[opensuse-factory] Tumbleweed 20160621: Thunderbird and Chromium are broken
Hello, This morning I installed the latest Tumbleweed updates on my laptop. The result is that Thunderbird and Chromium are broken. What I see is the following: * If Thunderbird is started as follows: "thunderbird --ProfileManager" and if I select "Create Profile" and "Start Thunderbird", Thunderbird crashes while I try to add an IMAP account. The following is displayed in the console from which I started thunderbird: -- Exception object -- + _message (string) '' + stack (string) 343 chars + code (number) 0 + uri (string) 'http://autoconfig.gmail.com/mail/config-v1.1.xml' + constructor (function) 6 lines + message (string) '' + toString (function) 4 lines * -- Stack Trace -- Exception@chrome://messenger/content/accountcreation/util.js:137:5 ServerException@chrome://messenger/content/accountcreation/fetchhttp.js :262:3 FetchHTTP.prototype._response@chrome://messenger/content/accountcreatio n/fetchhttp.js:190:19 FetchHTTP.prototype.start/request.onerror@chrome://messenger/content/ac countcreation/fetchhttp.js:117:36 * If I try to start chromium, it crashes, even if I remove all Chromium settings first. The topmost frames of the crash callstack are: Received signal 11 SEGV_MAPERR 000000000008 #0 0x7fc73317a44e base::debug::StackTrace::StackTrace() #1 0x7fc73317a87b base::debug::(anonymous namespace)::StackDumpSignalHandler() #2 0x7fc720486ef0 <unknown> #3 0x7fc71d79d018 blink::PaintLayer::enclosingSelfPaintingLayer() #4 0x7fc71d79dc63 blink::PaintLayer::insertOnlyThisLayerAfterStyleChange() #5 0x7fc71d11444a blink::LayoutBoxModelObject::styleDidChange() #6 0x7fc71d10e189 blink::LayoutBox::styleDidChange() #7 0x7fc71d0dcf07 blink::LayoutBlock::styleDidChange() #8 0x7fc71d0ea47b blink::LayoutBlockFlow::styleDidChange() #9 0x7fc71d17aaa5 blink::LayoutObject::setStyle() #10 0x7fc71d887689 blink::Document::attach() Sorry if this has already been reported. Bart. -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
On Thursday, 23 June 2016 08:30:39 BST Bart Van Assche wrote:
Hello,
This morning I installed the latest Tumbleweed updates on my laptop. The result is that Thunderbird and Chromium are broken. What I see is the following: * If Thunderbird is started as follows: "thunderbird --ProfileManager" and if I select "Create Profile" and "Start Thunderbird", Thunderbird crashes while I try to add an IMAP account. The following is displayed in the console from which I started thunderbird:
I agree with regards to chromium (i don't use thunderbird), same signal 11 SEGV_MAPERR.
-- Exception object -- + _message (string) '' + stack (string) 343 chars + code (number) 0 + uri (string) 'http://autoconfig.gmail.com/mail/config-v1.1.xml' + constructor (function) 6 lines + message (string) '' + toString (function) 4 lines * -- Stack Trace -- Exception@chrome://messenger/content/accountcreation/util.js:137:5 ServerException@chrome://messenger/content/accountcreation/fetchhttp.js
:262:3
FetchHTTP.prototype._response@chrome://messenger/content/accountcreatio n/fetchhttp.js:190:19 FetchHTTP.prototype.start/request.onerror@chrome://messenger/content/ac countcreation/fetchhttp.js:117:36
* If I try to start chromium, it crashes, even if I remove all Chromium settings first. The topmost frames of the crash callstack are:
Received signal 11 SEGV_MAPERR 000000000008 #0 0x7fc73317a44e base::debug::StackTrace::StackTrace() #1 0x7fc73317a87b base::debug::(anonymous namespace)::StackDumpSignalHandler() #2 0x7fc720486ef0 <unknown> #3 0x7fc71d79d018 blink::PaintLayer::enclosingSelfPaintingLayer() #4 0x7fc71d79dc63 blink::PaintLayer::insertOnlyThisLayerAfterStyleChange() #5 0x7fc71d11444a blink::LayoutBoxModelObject::styleDidChange() #6 0x7fc71d10e189 blink::LayoutBox::styleDidChange() #7 0x7fc71d0dcf07 blink::LayoutBlock::styleDidChange() #8 0x7fc71d0ea47b blink::LayoutBlockFlow::styleDidChange() #9 0x7fc71d17aaa5 blink::LayoutObject::setStyle() #10 0x7fc71d887689 blink::Document::attach()
Sorry if this has already been reported.
Bart.
-- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
HI On Thu, Jun 23, 2016 at 9:25 AM, ianseeks <ianseeks@btinternet.com> wrote:
On Thursday, 23 June 2016 08:30:39 BST Bart Van Assche wrote:
This morning I installed the latest Tumbleweed updates on my laptop. The result is that Thunderbird and Chromium are broken. What I see is the following:
Checking upstream it seems that this crash in blink was already reported as a GCC6 regression and that there is also actually a crash in V8 when compiled with GCC6. All are related to the same problem, which is described in an article on stackoverflow as: "Value range propagation now assumes that the this pointer of C++ member functions is non-null. This eliminates common null pointer checks but also breaks some non-conforming code-bases (such as Qt-5, Chromium, KDevelop). As a temporary work-around -fno-delete-null-pointer-checks can be used. Wrong code can be identified by using -fsanitize=undefined." The same issue appeared with the KDE Screenlocker which caused the issue noticed in openQA with this snapshot. So Qt5 has been resolved. I will make the same adjustment to Chromium and submit it to Factory/Tumbleweed. My feeling is that Thunderbird is suffering from the same and should also be build with the temporary work-around. Regards Raymond -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
On Thursday, 23 June 2016 10:00:25 BST Raymond Wooninck wrote:
HI
On Thu, Jun 23, 2016 at 9:25 AM, ianseeks <ianseeks@btinternet.com> wrote:
On Thursday, 23 June 2016 08:30:39 BST Bart Van Assche wrote:
This morning I installed the latest Tumbleweed updates on my laptop. The result is that Thunderbird and Chromium are broken. What I see is
the following: Checking upstream it seems that this crash in blink was already reported as a GCC6 regression and that there is also actually a crash in V8 when compiled with GCC6.
All are related to the same problem, which is described in an article on stackoverflow as:
"Value range propagation now assumes that the this pointer of C++ member functions is non-null. This eliminates common null pointer checks but also breaks some non-conforming code-bases (such as Qt-5, Chromium, KDevelop). As a temporary work-around -fno-delete-null-pointer-checks can be used. Wrong code can be identified by using -fsanitize=undefined."
The same issue appeared with the KDE Screenlocker which caused the issue noticed in openQA with this snapshot. So Qt5 has been resolved. I will make the same adjustment to Chromium and submit it to Factory/Tumbleweed.
My feeling is that Thunderbird is suffering from the same and should also be build with the temporary work-around.
Regards Raymond
Cheers -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
Am 23.06.2016 um 10:47 schrieb ianseeks:
On Thursday, 23 June 2016 10:00:25 BST Raymond Wooninck wrote:
HI
On Thu, Jun 23, 2016 at 9:25 AM, ianseeks <ianseeks@btinternet.com> wrote:
On Thursday, 23 June 2016 08:30:39 BST Bart Van Assche wrote:
This morning I installed the latest Tumbleweed updates on my laptop. The result is that Thunderbird and Chromium are broken. What I see is
the following: Checking upstream it seems that this crash in blink was already reported as a GCC6 regression and that there is also actually a crash in V8 when compiled with GCC6.
All are related to the same problem, which is described in an article on stackoverflow as:
"Value range propagation now assumes that the this pointer of C++ member functions is non-null. This eliminates common null pointer checks but also breaks some non-conforming code-bases (such as Qt-5, Chromium, KDevelop). As a temporary work-around -fno-delete-null-pointer-checks can be used. Wrong code can be identified by using -fsanitize=undefined."
The same issue appeared with the KDE Screenlocker which caused the issue noticed in openQA with this snapshot. So Qt5 has been resolved. I will make the same adjustment to Chromium and submit it to Factory/Tumbleweed.
My feeling is that Thunderbird is suffering from the same and should also be build with the temporary work-around.
Regards Raymond
Cheers
firefox is working fine here after the upgrade on a AMD A8 7600 APU, x64, packman repo active. -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
On 06/23/2016 10:00 AM, Raymond Wooninck wrote:
All are related to the same problem, which is described in an article on stackoverflow as:
"Value range propagation now assumes that the this pointer of C++ member functions is non-null. This eliminates common null pointer checks but also breaks some non-conforming code-bases (such as Qt-5, Chromium, KDevelop). As a temporary work-around -fno-delete-null-pointer-checks can be used. Wrong code can be identified by using -fsanitize=undefined."
The same issue appeared with the KDE Screenlocker which caused the issue noticed in openQA with this snapshot. So Qt5 has been resolved. I will make the same adjustment to Chromium and submit it to Factory/Tumbleweed.
My feeling is that Thunderbird is suffering from the same and should also be build with the temporary work-around.
Yes, RH and upstream bug reports suggest that. I've filled a bug: https://bugzilla.opensuse.org/show_bug.cgi?id=986162
Regards Raymond
-- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
On Thursday, June 23, 2016 10:00:25 AM WIB Raymond Wooninck wrote:
HI
On Thu, Jun 23, 2016 at 9:25 AM, ianseeks <ianseeks@btinternet.com> wrote:
On Thursday, 23 June 2016 08:30:39 BST Bart Van Assche wrote:
This morning I installed the latest Tumbleweed updates on my laptop. The result is that Thunderbird and Chromium are broken. What I see is
the following: Checking upstream it seems that this crash in blink was already reported as a GCC6 regression and that there is also actually a crash in V8 when compiled with GCC6.
All are related to the same problem, which is described in an article on stackoverflow as:
"Value range propagation now assumes that the this pointer of C++ member functions is non-null. This eliminates common null pointer checks but also breaks some non-conforming code-bases (such as Qt-5, Chromium, KDevelop). As a temporary work-around -fno-delete-null-pointer-checks can be used. Wrong code can be identified by using -fsanitize=undefined."
The same issue appeared with the KDE Screenlocker which caused the issue noticed in openQA with this snapshot. So Qt5 has been resolved. I will make the same adjustment to Chromium and submit it to Factory/Tumbleweed.
My feeling is that Thunderbird is suffering from the same and should also be build with the temporary work-around.
Interesting temporary work around but I did not understand the details and does not seem the recipe of an immediate work around. On one machine which I already updated I had chromium running but after the update it stopped working . I have an alternative in opera which does not seem affected so i am not directly affected.. On a second machine from my wife (who only wants to work with chromium) which is just updating I have blocked the update to the newest version of Chromium. Would that work also as an temporary solution? I will find out is an hour or two or three but I would like to know if I am getting in bigger troubles with the blocking the chromium update? -- Linux User 183145 using Tumbleweed on a Pentium IV System : Linux 4.6.2-1-default Distro : openSUSE 20160621 (x86_64) Desktop: KDE Frameworks: 5.22.0 and Plasma 5.6.4 -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
participants (6)
-
Bart Van Assche -
C. Brouerius van Nidek -
ianseeks -
Raymond Wooninck -
Thomas Langkamp -
Vlastimil Babka