EXC_BAD_ACCESS crash inside AppleIntelHD5000GraphicsGLDriver when starting 3.0a8+ #3359

Closed
radames opened this Issue Jun 7, 2015 · 23 comments

Comments

Projects
None yet
3 participants
@radames

radames commented Jun 7, 2015

I'm using the Mac OSX 10.10.3 and both last alpha versions Processing 3.0a8 and 3.0a9 are crashing on startup, although the version 3.0a5 works perfectly.

Both with same report,

...
Crashed Thread:        0  AppKit Thread  Dispatch queue: com.apple.main-thread

Exception Type:        EXC_BAD_ACCESS (SIGABRT)
Exception Codes:       KERN_INVALID_ADDRESS at 0x0000303030353150

VM Regions Near 0x303030353150:
    __LINKEDIT             000012340057b000-00001234006e7000 [ 1456K] r--/rwx SM=COW  /System/Library/Extensions/AppleIntelHD5000GraphicsGLDriver.bundle/Contents/MacOS/AppleIntelHD5000GraphicsGLDriver
--> 
    MALLOC_NANO            0000600000000000-0000600000200000 [ 2048K] rw-/rwx SM=COW  
...
@benfry

This comment has been minimized.

Show comment
Hide comment
@benfry

benfry Jun 7, 2015

Member

Can you try running it from the "guest" account on your machine and see if it works?

What type of Mac is this on?

(This looks like an Oracle or Apple bug, but we need to figure out what's going on...)

Member

benfry commented Jun 7, 2015

Can you try running it from the "guest" account on your machine and see if it works?

What type of Mac is this on?

(This looks like an Oracle or Apple bug, but we need to figure out what's going on...)

@prisonerjohn

This comment has been minimized.

Show comment
Hide comment
@prisonerjohn

prisonerjohn Jun 7, 2015

Member

@benfry Do you think this is the same issue I'm having?

My crash log is a little different:

Crashed Thread:        0  AppKit Thread  Dispatch queue: com.apple.main-thread

Exception Type:        EXC_BAD_ACCESS (SIGABRT)
Exception Codes:       EXC_I386_GPFLT

Application Specific Information:
abort() called
Member

prisonerjohn commented Jun 7, 2015

@benfry Do you think this is the same issue I'm having?

My crash log is a little different:

Crashed Thread:        0  AppKit Thread  Dispatch queue: com.apple.main-thread

Exception Type:        EXC_BAD_ACCESS (SIGABRT)
Exception Codes:       EXC_I386_GPFLT

Application Specific Information:
abort() called
@benfry

This comment has been minimized.

Show comment
Hide comment
@benfry

benfry Jun 7, 2015

Member

It's a suspicion, but doesn't seem like it's crashing in the same spot, so I wanted to see if the same thing was happening from another account (the way it worked on yours).

In both cases, I'm also wondering if it's a JDK bug—we've been moving to the latest JDK 8 with each release, so it may have more to do with that than anything we've actually changed.

On yours, try changing processing/build/build.xml to use:

<property name="jdk.version" value="8" />
<property name="jdk.update" value="25" />
<property name="jdk.build" value="17" />

Right now it's at 45 for the jdk.update value, and something else for jdk.build. Try changing it, then ant clean and ant run and see if it changes anything.

Member

benfry commented Jun 7, 2015

It's a suspicion, but doesn't seem like it's crashing in the same spot, so I wanted to see if the same thing was happening from another account (the way it worked on yours).

In both cases, I'm also wondering if it's a JDK bug—we've been moving to the latest JDK 8 with each release, so it may have more to do with that than anything we've actually changed.

On yours, try changing processing/build/build.xml to use:

<property name="jdk.version" value="8" />
<property name="jdk.update" value="25" />
<property name="jdk.build" value="17" />

Right now it's at 45 for the jdk.update value, and something else for jdk.build. Try changing it, then ant clean and ant run and see if it changes anything.

@benfry

This comment has been minimized.

Show comment
Hide comment
@benfry

benfry Jun 7, 2015

Member

Also for @prisonerjohn, where's your original bug report? I can't seem to find it. Does your laptop also have Intel HD 5000 graphics?

Member

benfry commented Jun 7, 2015

Also for @prisonerjohn, where's your original bug report? I can't seem to find it. Does your laptop also have Intel HD 5000 graphics?

