170390 2017-04-02 09:54:08 -0700 WebAssembly: several tests added in r214504 crash when building with GCC 2017-04-05 00:53:53 -0700 1 1 1 Unclassified WebKit JavaScriptCore WebKit Local Build Unspecified Unspecified RESOLVED FIXED https://bugs.webkit.org/show_bug.cgi?id=169187 Gtk P2 Normal --- 170391 1 cgarcia webkit-unassigned buildbot jfbastien keith_miller mark.lam msaboff saam ysuzuki oldest_to_newest 1293530 0 cgarcia 2017-04-02 09:54:08 -0700 The pattern foo->bar([f = WTFMove(foo)]); crashes when building with GCC, I assume the move happens before the foo is used to invoke the function. Thread 1 (Thread 0x7ff8237ff700 (LWP 9200)): #0 0x00007ffa7b5071e6 in std::_Function_handler<void (JSC::Wasm::Plan&), JSC::compileAndInstantiate(JSC::VM&, JSC::ExecState*, JSC::JSPromiseDeferred*, JSC::JSValue, JSC::JSObject*)::{lambda(JSC::Wasm::Plan&)#1}>::_M_invoke(std::_Any_data const&, JSC::Wasm::Plan&) () from /home/cgarcia/src/git/gnome/WebKit/WebKitBuild/Release/lib/libjavascriptcoregtk-4.0.so.18 #1 0x00007ffa7b4d101b in JSC::Wasm::Plan::complete(WTF::AbstractLocker const&) () from /home/cgarcia/src/git/gnome/WebKit/WebKitBuild/Release/lib/libjavascriptcoregtk-4.0.so.18 #2 0x00007ffa7b4d3e0b in JSC::Wasm::Plan::parseAndValidateModule() () from /home/cgarcia/src/git/gnome/WebKit/WebKitBuild/Release/lib/libjavascriptcoregtk-4.0.so.18 #3 0x00007ffa7b4f8098 in JSC::Wasm::Worklist::Thread::work() () from /home/cgarcia/src/git/gnome/WebKit/WebKitBuild/Release/lib/libjavascriptcoregtk-4.0.so.18 #4 0x00007ffa7b5760ef in std::_Function_handler<void (), WTF::AutomaticThread::start(WTF::AbstractLocker const&)::{lambda()#1}>::_M_invoke(std::_Any_data const&) () from /home/cgarcia/src/git/gnome/WebKit/WebKitBuild/Release/lib/libjavascriptcoregtk-4.0.so.18 #5 0x00007ffa7b589705 in WTF::threadEntryPoint(void*) () from /home/cgarcia/src/git/gnome/WebKit/WebKitBuild/Release/lib/libjavascriptcoregtk-4.0.so.18 #6 0x00007ffa7b5bb73a in WTF::wtfThreadEntryPoint(void*) () from /home/cgarcia/src/git/gnome/WebKit/WebKitBuild/Release/lib/libjavascriptcoregtk-4.0.so.18 #7 0x00007ffa77c0e424 in start_thread (arg=0x7ff8237ff700) at pthread_create.c:333 #8 0x00007ffa76a209bf in clone () at ../sysdeps/unix/sysv/linux/x86_64/clone.S:105 1293531 1 306054 cgarcia 2017-04-02 09:56:54 -0700 Created attachment 306054 Patch 1293532 2 jfbastien 2017-04-02 09:57:22 -0700 That's not good! Would you want to prepare a patch to fix it? 1293533 3 jfbastien 2017-04-02 09:58:36 -0700 (In reply to JF Bastien from comment #2) > That's not good! Would you want to prepare a patch to fix it? Ha, we raced! I think it would be better to get `VM& vm` from plan, and then use it, instead of doing p->vm(). 1293581 4 cgarcia 2017-04-02 22:45:47 -0700 (In reply to JF Bastien from comment #3) > (In reply to JF Bastien from comment #2) > > That's not good! Would you want to prepare a patch to fix it? > > Ha, we raced! > > > I think it would be better to get `VM& vm` from plan, and then use it, > instead of doing p->vm(). Why? p and plan are actually the same, plan is just a Ref created from p (which is a reference, we are not doing p->vm() but p.vm()) to pass it to the lambda. 