Re: [webkit-dev] 안광림 is out of office.
These out of office messages are getting out of hand. Is there anything we can do? El jue, 2 jun 2022 a las 10:43, 안광림 via webkit-dev (< webkit-dev@lists.webkit.org>) escribió: > > Period : 2022/05/23 00:00 ~ 2022/07/15 00:00 [Korea Standard Time] > > 부재중입니다. > ___ > webkit-dev mailing list > webkit-dev@lists.webkit.org > https://lists.webkit.org/mailman/listinfo/webkit-dev > ___ webkit-dev mailing list webkit-dev@lists.webkit.org https://lists.webkit.org/mailman/listinfo/webkit-dev
Re: [webkit-dev] [Help] WebProcess CRASHED
Hello Michael! Thanks for your prompt reply :) Okay that makes sense. How do I install debuginfo for webkit though? Looking into apt, all I can find is this: apt-cache search debuginfo debugedit - tools for handling build-ids and paths rewriting in DWARF data libdebuginfod-common - configuration to enable the Debian debug info server libdebuginfod-dev - libdebuginfod development libraries and header files libdebuginfod1 - library to interact with debuginfod (development files) debuginfod - debuginfo-related http file-server daemon And webkit: apt-cache search debug | grep -i webkit python3-pyqt5.qtwebkit-dbg - Python 3 bindings for Qt5's WebKit module (debug extensions) Looking around in google sends me a lot of python/QT-related links to some (old?) packages that I don't think are related: https://www.google.com/search?q=webkit+debuginfo Maybe I should install libdebuginfod1, libdebuginfod1-dev and debuginfod, then try again? Thanks again! -Alemar El vie, 20 ago 2021 a las 18:24, Michael Catanzaro () escribió: > > On Fri, Aug 20 2021 at 11:27:17 AM -0600, Alemar > wrote: > > #2 0x7fcfd172ff2b in () at > > /lib/x86_64-linux-gnu/libwebkit2gtk-4.0.so.37 > > #3 0x7fcfd37e0622 in () at > > /lib/x86_64-linux-gnu/libwebkit2gtk-4.0.so.37 > > #4 0x7fcfd37efef6 in () at > > /lib/x86_64-linux-gnu/libwebkit2gtk-4.0.so.37 > > #5 0x7fcfd31529a6 in () at > > /lib/x86_64-linux-gnu/libwebkit2gtk-4.0.so.37 > > #6 0x7fcfd315321a in () at > > /lib/x86_64-linux-gnu/libwebkit2gtk-4.0.so.37 > > #7 0x7fcfd2c5b9fd in () at > > /lib/x86_64-linux-gnu/libwebkit2gtk-4.0.so.37 > > #8 0x7fcfd2c5cf0f in () at > > /lib/x86_64-linux-gnu/libwebkit2gtk-4.0.so.37 > > #9 0x7fcfd3095412 in () at > > /lib/x86_64-linux-gnu/libwebkit2gtk-4.0.so.37 > > #10 0x7fcfd30954e6 in () at > > /lib/x86_64-linux-gnu/libwebkit2gtk-4.0.so.37 > > #11 0x7fcfd30c00b4 in () at > > /lib/x86_64-linux-gnu/libwebkit2gtk-4.0.so.37 > > #12 0x7fcfd073d025 in () at > > /lib/x86_64-linux-gnu/libjavascriptcoregtk-4.0.so.18 > > #13 0x7fcfd073d2a3 in () at > > /lib/x86_64-linux-gnu/libjavascriptcoregtk-4.0.so.18 > > Hi, it looks like you're missing debuginfo for WebKit. When you install > debuginfo and take the backtrace again, then you'll see function names, > variables, and even line numbers that will point to exactly what's > wrong. All we know from that is it crashes somewhere in WebKit, which > could be anywhere. :) > > ___ webkit-dev mailing list webkit-dev@lists.webkit.org https://lists.webkit.org/mailman/listinfo/webkit-dev
