Jump to content
buoo

RevoBoot

291 posts in this topic

Recommended Posts

The devs will move it to settings.h, just set the number and recompile.

With #define MAX_NUMBER_OF_P_STATES 19 in ssdt_pr_generator.h things go back to "normal":

MSRDumper PStatesReached: 16 22 26 30 33 41 42 43 44

Share this post


Link to post
Share on other sites
Advertisement
With #define MAX_NUMBER_OF_P_STATES 19 in ssdt_pr_generator.h things go back to "normal":

MSRDumper PStatesReached: 16 22 26 30 33 41 42 43 44

Great, now you can experiment and set it to 17, you should get higher mid range numbers.

 

That way we can tune the P-States we want to have.

Share this post


Link to post
Share on other sites

I'm trying to debug one-core fixed ratio, but I don't see the Turbo Ratio output by AICPUPM any more, why could that be?

Share this post


Link to post
Share on other sites
Why?

I'm currently using an efi-injection for my GPU and my audio chip and it works like a charm.

Because "my, my, my." Each user programmer?

Share this post


Link to post
Share on other sites
Because "my, my, my." Each user programmer?

Do you have an off-day?

I'm not a user programmer, is that a problem?

I simply ask you why do you think that efi strings are unreliable. You can't reply me, no problem, but don't be bossy, thanks ;)

 

Sorry for the Off-Topic.

Share this post


Link to post
Share on other sites

OK, did some testing. Setting all cores to a fixed ratio works, but RevoBoot will always output 59x, the reason being that the turbo ratio is always set to 1BBBA (for my CPU) if that setting is used. The CPU will actually use the multiplier set in UEFI, so the 59x is nothing to worry about.

Share this post


Link to post
Share on other sites

Interesting to see this posted here as a new topic… I agree with flAked that the text in the OP should be emphasised as being quoted from the RevoBoot wiki.

 

With regards to RevoBuilder, the original idea was to create a script which helped gather some of your system data for entering in to RevoBoot's settings.h file. As RevoBoot has grown, so have the options/directives in the settings.h file which in turn requires an understanding of exactly what each does, why you need it and how to obtain the required data. The comments within the settings.h file itself will go some way to explain things.

 

The last update I did to it was for RevoBoot v1.0.10 as that's the version I've been running. The releases after RevoBoot v1.0.10 have only happened in the last three days and I haven't yet looked at it properly or tried to get my head around all the changes that have been made so any updates to RevoBuilder will be as and when I get some time.

Share this post


Link to post
Share on other sites
With #define MAX_NUMBER_OF_P_STATES 19

MSRDumper PStatesReached: 16 22 26 30 33 41 42 43 44

#define MAX_NUMBER_OF_P_STATES 17

MSRDumper PStatesReached: 16 24 27 30 33 41 42 43 44

flAked: is the same for you?

Share this post


Link to post
Share on other sites

I haven't tested it, but it should be the same results for the same CPU, as it's just auto-generated.

 

Meanwhile, I added hibernation support based on Chameleon: https://github.com/flAked/RevoBoot

 

Testers?

 

You need to set

sudo pmset -a hibernatemode 1

and go to sleep. The computer will shut off. Power it on and start OSX, RevoBoot should pick up the sleepimage.

Share this post


Link to post
Share on other sites
I just saw the changes regarding hibernation support that you commit on GitHub.

Have you applied this changes at lastest revision of RevoBoot?

Yes, up-to-date. The latest.

 

Someone with Chameleon or Chimera available can test what

sudo purge

will output? Thanks!

Share this post


Link to post
Share on other sites

Quick question. In RevoBoot v1.0.18 I have the following in settings.h:

#define NUMBER_OF_TURBO_STATES 4 // Set to 4 by default.

 

Is this directive ignored by the code for my core2duo? in which case I can leave it set at 4?

To test, I've just compiled the latest version using my static cpu data and everything works fine while leaving it at 4 so I guess I've just answered my own question. :)

 

@flAked - I'll see if I get time this morning (only about 40 mins remaining before i leave for work) to test your wake from hibernation file.

Share this post


Link to post
Share on other sites
eanwhile, I added hibernation support based on Chameleon: https://github.com/flAked/RevoBoot

 

Testers?

 

You need to set

sudo pmset -a hibernatemode 1

and go to sleep. The computer will shut off. Power it on and start OSX, RevoBoot should pick up the sleepimage.

It doesn't work here.

Sleep-image is correctly created and then PC shutdown. But then I get a normal boot, image-sleep is not picked up.

Share this post


Link to post
Share on other sites

Make sure to have HIBERNATE_SUPPORT and DEBUG_BOOT set, what do you see before the kernel starts?

Also note that the sleepimage is quickly obsolete, so to test this you need to create a new sleepimage every time.

 

Quick question. In RevoBoot v1.0.18 I have the following in settings.h:

#define NUMBER_OF_TURBO_STATES 4 // Set to 4 by default.

 

Is this directive ignored by the code for my core2duo? in which case I can leave it set at 4?

This is used by the SSDT_PR generator, which only works for second generation i5/i7 with Turbo 2.0. This can be set via AUTOMATIC_SSDT_PR_CREATION.

Share this post


Link to post
Share on other sites
Make sure to have HIBERNATE_SUPPORT and DEBUG_BOOT set, what do you see before the kernel starts?

Also note that the sleepimage is quickly obsolete, so to test this you need to create a new sleepimage every time.

Both HIBERNATE_SUPPORT and DEBUG_BOOT are set to 1. Just cheked.

This is what I get before kernel starts:

sam4644.jpg

Share this post


Link to post
Share on other sites

Hmm seems the sleepimage is corrupt or incompatible for what ever reason. Can you try and remove it an generate a new one?

 

sudo rm /private/var/vm/sleepimage

 

Do you have FileVault / memory encryption enabled? That is not supported.

Share this post


Link to post
Share on other sites

It doesn't work.

Always get the same error regarding image signature before kernel starts.

Also this time the sleep-image (4,29GB) was correctly created and put in var/vm.

Don't know if could be help.

This is what I can see in /usr/bin/pmset -g log:

* Domain: hibernatestats
 - Message: Hibernate Statistics
 - Time: 14/06/11 13.37.54 GMT+02.00 
 - Signature: hibernatemode=0
 - UUID: 3AAADB21-378D-4F2A-9D4E-3B67CC5AB5CB

 * Domain: sleep
 - Message: Sleep: Platform Failure - AC
 - Time: 14/06/11 13.39.59 GMT+02.00 
 - Signature: Platform Failure
 - UUID: 742ADC8A-331A-4906-8B03-1DCEA23B44E7
 - Result: Failure

I also noted this on my kernel.log:

Sleep failure code 0x00000000 0x15000000

FileVault is currently disabled.

Share this post


Link to post
Share on other sites

You should have:

 * Domain: hibernatestats
- Message: Hibernate Statistics
- Time: 13/06/2011 22:01:36 CEST 
- Signature: hibernatemode=1
- UUID: 762E86E4-9BF2-4DCD-8404-893543B40016
- Time to write image (ms): undefined
- Time to read image (ms): 23774

 

Jun 13 22:00:00 slave kernel[0]: hibernate image path: /var/vm/sleepimage
Jun 13 22:00:00 slave kernel[0]: sizeof(IOHibernateImageHeader) == 512
Jun 13 22:00:00 slave kernel[0]: hibernate image major 14, minor 2, blocksize 512, pollers 5
Jun 13 22:00:00 slave kernel[0]: hibernate_alloc_pages flags 00000000, gobbling 0 pages
Jun 13 22:00:00 slave kernel[0]: hibernate_setup(0) took 0 ms
Jun 13 22:01:27 slave kernel[0]: hibernate_page_list_setall start 0xffffff8092b28000, 0xffffff8092b7e000
Jun 13 22:01:27 slave kernel[0]: hibernate_page_list_setall time: 182 ms
Jun 13 22:01:27 slave kernel[0]: hibernate_page_list_setall found pageCount 864332
Jun 13 22:01:27 slave kernel[0]: IOHibernatePollerOpen, ml_get_interrupts_enabled 0
Jun 13 22:01:27 slave kernel[0]: IOHibernatePollerOpen(0)
Jun 13 22:01:27 slave kernel[0]: hibernate_machine_init: state 2, image pages 293823, sum was 6fd7820b, image1Size 19fe0400, conflictCount 1572, nextFree 1385
Jun 13 22:01:27 slave kernel[0]: IOHibernatePollerOpen(), ml_get_interrupts_enabled 0
Jun 13 22:01:27 slave kernel[0]: IOHibernatePollerOpen(0)
Jun 13 22:01:27 slave kernel[0]: hibernate_machine_init reading
Jun 13 22:01:27 slave kernel[0]: PMStats: Hibernate read took 23774 ms
Jun 13 22:01:27 slave kernel[0]: hibernate_machine_init pagesDone 1010902 sum2 961f4117, time: 23774 ms

 

Check hibernatemode again, is it set to 1?

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

  • Recently Browsing   0 members

    No registered users viewing this page.

×