Clover General discussion

vector sigma · May 28, 2020

22 hours ago, MacNB said:

Hi @vector sigma looks like the 10.15.5 release update has changed since 10.15.5 Beta 2 when that patch was committed.

That patch does not work on Ivy Bridge anymore.

Any chance of doing your magic again ?

Just updated to the final version of 10.15.5 and the xcpm works as expected, except that still booting with r5117... later I'll try last commit ....just in case

Matgen84 · May 29, 2020

Hi @Slice

The latest commit d36f1ae introduce a new embedded patch. Can you explain its usefulness. Please.

Thanks

<key>EightApple</key>
<true/>

Slice · May 29, 2020

3 hours ago, Matgen84 said:
Hi @Slice

The latest commit d36f1ae introduce a new embedded patch. Can you explain its usefulness. Please.

Thanks
<key>EightApple</key>
<true/>

Sometimes, on some hardware with some OSes one can see apple with progress bar tears into 8 apples at some point of loading.

There was old methods to avoid this: find 01000075 and replace to 010000eb. But this method is OS version sensible.

I introduced system independent patch but I have no confirmation if it works.

chris1111 · May 29, 2020

7 hours ago, Slice said:

Sometimes, on some hardware with some OSes one can see apple with progress bar tears into 8 apples at some point of loading.

There was old methods to avoid this: find 01000075 and replace to 010000eb. But this method is OS version sensible.

I introduced system independent patch but I have no confirmation if it works.

Not working Mojave 10.14.6 and Catalina 10.5.5 HP ProBook 6570B CPU i5 3440m Intel HD 4000 Ivy Bridge

Edited May 29, 2020 by chris1111

Slice · May 29, 2020

1 hour ago, chris1111 said:

Not working Mojave 10.14.6 and Catalina 10.5.5 HP ProBook 6570B CPU i5 3440m Intel HD 4000 Ivy Bridge

Do you know working patch for this issue?

chris1111 · May 29, 2020

13 minutes ago, Slice said:

Do you know working patch for this issue?

No the patch inside CC not working here

Slice · May 29, 2020

23 hours ago, vector sigma said:

Just updated to the final version of 10.15.5 and the xcpm works as expected, except that still booting with r5117... later I'll try last commit ....just in case

I see no reason for concern.

9 minutes ago, chris1111 said:

No the patch inside CC not working here

As I expected. We need new ideas what to do.

Shaneee · May 29, 2020

@Slice I see the kernel patcher has had an overhaul. Would Procedure be the Clover version of OpenCore's Base in the kernel patcher? For example this OpenCore patch for AMD CPUs. How would you convert it to Clover? Thanks.

			<dict>
				<key>Base</key>
				<string>_cpu_topology_sort</string>
				<key>Comment</key>
				<string>algrey - cpu_topology_sort -disable _x86_validate_topology</string>
				<key>Count</key>
				<integer>1</integer>
				<key>Enabled</key>
				<true/>
				<key>Find</key>
				<data>
				6AAA//8=
				</data>
				<key>Identifier</key>
				<string>kernel</string>
				<key>Limit</key>
				<integer>0</integer>
				<key>Mask</key>
				<data>
				/wAA//8=
				</data>
				<key>MaxKernel</key>
				<string>19.99.99</string>
				<key>MinKernel</key>
				<string>17.0.0</string>
				<key>Replace</key>
				<data>
				Dx9EAAA=
				</data>
				<key>ReplaceMask</key>
				<data>
				</data>
				<key>Skip</key>
				<integer>0</integer>
			</dict>

apianti · May 29, 2020

5 hours ago, chris1111 said:

Not working Mojave 10.14.6 and Catalina 10.5.5 HP ProBook 6570B CPU i5 3440m Intel HD 4000 Ivy Bridge

4 hours ago, Slice said:

Do you know working patch for this issue?

3 hours ago, chris1111 said:

No the patch inside CC not working here

No patch yet, it looks like.

3 hours ago, Slice said:

I see no reason for concern.

As I expected. We need new ideas what to do.

I'll give it like maybe two days. Probably just find the original patch solution and follow it with the new kernel.

6 minutes ago, Shaneee said:

@Slice I see the kernel patcher has had an overhaul. Would Procedure be the Clover version of OpenCore's Base in the kernel patcher? For example this OpenCore patch for AMD CPUs. How would you convert it to Clover? Thanks.

