• After 15+ years, we've made a big change: Android Forums is now Early Bird Club. Learn more here.

Root [LINARO] AWEstruck v1.04 12/14/14 | [3.4.107] v1.05b6 5/30/15

I have a request #Spock... Is there anyway you could put the available kernel versions in your OP? Everything is all spread out and it's hard telling which are the best stable ones for KK and which are more for testing. Maybe just an arrangement of the latest JB and KK and then a section of newest test versions.
 
Can anyone recommend a stable stock rom kernel with the ability to OC? There's just too many stability problems with this kernel. I have to use a third party app for YouTube, after waking the device I have to force-quit the play store and sometimes other apps otherwise they can no longer connect to the internet and I've experienced general sluggishness and random app crashes.
 
So found something...funny(?) in a logcat on JB kernel. I've been trying very forcefully to sustain a 1.94ghz OC and I'm fairly certain it's possible to do.

Turns out it won't clock above 1.83 unless you lower the voltage to 1275mH at 1.89 or above. However upon doing this the screen will freeze up and flicker after about a minute of animations. Though funnily enough the phone remains entirely functionable and doesn't reset. This got me thinking...could this be an issue with the gpu driver?

So I ran LogcatX and tried again. The screen froze and I had to hard reset. Upon looking back at the log there were tons of errors referencing to height and width at the wrong resolution (960x480 or 960x540...can't remember, but we have 800x480) and other nonsense I couldn't distinguish but were all gpu related from what I COULD recognize.

Then I found something that completely threw me off. When I got to where the screen froze in the logcat, it read the gpu as an Adreno 200. We have an Adreno 305. It appears to me we have the wrong gpu driver(?) in the kernel.

What do you make of this theory spock1104? I'll upload the log and a screenshot when I get a chance later
 
Last edited:
So found something...funny(?) in a logcat on JB kernel. I've been trying very forcefully to sustain a 1.89ghz OC and I'm fairly certain it's possible to do.

Turns out it won't clock above 1.72 unless you lower the voltage to 1275mH at 1.83 or above. However upon doing this the screen will freeze up and flicker after about a minute of animations. Though funnily enough the phone remains entirely functionable and doesn't reset. This got me thinking...could this be an issue with the gpu driver?

So I ran LogcatX and tried again. The screen froze and I had to hard reset. Upon looking back at the log there were tons of errors referencing to height and width at the wrong resolution (960x480 or 960x540...can't remember, but we have 800x480) and other nonsense I couldn't distinguish but were all gpu related from what I COULD recognize.

Then I found something that completely threw me off. When I got to where the screen froze in the logcat, it read the gpu as an Adreno 200. We have an Adreno 305. It appears to me we have the wrong gpu driver(?) in the kernel.

What do you make of this theory spock1104? I'll upload the log and a screenshot when I get a chance later
that's strange I got it to go to 1.89 with no problems
 
that's strange I got it to go to 1.89 with no problems
JB or kitkat? I'm referring to the latest JB Beta 3. I can't get above 1.83 unless I UV 1.89 down to 1275mH. At 1275.1mH it does nothing. I've got screenshots of this with 3c Tuner too. I'll get what I can together when I have a chance tonight if I remember
 
Last edited:
JB or kitkat? I'm referring to the latest JB Beta 3. I can't get above 1.72 unless I UV 1.83 down to 1275mH. At 1275.1mH it does nothing. I've got screenshots of this with 3c Tuner too. I'll get what I can together when I have a chance tonight if I remember
I'm pm know and I believe I'm on same kernal
 
JB or kitkat? I'm referring to the latest JB Beta 3. I can't get above 1.72 unless I UV 1.83 down to 1275mH. At 1275.1mH it does nothing. I've got screenshots of this with 3c Tuner too. I'll get what I can together when I have a chance tonight if I remember
I'm pm know and I believe I'm on same kernal
 
I'm pm know and I believe I'm on same kernal
K, my mistake about the clocks I'm trying to reach. 1.83 is already voted at 1275, but anything above that needs UV to 1275 or it doesn't go that high. At 1.89 I get the freezing I mentioned and I've hit 1.94 without a reboot. Gonna edit my OG post properly
 
So found something...funny(?) in a logcat on JB kernel. I've been trying very forcefully to sustain a 1.94ghz OC and I'm fairly certain it's possible to do.

Turns out it won't clock above 1.83 unless you lower the voltage to 1275mH at 1.89 or above. However upon doing this the screen will freeze up and flicker after about a minute of animations. Though funnily enough the phone remains entirely functionable and doesn't reset. This got me thinking...could this be an issue with the gpu driver?

So I ran LogcatX and tried again. The screen froze and I had to hard reset. Upon looking back at the log there were tons of errors referencing to height and width at the wrong resolution (960x480 or 960x540...can't remember, but we have 800x480) and other nonsense I couldn't distinguish but were all gpu related from what I COULD recognize.

Then I found something that completely threw me off. When I got to where the screen froze in the logcat, it read the gpu as an Adreno 200. We have an Adreno 305. It appears to me we have the wrong gpu driver(?) in the kernel.

What do you make of this theory spock1104? I'll upload the log and a screenshot when I get a chance later

Very possible. Part of the rewrite I'm in the middle of is updating the video drivers to the latest CAF, so we'll see what happens with that. Unfortunately it's having a ripple effect right now, so it may be a while.

K, my mistake about the clocks I'm trying to reach. 1.83 is already voted at 1275, but anything above that needs UV to 1275 or it doesn't go that high. At 1.89 I get the freezing I mentioned and I've hit 1.94 without a reboot. Gonna edit my OG post properly

Confirmed, when I get some time I'll check if I missed a voltage limit.
 
this is more of a note to myself, but in the regulator code this looks like it needs to get bumped up:
Code:
/* SAW regulator constraints */
struct regulator_init_data msm8930_pm8917_saw_regulator_core0_pdata =
/* ID vreg_name min_uV max_uV */
SAW_VREG_INIT(S5, "8917_s5", 850000, 1300000);
struct regulator_init_data msm8930_pm8917_saw_regulator_core1_pdata =
SAW_VREG_INIT(S6, "8917_s6", 850000, 1300000);

I more or less need to make changes based on https://github.com/Grarak/grakernel-msm8930/commit/53e9f4bcf7e1878ad728aef10e451322ce259b79

which is where I updated the table from anyways...
 
Last edited:
Very possible. Part of the rewrite I'm in the middle of is updating the video drivers to the latest CAF, so we'll see what happens with that. Unfortunately it's having a ripple effect right now, so it may be a while.



Confirmed, when I get some time I'll check if I missed a voltage limit.
Awesome. Glad I could contribute my 2 cents that actually MAKES sense ^_^
 
Awesome. Glad I could contribute my 2 cents that actually MAKES sense ^_^

Keep it coming, I can't say enough how much I appreciate any and all feedback. It's the best way for this to get better.

New beta!
AWEstruck_JB_v1.05_beta4.zip - 9.43 MB
AWEstruck_KK_v1.05_beta4.zip - 9.51 MB

Changes:
Linaro 4.9.3 2014.12
disabled HID battery and hidraw
voltage corrections

My phone is still not going above 1.83, but I can confirm that msm_dvcs is no longer kicking in when going above 1.275V. Seriously, that was an awesome observation. May you guys have better luck than I do.
 
I rebooted and got a stable 1.89GHz run. 1.94 is a little flaky, 1.99 reboots right away.

GJV750Z.png


May your clocks be high and your temperatures be low.
 
Keep it coming, I can't say enough how much I appreciate any and all feedback. It's the best way for this to get better.

New beta!
AWEstruck_JB_v1.05_beta4.zip - 9.43 MB
AWEstruck_KK_v1.05_beta4.zip - 9.51 MB

Changes:
Linaro 4.9.3 2014.12
disabled HID battery and hidraw
voltage corrections

My phone is still not going above 1.83, but I can confirm that msm_dvcs is no longer kicking in when going above 1.275V. Seriously, that was an awesome observation. May you guys have better luck than I do.
Don't sweat it kid. Youve made significant strides for us. One of the most responsive devs I've had the pleasure to work with. Keep it up brosef ^_^
 
I got that too on my second attempt to bench at 1.94. The results I posted are from 1.89, my 1.94 score was 5000. I think we found the limit.
Not necessarily. Overvolt 12.5mV and slowly up both clocks in sequence to 1.94 and try again. Did this on CM11, scored 6079 on performance. I was able to hit 1.99 without crashing/rebooting ONCE by overvolting I think 25mV/37.5mV but after quadrant run realized I didnt have internet at the time so the score didnt post. Tried again and everything crashed til reboot after 2 minutes of playing around. Still working on volt settings for it. I'm sure its possible for a small amount of time
Btw...HID still not working in KK
 
Not necessarily. Overvolt 12.5mV and slowly up both clocks in sequence to 1.94 and try again. Did this on CM11, scored 6079 on performance. I was able to hit 1.99 without crashing/rebooting ONCE by overvolting I think 25mV/37.5mV but after quadrant run realized I didnt have internet at the time so the score didnt post. Tried again and everything crashed til reboot after 2 minutes of playing around. Still working on volt settings for it. I'm sure its possible for a small amount of time
Btw...HID still not working in KK

I over volted 25mV and still had some bad behavior. My phone even did some flashing of the screen on reboot. I won't be going past 1.89 any more.
 
I over volted 25mV and still had some bad behavior. My phone even did some flashing of the screen on reboot. I won't be going past 1.89 any more.
IMG_20150116_211125_861.jpg


Still can't understand why more people dont like AWEsome Rom...
This is at 1.94 performance overvolted by 12.5mV as i suggested btw. So...yeah...that just happened

Edit: lowered my entropy super low and just hit 7353...NUTS!!!
 
Back
Top Bottom