I've tested r4369 on InsydeH20 and the nvram is not working without the emulation driver :/ Let me know what info u need. This is the solution for Aptio UEFI BIOS (AptioFix), it is not for InsydeH20 BIOS. 4 Link to comment Share on other sites More sharing options...
Needy Posted January 7, 2018 Share Posted January 7, 2018 First of all thanks to vit9696. This is the solution for Aptio UEFI BIOS (AptioFix), it is not for InsydeH20 BIOS. Oh is it? My bad. Link to comment Share on other sites More sharing options...
mhaeuser Posted January 7, 2018 Share Posted January 7, 2018 This is the solution for Aptio UEFI BIOS (AptioFix), it is not for InsydeH20 BIOS. Except for your sleep fix, there is nothing explicitely Aptio-specifc in it. Link to comment Share on other sites More sharing options...
Funky frank Posted January 7, 2018 Share Posted January 7, 2018 Will the new clover version make my computer much faster and also fix the hd4600+nvidia video freeze? Link to comment Share on other sites More sharing options...
Slice Posted January 7, 2018 Share Posted January 7, 2018 Except for your sleep fix, there is nothing explicitely Aptio-specifc in it. I don't think so. There is a fix for bug introduced by AMI UEFI BIOS. As well whole AptioFix was developed to fight with AMI AllocateMemory() stupid function while EDK2 AllocateMemory is good. See legacy Clover == CloverEFI doesn't need AptioFix. Hm, may be we can do real NVRAM with CloverEFI in the case of using on UEFI machine?! 2 Link to comment Share on other sites More sharing options...
mhaeuser Posted January 7, 2018 Share Posted January 7, 2018 I don't think so. There is a fix for bug introduced by AMI UEFI BIOS. As well whole AptioFix was developed to fight with AMI AllocateMemory() stupid function while EDK2 AllocateMemory is good. See legacy Clover == CloverEFI doesn't need AptioFix. Yes, it because built because of Aptio, but is not exclusive to Aptio. Even today, people use AptioFix on non-AMI machines. Hm, may be we can do real NVRAM with CloverEFI in the case of using on UEFI machine?! You want UEFI to emulate BIOS so that you can load another UEFI environment? With some relocation magic, this really is a non-isse tbh Link to comment Share on other sites More sharing options...
Guest Posted January 7, 2018 Share Posted January 7, 2018 It is not working in my case (rig in signature) Last login: Sun Jan 7 17:36:41 on console fabios-Mac-Pro:~ fabio$ sudo nvram -p Password: fakesmc-key-MSWr-ui8 %00 flagstate %00%00%00%00%00%00%00%00%00%00%00%00%00%00%00%00%00%00%00%00%00%00%00%00%00%00%00%00%00%00%00%00 fakesmc-key-RBr -ch8* j90%00%00%00%00%00 fakesmc-key-EPCI-ui32 %06%f0%f0%00 fakesmc-key-MSAc-ui16 %00%00 efi-boot-device <array><dict><key>IOMatch</key><dict><key>IOProviderClass</key><string>IOMedia</string><key>IOPropertyMatch</key><dict><key>UUID</key><string>3A55787E-1275-45BE-87EB-0ACF96BC0343</string></dict></dict></dict></array> fakesmc-key-CLKH-{clh %00%00p%80%00%01%19@ bootercfg (%00 fakesmc-key-NATi-ui16 %00%1e LocationServicesEnabled %01 fakesmc-key-NTOK-ui8 %01 fakesmc-key-BATP-flag %00 fakesmc-key-$Num-ui8 %01 fakesmc-key-REV -ch8* %02 %0f%00%00%18 bluetoothActiveControllerInfo %0a%18%05%0b%00%00%00%00`%14%f0%03%8c<%b5%a3 fakesmc-key-HI0N-ui8 %10 fmm-computer-name fabio%e2%80%99s Mac Pro fakesmc-key-MSTc-ui8 %00 fakesmc-key-MSPS-ui16 %00%03 fakesmc-key-RMde-char A csr-active-config g%00%00%00 fakesmc-key-RPlt-ch8* j90%00%00%00%00%00 fakesmc-key-BEMB-flag %00 fakesmc-key-#KEY-ui32 %00%00%00%10 SystemAudioVolume A fakesmc-key-MSFW-ui8 %01%00 specialbootdevice %02%01%0c%00%d0A%03%0a%00%00%00%00%01%01%06%00%02%1f%03%12%0a%00%00%00%ff%ff%00%00%04%01*%00%03%00%00%00%00H%0a%00%00%00%00%00%88%ca%93;%00%00%00%00~xU:u%12%beE%87%eb%0a%cf%96%bc%03C%02%02%04%03$%00%f7%fct%be|%0b%f3I%91G%01%f4%04.hB%ea%00%1f.%994-E%bd%9e%da%ff%12}%c4%a6%7f%ff%04%00 security-mode none fakesmc-key-LsBA-ui8 %90 fakesmc-key-$Adr-ui32 %00%00%03%00 nvda_drv 1%00 EFILoginHiDPI %00%00%00%00 fakesmc-key-CLKT-ui32 %00%00%f7%aa fakesmc-key-MSDW-ui8 %01 fakesmc-key-BNum-ui8 %00 SystemAudioVolumeDB %f0 fakesmc-key-NATJ-ui8 %02 efi-boot-device-data %02%01%0c%00%d0A%03%0a%00%00%00%00%01%01%06%00%02%1f%03%12%0a%00%00%00%ff%ff%00%00%04%01*%00%03%00%00%00%00H%0a%00%00%00%00%00%88%ca%93;%00%00%00%00~xU:u%12%beE%87%eb%0a%cf%96%bc%03C%02%02%04%03$%00%f7%fct%be|%0b%f3I%91G%01%f4%04.hB%ea%00%1f.%994-E%bd%9e%da%ff%12}%c4%a6%7f%ff%04%00 fabios-Mac-Pro:~ fabio$ Link to comment Share on other sites More sharing options...
WinstonAce Posted January 7, 2018 Share Posted January 7, 2018 Hi currently using the new AptioFix V2 without slide value and everything seems to work fine should I still calculate slide value and use it? Type Start End # Pages Attributes available 0000000000000000-0000000000057FFF 0000000000000058 000000000000000F reserved 0000000000058000-0000000000058FFF 0000000000000001 000000000000000F available 0000000000059000-000000000009DFFF 0000000000000045 000000000000000F reserved 000000000009E000-000000000009FFFF 0000000000000002 000000000000000F available 0000000000100000-00000000A9845FFF 00000000000A9746 000000000000000F LoaderData 00000000A9846000-00000000A9945FFF 0000000000000100 000000000000000F available 00000000A9946000-00000000ABF1FFFF 00000000000025DA 000000000000000F LoaderData 00000000ABF20000-00000000ABF33FFF 0000000000000014 000000000000000F LoaderCode 00000000ABF34000-00000000AC000FFF 00000000000000CD 000000000000000F ACPI_NVS 00000000AC001000-00000000AC007FFF 0000000000000007 000000000000000F BS_data 00000000AC008000-00000000AC198FFF 0000000000000191 000000000000000F BS_code 00000000AC199000-00000000ACF0AFFF 0000000000000D72 000000000000000F BS_data 00000000ACF0B000-00000000ACF1BFFF 0000000000000011 000000000000000F BS_code 00000000ACF1C000-00000000ACF2DFFF 0000000000000012 000000000000000F BS_data 00000000ACF2E000-00000000ACF38FFF 000000000000000B 000000000000000F BS_code 00000000ACF39000-00000000ACF3AFFF 0000000000000002 000000000000000F RT_data 00000000ACF3B000-00000000AD7F6FFF 00000000000008BC 800000000000000F BS_data 00000000AD7F7000-00000000AD835FFF 000000000000003F 000000000000000F LoaderData 00000000AD836000-00000000AD844FFF 000000000000000F 000000000000000F available 00000000AD845000-00000000C0F8DFFF 0000000000013749 000000000000000F BS_data 