Re: [webkit-dev] [Help] WebProcess CRASHED
Hello Michael, Sorry for the late reply, now is when I had time to run the debugging instructions. Here's the full stack trace, I don't see any "???" frames so I think it's all good. Also, sorry for my lack of mailing-list-netiquette, I don't know if I should attach a file or send it all as text instead. Since it's not too large, I'm sending it here as text. Thanks, -Alemar --- dragon@linux:~/Desktop$ coredumpctl gdb PID: 15209 (WebKitWebProces) UID: 1000 (dragon) GID: 1000 (dragon) Signal: 6 (ABRT) Timestamp: Fri 2021-08-20 11:21:51 CST (1min 16s ago) Command Line: /usr/lib/x86_64-linux-gnu/webkit2gtk-4.0/WebKitWebProcess 12 24 Executable: /usr/lib/x86_64-linux-gnu/webkit2gtk-4.0/WebKitWebProcess Control Group: /user.slice/user-1000.slice/session-1.scope Unit: session-1.scope Slice: user-1000.slice Session: 1 Owner UID: 1000 (dragon) Boot ID: 3233f400f3314998b83493ae2db50127 Machine ID: e539c8b1c8d94274aca718766b6876d0 Hostname: linux Storage: /var/lib/systemd/coredump/core.WebKitWebProces.1000.3233f400f3314998b83493ae2db50127.15209.162948011100.zst Message: Process 15209 (WebKitWebProces) of user 1000 dumped core. Stack trace of thread 15209: #0 0x7fcfd0f42fbb raise (libc.so.6 + 0x40fbb) #1 0x7fcfd0f28864 abort (libc.so.6 + 0x26864) #2 0x7fcfd172ff2b n/a (libwebkit2gtk-4.0.so.37 + 0x641f2b) #3 0x7fcfd37e0622 n/a (libwebkit2gtk-4.0.so.37 + 0x26f2622) #4 0x7fcfd37efef6 n/a (libwebkit2gtk-4.0.so.37 + 0x2701ef6) #5 0x7fcfd31529a6 n/a (libwebkit2gtk-4.0.so.37 + 0x20649a6) #6 0x7fcfd315321a n/a (libwebkit2gtk-4.0.so.37 + 0x206521a) #7 0x7fcfd2c5b9fd n/a (libwebkit2gtk-4.0.so.37 + 0x1b6d9fd) #8 0x7fcfd2c5cf0f n/a (libwebkit2gtk-4.0.so.37 + 0x1b6ef0f) #9 0x7fcfd3095412 n/a (libwebkit2gtk-4.0.so.37 + 0x1fa7412) #10 0x7fcfd30954e6 n/a (libwebkit2gtk-4.0.so.37 + 0x1fa74e6) #11 0x7fcfd30c00b4 n/a (libwebkit2gtk-4.0.so.37 + 0x1fd20b4) #12 0x7fcfd073d025 n/a (libjavascriptcoregtk-4.0.so.18 + 0x1433025) #13 0x7fcfd073d2a3 n/a (libjavascriptcoregtk-4.0.so.18 + 0x14332a3) #14 0x7fcfd0b197ef g_main_context_dispatch (libglib-2.0.so.0 + 0x557ef) #15 0x7fcfd0b6cd28 n/a (libglib-2.0.so.0 + 0xa8d28) #16 0x7fcfd0b18e53 g_main_loop_run (libglib-2.0.so.0 + 0x54e53) #17 0x7fcfd073d400 _ZN3WTF7RunLoop3runEv (libjavascriptcoregtk-4.0.so.18 + 0x1433400) #18 0x7fcfd1e03269 n/a (libwebkit2gtk-4.0.so.37 + 0xd15269) #19 0x7fcfd0f2a565 __libc_start_main (libc.so.6 + 0x28565) #20 0x5632f41d670e n/a (WebKitWebProcess + 0x70e) Stack trace of thread 15212: #0 0x7fcfd100e22f __poll (libc.so.6 + 0x10c22f) #1 0x7fcfd0b6ccc6 n/a (libglib-2.0.so.0 + 0xa8cc6) #2 0x7fcfd0b18e53 g_main_loop_run (libglib-2.0.so.0 + 0x54e53) #3 0x7fcfd073d400 _ZN3WTF7RunLoop3runEv (libjavascriptcoregtk-4.0.so.18 + 0x1433400) #4 0x7fcfd06e _ZN3WTF6Thread10entryPointEPNS0_16NewThreadContextE (libjavascriptcoregtk-4.0.so.18 + 0x13e3ddd) #5 0x7fcfd073f66d n/a (libjavascriptcoregtk-4.0.so.18 + 0x143566d) #6 0x7fcfcd04a450 start_thread (libpthread.so.0 + 0x9450) #7 0x7fcfd1019d53 __clone (libc.so.6 + 0x117d53) Stack trace of thread 15257: #0 0x7fcfcd05731e __futex_abstimed_wait_cancelable64 (libpthread.so.0 + 0x1631e) #1 0x7fcfcd05099e pthread_cond_timedwait@@GLIBC_2.3.2 (libpthread.so.0 + 0xf99e) #2 0x7fcfd073ff4f _ZN3WTF15ThreadCondition9timedWaitERNS_5MutexENS_8WallTimeE (libjavascriptcoregtk-4.0.so.18 + 0x1435f4f) #3 0x7fcfd06e95a3 _ZN3WTF10ParkingLot21parkConditionallyImplEPKvRKNS_12ScopedLambdaIFbvEEERKNS3_IFvvEEERKNS_24TimeWithDynamicClockTypeE (libjavascriptcoregtk-4.0.so.18 + 0x13df5a3) #4 0x7fcfd06d5093 n/a (libjavascriptcoregtk-4.0.so.18 + 0x13cb093) #5 0x7fcfd06e _ZN3WTF6Thread10entryPointEPNS0_16NewThreadContextE (libjavascriptcoregtk-4.0.so.18 + 0x13e3ddd) #6 0x7fcfd073f66d n/a (libjavascriptcoregtk-4.0.so.18 + 0x143566d) #7 0x7fcfcd04a450 start_thread (libpthread.so.0 + 0x9450) #8 0x7fcfd1019d53 __clone (libc.so.6 + 0x117d53) Stack trace of thread 15211: #0 0x7fcfcd05731e __futex_abstimed_wait_cancelable64 (libpthread.so.0 + 0x1631e) #1
[webkit-dev] [Help] WebProcess CRASHED
Hello list! Currently I'm having a problem with the library when going to www.bing.com (it seems to happen only on that site though), where the browser process crashes. I thought it was my app, but when running it against the MiniBrowser, it also crashes! >From the developer tools, I'm suspecting it could be related to trying to load (and play?) an mp3 file (start.mp3). The current PC it's running on is a brand-new installed Ubuntu 21.04 up-to-date which seems to use the libwebkit2gtk-4.0-37 library. Such a PC does not have any audio drivers installed, I'm basically running off Ubuntu Server with X11, Openbox and NVIDIA drivers, nothing more, nothing less. I suspect it could be because it's trying to play the audio file but it can't. However, the log does not show much: (MiniBrowser:72279): dbind-WARNING **: 18:07:43.671: AT-SPI: Error retrieving accessibility bus address: org.freedesktop.DBus.Error.ServiceUnknown: The name org.a11y.Bus was not provided by any .service files (WebKitWebProcess:72294): dbind-WARNING **: 18:07:43.752: AT-SPI: Error retrieving accessibility bus address: org.freedesktop.DBus.Error.ServiceUnknown: The name org.a11y.Bus was not provided by any .service files (WebKitWebProcess:72393): dbind-WARNING **: 18:08:25.851: AT-SPI: Error retrieving accessibility bus address: org.freedesktop.DBus.Error.ServiceUnknown: The name org.a11y.Bus was not provided by any .service files (WebKitWebProcess:72420): dbind-WARNING **: 18:08:27.415: AT-SPI: Error retrieving accessibility bus address: org.freedesktop.DBus.Error.ServiceUnknown: The name org.a11y.Bus was not provided by any .service files ** (MiniBrowser:72279): WARNING **: 18:08:27.499: WebProcess CRASHED Is there any way I can try to get more debugging info for you guys to see if I can fix the problem on my own, but also without building the full webkit library as it's really heavy and time-consuming? Thanks, -Alemar ___ webkit-dev mailing list webkit-dev@lists.webkit.org https://lists.webkit.org/mailman/listinfo/webkit-dev
Re: [webkit-dev] Any way to get a debugging symbols build without compiling?
