Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.

saxfun

macrumors regular
Original poster
Mar 14, 2016
138
24
Germany
After hours of testing various linux distress on two of my old PowerBooks 5,6

(1. Debian 12 IS Working 2. Adelie is working too 3. T2SDE failing cause of Grub 4. Gentoo is too much trouble . . .)

I tried to install 10.5 on the second of the 5,6 PowerBook, 1,67Ghz, LowerRamFailure (1GB installed) and a 60GB HD, ATI Radeon 9700. 10.5.9 seems to work well.

I installed MacPorts 2.10 via Website, made a self update and changed legacy-support (even not on 10.6) to

Code:
sudo port deactivate legacy-support
sudo port install legacy-support-devel


I tried to install fastfetch, but without M4 it is not gonna happen. Also gcc7, icu or other ports are not installable. First macports complained about "Failed Checksum", then I tried the option to ignore the checksum, but it is not working.

At the end I tried to install icu, I got following error. Can you give me some advise what I did wrong? Is there a solution for an intermediate user like me?

Thanks, saxfun :)




Code:
--->  Cleaning python27
--->  Removing work directory for python27
--->  Fetching archive for m4
--->  m4-1.4.19_1.darwin_9.ppc.tbz2 doesn't seem to exist in /opt/local/var/macports/incoming/verified
--->  Attempting to fetch m4-1.4.19_1.darwin_9.ppc.tbz2 from http://packages.macports.org/m4

  % Total    % Received % Xferd  Average Speed   Time    Time     Time  Current
                                 Dload  Upload   Total   Spent    Left  Speed
  0     0    0     0    0     0      0      0 --:--:--  0:00:04 --:--:--     0--->  Attempting to fetch m4-1.4.19_1.darwin_9.ppc.tbz2 from http://nue.de.packages.macports.org/m4

  % Total    % Received % Xferd  Average Speed   Time    Time     Time  Current
                                 Dload  Upload   Total   Spent    Left  Speed
  0     0    0     0    0     0      0      0 --:--:--  0:00:04 --:--:--     0--->  Attempting to fetch m4-1.4.19_1.darwin_9.ppc.tbz2 from http://mirror.fcix.net/macports/packages/m4

  % Total    % Received % Xferd  Average Speed   Time    Time     Time  Current
                                 Dload  Upload   Total   Spent    Left  Speed
  0     0    0     0    0     0      0      0 --:--:--  0:00:19 --:--:--     0--->  Applying patches to m4
--->  Applying patch-clang-before-34.diff
Executing:  cd "/opt/local/var/macports/build/_opt_local_var_macports_sources_rsync.macports.org_macports_release_tarballs_ports_devel_m4/m4/work/m4-1.4.19" && /usr/bin/patch -t -N -p0 < '/opt/local/var/macports/sources/rsync.macports.org/macports/release/tarballs/ports/devel/m4/files/patch-clang-before-34.diff'
can't find file to patch at input line 8
Perhaps you used the wrong -p or --strip option?
The text leading up to this was:
--------------------------
|https://git.savannah.gnu.org/cgit/gnulib.git/commit/?id=605e20a698d3f7296cda18ac7dd84b1a8f388b50
|https://git.savannah.gnu.org/cgit/gnulib.git/commit/?id=d125d4f6d1a5d9845824aaf1d1c9dc69699bf2f1
|
|diff --git lib/cdefs.h lib/cdefs.h
|index 7d91b06..1cf2ebd 100644
|--- lib/cdefs.h
|+++ lib/cdefs.h
--------------------------
No file to patch.  Skipping patch.
1 out of 1 hunk ignored
can't find file to patch at input line 23
Perhaps you used the wrong -p or --strip option?
The text leading up to this was:
--------------------------
|diff --git lib/config.hin lib/config.hin
|index 1a9a70f..026b39c 100644
|--- lib/config.hin
|+++ lib/config.hin
--------------------------
No file to patch.  Skipping patch.
1 out of 1 hunk ignored
Command failed:  cd "/opt/local/var/macports/build/_opt_local_var_macports_sources_rsync.macports.org_macports_release_tarballs_ports_devel_m4/m4/work/m4-1.4.19" && /usr/bin/patch -t -N -p0 < '/opt/local/var/macports/sources/rsync.macports.org/macports/release/tarballs/ports/devel/m4/files/patch-clang-before-34.diff'
Exit code: 1
Error: Failed to patch m4: command execution failed
Error: See /opt/local/var/macports/logs/_opt_local_var_macports_sources_rsync.macports.org_macports_release_tarballs_ports_devel_m4/m4/main.log for details.
Error: Follow https://guide.macports.org/#project.tickets if you believe there is a bug.
Error: Processing of port icu failed
--->  Some of the ports you installed have notes:
  db48 has the following notes:
    The Java and Tcl bindings are now provided by the db48-java and
    db48-tcl subports.
  openssl11 has the following notes:
    This is the last release of OpenSSL 1.1. No further public security updates will be provided. Please migrate to the openssl3 port.
  python27 has the following notes:
    To make this the default Python or Python 2 (i.e., the version run by the 'python' or 'python2' commands), run one or both of:
   
        sudo port select --set python python27
        sudo port select --set python2 python27
 