1293582 5 306063 cgarcia 2017-04-02 23:03:36 -0700 Created attachment 306063 Updated patch Make it simpler by using makeRefPtr in the lambda capture 1293583 6 306063 saam 2017-04-02 23:17:42 -0700 Comment on attachment 306063 Updated patch View in context: https://bugs.webkit.org/attachment.cgi?id=306063&action=review > Source/JavaScriptCore/wasm/js/WebAssemblyPrototype.cpp:94 > + p.vm().promiseDeferredTimer->scheduleWorkSoon(promise, [source, promise, globalObject, plan = makeRefPtr(p)] () mutable { Why not plan=WTFMove(plan)? 1293617 7 cgarcia 2017-04-03 03:26:34 -0700 (In reply to Saam Barati from comment #6) > Comment on attachment 306063 [details] > Updated patch > > View in context: > https://bugs.webkit.org/attachment.cgi?id=306063&action=review > > > Source/JavaScriptCore/wasm/js/WebAssemblyPrototype.cpp:94 > > + p.vm().promiseDeferredTimer->scheduleWorkSoon(promise, [source, promise, globalObject, plan = makeRefPtr(p)] () mutable { > > Why not plan=WTFMove(plan)? Because I've removed the line that created the plan to simplify it. RefPtr<Plan> plan = makeRef(p); [plan = WTFMove(plan)] is equivalent to [plan = makeRefPtr(p)] I think this way is simpler, since we were only creating the plan to transfer it to the lambda capture (and to get the vm, but that is what caused the crash). 1293652 8 306063 saam 2017-04-03 09:11:31 -0700 Comment on attachment 306063 Updated patch View in context: https://bugs.webkit.org/attachment.cgi?id=306063&action=review >>> Source/JavaScriptCore/wasm/js/WebAssemblyPrototype.cpp:94 >>> + p.vm().promiseDeferredTimer->scheduleWorkSoon(promise, [source, promise, globalObject, plan = makeRefPtr(p)] () mutable { >> >> Why not plan=WTFMove(plan)? > > Because I've removed the line that created the plan to simplify it. > > RefPtr<Plan> plan = makeRef(p); > [plan = WTFMove(plan)] > > is equivalent to > > [plan = makeRefPtr(p)] > > I think this way is simpler, since we were only creating the plan to transfer it to the lambda capture (and to get the vm, but that is what caused the crash). Gotcha. I see what's going on. I still think you need a local variable for the plan, otherwise, won't the lambda capture the outermost plan? I think that'll lead to a reference cycle. Even if that assignment doesn't cause this to happen, I think it's harder to reason about than just having a local variable. 1293653 9 306063 saam 2017-04-03 09:14:36 -0700 Comment on attachment 306063 Updated patch View in context: https://bugs.webkit.org/attachment.cgi?id=306063&action=review >>>> Source/JavaScriptCore/wasm/js/WebAssemblyPrototype.cpp:94 >>>> + p.vm().promiseDeferredTimer->scheduleWorkSoon(promise, [source, promise, globalObject, plan = makeRefPtr(p)] () mutable { >>> >>> Why not plan=WTFMove(plan)? >> >> Because I've removed the line that created the plan to simplify it. >> >> RefPtr<Plan> plan = makeRef(p); >> [plan = WTFMove(plan)] >> >> is equivalent to >> >> [plan = makeRefPtr(p)] >> >> I think this way is simpler, since we were only creating the plan to transfer it to the lambda capture (and to get the vm, but that is what caused the crash). > > Gotcha. I see what's going on. > I still think you need a local variable for the plan, otherwise, won't the lambda capture the outermost plan? I think that'll lead to a reference cycle. > Even if that assignment doesn't cause this to happen, I think it's harder to reason about than just having a local variable. Seems like you could also keep the code