Oh also, sorry for the extra email, but I just noticed that my webkit2gtk build is like 3 GB of size (!) no wonder why it doesn't crash, with that size that's the least thing I'd expect haha. But I can't definitely distribute that. So yeah, does anyone know what is the build command line for a production build? The usual size is around 50 MB :) Thanks! -Alemar El jue, 8 abr 2021 a las 8:21, Alemar () escribió: > > Hey all! > > Thanks for the fast reply :) here I quote: > > --- > > Krzysztof, thanks for your response. Indeed, WebKit is awesome! You > have a point with the debug build, however, I'd say that for general > purpose it's valid to at least provide one (you'll see why below) like > they already did for Xenial (I didn't know until now though) > https://packages.ubuntu.com/xenial/libwebkit2gtk-3.0-25-dbg . Sadly, > there's no package for 4.0. I don't see it as a burden or a problem, I > mean, I was able to compile one, so sharing it should be a piece of > cake, no? > > Oh, I didn't know about Slack, thanks! I only looked at the homepage > to see the communication channels, and Slack wasn't mentioned anywhere > :) I thought about IRC but yeah, it doesn't surprise me that it's > quiet so that's why I gave the mailing list a try. > > --- > > Guillaume, thanks a lot for your detailed explanation! It did help! I > think you were right - I honestly don't know if it was due to OOM (I > was definitely getting reboots though), but using -j4 with Ninja fixed > the issue and I was able to have a debug build, yay! Yeah I went into > full console mode, took around 70 minutes or so. Thank you! > > --- > > Now, I have another question (maybe unrelated but now that we're here...): > > I've noticed that with the debug build, I'm not experiencing the stack > smashing issue I used to before. How come? I'm thinking that maybe the > debug version has some kind of protections against buffer overflows so > the issue doesn't happen? I'm a bit concerned about having to > distribute my app along with custom-built libraries instead of using > each distro's official packages. > > So, my question is: What CLI arguments are used for building the > release version posted on the website? I'm assuming it's not just: > > cmake -DPORT=GTK -DCMAKE_BUILD_TYPE=RelWithDebInfo -DENABLE_GAMEPAD=NO > -DENABLE_INTROSPECTION=NO -GNinja > > Because that's what I ran (RelWithDebInfo is shown on the Trac site as > an example, but I want to follow the official build, not an example). > I'm very new to C++/CMake stuff so I'd like to know if there's a place > where I can figure out what is the final command line used to build > the same release package? My concern is to reproduce the same issue > I'm experiencing with my distro's package and the debug build doesn't > crash, so how would I go into doing that? > > My distro right now is Manjaro 5.11.6.1-MANJARO using using webkit2gtk > 2.30.5-1 from the official repo, although the issue also happens with > Ubuntu Focal under Docker. > > Thanks again! > -Alemar > > -- Forwarded message - > De: Guillaume Emont > Date: jue, 8 abr 2021 a las 5:04 > Subject: Re: [webkit-dev] Any way to get a debugging symbols build > without compiling? > To: Alemar > > > Hi Alemar, > > Could you be running out of memory? Gcc (assuming that's the compiler > you're using) takes a lot of RAM (IIRC usually around 2GB per process), > and the linking phase takes a lot of RAM too, especially in debug, so > if you are using many > cores for compilation, you could be running into OOM territory, > especially if you have other memory-intensive processes running on the > host. I suggest you try to use less processes in parallel (with e.g -j4 > passed to ninja), and try to close other memory-hungry applications > (e.g. web browsers) while compiling. > > Hope this helps, > > Guillaume > > Quoting Alemar via webkit-dev (2021-04-08 06:29:01) > > Hello list, > > > > I just joined here because I have a question/problem related to > > WebKitGTK development. I wanted to know if it's possible for you devs > > to also publish or provide builds with debugging symbols? > > > > To give some context: I've spent around 2 days worth of free time this > > week trying to get the standard build to work, and it always > > hangs/crashes my computers when building the JavaSscriptCore files > > (not to mention they seem rather heavy). > > > > This has happened to me on my i7 4790K w/ 16 GB RAM (4C/8T) under both > > Manjaro Linux and Ubuntu 20.10 (I even switched to Manjaro to see if > > it was a lack of resources, but no) and it also happened on a VPS > > server I have rented on SSDNodes
[webkit-dev] Fwd: Any way to get a debugging symbols build without compiling?
Hey all! Thanks for the fast reply :) here I quote: --- Krzysztof, thanks for your response. Indeed, WebKit is awesome! You have a point with the debug build, however, I'd say that for general purpose it's valid to at least provide one (you'll see why below) like they already did for Xenial (I didn't know until now though) https://packages.ubuntu.com/xenial/libwebkit2gtk-3.0-25-dbg . Sadly, there's no package for 4.0. I don't see it as a burden or a problem, I mean, I was able to compile one, so sharing it should be a piece of cake, no? Oh, I didn't know about Slack, thanks! I only looked at the homepage to see the communication channels, and Slack wasn't mentioned anywhere :) I thought about IRC but yeah, it doesn't surprise me that it's quiet so that's why I gave the mailing list a try. --- Guillaume, thanks a lot for your detailed explanation! It did help! I think you were right - I honestly don't know if it was due to OOM (I was definitely getting reboots though), but using -j4 with Ninja fixed the issue and I was able to have a debug build, yay! Yeah I went into full console mode, took around 70 minutes or so. Thank you! --- Now, I have another question (maybe unrelated but now that we're here...): I've noticed that with the debug build, I'm not experiencing the stack smashing issue I used to before. How come? I'm thinking that maybe the debug version has some kind of protections against buffer overflows so the issue doesn't happen? I'm a bit concerned about having to distribute my app along with custom-built libraries instead of using each distro's official packages. So, my question is: What CLI arguments are used for building the release version posted on the website? I'm assuming it's not just: cmake -DPORT=GTK -DCMAKE_BUILD_TYPE=RelWithDebInfo -DENABLE_GAMEPAD=NO -DENABLE_INTROSPECTION=NO -GNinja Because that's what I ran (RelWithDebInfo is shown on the Trac site as an example, but I want to follow the official build, not an example). I'm very new to C++/CMake stuff so I'd like to know if there's a place where I can figure out what is the final command line used to build the same release package? My concern is to reproduce the same issue I'm experiencing with my distro's package and the debug build doesn't crash, so how would I go into doing that? My distro right now is Manjaro 5.11.6.1-MANJARO using using webkit2gtk 2.30.5-1 from the official repo, although the issue also happens with Ubuntu Focal under Docker. Thanks again! -Alemar -- Forwarded message - De: Guillaume Emont Date: jue, 8 abr 2021 a las 5:04 Subject: Re: [webkit-dev] Any way to get a debugging symbols build without compiling? To: Alemar Hi Alemar, Could you be running out of memory? Gcc (assuming that's the compiler you're using) takes a lot of RAM (IIRC usually around 2GB per process), and the linking phase takes a lot of RAM too, especially in debug, so if you are using many cores for compilation, you could be running into OOM territory, especially if you have other memory-intensive processes running on the host. I suggest you try to use less processes in parallel (with e.g -j4 passed to ninja), and try to close other memory-hungry applications (e.g. web browsers) while compiling. Hope this helps, Guillaume Quoting Alemar via webkit-dev (2021-04-08 06:29:01) > Hello list, > > I just joined here because I have a question/problem related to > WebKitGTK development. I wanted to know if it's possible for you devs > to also publish or provide builds with debugging symbols? > > To give some context: I've spent around 2 days worth of free time this > week trying to get the standard build to work, and it always > hangs/crashes my computers when building the JavaSscriptCore files > (not to mention they seem rather heavy). > > This has happened to me on my i7 4790K w/ 16 GB RAM (4C/8T) under both > Manjaro Linux and Ubuntu 20.10 (I even switched to Manjaro to see if > it was a lack of resources, but no) and it also happened on a VPS > server I have rented on SSDNodes (12 vCPUs and 48 GB RAM), it hanged > so bad that I had to hard-reboot it. I dunno what the code is doing, > but I don't think I need to have a mainframe to build this, do I? > > The question also comes out of frustration: I'm trying to debug a > "stack smashing detected" issue and to see the call stack I need the > debugging symbols, otherwise all I see is undefined symbols all > around. > > So, how hard/possible it is for you to provide a build with the > debugging symbols (since you're the devs and might have already > reached a checkpoint where you don't have to build the JavaScriptCore > anymore since builds are incremental) for us poor devs who can't > afford a more powerful rig and are actually developing apps to bring > some bread to the table? > > Thanks and sorry for my lack of netiquette regarding mailing lists, >
[webkit-dev] Any way to get a debugging symbols build without compiling?
Hello list, I just joined here because I have a question/problem related to WebKitGTK development. I wanted to know if it's possible for you devs to also publish or provide builds with debugging symbols? To give some context: I've spent around 2 days worth of free time this week trying to get the standard build to work, and it always hangs/crashes my computers when building the JavaSscriptCore files (not to mention they seem rather heavy). This has happened to me on my i7 4790K w/ 16 GB RAM (4C/8T) under both Manjaro Linux and Ubuntu 20.10 (I even switched to Manjaro to see if it was a lack of resources, but no) and it also happened on a VPS server I have rented on SSDNodes (12 vCPUs and 48 GB RAM), it hanged so bad that I had to hard-reboot it. I dunno what the code is doing, but I don't think I need to have a mainframe to build this, do I? The question also comes out of frustration: I'm trying to debug a "stack smashing detected" issue and to see the call stack I need the debugging symbols, otherwise all I see is undefined symbols all around. So, how hard/possible it is for you to provide a build with the debugging symbols (since you're the devs and might have already reached a checkpoint where you don't have to build the JavaScriptCore anymore since builds are incremental) for us poor devs who can't afford a more powerful rig and are actually developing apps to bring some bread to the table? Thanks and sorry for my lack of netiquette regarding mailing lists, I'm pretty new to using these. I wonder why don't you guys have a forum, a discourse, or a way to submit questions/bug that's not an email list? Thanks, -Alemar ___ webkit-dev mailing list webkit-dev@lists.webkit.org https://lists.webkit.org/mailman/listinfo/webkit-dev
