CDMA Touch Pro XIPs 20764 20771 20954 21015 21018 21028 21032 21109

[scrosler]

Originally Posted by gguruusa

You do know you can patch up any kitchen to work, right? It can be done easiest, however using PPCkitchen.

ok, I've "stopped realizing that the PPCK is NOT FOR EVERYONE". Now I'm onto it IS for everyone.
lol - You realize that you are trying to argue in favor of using tools that even the author acknowledges as being antiquated? Essentially, you're arguing that using ancient versions of software is "innovative and progressive". Nice logic.

------------------------------

Back to the thread subject, you guys completely lost me in your xip discussion, and if I'm confused I'm sure everyone else is too.

Things are taken out of xip for two reasons: not enough xip space allocated (ram problem), or, xip partition isn't sized large enough (rom problem). Both are resizeable, but it's just plain easier to work within the existing layout than restructure stuff - so that's what most people do. Most of the older tools don't react well to non-standard layouts, anyway.

Things execute faster when they are in XIP form because they don't have to be rebased on load (rebasing is where you change all the jump tables to reflect where the module was actually loaded in memory). They also only load once (obviously, since they only load at a single fixed address). Unfortunately, nothing is free - not everything is pure and reentrant (which you must be to work right in xip form), and xip modules aren't unloadable - they permanently consume resources.

Well thank you for appreciating my logic ;c)

Question Guru, Where did you start? Just decompiled a ROM then created the PPC Kitchen? No intermediates? No learning curve? Just happened? When I am talking about innovation and growth, per se, Im reffering to my knowledge of how ROM's work and others knowledge. This is inovating (and asdly I must admit even exciting) to me to be able to understand XIP's, ROM folders, etc. Never done anything on this level before. Why stiffle it? Why try to stop the learning? Im not talking about inovation on a physical product level. IM talking about the human brain and what it can do when it starts learning something. That my friend is a form of innovation. But pelase lets not turn this into a game of symantecs, my terminology may not be perfect, nor am I.

I know the bepe style ktichens are old news to you. No doubt. But they work regardless of what the author says about there age. Fact of the matter is I would certainly put my ROM up on any benchmark test you have or can think of in comparison with a ROM produced by the PPCK.

BTW: I would love to work with someone on importing the correct GPS drivers so the ktichen would benefit the rest of us. As of now it doesnt. Even posted about in a thread or two. ASked a few people here and there. Wanted it even more than life itself at the time. But I never found what I was looking for. Had to move on. Would you like to help me undertand the PPC Kitchen better? If the kitchen is as good as you beleive it is maybe you will help instead of responding to me in a condesending matter of factly way. The balls is in your court now. You gonna pony and really help the XV6850 community or spew more condesension and rude comments towards me?

Why _ARE_ you so disrespectful to me? I give you guys nothing but respect. Its all over my own personal website. Here is a direct quote from my thank you post on my site. "The PPC Kitchen Team - For all the work they do! They are way under appreciated. Without the work and dedication of these people this ROM would not exist." I gave you and your team credit for helping me even though you had no direct impact on what I did. You repay me with total disrespect because I choose to use a differnt kitchen? WOW! I promtoe YOUR team and prodcut on MY website (which I dont even use because it doesnt fully support MY phone)
You think if someone paid you respect you would do it back.

Anyways... Ball is in your court now. Are you really here to help?
Thank you for your time, your conversation certainly broke the boredom of flashing all night again ;c)

-Scott


EDIT1: I even give your team props in my sig block man. Come on, lets work togeth instead of argue. Are you willing?

[lllboredlll]

Originally Posted by gguruusa

You do know you can patch up any kitchen to work, right? It can be done easiest, however using PPCkitchen.

ok, I've "stopped realizing that the PPCK is NOT FOR EVERYONE". Now I'm onto it IS for everyone.
lol - You realize that you are trying to argue in favor of using tools that even the author acknowledges as being antiquated? Essentially, you're arguing that using ancient versions of software is "innovative and progressive". Nice logic.

------------------------------

Back to the thread subject, you guys completely lost me in your xip discussion, and if I'm confused I'm sure everyone else is too.

Things are taken out of xip for two reasons: not enough xip space allocated (ram problem), or, xip partition isn't sized large enough (rom problem). Both are resizeable, but it's just plain easier to work within the existing layout than restructure stuff - so that's what most people do. Most of the older tools don't react well to non-standard layouts, anyway.

Things execute faster when they are in XIP form because they don't have to be rebased on load (rebasing is where you change all the jump tables to reflect where the module was actually loaded in memory). They also only load once (obviously, since they only load at a single fixed address). Unfortunately, nothing is free - not everything is pure and reentrant (which you must be to work right in xip form), and xip modules aren't unloadable - they permanently consume resources.