as is, and just use p.vm() instead of plan->vm(). Also, I'm not sure why we make a RefPtr instead of a Ref here. 1293657 10 jfbastien 2017-04-03 09:30:17 -0700 (In reply to Saam Barati from comment #9) > Comment on attachment 306063 [details] > Updated patch > > View in context: > https://bugs.webkit.org/attachment.cgi?id=306063&action=review > > >>>> Source/JavaScriptCore/wasm/js/WebAssemblyPrototype.cpp:94 > >>>> + p.vm().promiseDeferredTimer->scheduleWorkSoon(promise, [source, promise, globalObject, plan = makeRefPtr(p)] () mutable { > >>> > >>> Why not plan=WTFMove(plan)? > >> > >> Because I've removed the line that created the plan to simplify it. > >> > >> RefPtr<Plan> plan = makeRef(p); > >> [plan = WTFMove(plan)] > >> > >> is equivalent to > >> > >> [plan = makeRefPtr(p)] > >> > >> I think this way is simpler, since we were only creating the plan to transfer it to the lambda capture (and to get the vm, but that is what caused the crash). > > > > Gotcha. I see what's going on. > > I still think you need a local variable for the plan, otherwise, won't the lambda capture the outermost plan? I think that'll lead to a reference cycle. > > Even if that assignment doesn't cause this to happen, I think it's harder to reason about than just having a local variable. > > Seems like you could also keep the code as is, and just use p.vm() instead > of plan->vm(). > Also, I'm not sure why we make a RefPtr instead of a Ref here. This weirdness is why I suggest getting VM, and then using it on this line. The entire problem is using plan two ways on one line of code: once to get to VM, once to lambda capture. Getting rid of one is an easy fix, lambda capture can't go, so VM on another line seems like the right fix. 1293661 11 cgarcia 2017-04-03 09:40:38 -0700 (In reply to Saam Barati from comment #8) > Comment on attachment 306063 [details] > Updated patch > > View in context: > https://bugs.webkit.org/attachment.cgi?id=306063&action=review > > >>> Source/JavaScriptCore/wasm/js/WebAssemblyPrototype.cpp:94 > >>> + p.vm().promiseDeferredTimer->scheduleWorkSoon(promise, [source, promise, globalObject, plan = makeRefPtr(p)] () mutable { > >> > >> Why not plan=WTFMove(plan)? > > > > Because I've removed the line that created the plan to simplify it. > > > > RefPtr<Plan> plan = makeRef(p); > > [plan = WTFMove(plan)] > > > > is equivalent to > > > > [plan = makeRefPtr(p)] > > > > I think this way is simpler, since we were only creating the plan to transfer it to the lambda capture (and to get the vm, but that is what caused the crash). > > Gotcha. I see what's going on. > I still think you need a local variable for the plan, otherwise, won't the > lambda capture the outermost plan? I think that'll lead to a reference cycle. I think it's the same, we are taking a reference of p and transferring it to the lambda. > Even if that assignment doesn't cause this to happen, I think it's harder to > reason about than just having a local variable. I don't mind leaving the local variable, I just thought it was simpler this way, in one step. 1293662 12 cgarcia 2017-04-03 09:43:21 -0700 (In reply to Saam Barati from comment #9) > Comment on attachment 306063 [details] > Updated patch > > View in context: > https://bugs.webkit.org/attachment.cgi?id=306063&action=review > > >>>> Source/JavaScriptCore/wasm/js/WebAssemblyPrototype.cpp:94 > >>>> + p.vm().promiseDeferredTimer->scheduleWorkSoon(promise, [source, promise, globalObject, plan = makeRefPtr(p)] () mutable { > >>> > >>> Why not plan=WTFMove(plan)? > >> > >> Because I've removed the line that created the plan to simplify it. > >> > >> RefPtr<Plan> plan = makeRef(p); > >> [plan = WTFMove(plan)] > >> > >> is equivalent to > >> > >> [plan = makeRefPtr(p)] > >> > >> I think this way is simpler, since we were only creating the plan to transfer it to the lambda capture (and to get the vm, but that is what caused the crash). > > > > Gotcha. I see what's going on. > > I still think you need a local variable for the plan, otherwise, won't the lambda capture the outermost plan? I think that'll lead to a reference cycle. > > Even if that assignment doesn't cause this to happen, I think it's harder to reason about than just having a local variable. > > Seems like you could also keep the code as is, and just use p.vm() instead > of plan->vm(). Yes, that's my first patch see comment #1 > Also, I'm not sure why we make a RefPtr instead of a Ref here. I think it's because then we end up doing Ref<> foo = Ref(); and Ref(const Ref& other) is deleted. 1293663 13 cgarcia 2017-04-03 09:45:01 -0700 (In reply to JF Bastien from comment #10) > (In reply to Saam Barati from comment #9) > > Comment on attachment 306063 [details] > > Updated patch > > > > View in context: > > https://bugs.webkit.org/attachment.cgi?id=306063&action=review > > > > >>>> Source/JavaScriptCore/wasm/js/WebAssemblyPrototype.cpp:94 > > >>>> + p.vm().promiseDeferredTimer->scheduleWorkSoon(promise, [source, promise, globalObject, plan = makeRefPtr(p)] () mutable { > > >>> > > >>> Why not plan=WTFMove(plan)? > > >> > > >> Because I've removed the line that created the plan to simplify it. > > >> > > >> RefPtr<Plan> plan = makeRef(p); > > >> [plan = WTFMove(plan)] > > >> > > >> is equivalent to > > >> > > >> [plan = makeRefPtr(p)] > > >> > > >> I think this way is simpler, since we were only creating the plan to transfer it to the lambda capture (and to get the vm, but that is what caused the crash). > > > > > > Gotcha. I see what's going on. > > > I still think you need a local variable for the plan, otherwise, won't the lambda capture the outermost plan? I think that'll lead to a reference cycle. > > > Even if that assignment doesn't cause this to happen, I think it's harder to reason about than just having a local variable. > > > > Seems like you could also keep the code as is, and just use p.vm() instead > > of plan->vm(). > > Also, I'm not sure why we make a RefPtr instead of a Ref here. > > This weirdness is why I suggest getting VM, and then using it on this line. > The entire problem is using plan two ways on one line of code: once to get > to VM, once to lambda capture. Getting rid of one is an easy fix, lambda > capture can't go, so VM on another line seems like the right fix. I don't see anything weird in the current patch, TBH. But I'm fine with any other solution, either my first patch or adding another local variable for the VM or a combination of both using a local variable for VM, but using makeRefPtr directly in the lambda capture. All of them should work. 1293717 14 saam 2017-04-03 11:42:36 -0700 (In reply to Carlos Garcia