Last edited:

saxfun

macrumors regular
Original poster
Mar 14, 2016
138
24
Germany
I solved it myself, after some trial and error.

at the end the repo from @doctor_dog was very helpful , KUDOS! I manually downloaded m4-1.4.19_1.darwin_9.ppc from his repo and put it in the DIR

/opt/local/var/macports/incoming/verified

and installed it via port install m4
 

barracuda156

macrumors 68020
Sep 3, 2021
2,295
1,514
After hours of testing various linux distress on two of my old PowerBooks 5,6

(1. Debian 12 IS Working 2. Adelie is working too 3. T2SDE failing cause of Grub 4. Gentoo is too much trouble . . .)

I tried to install 10.5 on the second of the 5,6 PowerBook, 1,67Ghz, LowerRamFailure (1GB installed) and a 60GB HD, ATI Radeon 9700. 10.5.9 seems to work well.

I installed MacPorts 2.10 via Website, made a self update and changed legacy-support (even not on 10.6) to

Code:
sudo port deactivate legacy-support
sudo port install legacy-support-devel


I tried to install fastfetch, but without M4 it is not gonna happen. Also gcc7, icu or other ports are not installable. First macports complained about "Failed Checksum", then I tried the option to ignore the checksum, but it is not working.

At the end I tried to install icu, I got following error. Can you give me some advise what I did wrong? Is there a solution for an intermediate user like me?

Thanks, saxfun :)




Code:
--->  Cleaning python27
--->  Removing work directory for python27
--->  Fetching archive for m4
--->  m4-1.4.19_1.darwin_9.ppc.tbz2 doesn't seem to exist in /opt/local/var/macports/incoming/verified
--->  Attempting to fetch m4-1.4.19_1.darwin_9.ppc.tbz2 from http://packages.macports.org/m4

  % Total    % Received % Xferd  Average Speed   Time    Time     Time  Current
                                 Dload  Upload   Total   Spent    Left  Speed
  0     0    0     0    0     0      0      0 --:--:--  0:00:04 --:--:--     0--->  Attempting to fetch m4-1.4.19_1.darwin_9.ppc.tbz2 from http://nue.de.packages.macports.org/m4

  % Total    % Received % Xferd  Average Speed   Time    Time     Time  Current
                                 Dload  Upload   Total   Spent    Left  Speed
  0     0    0     0    0     0      0      0 --:--:--  0:00:04 --:--:--     0--->  Attempting to fetch m4-1.4.19_1.darwin_9.ppc.tbz2 from http://mirror.fcix.net/macports/packages/m4