Well unless things have changed ALOT since I last used PPCKitchen I will have to say it's NOT for everyone. You can argue that point all day long til you are blue in the face and I will not agree with you. One of the biggest problems with the kitchen is the poorly compiled OEM Pack. Vague descriptions as well as dependencies that are not identified unless you do trial and error ROM building, which is a bunch of BS. I dont have all day to test a compilation because someone was too lazy to author their packages correctly. Last but not least, you guys make it a pain in the a$$ to compile a ROM for the kitchen. Constant changes get annoying after awhile. Especially when you supply NO resource for the cooks to use in order to integrate their ROM's into "Your" kitchen. It's personally traded information based on who kisses who's a$$. Plus I never have cared for "Glamware" and the kitchen is exactly that. Tell me your kitchen can give me the resources that Ervius's Kitchen supplies me and I'll tell you that you're full of ****. I can cook my ROM with the "Caveman" style kitchen and get more stability, less bugs and better ram performance but as always anything you support is perfect and you the only one that knows anything. We're just a bunch of dumb sob's.

[jakdillard]

Originally Posted by FormerPalmOS

First, I have to disagree that pagepool is the only thing that speeds up a ROM. I do agree that it is the biggest bang for the buck. But a correctly implemented file filter cache does seem to be helping. Then there's making modules out of static DLLs and EXEs (well, speeds up load but not execute).

And the XIP port I did was basically exactly as you describe, however there were more than one or two modules overlapping, and quite a bit of free space - it took a lot of time to remove the overlap and decompress out the free space. In the end I don't think it was worth it - not that much was gained in terms of recovered RAM. BTW - I did remove the debug drivers and other items that the experts remove - just kept mencfilt.dll and cachefilt.dll.

My XIP map attached so you can see what is left and where, if interested.

I looked at your xip and you did not move nk.exe and you still have nul spaces so you did gain anything

[jakdillard]

Originally Posted by lllboredlll

Well unless things have changed ALOT since I last used PPCKitchen I will have to say it's NOT for everyone. You can argue that point all day long til you are blue in the face and I will not agree with you. One of the biggest problems with the kitchen is the poorly compiled OEM Pack. Vague descriptions as well as dependencies that are not identified unless you do trial and error ROM building, which is a bunch of BS. I dont have all day to test a compilation because someone was too lazy to author their packages correctly. Last but not least, you guys make it a pain in the a$$ to compile a ROM for the kitchen. Constant changes get annoying after awhile. Especially when you supply NO resource for the cooks to use in order to integrate their ROM's into "Your" kitchen. It's personally traded information based on who kisses who's a$$. Plus I never have cared for "Glamware" and the kitchen is exactly that. Tell me your kitchen can give me the resources that Ervius's Kitchen supplies me and I'll tell you that you're full of ****. I can cook my ROM with the "Caveman" style kitchen and get more stability, less bugs and better ram performance but as always anything you support is perfect and you the only one that knows anything. We're just a bunch of dumb sob's.

I have noticed that the ppckitchen has slowed down on releasing any updates .... I honestly am beginning to wonder if it has anything to do with how you guys constantly gripe about it .... thats not cool to my knowledge the ppckitchen is freeware and all of this stuff that we do to these phones are beta so expect some bugs most people complain about stability speed and what have you but think of this the roms that you say are the most stable if you actually go behind the scenes of them most just play in the registry there is nothing actually done to the actuall image to change it .... like I posted back there is a dummies way to port over the numbers and thats as far as most go .....

[jakdillard]

Originally Posted by gguruusa

Ding. Exactly why I suggest everyone use the same one.

I have watched alot of threads and I see that alot of some of these chefs use hybrid builds which is not a good idea ..... also isn't 21109 a hexed build

[cornelious2]

I am going to throw my two cents into this argument. personally I have made a few roms with ppckitchen and its pretty nice, but I almost always invariably went back to someones cooked rom because of the fact that its easy to make a rom in easybake mode, but making the kitchen your own, I never figured it out fully. now using an old style (I think it was dogguy's kitchen) I was able to throw in my own mix of goodies and make a few custom roms that I was pretty darn happy with.

Gguruusa - I am sure the ppckitchen is probably the best way, but the best way i can think to describe it is quirky. I have been reading through the faqs lately to try and give it another shot, but havent gotten anyware really. Is there some good tutorial for how to get out of easy bake mode and into real rom dev?

[lllboredlll]

Originally Posted by jakdillard

I have noticed that the ppckitchen has slowed down on releasing any updates .... I honestly am beginning to wonder if it has anything to do with how you guys constantly gripe about it .... thats not cool to my knowledge the ppckitchen is freeware and all of this stuff that we do to these phones are beta so expect some bugs most people complain about stability speed and what have you but think of this the roms that you say are the most stable if you actually go behind the scenes of them most just play in the registry there is nothing actually done to the actuall image to change it .... like I posted back there is a dummies way to port over the numbers and thats as far as most go .....

Well for the record, I've never made a comment about the kitchen until now. I normally keep my opinions to myself. They can do what the want with it. I wont use it nor will I comment but what does get under my skin is the approach he takes about the kitchen. Just because he thinks it's great don't mean anyone else has to agree. It's called personal opinion and I'm allowed to have it no matter what he says. So if he will keep his personal opinion out of it, so would I. Not long ago I come across a thread where he flamed someone because they gave a tutorial on XIP's. He told them the were ego surfing or some dumb crap like that. Why not contribute instead of retaining that information just to make ones self seem supreme. I just think it's ignorant as hell to act like that and kinda childish too. One thing I loved about xda is everyone could disagree but yet still get along and not degrade each other. Well except when guru showed up then it got testy. I really wish this site could be like that. Further more I think the mod's need to hold accountability for actions such as those.


I once asked Calcu about XIP's. Instead of making me feel like an idiot he done one simple thing. It's called ctrl+v and I appreciated the hell out of it.

http://forum.xda-developers.com/showthread.php?t=438676


FormerPalmOS.... it's nice to see you making progress on how to work XIP's. Maybe more people like JD will come along and give you some more creative input instead of telling you that you don't know what your doing.


I will comment not further in this thread. It's become completely pointless.

[FormerPalmOS]

Originally Posted by jakdillard

I looked at your xip and you did not move nk.exe and you still have nul spaces so you did gain anything

I didn't check the physical map when I was done - just that there was no overlap and no space in the DLL start and virtual base addresses. Making it smaller wasn't actually one of my goals though - my primary goals were restoring menfilt.dll and cachefilt.dll and properly applying the cache. I accomplished my goals for my first try at this. For my second attempt I'll go back and look at this again. I would consider moving stuff out of XIP into IMGFS - but I just don't have time to figure out what can safely be moved out by trial and error - and a previous comment about a ROM with two modules removed from XIP not booting on a TP scared me off.

As for further compressing out space, I'm looking at my physical RAM map and see a ROM partition header from 0x000000 - 0x001000, then the start of NK.EXE region 0, etc. There are null spaces here and there, mostly sized < 0x1000 bytes. Is there any page alignment requirement for the data regions or module entry points? In other words, can I compact stuff to get rid of all null spaces? If so, then I assume that in the end I will have free space at the end of the image and can change the XIP partition size to get rid of that free space, which will return that RAM back to the OS. Am I on the right track here? Any pitfalls to look out for?

P.S. Can you guys arguing over the kitchen please move it to a different thread?

[kdj67f]

No offense, but moving the kitchen argument to a different thread would be great. I am kinda interested in an experienced member sharing some knowledge about XIP too, even if it is antiquated or whatever. Its still useful knowledge.

[jakdillard]

Originally Posted by FormerPalmOS

I didn't check the physical map when I was done - just that there was no overlap and no space in the DLL start and virtual base addresses. Making it smaller wasn't actually one of my goals though - my primary goals were restoring menfilt.dll and cachefilt.dll and properly applying the cache. I accomplished my goals for my first try at this. For my second attempt I'll go back and look at this again. I would consider moving stuff out of XIP into IMGFS - but I just don't have time to figure out what can safely be moved out by trial and error - and a previous comment about a ROM with two modules removed from XIP not booting on a TP scared me off.

As for further compressing out space, I'm looking at my physical RAM map and see a ROM partition header from 0x000000 - 0x001000, then the start of NK.EXE region 0, etc. There are null spaces here and there, mostly sized < 0x1000 bytes. Is there any page alignment requirement for the data regions or module entry points? In other words, can I compact stuff to get rid of all null spaces? If so, then I assume that in the end I will have free space at the end of the image and can change the XIP partition size to get rid of that free space, which will return that RAM back to the OS. Am I on the right track here? Any pitfalls to look out for?

P.S. Can you guys arguing over the kitchen please move it to a different thread?

Later tonight I'll upload a XIP that I'll be using in a coming rom look I actually left cachefilter and menchfilt in as well but check it out look at what I moved and deleted and go from there believe me once it is not very complicated