Yes, works for both kernel and kexts.

https://github.com/CloverHackyColor/CloverBootloader/blob/master/rEFIt_UEFI/Platform/kernel_patcher.cpp#L179

https://github.com/CloverHackyColor/CloverBootloader/blob/master/rEFIt_UEFI/Platform/kernel_patcher.cpp#L100

Slice · May 30, 2020

7 hours ago, Shaneee said:

@Slice I see the kernel patcher has had an overhaul. Would Procedure be the Clover version of OpenCore's Base in the kernel patcher? For example this OpenCore patch for AMD CPUs. How would you convert it to Clover? Thanks.


			<dict>
				<key>Base</key>
				<string>_cpu_topology_sort</string>
				<key>Comment</key>
				<string>algrey - cpu_topology_sort -disable _x86_validate_topology</string>
				<key>Count</key>
				<integer>1</integer>
				<key>Enabled</key>
				<true/>
				<key>Find</key>
				<data>
				6AAA//8=
				</data>
				<key>Identifier</key>
				<string>kernel</string>
				<key>Limit</key>
				<integer>0</integer>
				<key>Mask</key>
				<data>
				/wAA//8=
				</data>
				<key>MaxKernel</key>
				<string>19.99.99</string>
				<key>MinKernel</key>
				<string>17.0.0</string>
				<key>Replace</key>
				<data>
				Dx9EAAA=
				</data>
				<key>ReplaceMask</key>
				<data>
				</data>
				<key>Skip</key>
				<integer>0</integer>
			</dict>

Base -> Procedure

Enable=true -> Disable=false

Identifier -> not needed, there is different sections

Count, Limit, Skip -> these values are default behaviour for Clover

MaxKernel,MinKernel -> MatchOS= All

Badruzeus · May 30, 2020

Anyway, I got this build error under Ubuntu 20.04 from recent commit(s). But I remember previously it did work (also on a same host). Thanks.


......

[CC] AutoGen
[CC] AutoGen
[CC] Measurement
[CC] fsw_efi
cc1plus: all warnings being treated as errors
make: *** [GNUmakefile:22595: /home/badruzeus/src/CloverBootloader/Build/Clover/RELEASE_GCC53/X64/rEFIt_UEFI/refit/OUTPUT/Platform/Events.obj] Error 1


build.py...
 : error 7000: Failed to execute command
    make tbuild [/home/badruzeus/src/CloverBootloader/Build/Clover/RELEASE_GCC53/X64/rEFIt_UEFI/refit]


build.py...
 : error F002: Failed to build module
    /home/badruzeus/src/CloverBootloader/rEFIt_UEFI/refit.inf [X64, GCC53, RELEASE]

- Failed -
Build end time: 13:31:21, May.30 2020
Build total time: 00:07:39

badruzeus@AO-HAPPY:~/src/CloverBootloader$

buildme_failed_AOH_UB20.04_GCC53.zip

LockDown · May 30, 2020

OS: 10.13.6

Xcode 10.1

Using buildme XCODE8 toolchain

Please enter your choice: 1
[CHECK XCODE]
[BUILD CLOVER]
Running custom build script
TOOLCHAIN_DIR: /Users/ella/src/CloverBootloader/toolchain
MTOC_PREFIX: /Users/ella/src/CloverBootloader/toolchain/bin/

Status: nasm-2.14.02.tar.xz not found.
  % Total    % Received % Xferd  Average Speed   Time    Time     Time  Current
                                 Dload  Upload   Total   Spent    Left  Speed
  0     0    0     0    0     0      0      0 --:--:-- --:--:-- --:--:--     0
curl: (60) SSL certificate problem: certificate has expired
More details here: https://curl.haxx.se/docs/sslcerts.html

curl performs SSL certificate verification by default, using a "bundle"
of Certificate Authority (CA) public keys (CA certs). If the default
bundle file isn't adequate, you can specify an alternate file
using the --cacert option.
If this HTTPS server uses a certificate signed by a CA represented in
the bundle, the certificate verification probably failed due to a
problem with the certificate (it might be expired, or the name might
not match the domain name in the URL).
If you'd like to turn off curl's verification of the certificate, use
the -k (or --insecure) option.
HTTPS-proxy has similar options --proxy-cacert and --proxy-insecure.
TOOLCHAIN_DIR: /Users/ella/src/CloverBootloader/toolchain
MTOC_PREFIX: /Users/ella/src/CloverBootloader/toolchain/bin/

Status: nasm-2.14.02.tar.xz not found.
  % Total    % Received % Xferd  Average Speed   Time    Time     Time  Current
                                 Dload  Upload   Total   Spent    Left  Speed
  0     0    0     0    0     0      0      0 --:--:-- --:--:-- --:--:--     0
curl: (60) SSL certificate problem: certificate has expired
More details here: https://curl.haxx.se/docs/sslcerts.html

curl performs SSL certificate verification by default, using a "bundle"
of Certificate Authority (CA) public keys (CA certs). If the default
bundle file isn't adequate, you can specify an alternate file
using the --cacert option.
If this HTTPS server uses a certificate signed by a CA represented in
the bundle, the certificate verification probably failed due to a
problem with the certificate (it might be expired, or the name might
not match the domain name in the URL).
If you'd like to turn off curl's verification of the certificate, use
the -k (or --insecure) option.
HTTPS-proxy has similar options --proxy-cacert and --proxy-insecure.


------------------------------------------------------------------------
buildme, Clover r5118 (SHA: d36f1ae23)
TOOLCHAIN: XCODE8 (override example: './buildme XCODE8')

1) build Clover
2) build Clover with HFSPlus
3) make pkg
4) make app
5) make app (with Clover)
6) make iso
7) build all
8) test build (no autogen, no boot files)
9) status
10) update Clover
11) show diff
12) open drivers directory
13) update Clover (reset changes)
14) clean BaseTools
15) quit
Please enter your choice:

Edited May 30, 2020 by ellaosx

vector sigma · May 30, 2020

1 hour ago, ellaosx said:

OS: 10.13.6

Xcode 10.1

Using buildme XCODE8 toolchain


Please enter your choice: 1
[CHECK XCODE]
[BUILD CLOVER]
Running custom build script
TOOLCHAIN_DIR: /Users/ella/src/CloverBootloader/toolchain
MTOC_PREFIX: /Users/ella/src/CloverBootloader/toolchain/bin/

Status: nasm-2.14.02.tar.xz not found.
  % Total    % Received % Xferd  Average Speed   Time    Time     Time  Current
                                 Dload  Upload   Total   Spent    Left  Speed
  0     0    0     0    0     0      0      0 --:--:-- --:--:-- --:--:--     0
curl: (60) SSL certificate problem: certificate has expired
More details here: https://curl.haxx.se/docs/sslcerts.html

curl performs SSL certificate verification by default, using a "bundle"
of Certificate Authority (CA) public keys (CA certs). If the default
bundle file isn't adequate, you can specify an alternate file
using the --cacert option.
If this HTTPS server uses a certificate signed by a CA represented in
the bundle, the certificate verification probably failed due to a
problem with the certificate (it might be expired, or the name might
not match the domain name in the URL).
If you'd like to turn off curl's verification of the certificate, use
the -k (or --insecure) option.
HTTPS-proxy has similar options --proxy-cacert and --proxy-insecure.
TOOLCHAIN_DIR: /Users/ella/src/CloverBootloader/toolchain
MTOC_PREFIX: /Users/ella/src/CloverBootloader/toolchain/bin/

Status: nasm-2.14.02.tar.xz not found.
  % Total    % Received % Xferd  Average Speed   Time    Time     Time  Current
                                 Dload  Upload   Total   Spent    Left  Speed
  0     0    0     0    0     0      0      0 --:--:-- --:--:-- --:--:--     0
curl: (60) SSL certificate problem: certificate has expired
More details here: https://curl.haxx.se/docs/sslcerts.html

curl performs SSL certificate verification by default, using a "bundle"
of Certificate Authority (CA) public keys (CA certs). If the default
bundle file isn't adequate, you can specify an alternate file
using the --cacert option.
If this HTTPS server uses a certificate signed by a CA represented in
the bundle, the certificate verification probably failed due to a
problem with the certificate (it might be expired, or the name might
not match the domain name in the URL).
If you'd like to turn off curl's verification of the certificate, use
the -k (or --insecure) option.
HTTPS-proxy has similar options --proxy-cacert and --proxy-insecure.


------------------------------------------------------------------------
buildme, Clover r5118 (SHA: d36f1ae23)
TOOLCHAIN: XCODE8 (override example: './buildme XCODE8')

1) build Clover
2) build Clover with HFSPlus
3) make pkg
4) make app
5) make app (with Clover)
6) make iso
7) build all
8) test build (no autogen, no boot files)
9) status
10) update Clover
11) show diff
12) open drivers directory
13) update Clover (reset changes)
14) clean BaseTools
15) quit
Please enter your choice:

Sorry but this is your problem. We provide no certificate for your SSH.