  % Total    % Received % Xferd  Average Speed   Time    Time     Time  Current
                                 Dload  Upload   Total   Spent    Left  Speed
  0     0    0     0    0     0      0      0 --:--:--  0:00:19 --:--:--     0--->  Applying patches to m4
--->  Applying patch-clang-before-34.diff
Executing:  cd "/opt/local/var/macports/build/_opt_local_var_macports_sources_rsync.macports.org_macports_release_tarballs_ports_devel_m4/m4/work/m4-1.4.19" && /usr/bin/patch -t -N -p0 < '/opt/local/var/macports/sources/rsync.macports.org/macports/release/tarballs/ports/devel/m4/files/patch-clang-before-34.diff'
can't find file to patch at input line 8
Perhaps you used the wrong -p or --strip option?
The text leading up to this was:
--------------------------
|https://git.savannah.gnu.org/cgit/gnulib.git/commit/?id=605e20a698d3f7296cda18ac7dd84b1a8f388b50
|https://git.savannah.gnu.org/cgit/gnulib.git/commit/?id=d125d4f6d1a5d9845824aaf1d1c9dc69699bf2f1
|
|diff --git lib/cdefs.h lib/cdefs.h
|index 7d91b06..1cf2ebd 100644
|--- lib/cdefs.h
|+++ lib/cdefs.h
--------------------------
No file to patch.  Skipping patch.
1 out of 1 hunk ignored
can't find file to patch at input line 23
Perhaps you used the wrong -p or --strip option?
The text leading up to this was:
--------------------------
|diff --git lib/config.hin lib/config.hin
|index 1a9a70f..026b39c 100644
|--- lib/config.hin
|+++ lib/config.hin
--------------------------
No file to patch.  Skipping patch.
1 out of 1 hunk ignored
Command failed:  cd "/opt/local/var/macports/build/_opt_local_var_macports_sources_rsync.macports.org_macports_release_tarballs_ports_devel_m4/m4/work/m4-1.4.19" && /usr/bin/patch -t -N -p0 < '/opt/local/var/macports/sources/rsync.macports.org/macports/release/tarballs/ports/devel/m4/files/patch-clang-before-34.diff'
Exit code: 1
Error: Failed to patch m4: command execution failed
Error: See /opt/local/var/macports/logs/_opt_local_var_macports_sources_rsync.macports.org_macports_release_tarballs_ports_devel_m4/m4/main.log for details.
Error: Follow https://guide.macports.org/#project.tickets if you believe there is a bug.
Error: Processing of port icu failed
--->  Some of the ports you installed have notes:
  db48 has the following notes:
    The Java and Tcl bindings are now provided by the db48-java and
    db48-tcl subports.
  openssl11 has the following notes:
    This is the last release of OpenSSL 1.1. No further public security updates will be provided. Please migrate to the openssl3 port.
  python27 has the following notes:
    To make this the default Python or Python 2 (i.e., the version run by the 'python' or 'python2' commands), run one or both of:
 
        sudo port select --set python python27
        sudo port select --set python2 python27

I cannot reproduce the problem. The patch applies cleanly to m4. Which version of MacPorts do you have installed?

Code:
--->  Applying patches to m4
--->  Applying patch-clang-before-34.diff
Executing:  cd "/opt/local/var/macports/build/_opt_local_var_macports_sources_rsync.macports.org_macports_release_tarballs_ports_devel_m4/m4/work/m4-1.4.19" && /usr/bin/patch -p0 < '/opt/local/var/macports/sources/rsync.macports.org/macports/release/tarballs/ports/devel/m4/files/patch-clang-before-34.diff'
patching file 'lib/cdefs.h'
patching file 'lib/config.hin'

(Notice, this patch is not specific to 10.5, it applies for every system.)
 

barracuda156

macrumors 68020
Sep 3, 2021
2,295
1,514
Newest Version for Leopard, 2.10.1.
Some ports failed to compile, like icu, too.

I discovered @doctor_dog 's unofficial repo and atm i am installing stuff from there.


And you have Xcode installed? I just can’t see how a patch can fail to apply: if it did not, if will break all systems, not only 10.5.
There are no failures though: https://ports.macports.org/port/m4/details

So if it fails, something should be wrong locally. If you could run sudo port clean m4 and build again, it will be helpful.
If we have an error specific to 10.5, I will fix it.

icu is a different story, but last time I built it with gcc7, it worked. Notice, it will not build for ppc64 with gcc7, you will need a newer gcc.
On 10.6 ppc I use gcc14, so everything builds fine (and available as pre-built packages).
 

saxfun

macrumors regular
Original poster
Mar 14, 2016
138
24
Germany
Xcode was installed, 3.1.4. I tried many times to uninstall m4, made a port selfupdate, tried to install it again. Failed again. Etc.pp.

ATM I’m building git, after gcc7 etc. frommte unofficial repo.
 

barracuda156

macrumors 68020
Sep 3, 2021
2,295
1,514
Xcode was installed, 3.1.4. I tried many times to uninstall m4, made a port selfupdate, tried to install it again. Failed again. Etc.pp.

Well, okay, I will try this specifically on 10.5.8 today.

On a general note, please report failures on https://trac.macports.org – this is in your interests. There is already a lot of stuff we are forced to support unofficially, it is better to have breakages on supported systems to be fixed in upstream rather than hacked locally. Otherwise every next user has to waste time repeating the same fixes over again, and many will simply give up.
 

saxfun

macrumors regular
Original poster
Mar 14, 2016
138
24
Germany
I hear you - I will report failures. Maybe it’s me or m old PowerBook. It’s a real adventure using Macports , even with the doctor_dogs repo. Some ports are failing, I have to compile them with the -s or without -b option. It is much worse than on e.g. Debian. Maybe the HD is getting worse or the lower ram failure is the reason of all trouble . . . I don’t know.
Thanks!
 

barracuda156

macrumors 68020
Sep 3, 2021
2,295
1,514
I hear you - I will report failures. Maybe it’s me or m old PowerBook. It’s a real adventure using Macports , even with the doctor_dogs repo. Some ports are failing, I have to compile them with the -s or without -b option. It is much worse than on e.g. Debian. Maybe the HD is getting worse or the lower ram failure is the reason of all trouble . . . I don’t know.
Thanks!

There are two systemic issues with MacPorts on 10.5 (and PowerPC in general):

1. When ports get updated, nothing is tested beyond three latest macOS versions (unless someone is really motivated to and has hardware). So literally every update can break something for most supported systems, not just PowerPC, and it gets known only from buildbots status, if someone cares to check. And then seldom something gets fixed, unless users ask for that. As much regrettable, this is a very understandable situation, because everyone has just that much time, and most developers do not prioritize legacy systems and exotic archs.
Most of these breakages are accidental and trivial (like C++ standard should be set or legacysupport port group added to a portfile), anyone could fix them, just no one care to. Some require hours or days of work to fix, and some are unfixable with reasonable efforts, and then a port has to be pegged to the last supported version instead (conditional on macOS version). For example, when some Python ports began using Rust, they were automatically broken for PowerPC, so I had to find versions which worked and restore that into portfiles. Another typical case is when Qt is switched from Qt4 to Qt5/Qt6, which are also broken, so that ports stop building (what should be done is to have a separate case for systems restricted to Qt4).

How many versions of Debian its community supports, how different are they and how many people they got working on it? )
It is non-trivial with macOS, where we got more than a dozen OS versions, which differ substantially, and with PowerPC they use different C++ runtime and different compiler, and perhaps 2 or 3 people do something for those PowerPC systems. Literally, the only reason why something still builds is that I keep updating about 10k or whatever installed ports and handling everything that gets broken. But I only have just that much of time (I only one Quad, so I cannot run multiple gcc builds in parallel) and zero support for this work. So it is to be expected that not everything works as supposed to. That sucks, but it is a realistic take.

2. A purely technical problem: unfortunately, 10.5 in MacPorts is stuck at gcc7 for no good reason, and that necessarily implies that many ports cannot be built. gcc7 is an old, unsupported compiler. Recent fixes are not backported, it does not support modern C++ standards. It should just not have been used at all. I have been using a modern gcc since gcc11 on PowerPC systems (mostly 10.6 but also 10.5 earlier), but I could not do anything about inertia in MacPorts about changing something in this regard. This could have been easily fixed, but as long as folks with commit access do not want to do that, the only thing left is unofficial support.
What should be kept in mind here: as things stand, most of testing, work and fixes for PowerPC are done for 10.6 with the current gcc version. I cannot guarantee anything for 10.4–10.5, though I have a reasonable idea what is expected to work or not. But strictly speaking, I just don’t know. I can only advise to use 10.6 ppc and have everything what works to work for sure, or otherwise report breakages on 10.5 in a hope someone cares to fix those. I cannot do much for 10.5 presently due to lack of time and extra and fast hardware, but I can try to help at least with some fundamental stuff.
 
  • Like
Reactions: Yumiko Noua

saxfun

macrumors regular
Original poster
Mar 14, 2016
138
24
Germany
Point taken. But 10.6A190 breaks suspend on my PowerBook, a real showstopper. Also the non-working support for the Radeon ATI 9700.

Networking seemed to work, then disappeared. I do not remember Wifi.

So I’m back on 10.5.
@barracuda156 : on 10.5 with the unofficial repo I got also many problems installing binaries or compiling from source. I will report soon in this thread.
 

barracuda156

macrumors 68020
Sep 3, 2021
2,295
1,514
Point taken. But 10.6A190 breaks suspend on my PowerBook, a real showstopper. Also the non-working support for the Radeon ATI 9700.

Networking seemed to work, then disappeared. I do not remember Wifi.

So I’m back on 10.5.
@barracuda156 : on 10.5 with the unofficial repo I got also many problems installing binaries or compiling from source. I will report soon in this thread.

For 10.5 repo, perhaps rather report in a thread dedicated to it. So it is trackable later.

10a190 has issues, admittedly, it was a developer build after all. WiFi is totally fixable though, there are some instructions in 10.6 PPC thread (basically, several files should be replaced with those from 10.5.8).
I am a bit surprised there are differences in speed between 10.5 vs 10.6 on this machine; 10.5 is just as much resource-hungry, AFAIK. 10.4 will probably be more responsive, but then has a lot more stuff broken.
 
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.