00000000C0F8E000-00000000C128EFFF 0000000000000301 000000000000000F available 00000000C128F000-00000000C158FFFF 0000000000000301 000000000000000F BS_data 00000000C1590000-00000000C16BCFFF 000000000000012D 000000000000000F available 00000000C16BD000-00000000C174AFFF 000000000000008E 000000000000000F BS_data 00000000C174B000-00000000C58FAFFF 00000000000041B0 000000000000000F available 00000000C58FB000-00000000C6C28FFF 000000000000132E 000000000000000F BS_code 00000000C6C29000-00000000C6FF2FFF 00000000000003CA 000000000000000F reserved 00000000C6FF3000-00000000C7023FFF 0000000000000031 000000000000000F reserved 00000000C7024000-00000000C7084FFF 0000000000000061 000000000000000F available 00000000C7085000-00000000C70ECFFF 0000000000000068 000000000000000F ACPI_NVS 00000000C70ED000-00000000C7215FFF 0000000000000129 000000000000000F ACPI_NVS 00000000C7216000-00000000C7219FFF 0000000000000004 000000000000000F ACPI_NVS 00000000C721A000-00000000C7231FFF 0000000000000018 000000000000000F RT_data 00000000C7232000-00000000C9C3CFFF 0000000000002A0B 800000000000000F RT_data 00000000C9C3D000-00000000C9E95FFF 0000000000000259 800000000000000F RT_data 00000000C9E96000-00000000C9E97FFF 0000000000000002 800000000000000F RT_data 00000000C9E98000-00000000C9F16FFF 000000000000007F 800000000000000F RT_code 00000000C9F17000-00000000C9F60FFF 000000000000004A 800000000000000F RT_code 00000000C9F61000-00000000C9FFEFFF 000000000000009E 800000000000000F BS_data 00000000C9FFF000-00000000C9FFFFFF 0000000000000001 000000000000000F available 0000000100000000-000000042FDFFFFF 000000000032FE00 000000000000000F reserved 00000000CB000000-00000000CF1FFFFF 0000000000004200 8000000000000000 MemMapIO 00000000F8000000-00000000FBFFFFFF 0000000000004000 8000000000000001 MemMapIO 00000000FEC00000-00000000FEC00FFF 0000000000000001 8000000000000001 MemMapIO 00000000FED00000-00000000FED03FFF 0000000000000004 8000000000000001 MemMapIO 00000000FED1C000-00000000FED1FFFF 0000000000000004 8000000000000001 MemMapIO 00000000FEE00000-00000000FEE00FFF 0000000000000001 8000000000000001 MemMapIO 00000000FF000000-00000000FFFFFFFF 0000000000001000 8000000000000001 reserved : 17,045 Pages (69,816,320) LoaderCode: 205 Pages (839,680) LoaderData: 291 Pages (1,191,936) BS_code : 4,432 Pages (18,153,472) BS_data : 18,379 Pages (75,280,384) RT_code : 232 Pages (950,272) RT_data : 13,729 Pages (56,233,984) available : 4,131,371 Pages (16,922,095,616) ACPI_NVS : 332 Pages (1,359,872) MemMapIO : 20,490 Pages (83,927,040) Total Memory: 16,285 MB (17,076,105,216) Bytes Link to comment Share on other sites More sharing options...
Zenith432 Posted January 7, 2018 Share Posted January 7, 2018 There is a fix for bug introduced by AMI UEFI BIOS. As well whole AptioFix was developed to fight with AMI AllocateMemory() stupid function while EDK2 AllocateMemory is good. See legacy Clover == CloverEFI doesn't need AptioFix. I have AMI firmware, and it needs either of the AptioFixDrv to work. I don't know which "Aptio" it is because there's no indication. It's made by MB manufacturer. Is there any way to identify the "Aptio" version? Link to comment Share on other sites More sharing options...
Matgen84 Posted January 7, 2018 Share Posted January 7, 2018 It is not working in my case (rig in signature) Last login: Sun Jan 7 17:36:41 on console fabios-Mac-Pro:~ fabio$ sudo nvram -p Password: fakesmc-key-MSWr-ui8 %00 flagstate %00%00%00%00%00%00%00%00%00%00%00%00%00%00%00%00%00%00%00%00%00%00%00%00%00%00%00%00%00%00%00%00 fakesmc-key-RBr -ch8* j90%00%00%00%00%00 fakesmc-key-EPCI-ui32 %06%f0%f0%00 fakesmc-key-MSAc-ui16 %00%00 efi-boot-device <array><dict><key>IOMatch</key><dict><key>IOProviderClass</key><string>IOMedia</string><key>IOPropertyMatch</key><dict><key>UUID</key><string>3A55787E-1275-45BE-87EB-0ACF96BC0343</string></dict></dict></dict></array> fakesmc-key-CLKH-{clh %00%00p%80%00%01%19@ bootercfg (%00 fakesmc-key-NATi-ui16 %00%1e LocationServicesEnabled %01 fakesmc-key-NTOK-ui8 %01 fakesmc-key-BATP-flag %00 fakesmc-key-$Num-ui8 %01 fakesmc-key-REV -ch8* %02 %0f%00%00%18 bluetoothActiveControllerInfo %0a%18%05%0b%00%00%00%00`%14%f0%03%8c<%b5%a3 fakesmc-key-HI0N-ui8 %10 fmm-computer-name fabio%e2%80%99s Mac Pro fakesmc-key-MSTc-ui8 %00 fakesmc-key-MSPS-ui16 %00%03 fakesmc-key-RMde-char A csr-active-config g%00%00%00 fakesmc-key-RPlt-ch8* j90%00%00%00%00%00 fakesmc-key-BEMB-flag %00 fakesmc-key-#KEY-ui32 %00%00%00%10 SystemAudioVolume A fakesmc-key-MSFW-ui8 %01%00 specialbootdevice %02%01%0c%00%d0A%03%0a%00%00%00%00%01%01%06%00%02%1f%03%12%0a%00%00%00%ff%ff%00%00%04%01*%00%03%00%00%00%00H%0a%00%00%00%00%00%88%ca%93;%00%00%00%00~xU:u%12%beE%87%eb%0a%cf%96%bc%03C%02%02%04%03$%00%f7%fct%be|%0b%f3I%91G%01%f4%04.hB%ea%00%1f.%994-E%bd%9e%da%ff%12}%c4%a6%7f%ff%04%00 security-mode none fakesmc-key-LsBA-ui8 %90 fakesmc-key-$Adr-ui32 %00%00%03%00 nvda_drv 1%00 EFILoginHiDPI %00%00%00%00 fakesmc-key-CLKT-ui32 %00%00%f7%aa fakesmc-key-MSDW-ui8 %01 fakesmc-key-BNum-ui8 %00 SystemAudioVolumeDB %f0 fakesmc-key-NATJ-ui8 %02 efi-boot-device-data %02%01%0c%00%d0A%03%0a%00%00%00%00%01%01%06%00%02%1f%03%12%0a%00%00%00%ff%ff%00%00%04%01*%00%03%00%00%00%00H%0a%00%00%00%00%00%88%ca%93;%00%00%00%00~xU:u%12%beE%87%eb%0a%cf%96%bc%03C%02%02%04%03$%00%f7%fct%be|%0b%f3I%91G%01%f4%04.hB%ea%00%1f.%994-E%bd%9e%da%ff%12}%c4%a6%7f%ff%04%00 fabios-Mac-Pro:~ fabio$ I had the same problem: I delete rcscript and type sudo nvram -c in terminal. After reboot, I do the test and it works. Link to comment Share on other sites More sharing options...
Cyberdevs Posted January 7, 2018 Share Posted January 7, 2018 Hi guys, Is there any specific notes on the SIP values regarding the new AptioFixDrv? Thanks in advance Link to comment Share on other sites More sharing options...
Guest Posted January 7, 2018 Share Posted January 7, 2018 I will try it Thanks I had the same problem: I delete rcscript and type sudo nvram -c in terminal. After reboot, I do the test and it works. without slide=128 with clover R.4369 I have always printf error also with it there is error message with -x flag (slide=128 -s instead is good) in printf error condition after waiting a couple of seconds or more if I press enter on keyboard system enters in bios configuration from there I have force to override boot from my osx ssd disk but system freeze on bios Link to comment Share on other sites More sharing options...
Slice Posted January 7, 2018 Share Posted January 7, 2018 I have AMI firmware, and it needs either of the AptioFixDrv to work. I don't know which "Aptio" it is because there's no indication. It's made by MB manufacturer. Is there any way to identify the "Aptio" version? I don't know. I also have AMI UEFI without "Aptio" word. And AptioFix is exactly for me. Yes, it because built because of Aptio, but is not exclusive to Aptio. Even today, people use AptioFix on non-AMI machines. You want UEFI to emulate BIOS so that you can load another UEFI environment? With some relocation magic, this really is a non-isse tbh May be other manufacturers uses same codebase. CloverEFI somehow works better then OEM UEFI so why I asked. Hi guys, Is there any specific notes on the SIP values regarding the new AptioFixDrv? Thanks in advance Why not using Clover default value? Something wrong? Link to comment Share on other sites More sharing options...
Ramalama Posted January 7, 2018 Share Posted January 7, 2018 @apianti I have a small problem with the Calculation of the Slide Value... (Annd i seems like i need one, because i cant boot without) i have this Region Available: Available 0000000000100000-0000000027D67FFF 0000000000027C68 000000000000000F But if i start to Calculate: 0x100000 - 0x100000 = 0x0 The Next Available Big Region is: Available 0000000100000000-000000109FFFFFFF 0000000000FA0000 000000000000000F But thats above 0x100000000 ... I have attached my Memmap, but i dont understand which region i need.... Maybe anyone can help me... Cheers Link to comment Share on other sites More sharing options...
Cyberdevs Posted January 7, 2018 Share Posted January 7, 2018 Why not using Clover default value? Something wrong? I was just curious to see if we can fully enable the SIP with the new AptioFixDrv or not. Link to comment Share on other sites More sharing options...
mhaeuser Posted January 7, 2018 Share Posted January 7, 2018 i have this Region Available: Available 0000000000100000-0000000027D67FFF 0000000000027C68 000000000000000F But if i start to Calculate: 0x100000 - 0x100000 = 0x0 slide=0 2 Link to comment Share on other sites More sharing options...
Guest Posted January 7, 2018 Share Posted January 7, 2018 tried (I have had not rescript installed ) for me it is not working (X99 rig) After reboot test call is not there I had the same problem: I delete rcscript and type sudo nvram -c in terminal. After reboot, I do the test and it works. Link to comment Share on other sites More sharing options...
apianti Posted January 7, 2018 Share Posted January 7, 2018 Please don't kill "random KASLR", thx ;p That's not what I meant because it obviously works for some. I meant trying to figure out some way to determine when the slide needs to be figured out because it will fail. Now i tried newest ..69 Clover.pkg (DLed from SF, installed for test on USB Stick). Used aptiofix2drv-64.efi - before i must use the aptiofixdrv-64.efi on my GA-Z77 MB. Worked! So no need for me to calc and use an slide value. Thanks Q: The size of that ..2..efi coming within the .pkg is 23KB, that test version from page 800 is about 52 KB size. Same stuff / code - only bigger by debug code / not size optimized? Or should i try also that test ..2fix.. version from page 800? The toolchain and debug information. The package is built for release usually everything else (especially by me) is debug because we want information. It looks like the fix is good so you should be fine with the one from the package. It's obviously better to use the smaller more optimized one. I don't know why my bin is larger, the toolchain is what makes the diff He was referring to the fix I posted in #15994. I think yours was built release, no? It's like half the size. Link to comment Share on other sites More sharing options...
mhaeuser Posted January 7, 2018 Share Posted January 7, 2018 That's not what I meant because it obviously works for some. I meant trying to figure out some way to determine when the slide needs to be figured out because it will fail. Ugh, you would need to scan the kernel image before it is extracted by boot.efi... I still prefer conditional relocation He was referring to the fix I posted in #15994. I think yours was built release, no? It's like half the size. Mine is RELEASE and 53 KB... so yes, it's a lot larger Link to comment Share on other sites More sharing options...
apianti Posted January 7, 2018 Share Posted January 7, 2018 Hi there, I've tested r4369 on InsydeH20 and the nvram is not working without the emulation driver :/ Let me know what info u need. Ah.... Ummm.... Why you bought a board with an Insyde firmware? Then how you determined it's not working. Also what else are you doing/using? -s single mode is working. -x safe mode is not working. Only shown "does printf work?" 나의 LG-F800S 의 Tapatalk에서 보냄 Hasn't -x had problems for a while now though? Probably need to do some more debugging, or maybe it just doesn't exist anymore.... I think it is from apfs.efi. My compilation is 23kb. The toolchain is XCODE5. The sources are from sf.net. RELEASE. Are you sure you have the same sources? Yeah I've never seen that line before. There's confusion because I posted a DEBUG driver and it's twice the size. @apianti @vit9696 Thank you Works here HP ProBook 6570b WithOut EmuV Was mostly vit9696, Download-Fritz, and some others. I just do what I do best and burned the forest down.... First of all thanks to vit9696. This is the solution for Aptio UEFI BIOS (AptioFix), it is not for InsydeH20 BIOS. Yeah, but it could work, it just doesn't. Remember when Insyde firmware actually just worked really well? Neither do I.... HAHAHA Oh is it? My bad. It's not implausible it could have worked. It is not working in my case (rig in signature) Oh don't tell me you have the holy grail of the worst firmware ever made, is that the ASUS X99 that has the whitelist, the memory fragmentation issue, and the SMM issue? Ummm... Hmmmm...... I am unsure at this moment. Hi currently using the new AptioFix V2 without slide value and everything seems to work fine should I still calculate slide value and use it? You can but why? If it's working you're good. I have AMI firmware, and it needs either of the AptioFixDrv to work. I don't know which "Aptio" it is because there's no indication. It's made by MB manufacturer. Is there any way to identify the "Aptio" version? Use the newest version of AptioFix2 it does not matter it SHOULD be working for all Aptio firmwares now.... I had the same problem: I delete rcscript and type sudo nvram -c in terminal. After reboot, I do the test and it works. You have to reboot after you set a variable or it is not guaranteed to be set because the nvram app says that it won't necessarily write the value until a successful restart or shutdown. Hi guys, Is there any specific notes on the SIP values regarding the new AptioFixDrv? Thanks in advance Use it? Also, AptioFix2. without slide=128 with clover R.4369 I have always printf error also with it there is error message with -x flag (slide=128 -s instead is good) in printf error condition after waiting a couple of seconds or more if I press enter on keyboard system enters in bios configuration from there I have force to override boot from my osx ssd disk but system freeze on bios If you need slide=128 then you need it for every boot including -x. I don't know. I also have AMI UEFI without "Aptio" word. And AptioFix is exactly for me. May be other manufacturers uses same codebase. CloverEFI somehow works better then OEM UEFI so why I asked. Why not using Clover default value? Something wrong? Yeah the fix is just badly named because it was originally meant just for Aptio firmwares since that's what was causing us trouble ~7 years ago. @apianti I have a small problem with the Calculation of the Slide Value... (Annd i seems like i need one, because i cant boot without) i have this Region Available: Available 0000000000100000-0000000027D67FFF 0000000000027C68 000000000000000F But if i start to Calculate: 0x100000 - 0x100000 = 0x0 The Next Available Big Region is: Available 0000000100000000-000000109FFFFFFF 0000000000FA0000 000000000000000F But thats above 0x100000000 ... I have attached my Memmap, but i dont understand which region i need.... Maybe anyone can help me... Cheers Lol that is slide=0. slide=0 Yeah. I was just curious to see if we can fully enable the SIP with the new AptioFixDrv or not. Use AptioFix2. You should be able. tried (I have had not rescript installed ) for me it is not working (X99 rig) After reboot test call is not there Some other issue then. X99, gah. Ugh, you would need to scan the kernel image before it is extracted by boot.efi... I still prefer conditional relocation Why would the kernel need scanned? What do you mean by conditional relocation? I have a feeling that there is another bug because why would it not say it could not allocate a region? It appears to have been happy to place the kernel somewhere and it got it. So what's causing the hang? Did it overwrite other areas? Mine is RELEASE and 53 KB... so yes, it's a lot larger Oh I was looking at the compressed size. It's the same size as mine and mine was actually RELEASE too by accident, lol... So don't see how it got half the size in the package. Link to comment Share on other sites More sharing options...
mhaeuser Posted January 7, 2018 Share Posted January 7, 2018 Why would the kernel need scanned? What do you mean by conditional relocation? How do you know how much space the kernel will need without scanning it? Conditional relocation: Try to allocate, see if it worked, if not, relocate just that block. About the rest: Not sure what you are refering to tbh. So don't see how it got half the size in the package. XCODE5 vs VS2015x86 :/ Link to comment Share on other sites More sharing options...
apianti Posted January 7, 2018 Share Posted January 7, 2018 How do you know how much space the kernel will need without scanning it? Well picking the largest area is all that matters...? If it can't fit there then it can't fit anywhere... EDIT: Can also keep a semi KASLR by getting all the regions that seem big enough, usually 0x12000 pages is fine. Then choose one by some randomization just like KASLR but with only those slide values. Conditional relocation: Try to allocate, see if it worked, if not, relocate just that block. Might as well use AptioFix then. I do not like that. About the rest: Not sure what you are refering to tbh. Before, whenever it failed to allocate a region because it chose a wrong slide value it very clearly printed that it couldn't allocate the region and gave it's address and page count. XCODE5 vs VS2015x86 :/ Oh, you're using VS? Me too. Then that explains it, maybe the optimization is not as high for VS then... Maybe we should look into that. Link to comment Share on other sites More sharing options...
mhaeuser Posted January 7, 2018 Share Posted January 7, 2018 Well picking the largest area is all that matters...? If it can't fit there then it can't fit anywhere... So, you plan to entirely ditch the original purpose of KASLR? Before, whenever it failed to allocate a region because it chose a wrong slide value it very clearly printed that it couldn't allocate the region and gave it's address and page count. Probably just killed that code? Scanned boot.efi for it? Might as well use AptioFix then. I do not like that. Lol yeah, may just use a method that is way more error-prone and disables KASLR ;p Oh, you're using VS? Me too. Then that explains it, maybe the optimization is not as high for VS then... Maybe we should look into that. A bud also reported dramatic increase in size in bin size with an edk2 update, not sure which... so yeah, idk if they work around some bugs or whatever Link to comment Share on other sites More sharing options...
apianti Posted January 7, 2018 Share Posted January 7, 2018 So, you plan to entirely ditch the original purpose of KASLR? You didn't read my edit. And that's already the case if you use AptioFix. Probably just killed that code? Scanned boot.efi for it? I saw it like a few days ago, on my own screen. Lol yeah, may just use a method that is way more error-prone and disables KASLR ;p AptioFix disables KASLR to use relocation.................... A bud also reported dramatic increase in size in bin size with an edk2 update, not sure which... so yeah, idk if they work around some bugs or whatever Yeah possible they made some changes to build flags or to code that bloated. EDIT: It's probably just a different optimization flag. Like full for him compared to speed for VS or something. Who knows. Link to comment Share on other sites More sharing options...
mhaeuser Posted January 7, 2018 Share Posted January 7, 2018 You didn't read my edit. And that's already the case if you use AptioFix. [...] AptioFix disables KASLR to use relocation.................... Uhm, api.. that is my point lol Link to comment Share on other sites More sharing options...
Recommended Posts