@prisonerjohn

This comment has been minimized.

Show comment
Hide comment
@prisonerjohn

prisonerjohn Jun 7, 2015

Member

#3285 is my original report, but it looks like it was a different issue.

My laptop has an Intel HD Graphics 4000, but it's never used. I'm always on the NVIDIA GeForce GT 650M.

Member

prisonerjohn commented Jun 7, 2015

#3285 is my original report, but it looks like it was a different issue.

My laptop has an Intel HD Graphics 4000, but it's never used. I'm always on the NVIDIA GeForce GT 650M.

@prisonerjohn

This comment has been minimized.

Show comment
Hide comment
@prisonerjohn

prisonerjohn Jun 7, 2015

Member

Tried changing the JDK version values but still crashing:

#
# A fatal error has been detected by the Java Runtime Environment:
#
#  SIGSEGV (0xb) at pc=0x00007fff8cf18152, pid=6476, tid=1299
#
# JRE version: Java(TM) SE Runtime Environment (8.0_25-b17) (build 1.8.0_25-b17)
# Java VM: Java HotSpot(TM) 64-Bit Server VM (25.25-b02 mixed mode bsd-amd64 compressed oops)
# Problematic frame:
# C  [libsystem_c.dylib+0x1152]  strlen+0x12
#
# Failed to write core dump. Core dumps have been disabled. To enable core dumping, try "ulimit -c unlimited" before starting Java again
#
# An error report file with more information is saved as:
# /Users/elias/hs_err_pid6476.log
#
# If you would like to submit a bug report, please visit:
#   http://bugreport.sun.com/bugreport/crash.jsp
# The crash happened outside the Java Virtual Machine in native code.
# See problematic frame for where to report the bug.
#

Crash log is here.

Member

prisonerjohn commented Jun 7, 2015

Tried changing the JDK version values but still crashing:

#
# A fatal error has been detected by the Java Runtime Environment:
#
#  SIGSEGV (0xb) at pc=0x00007fff8cf18152, pid=6476, tid=1299
#
# JRE version: Java(TM) SE Runtime Environment (8.0_25-b17) (build 1.8.0_25-b17)
# Java VM: Java HotSpot(TM) 64-Bit Server VM (25.25-b02 mixed mode bsd-amd64 compressed oops)
# Problematic frame:
# C  [libsystem_c.dylib+0x1152]  strlen+0x12
#
# Failed to write core dump. Core dumps have been disabled. To enable core dumping, try "ulimit -c unlimited" before starting Java again
#
# An error report file with more information is saved as:
# /Users/elias/hs_err_pid6476.log
#
# If you would like to submit a bug report, please visit:
#   http://bugreport.sun.com/bugreport/crash.jsp
# The crash happened outside the Java Virtual Machine in native code.
# See problematic frame for where to report the bug.
#

Crash log is here.

@radames

This comment has been minimized.

Show comment
Hide comment
@radames

radames Jun 7, 2015

Hi @benfry , I've tried to open with the Guest account and had the same problem. I'm on osx 10.10.3 using Java 7 Update 75, my graphics card is Intel Iris Pro 1536 MB

I'll try to compile it later

radames commented Jun 7, 2015

Hi @benfry , I've tried to open with the Guest account and had the same problem. I'm on osx 10.10.3 using Java 7 Update 75, my graphics card is Intel Iris Pro 1536 MB

I'll try to compile it later

@benfry

This comment has been minimized.

Show comment
Hide comment
@benfry

benfry Jun 7, 2015

Member

Alright, @prisonerjohn I've re-filed yours here: #3360. The error message shown in that other one has nothing to do with the hard crash, so I don't want to confuse things further. Let's continue the discussion over there.

This appears to be a separate issue.

Member

benfry commented Jun 7, 2015

Alright, @prisonerjohn I've re-filed yours here: #3360. The error message shown in that other one has nothing to do with the hard crash, so I don't want to confuse things further. Let's continue the discussion over there.

This appears to be a separate issue.

@benfry benfry changed the title from Processing 3.0a8 and Processing 3.0a9 crashing on startup on Mac OSX to EXC_BAD_ACCESS crash inside AppleIntelHD5000GraphicsGLDriver when starting 3.0a8+ Jun 7, 2015

@benfry benfry added the macosx label Jun 7, 2015

@benfry

This comment has been minimized.

Show comment
Hide comment
@benfry

benfry Jun 7, 2015

Member

@radames Don't worry about building, it's not necessary. But out of curiosity, can you try to replace the Processing.app/Contents/MacOS/Processing file in 3.0a9 with the one from 3.0a5 and then try running 3.0a9 again?

Member

benfry commented Jun 7, 2015

@radames Don't worry about building, it's not necessary. But out of curiosity, can you try to replace the Processing.app/Contents/MacOS/Processing file in 3.0a9 with the one from 3.0a5 and then try running 3.0a9 again?

@radames

This comment has been minimized.

Show comment
Hide comment
@radames

radames Jun 7, 2015

@benfry yes I did the replacement and the 3.0a9 seems working fine, I've tried some examples using shaders and seems compiling and working.

radames commented Jun 7, 2015

@benfry yes I did the replacement and the 3.0a9 seems working fine, I've tried some examples using shaders and seems compiling and working.

@radames

This comment has been minimized.

Show comment
Hide comment
@radames

radames Jun 7, 2015

even the debug mode works fine!

radames commented Jun 7, 2015

even the debug mode works fine!

@benfry

This comment has been minimized.

Show comment
Hide comment
@benfry

benfry Jun 7, 2015

Member

Here are two versions of that Processing file. Can you try each and see if either or both work properly?
https://www.dropbox.com/s/t0eggec8o4s6v9s/alternates.zip?dl=0

Unzip that file, and use Processing-main to replace the Processing file at the location listed earlier. Try running, see if it works. Then do the same with the other.

Thanks!

Member

benfry commented Jun 7, 2015

Here are two versions of that Processing file. Can you try each and see if either or both work properly?
https://www.dropbox.com/s/t0eggec8o4s6v9s/alternates.zip?dl=0

Unzip that file, and use Processing-main to replace the Processing file at the location listed earlier. Try running, see if it works. Then do the same with the other.

Thanks!

@radames

This comment has been minimized.

Show comment
Hide comment
@radames

radames Jun 8, 2015

@benfry I did try for both files neither of then worked, had the same crash message.
I assured the change the execute permissions.

radames commented Jun 8, 2015

@benfry I did try for both files neither of then worked, had the same crash message.
I assured the change the execute permissions.

@benfry

This comment has been minimized.

Show comment
Hide comment
@benfry

benfry Jun 8, 2015

Member

Ouch, ok. That's the worst of the scenarios that could have happened. One version is the exact same code as 3.0a5 (just recompiled on 10.10.3). In the meantime, we'll roll back to the 3.0a5 version, at any rate.

Member

benfry commented Jun 8, 2015

Ouch, ok. That's the worst of the scenarios that could have happened. One version is the exact same code as 3.0a5 (just recompiled on 10.10.3). In the meantime, we'll roll back to the 3.0a5 version, at any rate.

@benfry

This comment has been minimized.

Show comment
Hide comment
@benfry

benfry Jun 8, 2015

Member

Thanks for checking into it. On further digging, it looks like there were some other edits (and a likely culprit) since 3.0a5. Can you try this version?

Member

benfry commented Jun 8, 2015

Thanks for checking into it. On further digging, it looks like there were some other edits (and a likely culprit) since 3.0a5. Can you try this version?

@radames

This comment has been minimized.

Show comment
Hide comment
@radames

radames Jun 8, 2015

Same, problem with your last version.

Crashed Thread:        0  AppKit Thread  Dispatch queue: com.apple.main-thread

Exception Type:        EXC_BAD_ACCESS (SIGABRT)
Exception Codes:       KERN_INVALID_ADDRESS at 0x0000303030353150

VM Regions Near 0x303030353150:
    __LINKEDIT             000012340057b000-00001234006e7000 [ 1456K] r--/rwx SM=COW  /System/Library/Extensions/AppleIntelHD5000GraphicsGLDriver.bundle/Contents/MacOS/AppleIntelHD5000GraphicsGLDriver
--> 
    MALLOC_NANO            0000600000000000-0000600000200000 [ 2048K] rw-/rwx SM=COW 

Anyway it still working with the one file taken from Processing 3.0a5

radames commented Jun 8, 2015

Same, problem with your last version.

Crashed Thread:        0  AppKit Thread  Dispatch queue: com.apple.main-thread

Exception Type:        EXC_BAD_ACCESS (SIGABRT)
Exception Codes:       KERN_INVALID_ADDRESS at 0x0000303030353150

VM Regions Near 0x303030353150:
    __LINKEDIT             000012340057b000-00001234006e7000 [ 1456K] r--/rwx SM=COW  /System/Library/Extensions/AppleIntelHD5000GraphicsGLDriver.bundle/Contents/MacOS/AppleIntelHD5000GraphicsGLDriver
--> 
    MALLOC_NANO            0000600000000000-0000600000200000 [ 2048K] rw-/rwx SM=COW 

Anyway it still working with the one file taken from Processing 3.0a5

@benfry

This comment has been minimized.

Show comment
Hide comment
@benfry

benfry Jun 8, 2015

Member

Nuts, ok. Thought we had it. Will continue looking...

Member

benfry commented Jun 8, 2015

Nuts, ok. Thought we had it. Will continue looking...

@radames

This comment has been minimized.

Show comment
Hide comment
@radames

radames Jun 8, 2015

Let me know how can I help. One question, I'll be starting teaching some class this week and I'd like to use this last alpha version, do you think the trick of moving the file from the 3.0a5 is a good idea in order to try the new 3.0a9?

radames commented Jun 8, 2015

Let me know how can I help. One question, I'll be starting teaching some class this week and I'd like to use this last alpha version, do you think the trick of moving the file from the 3.0a5 is a good idea in order to try the new 3.0a9?

@benfry

This comment has been minimized.

Show comment
Hide comment
@benfry

benfry Jun 8, 2015

Member

That's definitely a fix—I was hoping to have a new release out this past weekend but that was held up in part by this issue, since it's obviously quite serious (though only causes trouble for a small percentage of people). I'm still not sure why this is showing up on some machines and not others, but am hoping to find a better fix in the next 24-48 hours.

Member

benfry commented Jun 8, 2015

That's definitely a fix—I was hoping to have a new release out this past weekend but that was held up in part by this issue, since it's obviously quite serious (though only causes trouble for a small percentage of people). I'm still not sure why this is showing up on some machines and not others, but am hoping to find a better fix in the next 24-48 hours.

@radames

This comment has been minimized.

Show comment
Hide comment
@radames

radames Jun 8, 2015

that's great! thanks!

radames commented Jun 8, 2015

that's great! thanks!

@benfry

This comment has been minimized.

Show comment
Hide comment
@benfry

benfry Jun 16, 2015

Member

There's a workaround in 3.0a10 for this (rolling back to an older, buggier native app), but we still need to figure out what's going on here.

The same code as what shipped with 3.0a5, but recompiled on a later OS X, was still causing the crash (see #3359 (comment)). So it's not clear if 1) we have a compilation problem, 2) the recompile reveals a bug that's in the earlier code, or 3) somebody in heaven hates me.

Member

benfry commented Jun 16, 2015

There's a workaround in 3.0a10 for this (rolling back to an older, buggier native app), but we still need to figure out what's going on here.

The same code as what shipped with 3.0a5, but recompiled on a later OS X, was still causing the crash (see #3359 (comment)). So it's not clear if 1) we have a compilation problem, 2) the recompile reveals a bug that's in the earlier code, or 3) somebody in heaven hates me.

@radames

This comment has been minimized.

Show comment
Hide comment
@radames

radames Jun 17, 2015

yes, I'm already using it, thanks! so far it's been ok! some of my students are using the same version for windows and they're facing some problems, but still need to help them to report it here!

radames commented Jun 17, 2015

yes, I'm already using it, thanks! so far it's been ok! some of my students are using the same version for windows and they're facing some problems, but still need to help them to report it here!

@benfry

This comment has been minimized.

Show comment
Hide comment
@benfry

benfry Sep 11, 2015

Member

Fixed in that earlier alpha, but new issues introduced; see #3790.

Member

benfry commented Sep 11, 2015

Fixed in that earlier alpha, but new issues introduced; see #3790.

@benfry benfry closed this Sep 11, 2015

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment