I'm recently facing the same Bus error and Seg faults frequently on my linux machine during chromium builds. My machine is a Dell desktop with core i7 vPro and having 4 sticks of RAM (4+4+8+8). Can you please help me- how to find out which RAM is faulty while running the build?
On Wednesday, July 23, 2014 at 12:21:40 AM UTC+5:30, Tom Deering wrote: > > I think I'm all set. I tested my ram sticks individually in each slot > using the build. > > | Slot 1 | Slot 2 | > Stick 1 | Fail | Fail | > --------------------------- > Stick 2 | Pass | Pass | > > It would seem that stick 1 is bad. Hopefully this helps someone in the > future, and is evidence that builds may fail even though memtest passes! > > Tom Deering | PhD Student | ISU ECpE Software Systems | (563) 249-9277 > > > On Tue, Jul 22, 2014 at 11:34 AM, Tom Deering <[email protected] > <javascript:>> wrote: > >> Disregard my last. Looks like a bus error is another kind of hardware >> fault which is a lesser-known cousin of the seg fault. >> >> http://en.wikipedia.org/wiki/Bus_error >> >> >> >> >> >> Tom Deering | PhD Student | ISU ECpE Software Systems | (563) 249-9277 >> >> >> On Tue, Jul 22, 2014 at 11:28 AM, Tom Deering <[email protected] >> <javascript:>> wrote: >> >>> Does "Bus error" mean anything to anyone? To me, it sounds like it >>> implicates the memory bus on the motherboard. >>> >>> In file included from external/chromium_org/v8/src/v8.h:52:0, >>> from external/chromium_org/v8/src/assembler.h:38, >>> from external/chromium_org/v8/src/assembler.cc:35: >>> external/chromium_org/v8/src/../include/v8.h:3233:8: internal compiler >>> error: Bus error >>> Please submit a full bug report, >>> with preprocessed source if appropriate. >>> See <http://source.android.com/source/report-bugs.html> for >>> instructions. >>> make: *** >>> [out/target/product/generic/obj/STATIC_LIBRARIES/v8_tools_gyp_v8_base_arm_gyp_intermediates/v8/src/assembler.o] >>> >>> Error 1 >>> make: *** Waiting for unfinished jobs.... >>> >>> >>> Tom Deering | PhD Student | ISU ECpE Software Systems | (563) 249-9277 >>> >>> >>> On Tue, Jul 22, 2014 at 10:59 AM, Tom Deering <[email protected] >>> <javascript:>> wrote: >>> >>>> I tried to give my laptop some tender loving care by cleaning out dust, >>>> upgrading thermal paste between the heat sink and cpu/gpu, and re-seating >>>> the ram. No luck on a subsequent build. This time it was g++ which died, >>>> not gcc. :( >>>> >>>> > make -j10 >>>> >>>> ============================================ >>>> PLATFORM_VERSION_CODENAME=REL >>>> PLATFORM_VERSION=4.4.3 >>>> TARGET_PRODUCT=full >>>> TARGET_BUILD_VARIANT=eng >>>> TARGET_BUILD_TYPE=release >>>> TARGET_BUILD_APPS= >>>> TARGET_ARCH=arm >>>> TARGET_ARCH_VARIANT=armv7-a >>>> TARGET_CPU_VARIANT=generic >>>> HOST_ARCH=x86 >>>> HOST_OS=linux >>>> HOST_OS_EXTRA=Linux-3.13.0-32-generic-x86_64-with-Ubuntu-14.04-trusty >>>> HOST_BUILD_TYPE=release >>>> BUILD_ID=KTU84L >>>> OUT_DIR=out >>>> ============================================ >>>> >>>> ... >>>> target thumb C++: sandbox_sandbox_services_gyp <= >>>> external/chromium_org/sandbox/linux/services/init_process_reaper.cc >>>> arm-linux-androideabi-g++: internal compiler error: Segmentation fault >>>> (program as) >>>> Please submit a full bug report, >>>> with preprocessed source if appropriate. >>>> See <http://source.android.com/source/report-bugs.html> for >>>> instructions. >>>> make: *** >>>> [out/target/product/generic/obj/STATIC_LIBRARIES/base_base_prefs_gyp_intermediates/base/prefs/pref_service_factory.o] >>>> >>>> Error 4 >>>> make: *** Deleting file >>>> `out/target/product/generic/obj/STATIC_LIBRARIES/base_base_prefs_gyp_intermediates/base/prefs/pref_service_factory.o' >>>> make: *** Waiting for unfinished jobs.... >>>> >>>> It seems very clear now that it *must* be a hardware problem. In fact, >>>> some other software on my system has been a bit unstable as well. I >>>> definitely want to diagnose and fix this, but that's an irritating task >>>> when I can't pin down the problem. I know it's not the purpose of this >>>> group, but do you have suggestions for how to proceed in debugging flaky >>>> hardware? Given that memtest and prime95 were unhelpful, I was thinking of >>>> trying another set of ram and seeing if the problem goes away. If it does, >>>> the ram must be to blame. If it doesn't... the mobo or cpu are probably >>>> bad, suggesting I should simply replace the laptop. >>>> >>>> Any other thoughts? If not, I'll stop asking :) >>>> >>>> On Friday, July 18, 2014 12:59:54 PM UTC-5, JBQ wrote: >>>> >>>>> On a laptop, you might have opportunities to force the system to lower >>>>> power modes, especially reduced clock speeds, though there's a risk that >>>>> those would imply lower voltages which in turn could make things worse. >>>>> >>>>> It's hard to imagine that you'll find options that make the build >>>>> significantly more reliable without also deeply slowing it down. >>>>> >>>>> JBQ >>>>> >>>>> -- >>>>> >>>>> Jean-Baptiste M. "JBQ" Quéru >>>>> Architect, Mobile, Yahoo >>>>> >>>>> >>>>> On Friday, July 18, 2014 10:13 AM, Tom Deering <[email protected]> >>>>> wrote: >>>>> >>>>> >>>>> Good to know. The SSD passed an extensive SMART test anyway. Prime95 >>>>> didn't turn up any errors in any of its three stress test modes either. >>>>> Not >>>>> sure what to say. >>>>> >>>>> This certainly reeks of a hardware problem, even though I can't >>>>> demonstrate it. I figured there was no way it could be a widespread >>>>> problem >>>>> affecting many devs, or else there'd be hoards of unhappy people. Thank >>>>> you >>>>> for your time and responses. For now I'm inclined to shrug and use a >>>>> different machine, unless you have other suggestions. >>>>> >>>>> Tom Deering | PhD Student | ISU ECpE Software Systems | (563) 249-9277 >>>>> >>>>> >>>>> On Fri, Jul 18, 2014 at 11:47 AM, 'Jean-Baptiste Queru' via Android >>>>> Building <[email protected]> wrote: >>>>> >>>>> If the issue was related to the SSD holding bad data, the crashes >>>>> would be more consistent since they'd be related to a persistent state >>>>> (and >>>>> you'd probably see fewer crashes as gcc does a good job at detecting bad >>>>> input). >>>>> >>>>> If the issue was related to the SSD holding correct data but bad data >>>>> being read in, you'd probably see crashes or issues in javac as well, >>>>> since >>>>> that also reads a whole lot of data (more than gcc in fact). >>>>> >>>>> JBQ >>>>> >>>>> -- >>>>> >>>>> Jean-Baptiste M. "JBQ" Quéru >>>>> Architect, Mobile, Yahoo >>>>> >>>>> >>>>> On Friday, July 18, 2014 8:53 AM, Tom Deering <[email protected]> >>>>> wrote: >>>>> >>>>> >>>>> My laptop's got an SSD rather than HDD, but you two could be right >>>>> about the CPU+RAM or ram slot combination. I'll see if Prime95 or >>>>> something >>>>> can show problems. >>>>> >>>>> Would it make any sense for the SSD to be responsible? I recently got >>>>> a new SSD. Perhaps intermediate binaries and libraries were corrupted as >>>>> they were written to disk, resulting in later errors as they were used? >>>>> >>>>> The laptop gets warm but not hot when building. The fan heads up to >>>>> about its 80% speed/loudness. >>>>> >>>>> Tom Deering | PhD Student | ISU ECpE Software Systems | (563) 249-9277 >>>>> >>>>> >>>>> On Fri, Jul 18, 2014 at 10:32 AM, Al Sutton <[email protected]> >>>>> wrote: >>>>> >>>>> Does the laptop get warm when building? >>>>> >>>>> I don’t think memtest stresses the CPU and the RAM at the same time, >>>>> so it could be CPU+RAM heat which is going beyond what your machine can >>>>> take, and if you have a HDD rather than an SSD that will add to the >>>>> problem. >>>>> >>>>> Al. >>>>> -- >>>>> Al Sutton - Funky Android Ltd. >>>>> (funkyandroid.com) >>>>> >>>>> The views expressed in this email are those of the author and not >>>>> necessarily those of Funky Android Limited, it's associates, or it's >>>>> subsidiaries. >>>>> >>>>> >>>>> >>>>> On 18 Jul 2014, at 16:11, Tom Deering <[email protected]> wrote: >>>>> >>>>> My laptop has a 4-core, 8-thread Intel Core i7-3610QM, and 16GB of >>>>> DDR3. I ran with -j16 to keep all cores busy even when some jobs were >>>>> blocking for disk I/O. I've put the ram through memtest86+ several times >>>>> and never observed any errors. Is that sufficient to eliminate that >>>>> hypothesis? I should mention... if I run the build repeatedly, it does >>>>> eventually succeed. That indicates to me that somehow the failure is >>>>> non-deterministic or intermittent. This non-determinism occurs even with >>>>> -j1. >>>>> >>>>> Ideas for what to investigate next? >>>>> >>>>> Tom Deering | PhD Student | ISU ECpE Software Systems | (563) 249-9277 >>>>> >>>>> >>>>> On Fri, Jul 18, 2014 at 10:06 AM, 'Jean-Baptiste Queru' via Android >>>>> Building <[email protected]> wrote: >>>>> >>>>> Those typically indicate a RAM problem, i.e. a faulty chip, poor >>>>> cooling, or overclocked hardware, especially when using non-ECC RAM. gcc >>>>> pushes RAM especially hard and it is known to trigger RAM errors on >>>>> hardware that otherwise seems rock solid. >>>>> >>>>> JBQ >>>>> >>>>> -- >>>>> >>>>> Jean-Baptiste M. "JBQ" Quéru >>>>> Architect, Mobile, Yahoo >>>>> >>>>> >>>>> On Friday, July 18, 2014 7:59 AM, Tom Deering <[email protected]> >>>>> wrote: >>>>> >>>>> >>>>> Ok, I just ran a build and produced an example of what I am seeing: >>>>> >>>>> >make -j16 >>>>> >>>>> ============================================ >>>>> PLATFORM_VERSION_CODENAME=REL >>>>> PLATFORM_VERSION=4.4.3 >>>>> TARGET_PRODUCT=full >>>>> TARGET_BUILD_VARIANT=eng >>>>> TARGET_BUILD_TYPE=release >>>>> TARGET_BUILD_APPS= >>>>> TARGET_ARCH=arm >>>>> TARGET_ARCH_VARIANT=armv7-a >>>>> TARGET_CPU_VARIANT=generic >>>>> HOST_ARCH=x86 >>>>> HOST_OS=linux >>>>> HOST_OS_EXTRA=Linux-3.13.0-32-generic-x86_64-with-Ubuntu-14.04-trusty >>>>> HOST_BUILD_TYPE=release >>>>> BUILD_ID=KTU84L >>>>> OUT_DIR=out >>>>> ============================================ >>>>> >>>>> ... >>>>> >>>>> external/sqlite/dist/sqlite3.c: In function 'sqlite3VXPrintf': >>>>> external/sqlite/dist/sqlite3.c:19527:21: internal compiler error: >>>>> Segmentation fault >>>>> Please submit a full bug report, >>>>> with preprocessed source if appropriate. >>>>> See <http://gcc.gnu.org/bugs.html> for instructions. >>>>> target thumb C++: libandroidfw <= frameworks/base/libs/androidfw/ >>>>> StreamingZipInflater.cpp >>>>> make: *** >>>>> [out/host/linux-x86/obj/SHARED_LIBRARIES/libsqlite_intermediates/sqlite3.o] >>>>> >>>>> Error 1 >>>>> >>>>> >>>>> >>>>> >>>>> >>>>> Tom Deering | PhD Student | ISU ECpE Software Systems | (563) 249-9277 >>>>> >>>>> >>>>> On Fri, Jul 18, 2014 at 9:09 AM, Tom Deering <[email protected]> >>>>> wrote: >>>>> >>>>> I've got gcc 4.8.2: >>>>> >>>>> > gcc --version >>>>> gcc (Ubuntu 4.8.2-19ubuntu1) 4.8.2 >>>>> >>>>> I think I've seen it happen on multiple files, though something in the >>>>> Chromium project seems particularly good at triggering it. I'll >>>>> double-check. Yes, the checkouts of Android are completely clean (I've >>>>> made >>>>> no changes). Yes, I'm using parallel builds. >>>>> >>>>> Tom Deering | PhD Student | ISU ECpE Software Systems | (563) 249-9277 >>>>> >>>>> >>>>> On Fri, Jul 18, 2014 at 4:54 AM, allicoder redocilla < >>>>> [email protected]> wrote: >>>>> >>>>> I've never seen this happening before. Does it always happen at the >>>>> same file? What is your gcc version (output of $ gcc --version)? Is this >>>>> with completely clean checkouts? Do you have parallel builds enabled? >>>>> >>>>> alli >>>>> >>>>> >>>>> On Wednesday, 16 July 2014 19:41:17 UTC+1, Tom Deering wrote: >>>>> >>>>> Hello Android Wizards, >>>>> >>>>> I am trying to build Android and have followed the instructions on the >>>>> 'Downloading and Building' page. No matter which branch or configuration >>>>> I >>>>> build, or whether I used Ubuntu 12.04 or 14.04, I encounter the same >>>>> problem. Deep into the build, gcc always has an internal segmentation >>>>> fault, stopping all progress before the build can complete. >>>>> >>>>> Does everyone else encounter gcc segfaults when building Android? If >>>>> so, how do you deal with them? Thanks in advance. >>>>> >>>>> -- >>>>> -- >>>>> You received this message because you are subscribed to the "Android >>>>> Building" mailing list. >>>>> To post to this group, send email to [email protected] >>>>> >>>>> To unsubscribe from this group, send email to >>>>> [email protected] >>>>> >>>>> For more options, visit this group at >>>>> http://groups.google.com/group/android-building?hl=en >>>>> >>>>> --- >>>>> You received this message because you are subscribed to the Google >>>>> Groups "Android Building" group. >>>>> To unsubscribe from this group and stop receiving emails from it, send >>>>> an email to [email protected]. >>>>> >>>>> For more options, visit https://groups.google.com/d/optout. >>>>> >>>>> >>>>> >>>>> -- >>>>> -- >>>>> You received this message because you are subscribed to the "Android >>>>> Building" mailing list. >>>>> To post to this group, send email to [email protected] >>>>> >>>>> To unsubscribe from this group, send email to >>>>> [email protected] >>>>> >>>>> For more options, visit this group at >>>>> http://groups.google.com/group/android-building?hl=en >>>>> >>>>> --- >>>>> You received this message because you are subscribed to the Google >>>>> Groups "Android Building" group. >>>>> To unsubscribe from this group and stop receiving emails from it, send >>>>> an email to [email protected]. >>>>> >>>>> For more options, visit https://groups.google.com/d/optout. >>>>> >>>>> >>>>> >>>>> -- >>>>> -- >>>>> You received this message because you are subscribed to the "Android >>>>> Building" mailing list. >>>>> To post to this group, send email to [email protected] >>>>> >>>>> To unsubscribe from this group, send email to >>>>> [email protected] >>>>> >>>>> For more options, visit this group at >>>>> http://groups.google.com/group/android-building?hl=en >>>>> >>>>> --- >>>>> You received this message because you are subscribed to the Google >>>>> Groups "Android Building" group. >>>>> To unsubscribe from this group and stop receiving emails from it, send >>>>> an email to [email protected]. >>>>> >>>>> For more options, visit https://groups.google.com/d/optout. >>>>> >>>>> >>>>> >>>>> -- >>>>> -- >>>>> You received this message because you are subscribed to the "Android >>>>> Building" mailing list. >>>>> To post to this group, send email to [email protected] >>>>> >>>>> To unsubscribe from this group, send email to >>>>> [email protected] >>>>> >>>>> For more options, visit this group at >>>>> http://groups.google.com/group/android-building?hl=en >>>>> >>>>> --- >>>>> You received this message because you are subscribed to the Google >>>>> Groups "Android Building" group. >>>>> To unsubscribe from this group and stop receiving emails from it, send >>>>> an email to [email protected]. >>>>> >>>>> For more options, visit https://groups.google.com/d/optout. >>>>> >>>>> >>>>> -- >>>>> -- >>>>> You received this message because you are subscribed to the "Android >>>>> Building" mailing list. >>>>> To post to this group, send email to [email protected] >>>>> >>>>> To unsubscribe from this group, send email to >>>>> [email protected] >>>>> >>>>> For more options, visit this group at >>>>> http://groups.google.com/group/android-building?hl=en >>>>> >>>>> --- >>>>> You received this message because you are subscribed to the Google >>>>> Groups "Android Building" group. >>>>> To unsubscribe from this group and stop receiving emails from it, send >>>>> an email to [email protected]. >>>>> >>>>> For more options, visit https://groups.google.com/d/optout. >>>>> >>>>> >>>>> -- >>>>> -- >>>>> You received this message because you are subscribed to the "Android >>>>> Building" mailing list. >>>>> To post to this group, send email to [email protected] >>>>> >>>>> To unsubscribe from this group, send email to >>>>> [email protected] >>>>> >>>>> For more options, visit this group at >>>>> http://groups.google.com/group/android-building?hl=en >>>>> >>>>> --- >>>>> You received this message because you are subscribed to the Google >>>>> Groups "Android Building" group. >>>>> To unsubscribe from this group and stop receiving emails from it, send >>>>> an email to [email protected]. >>>>> >>>>> For more options, visit https://groups.google.com/d/optout. >>>>> >>>>> >>>>> -- >>>>> -- >>>>> You received this message because you are subscribed to the "Android >>>>> Building" mailing list. >>>>> To post to this group, send email to [email protected] >>>>> >>>>> To unsubscribe from this group, send email to >>>>> [email protected] >>>>> >>>>> For more options, visit this group at >>>>> http://groups.google.com/group/android-building?hl=en >>>>> >>>>> --- >>>>> You received this message because you are subscribed to the Google >>>>> Groups "Android Building" group. >>>>> To unsubscribe from this group and stop receiving emails from it, send >>>>> an email to [email protected]. >>>>> >>>>> For more options, visit https://groups.google.com/d/optout. >>>>> >>>>> >>>>> -- >>>>> -- >>>>> You received this message because you are subscribed to the "Android >>>>> Building" mailing list. >>>>> To post to this group, send email to [email protected] >>>>> >>>>> To unsubscribe from this group, send email to >>>>> [email protected] >>>>> >>>>> For more options, visit this group at >>>>> http://groups.google.com/group/android-building?hl=en >>>>> >>>>> --- >>>>> You received this message because you are subscribed to the Google >>>>> Groups "Android Building" group. >>>>> To unsubscribe from this group and stop receiving emails from it, send >>>>> an email to [email protected]. >>>>> >>>>> For more options, visit https://groups.google.com/d/optout. >>>>> >>>>> >>>>> -- >>>> -- >>>> You received this message because you are subscribed to the "Android >>>> Building" mailing list. >>>> To post to this group, send email to [email protected] >>>> <javascript:> >>>> To unsubscribe from this group, send email to >>>> [email protected] <javascript:> >>>> For more options, visit this group at >>>> http://groups.google.com/group/android-building?hl=en >>>> >>>> --- >>>> You received this message because you are subscribed to a topic in the >>>> Google Groups "Android Building" group. >>>> To unsubscribe from this topic, visit >>>> https://groups.google.com/d/topic/android-building/p7_JAKSpO1M/unsubscribe >>>> . >>>> To unsubscribe from this group and all its topics, send an email to >>>> [email protected] <javascript:>. >>>> >>>> For more options, visit https://groups.google.com/d/optout. >>>> >>> >>> >> > -- -- You received this message because you are subscribed to the "Android Building" mailing list. To post to this group, send email to [email protected] To unsubscribe from this group, send email to [email protected] For more options, visit this group at http://groups.google.com/group/android-building?hl=en --- You received this message because you are subscribed to the Google Groups "Android Building" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. For more options, visit https://groups.google.com/d/optout.