EDIT

anyway I added the -k option to ignore it

Edited May 30, 2020 by vector sigma

vector sigma · May 30, 2020

5 hours ago, Badruzeus said:

Anyway, I got this build error under Ubuntu 20.04 from recent commit(s). But I remember previously it did work (also on a same host). Thanks.


......

[CC] AutoGen
[CC] AutoGen
[CC] Measurement
[CC] fsw_efi
cc1plus: all warnings being treated as errors
make: *** [GNUmakefile:22595: /home/badruzeus/src/CloverBootloader/Build/Clover/RELEASE_GCC53/X64/rEFIt_UEFI/refit/OUTPUT/Platform/Events.obj] Error 1


build.py...
 : error 7000: Failed to execute command
    make tbuild [/home/badruzeus/src/CloverBootloader/Build/Clover/RELEASE_GCC53/X64/rEFIt_UEFI/refit]


build.py...
 : error F002: Failed to build module
    /home/badruzeus/src/CloverBootloader/rEFIt_UEFI/refit.inf [X64, GCC53, RELEASE]

- Failed -
Build end time: 13:31:21, May.30 2020
Build total time: 00:07:39

badruzeus@AO-HAPPY:~/src/CloverBootloader$

buildme_failed_AOH_UB20.04_GCC53.zip

it says:

/home/badruzeus/src/CloverBootloader/rEFIt_UEFI/Platform/kernel_patcher.h:54:9: error: ISO C++11 requires whitespace after the macro name [-Werror]
   54 | #define ID_SEС_CONST                           0x070f
      |         ^~~~~

but in code is ok:

#define ID_SEG_TEXT_CONST                      0x030e
#define ID_SEG_DATA_DATA                       0x080e
#define ID_SEG_DATA_DATA2                      0x081e
#define ID_SEC_BSS                             0x0a0e
#define ID_SEС_CONST                           0x070f
#define ID_SEG_DATA_COMMON                     0x090f
#define ID_SEG_DATA                            0x0f0f
#define ID_SEG_DATA_CONST                      0x110f
#define ID_SEG_KLD                             0x180f
#define ID_SEG_KLD2                            0x1a0f

...unless your compiler is not smart enough and wants only one space..

Edited May 30, 2020 by vector sigma

LockDown · May 30, 2020

54 minutes ago, vector sigma said:

Sorry but this is your problem. We provide no certificate for your SSH.

EDIT

anyway I added the -k option to ignore it

Hi @vector sigma

That fixed it. Problem solved!

Thanks

arsradu · May 30, 2020

Is it just me or that C in ID_SEC <-this one...looks like a different one from all the other ones? :)) Looks like different font. Aren't all capital letters? Cause that C definitely looks more "capital" than the others.

And coincidentally it's exactly the one for 0x070f :))

Anyway, sorry I just found myself typing non-sense again... Please, ignore my post if it's actually not related in any way. And sorry for wasting your time reading this.

apianti · May 30, 2020

7 hours ago, Badruzeus said:

Anyway, I got this build error under Ubuntu 20.04 from recent commit(s). But I remember previously it did work (also on a same host). Thanks.


......

[CC] AutoGen
[CC] AutoGen
[CC] Measurement
[CC] fsw_efi
cc1plus: all warnings being treated as errors
make: *** [GNUmakefile:22595: /home/badruzeus/src/CloverBootloader/Build/Clover/RELEASE_GCC53/X64/rEFIt_UEFI/refit/OUTPUT/Platform/Events.obj] Error 1


build.py...
 : error 7000: Failed to execute command
    make tbuild [/home/badruzeus/src/CloverBootloader/Build/Clover/RELEASE_GCC53/X64/rEFIt_UEFI/refit]


build.py...
 : error F002: Failed to build module
    /home/badruzeus/src/CloverBootloader/rEFIt_UEFI/refit.inf [X64, GCC53, RELEASE]

- Failed -
Build end time: 13:31:21, May.30 2020
Build total time: 00:07:39

badruzeus@AO-HAPPY:~/src/CloverBootloader$

2 hours ago, vector sigma said:

it says:


/home/badruzeus/src/CloverBootloader/rEFIt_UEFI/Platform/kernel_patcher.h:54:9: error: ISO C++11 requires whitespace after the macro name [-Werror]
   54 | #define ID_SEС_CONST                           0x070f
      |         ^~~~~

but in code is ok:


#define ID_SEG_TEXT_CONST                      0x030e
#define ID_SEG_DATA_DATA                       0x080e
#define ID_SEG_DATA_DATA2                      0x081e
#define ID_SEC_BSS                             0x0a0e
#define ID_SEС_CONST                           0x070f
#define ID_SEG_DATA_COMMON                     0x090f
#define ID_SEG_DATA                            0x0f0f
#define ID_SEG_DATA_CONST                      0x110f
#define ID_SEG_KLD                             0x180f
#define ID_SEG_KLD2                            0x1a0f

...unless your compiler is not smart enough and wants only one space..

1 hour ago, arsradu said:

Is it just me or that C in ID_SEC <-this one...looks like a different one from all the other ones? :)) Looks like different font. Aren't all capital letters? Cause that C definitely looks more "capital" than the others.

And coincidentally it's exactly the one for 0x070f :))

Anyway, sorry I just found myself typing non-sense again... Please, ignore my post if it's actually not related in any way. And sorry for wasting your time reading this.

There is an error in the file encoding. The C is encoded in some combined form in UTF-8 or something, see attached screenshot from hex editor showing invalid characters.

EDIT: Just so you see this is actually the cyrillic character С (es) not the latin character C (se):

0xD0 = b1101 0000

0xA1 = b1010 0001

b100 0010 0001 = 0x421

U+0421	С	Cyrillic Capital Letter Es	0489

Edited May 30, 2020 by apianti

vector sigma · May 30, 2020

2 hours ago, apianti said:

EDIT: Just so you see this is actually the cyrillic character С (es) not the latin character C (se):

That make sense, thanks. I committed the change.

arsradu · May 30, 2020

3 hours ago, apianti said:
EDIT: Just so you see this is actually the cyrillic character С (es) not the latin character C (se):

0xD0 = b1101 0000

0xA1 = b1010 0001

b100 0010 0001 = 0x421
U+0421	С	Cyrillic Capital Letter Es	0489

Yeeep, that also crossed my mind. And...according to Google, that is indeed the corresponding letter S from the Latin alphabet.

cyrilic-latin.jpeg.b79161752aa7ec43290e2c8e674fa81e.jpeg

Edited May 30, 2020 by arsradu

apianti · May 30, 2020

15 minutes ago, arsradu said:

Yeeep, that also crossed my mind. And...according to Google, that is indeed the corresponding letter S from the Latin alphabet.

Yeah, I know because I can read/write/speak Russian... Barely. Haha.

arsradu · May 30, 2020

Just now, apianti said:

Yeah, I know because I can read/write/speak Russian... Barely. Haha.

Knowing all the curse words doesn't count, man. :))

Joking, of course.

Slice · May 31, 2020

Oh, sorry! I have C and С at one button and they looks exactly the same. But strange that I had no build error while it must be.

Slice · May 31, 2020

AFAIK gcc10 allows to use cyrillic characters inside identifiers.

arsradu · May 31, 2020

I'm guessing you probably forgot to switch keyboard layout from RU to EN. And when you look at all that code, it's not really obvious which letter is wrong. The difference is really small. It can happen to anyone. Don't worry about it! But yeah, kinda weird that you haven't gotten any errors while compiling.

Also, I wanted to ask you something.

A few pages ago, I mentioned an issue with one of the older themes (Minimalism) when using newer Clover. Now...I checked pointer.icns and it's not present. There's only pointer.png (which is intended). So...any idea how or if this theme can still run with newer Clover? Or do I need to switch to a newer theme entirely? Right now, I just switched to Clovy and that works perfectly fine. I was just wondering if there's still a way to fix that particular one.

If it requires too much effort on Clover side, don't worry about it. It's not a big deal. I can use Clovy or another theme and that's it. I feel it might not be worth the trouble, just to get these old themes to work. I just love that simple black and white one.

Edited May 31, 2020 by arsradu

jsl2000 · May 31, 2020

On 5/30/2020 at 1:40 PM, Slice said:

Base -> Procedure

Enable=true -> Disable=false

Identifier -> not needed, there is different sections

Count, Limit, Skip -> these values are default behaviour for Clover

MaxKernel,MinKernel -> MatchOS= All

Thanks for your explanation.

Shaneee, XLNC:

Can you confirm these work for AMD CPUs at 10.15.x ?

Or can you provide the new kernel patches for Clover users ?

Clover General discussion

Recommended Posts

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites