@HunoPPC
Continue the good and hard work! Seeing evolution hour after hour was impressive!
And one day, think about upgrading to a more recent SDK (even if I understand you prefer to change nothing for now when that works).
You know that we will support you. Tell me if you need help to profile your games.

@abalaban @HunoPPC
Oh my god, oh my friends, what a week-end!!

One more general question: what is the purpose of this clib2 implementation? Will it be merged into the official clib2 project?

@mr2 I don't know if Grim Fandango Remastered works ... I have a version from CD.

grim-patch.lab is provided in the ScummVM archive, in the "extras" directory. Just copy it in the main directory of GF.

Do you have MiniGL?

As GF is a 3D game, you need MiniGL, what is not the case of other games (except Myst 3, Longest Journey, Escape From Monkey Island).

@kas1e
When I submitted it, that seemed to be good ... anyway, I sent a message to orgin.

@walkero
Thanks for the idea about spam, I checked.
About the option to track a single program, this feature was not in the previous version ... even if I think about it for a long time. Now that I started to move using the performance monitor by default, it should not be so difficult, the mechanism is already used for the gcc profiling. For other platforms, I suppose I will have to be creative, one more time (like for the current X1000 support).
That's one of the feature for a future version, in my todo list.

Note that after 3 days, Hieronymus is still in the Uploads page

Do I have something to do? What happens?

@Raziel
I don't see what would prevent U-Boot running on X1000. It exists for very different devices. I am not aware of a constraint coming from a BIOS or something else.
Anyway, porting U-Boot would be a huge work, to be done by someone who knows U-Boot, the platform ... and has time!

@kas1e
Thanks for the news. Here the file still does not appear in the recent files.

@walkero
I hope you will enjoy this new version!
You can change the default sampling duration with the option "duration", for example "duration=60" for one minute (and you have the possibility to stop in that time with CTRL-C). I may have to clarify the command "help" where the information appears.
About debug option, it seems that -g is enough to access function symbols.

@salass0
I just see your old comment. I know more details about the crash in CloseElfTags(). I will create a test file and see if I can fix it or create a Bugzilla ticket.

@all
This release comes after a huge rework of the tool. Reporting is improved and among new features, A1222/X5000 support is added (at the moment, limited to OS4 beta users due to a change in the kernel) ... and finally, after several approaches, I found a solution to make profiling working on X1000 (not the best implementation but seems to do the job though).

@alfkil
Very good to see such progress on Qt6 port!

@asymetrix
I am working on an update of my statistical profiler Hieronymus, that is inspired by perf.

@msteed I haven't tried yet but that's great to see a new tool like that!

@sTix

I was a not aware about AxRuntime. I just had a quick look at it, so I don't know really how that works but thanks, the approach seems to be very good. Could be really useful.

Your advice about trying to design software in order to have some parts independent and testable / able to work on Linux for example, is very good (having testing in mind also helps).

To come back to Valgrind, I agree, this Linux tool clearly relies on mechanisms we won't have on OS4.

@all

To avoid confusion, Valgrind is not a debugger like GDB. Valgrind (and more exactly its tool/option memcheck) is used to track bad memory behaviors. Note that gcc and clang "address sanitizer" is very good on that point (but not accessible for us).

Personal note: Please, don't reopen 10-year old threads

@kas1e
Thanks for the feedback and this summary. Very good!

Note that I also have setups to build and test on Linux PPC if necessary.

Thank you guys for the continuous and hard work!

Good to see encouraging results for the 3D games supported by ScummVM.

If we had to focus on a reduced subset of games and technologies, depending on results:

- if performance with MiniGL is low, and knowing that it won't support Longest Journey, it may be better to work on GL4ES support (or OGLES2)

- I know that not all games work well so, among these games, I think it is would be good to get one working well, and then another one ... Grim Fandango would be great first, then Escape From Monkey Island, and finally Myst 3 or Longest Journey (I would prefer this one personally).

Count on me to test (I own the 4 games) and maybe to try fixing things.

That becomes off topic but the current situation is a nonsense (components moving from one side to the other, rewrite of commands, ...) and I don't see a good solution. And note that writing graphics library from scratch is another challenge than the copy command.

@kas1e
NovaBridge sounds great and seems to be the right/clean solution to provide MiniGL support for recent cards.

@kas1e I'am afraid that would be a lot of work for a small benefit.

If it is really considered as necessary, I think this is the kind of feature to work on in an updated set of development tools (compiler, clib, binutils, ...).

Just to mention that I've just pushed a pull request in clib2 project to fix the crash. At least, that must be done whatever the status of gprof is.

kas1e summarized well the situation. gprof relies on many elements: compiler, libc (and with have 2 of them), performance monitor support, signals (?), gprof tool ... At the end, it seems that gprof worked a long time ago. We don't know when it stopped to work even on machines that are supposed to get it running well (I tested yesterday on my MicroAOne).

About the crash, Fredrik fixed it in newlib. The same should be done in clib2.

Let's collect more information but I think all that is not in a good shape.

But what we need, after all, is the profiling feature. For me, I can help on some points about gprof but I am clearly going to focus on improving my profiling tool, Hieronymus (yes, I know that in the past, I already said that and nothing was released).

@Raziel
Quote:

I tested with a fresh SDK, including the default SDL1 library and MiniGL.

I haven't tried any investigations since.

@ amikit

Thanks for the info about Enhancer Pack V2, that will have only 2 variants. That's a good point.

A shop in amisphere could replace AmiStore that is not very convenient to use (and need a web browser to pay, what fails here with Odyssey). In this situation, Updated should be available from outside Enhancer Pack, to install and update software bought online.

The crash looks like the one I got 1 or 2 weeks ago on glGetIntegerv. I didn't investigate as I thought it was due to a lack of 3D capabilities (when I wanted to buy Warp3DSI drivers).

@Spectre660

Thank you for your explanation. I really appreciate. But even if that clarifies some things ... that should be simple by default.

We have to admit users are confused and that harms our small market and companies that aim to sell more and make users satisfied.

Things have to be simple. I insist on this point because there are some actions to take about that to improve communication (our topic here).