Campos from comment #12) > (In reply to Saam Barati from comment #9) > > Comment on attachment 306063 [details] > > Updated patch > > > > View in context: > > https://bugs.webkit.org/attachment.cgi?id=306063&action=review > > > > >>>> Source/JavaScriptCore/wasm/js/WebAssemblyPrototype.cpp:94 > > >>>> + p.vm().promiseDeferredTimer->scheduleWorkSoon(promise, [source, promise, globalObject, plan = makeRefPtr(p)] () mutable { > > >>> > > >>> Why not plan=WTFMove(plan)? > > >> > > >> Because I've removed the line that created the plan to simplify it. > > >> > > >> RefPtr<Plan> plan = makeRef(p); > > >> [plan = WTFMove(plan)] > > >> > > >> is equivalent to > > >> > > >> [plan = makeRefPtr(p)] > > >> > > >> I think this way is simpler, since we were only creating the plan to transfer it to the lambda capture (and to get the vm, but that is what caused the crash). > > > > > > Gotcha. I see what's going on. > > > I still think you need a local variable for the plan, otherwise, won't the lambda capture the outermost plan? I think that'll lead to a reference cycle. > > > Even if that assignment doesn't cause this to happen, I think it's harder to reason about than just having a local variable. > > > > Seems like you could also keep the code as is, and just use p.vm() instead > > of plan->vm(). > > Yes, that's my first patch see comment #1 I like your first patch. I'm not sure what the semantics are here of your capture of plan, since you don't define another variable named "plan". Anyways, I'm worried about a reference cycle, which is why I like an explicit local variable. See the comment in WasmPlan.h. > > > Also, I'm not sure why we make a RefPtr instead of a Ref here. > > I think it's because then we end up doing Ref<> foo = Ref(); and Ref(const > Ref& other) is deleted. Why not just do: Ref<Plan> plan(p); ? 1294005 15 cgarcia 2017-04-04 01:09:22 -0700 (In reply to Saam Barati from comment #14) > (In reply to Carlos Garcia Campos from comment #12) > > (In reply to Saam Barati from comment #9) > > > Comment on attachment 306063 [details] > > > Updated patch > > > > > > View in context: > > > https://bugs.webkit.org/attachment.cgi?id=306063&action=review > > > > > > >>>> Source/JavaScriptCore/wasm/js/WebAssemblyPrototype.cpp:94 > > > >>>> + p.vm().promiseDeferredTimer->scheduleWorkSoon(promise, [source, promise, globalObject, plan = makeRefPtr(p)] () mutable { > > > >>> > > > >>> Why not plan=WTFMove(plan)? > > > >> > > > >> Because I've removed the line that created the plan to simplify it. > > > >> > > > >> RefPtr<Plan> plan = makeRef(p); > > > >> [plan = WTFMove(plan)] > > > >> > > > >> is equivalent to > > > >> > > > >> [plan = makeRefPtr(p)] > > > >> > > > >> I think this way is simpler, since we were only creating the plan to transfer it to the lambda capture (and to get the vm, but that is what caused the crash). > > > > > > > > Gotcha. I see what's going on. > > > > I still think you need a local variable for the plan, otherwise, won't the lambda capture the outermost plan? I think that'll lead to a reference cycle. > > > > Even if that assignment doesn't cause this to happen, I think it's harder to reason about than just having a local variable. > > > > > > Seems like you could also keep the code as is, and just use p.vm() instead > > > of plan->vm(). > > > > Yes, that's my first patch see comment #1 > I like your first patch. I'm not sure what the semantics are here of your > capture of plan, since you don't define another variable named "plan". > Anyways, I'm worried about a reference cycle, which is why I like an > explicit local variable. See the comment in WasmPlan.h. But this lambda is not the completion task one. That one is kept as a member of Plan, that's why it can cause a cycle. In this case we are keeping a ref in the lambda even if we use a local variable, because create the ref and then move to the lambda is the same as creating the ref in the lambda capture directly. > > > > > Also, I'm not sure why we make a RefPtr instead of a Ref here. > > > > I think it's because then we end up doing Ref<> foo = Ref(); and Ref(const > > Ref& other) is deleted. > Why not just do: > Ref<Plan> plan(p); > ? JSWebAssemblyModule::createStub() expects a RefPtr<>&& anyway, so it's simpler this way. 1294420 16 cgarcia 2017-04-05 00:32:11 -0700 Saam, anything else you want me to change in the patch? 1294426 17 306054 saam 2017-04-05 00:45:49 -0700 Comment on attachment 306054 Patch r=me 1294428 18 cgarcia 2017-04-05 00:53:53 -0700 Committed r214935: <http://trac.webkit.org/changeset/214935> 306054 2017-04-02 09:56:54 -0700 2017-04-05 00:45:49 -0700 Patch jsc-wasm-crashes.diff text/plain 3422 cgarcia ZGlmZiAtLWdpdCBhL1NvdXJjZS9KYXZhU2NyaXB0Q29yZS9DaGFuZ2VMb2cgYi9Tb3VyY2UvSmF2 YVNjcmlwdENvcmUvQ2hhbmdlTG9nCmluZGV4IDBkNWI3ZmMyOTk5Li5jOWE5NmY1YzcyYiAxMDA2 NDQKLS0tIGEvU291cmNlL0phdmFTY3JpcHRDb3JlL0NoYW5nZUxvZworKysgYi9Tb3VyY2UvSmF2 YVNjcmlwdENvcmUvQ2hhbmdlTG9nCkBAIC0xLDMgKzEsMTggQEAKKzIwMTctMDQtMDIgIENhcmxv cyBHYXJjaWEgQ2FtcG9zICA8Y2dhcmNpYUBpZ2FsaWEuY29tPgorCisgICAgICAgIFdlYkFzc2Vt Ymx5OiBzZXZlcmFsIHRlc3RzIGFkZGVkIGluIHIyMTQ1MDQgY3Jhc2ggd2hlbiBidWlsZGluZyB3 aXRoIEdDQworICAgICAgICBodHRwczovL2J1Z3Mud2Via2l0Lm9yZy9zaG93X2J1Zy5jZ2k/aWQ9 MTcwMzkwCisKKyAgICAgICAgUmV2aWV3ZWQgYnkgTk9CT0RZIChPT1BTISkuCisKKyAgICAgICAg VGhlIHBhdHRlcm4gZm9vLT5iYXIoW2YgPSBXVEZNb3ZlKGZvbylde30pOyBjcmFzaGVzIHdoZW4g YnVpbGRpbmcgd2l0aCBHQ0MsIEkgYXNzdW1lIHRoZSBtb3ZlIGhhcHBlbnMgYmVmb3JlIHRoZQor ICAgICAgICBmb28gaXMgdXNlZCB0byBpbnZva2UgdGhlIGZ1bmN0aW9uLgorCisgICAgICAgICog d2FzbS9qcy9XZWJBc3NlbWJseVByb3RvdHlwZS5jcHA6CisgICAgICAgIChKU0M6OndlYkFzc2Vt Ymx5Q29tcGlsZUZ1bmMpOiBVc2UgcC52bSgpIGluc3RlYWQgb2YgcGxhbi0+dm0oKSwgYmVjYXVz ZSBwbGFuIGlzIG1vdmVkIGJ5IHRoZSBsYW1iZGEuCisgICAgICAgIChKU0M6Omluc3RhbnRpYXRl KTogRGl0dG8uCisgICAgICAgIChKU0M6OmNvbXBpbGVBbmRJbnN0YW50aWF0ZSk6IERpdHRvLgor CiAyMDE3LTA0LTAxICBPbGVrc2FuZHIgU2thY2hrb3YgIDxnc2thY2hrb3ZAZ21haWwuY29tPgog CiAgICAgICAgIE9iamVjdCB3aXRoIG51bWVyaWNhbCBrZXlzIHdpdGggZ2FwcyBnZXRzIGZpbGxl ZCBieSBOYU4gdmFsdWVzCmRpZmYgLS1naXQgYS9Tb3VyY2UvSmF2YVNjcmlwdENvcmUvd2FzbS9q cy9XZWJBc3NlbWJseVByb3RvdHlwZS5jcHAgYi9Tb3VyY2UvSmF2YVNjcmlwdENvcmUvd2FzbS9q cy9XZWJBc3NlbWJseVByb3RvdHlwZS5jcHAKaW5kZXggNzlmNDZiZmRjNzIuLjU4MThhYjhiNDYw IDEwMDY0NAotLS0gYS9Tb3VyY2UvSmF2YVNjcmlwdENvcmUvd2FzbS9qcy9XZWJBc3NlbWJseVBy b3RvdHlwZS5jcHAKKysrIGIvU291cmNlL0phdmFTY3JpcHRDb3JlL3dhc20vanMvV2ViQXNzZW1i bHlQcm90b3R5cGUuY3BwCkBAIC05Miw3ICs5Miw3IEBAIHN0YXRpYyBFbmNvZGVkSlNWYWx1ZSBK U0NfSE9TVF9DQUxMIHdlYkFzc2VtYmx5Q29tcGlsZUZ1bmMoRXhlY1N0YXRlKiBleGVjKQogCiAg ICAgUmVmPFBsYW4+IHBsYW4gPSBhZG9wdFJlZigqbmV3IFBsYW4odm0sICpzb3VyY2UsIFBsYW46 OlZhbGlkYXRpb24sIFtzb3VyY2UsIHByb21pc2UsIGdsb2JhbE9iamVjdF0gKFBsYW4mIHApIG11 dGFibGUgewogICAgICAgICBSZWZQdHI8UGxhbj4gcGxhbiA9IG1ha2VSZWYocCk7Ci0gICAgICAg IHBsYW4tPnZtKCkucHJvbWlzZURlZmVycmVkVGltZXItPnNjaGVkdWxlV29ya1Nvb24ocHJvbWlz ZSwgW3NvdXJjZSwgcHJvbWlzZSwgZ2xvYmFsT2JqZWN0LCBwbGFuID0gV1RGTW92ZShwbGFuKV0g KCkgbXV0YWJsZSB7CisgICAgICAgIHAudm0oKS5wcm9taXNlRGVmZXJyZWRUaW1lci0+c2NoZWR1 bGVXb3JrU29vbihwcm9taXNlLCBbc291cmNlLCBwcm9taXNlLCBnbG9iYWxPYmplY3QsIHBsYW4g PSBXVEZNb3ZlKHBsYW4pXSAoKSBtdXRhYmxlIHsKICAgICAgICAgICAgIFZNJiB2bSA9IHBsYW4t PnZtKCk7CiAgICAgICAgICAgICBhdXRvIHNjb3BlID0gREVDTEFSRV9DQVRDSF9TQ09QRSh2bSk7 CiAgICAgICAgICAgICBFeGVjU3RhdGUqIGV4ZWMgPSBnbG9iYWxPYmplY3QtPmdsb2JhbEV4ZWMo KTsKQEAgLTE2OSw3ICsxNjksNyBAQCBzdGF0aWMgdm9pZCBpbnN0YW50aWF0ZShWTSYgdm0sIEV4 ZWNTdGF0ZSogZXhlYywgSlNQcm9taXNlRGVmZXJyZWQqIHByb21pc2UsIEpTVwogICAgIC8vIGh0 dHBzOi8vYnVncy53ZWJraXQub3JnL3Nob3dfYnVnLmNnaT9pZD0xNzAyMDUKICAgICBSZWY8UGxh bj4gcGxhbiA9IGFkb3B0UmVmKCpuZXcgUGxhbih2bSwgbW9kdWxlLT5zb3VyY2UoKSwgUGxhbjo6 RnVsbENvbXBpbGUsIFtwcm9taXNlLCBpbnN0YW5jZSwgbW9kdWxlLCBlbnRyaWVzXSAoUGxhbiYg cCkgewogICAgICAgICBSZWZQdHI8UGxhbj4gcGxhbiA9IG1ha2VSZWYocCk7Ci0gICAgICAgIHBs YW4tPnZtKCkucHJvbWlzZURlZmVycmVkVGltZXItPnNjaGVkdWxlV29ya1Nvb24ocHJvbWlzZSwg W3Byb21pc2UsIGluc3RhbmNlLCBtb2R1bGUsIGVudHJpZXMsIHBsYW4gPSBXVEZNb3ZlKHBsYW4p XSAoKSB7CisgICAgICAgIHAudm0oKS5wcm9taXNlRGVmZXJyZWRUaW1lci0+c2NoZWR1bGVXb3Jr U29vbihwcm9taXNlLCBbcHJvbWlzZSwgaW5zdGFuY2UsIG1vZHVsZSwgZW50cmllcywgcGxhbiA9 IFdURk1vdmUocGxhbildICgpIHsKICAgICAgICAgICAgIFZNJiB2bSA9IHBsYW4tPnZtKCk7CiAg ICAgICAgICAgICBFeGVjU3RhdGUqIGV4ZWMgPSBpbnN0YW5jZS0+Z2xvYmFsT2JqZWN0KCktPmds b2JhbEV4ZWMoKTsKICAgICAgICAgICAgIHJlc29sdmUodm0sIGV4ZWMsIHByb21pc2UsIGluc3Rh bmNlLCBtb2R1bGUsIGVudHJpZXMpOwpAQCAtMTk1LDcgKzE5NSw3IEBAIHN0YXRpYyB2b2lkIGNv bXBpbGVBbmRJbnN0YW50aWF0ZShWTSYgdm0sIEV4ZWNTdGF0ZSogZXhlYywgSlNQcm9taXNlRGVm ZXJyZWQqIHByCiAKICAgICBSZWY8UGxhbj4gcGxhbiA9IGFkb3B0UmVmKCpuZXcgUGxhbih2bSwg KnNvdXJjZSwgUGxhbjo6VmFsaWRhdGlvbiwgW3NvdXJjZSwgcHJvbWlzZSwgaW1wb3J0T2JqZWN0 LCBnbG9iYWxPYmplY3RdIChQbGFuJiBwKSBtdXRhYmxlIHsKICAgICAgICAgUmVmUHRyPFBsYW4+ IHBsYW4gPSBtYWtlUmVmKHApOwotICAgICAgICBwbGFuLT52bSgpLnByb21pc2VEZWZlcnJlZFRp bWVyLT5zY2hlZHVsZVdvcmtTb29uKHByb21pc2UsIFtzb3VyY2UsIHByb21pc2UsIGltcG9ydE9i amVjdCwgZ2xvYmFsT2JqZWN0LCBwbGFuID0gV1RGTW92ZShwbGFuKV0gKCkgbXV0YWJsZSB7Cisg ICAgICAgIHAudm0oKS5wcm9taXNlRGVmZXJyZWRUaW1lci0+c2NoZWR1bGVXb3JrU29vbihwcm9t aXNlLCBbc291cmNlLCBwcm9taXNlLCBpbXBvcnRPYmplY3QsIGdsb2JhbE9iamVjdCwgcGxhbiA9 IFdURk1vdmUocGxhbildICgpIG11dGFibGUgewogICAgICAgICAgICAgVk0mIHZtID0gcGxhbi0+ dm0oKTsKICAgICAgICAgICAgIGF1dG8gc2NvcGUgPSBERUNMQVJFX0NBVENIX1NDT1BFKHZtKTsK ICAgICAgICAgICAgIEV4ZWNTdGF0ZSogZXhlYyA9IGdsb2JhbE9iamVjdC0+Z2xvYmFsRXhlYygp Owo= 306063 2017-04-02 23:03:36 -0700 2017-04-04 01:10:53 -0700 Updated patch jsc-wasm-crashes.diff text/plain 3712 cgarcia ZGlmZiAtLWdpdCBhL1NvdXJjZS9KYXZhU2NyaXB0Q29yZS9DaGFuZ2VMb2cgYi9Tb3VyY2UvSmF2 YVNjcmlwdENvcmUvQ2hhbmdlTG9nCmluZGV4IGExYzhiMmIwOTM2Li41MWRiNjcyNTlmNCAxMDA2 NDQKLS0tIGEvU291cmNlL0phdmFTY3JpcHRDb3JlL0NoYW5nZUxvZworKysgYi9Tb3VyY2UvSmF2 YVNjcmlwdENvcmUvQ2hhbmdlTG9nCkBAIC0xLDUgKzEsMjAgQEAKIDIwMTctMDQtMDIgIENhcmxv cyBHYXJjaWEgQ2FtcG9zICA8Y2dhcmNpYUBpZ2FsaWEuY29tPgogCisgICAgICAgIFdlYkFzc2Vt Ymx5OiBzZXZlcmFsIHRlc3RzIGFkZGVkIGluIHIyMTQ1MDQgY3Jhc2ggd2hlbiBidWlsZGluZyB3 aXRoIEdDQworICAgICAgICBodHRwczovL2J1Z3Mud2Via2l0Lm9yZy9zaG93X2J1Zy5jZ2k/aWQ9 MTcwMzkwCisKKyAgICAgICAgUmV2aWV3ZWQgYnkgTk9CT0RZIChPT1BTISkuCisKKyAgICAgICAg VGhlIHBhdHRlcm4gZm9vLT5iYXIoW2YgPSBXVEZNb3ZlKGZvbylde30pOyBjcmFzaGVzIHdoZW4g YnVpbGRpbmcgd2l0aCBHQ0MsIEkgYXNzdW1lIHRoZSBtb3ZlIGhhcHBlbnMgYmVmb3JlIHRoZQor ICAgICAgICBmb28gaXMgdXNlZCB0byBpbnZva2UgdGhlIGZ1bmN0aW9uLgorCisgICAgICAgICog d2FzbS9qcy9XZWJBc3NlbWJseVByb3RvdHlwZS5jcHA6CisgICAgICAgIChKU0M6OndlYkFzc2Vt Ymx5Q29tcGlsZUZ1bmMpOiBVc2UgcC52bSgpIGluc3RlYWQgb2YgcGxhbi0+dm0oKSwgYmVjYXVz ZSBwbGFuIGlzIG1vdmVkIGJ5IHRoZSBsYW1iZGEuCisgICAgICAgIChKU0M6Omluc3RhbnRpYXRl KTogRGl0dG8uCisgICAgICAgIChKU0M6OmNvbXBpbGVBbmRJbnN0YW50aWF0ZSk6IERpdHRvLgor CisyMDE3LTA0LTAyICBDYXJsb3MgR2FyY2lhIENhbXBvcyAgPGNnYXJjaWFAaWdhbGlhLmNvbT4K KwogICAgICAgICBTaGFyZSBpbXBsZW1lbnRhdGlvbiBvZiBKU1J1bkxvb3BUaW1lcjo6dGltZXJE aWRGaXJlCiAgICAgICAgIGh0dHBzOi8vYnVncy53ZWJraXQub3JnL3Nob3dfYnVnLmNnaT9pZD0x NzAzOTIKIApkaWZmIC0tZ2l0IGEvU291cmNlL0phdmFTY3JpcHRDb3JlL3dhc20vanMvV2ViQXNz ZW1ibHlQcm90b3R5cGUuY3BwIGIvU291cmNlL0phdmFTY3JpcHRDb3JlL3dhc20vanMvV2ViQXNz ZW1ibHlQcm90b3R5cGUuY3BwCmluZGV4IDc5ZjQ2YmZkYzcyLi5iNDExNmM3NDJlNyAxMDA2NDQK LS0tIGEvU291cmNlL0phdmFTY3JpcHRDb3JlL3dhc20vanMvV2ViQXNzZW1ibHlQcm90b3R5cGUu Y3BwCisrKyBiL1NvdXJjZS9KYXZhU2NyaXB0Q29yZS93YXNtL2pzL1dlYkFzc2VtYmx5UHJvdG90 eXBlLmNwcApAQCAtOTEsOCArOTEsNyBAQCBzdGF0aWMgRW5jb2RlZEpTVmFsdWUgSlNDX0hPU1Rf Q0FMTCB3ZWJBc3NlbWJseUNvbXBpbGVGdW5jKEV4ZWNTdGF0ZSogZXhlYykKICAgICB2bS5wcm9t aXNlRGVmZXJyZWRUaW1lci0+YWRkUGVuZGluZ1Byb21pc2UocHJvbWlzZSwgV1RGTW92ZShkZXBl bmRlbmNpZXMpKTsKIAogICAgIFJlZjxQbGFuPiBwbGFuID0gYWRvcHRSZWYoKm5ldyBQbGFuKHZt LCAqc291cmNlLCBQbGFuOjpWYWxpZGF0aW9uLCBbc291cmNlLCBwcm9taXNlLCBnbG9iYWxPYmpl Y3RdIChQbGFuJiBwKSBtdXRhYmxlIHsKLSAgICAgICAgUmVmUHRyPFBsYW4+IHBsYW4gPSBtYWtl UmVmKHApOwotICAgICAgICBwbGFuLT52bSgpLnByb21pc2VEZWZlcnJlZFRpbWVyLT5zY2hlZHVs ZVdvcmtTb29uKHByb21pc2UsIFtzb3VyY2UsIHByb21pc2UsIGdsb2JhbE9iamVjdCwgcGxhbiA9 IFdURk1vdmUocGxhbildICgpIG11dGFibGUgeworICAgICAgICBwLnZtKCkucHJvbWlzZURlZmVy cmVkVGltZXItPnNjaGVkdWxlV29ya1Nvb24ocHJvbWlzZSwgW3NvdXJjZSwgcHJvbWlzZSwgZ2xv YmFsT2JqZWN0LCBwbGFuID0gbWFrZVJlZlB0cihwKV0gKCkgbXV0YWJsZSB7CiAgICAgICAgICAg ICBWTSYgdm0gPSBwbGFuLT52bSgpOwogICAgICAgICAgICAgYXV0byBzY29wZSA9IERFQ0xBUkVf Q0FUQ0hfU0NPUEUodm0pOwogICAgICAgICAgICAgRXhlY1N0YXRlKiBleGVjID0gZ2xvYmFsT2Jq ZWN0LT5nbG9iYWxFeGVjKCk7CkBAIC0xNjgsOCArMTY3LDcgQEAgc3RhdGljIHZvaWQgaW5zdGFu dGlhdGUoVk0mIHZtLCBFeGVjU3RhdGUqIGV4ZWMsIEpTUHJvbWlzZURlZmVycmVkKiBwcm9taXNl LCBKU1cKICAgICAvLyBGSVhNRTogVGhpcyByZS1wYXJzZXMgdGhlIG1vZHVsZSBoZWFkZXIsIHdo aWNoIHNob3VsZG4ndCBiZSBuZWNlc3NhcnkuCiAgICAgLy8gaHR0cHM6Ly9idWdzLndlYmtpdC5v cmcvc2hvd19idWcuY2dpP2lkPTE3MDIwNQogICAgIFJlZjxQbGFuPiBwbGFuID0gYWRvcHRSZWYo Km5ldyBQbGFuKHZtLCBtb2R1bGUtPnNvdXJjZSgpLCBQbGFuOjpGdWxsQ29tcGlsZSwgW3Byb21p c2UsIGluc3RhbmNlLCBtb2R1bGUsIGVudHJpZXNdIChQbGFuJiBwKSB7Ci0gICAgICAgIFJlZlB0 cjxQbGFuPiBwbGFuID0gbWFrZVJlZihwKTsKLSAgICAgICAgcGxhbi0+dm0oKS5wcm9taXNlRGVm ZXJyZWRUaW1lci0+c2NoZWR1bGVXb3JrU29vbihwcm9taXNlLCBbcHJvbWlzZSwgaW5zdGFuY2Us IG1vZHVsZSwgZW50cmllcywgcGxhbiA9IFdURk1vdmUocGxhbildICgpIHsKKyAgICAgICAgcC52 bSgpLnByb21pc2VEZWZlcnJlZFRpbWVyLT5zY2hlZHVsZVdvcmtTb29uKHByb21pc2UsIFtwcm9t aXNlLCBpbnN0YW5jZSwgbW9kdWxlLCBlbnRyaWVzLCBwbGFuID0gbWFrZVJlZlB0cihwKV0gKCkg ewogICAgICAgICAgICAgVk0mIHZtID0gcGxhbi0+dm0oKTsKICAgICAgICAgICAgIEV4ZWNTdGF0 ZSogZXhlYyA9IGluc3RhbmNlLT5nbG9iYWxPYmplY3QoKS0+Z2xvYmFsRXhlYygpOwogICAgICAg ICAgICAgcmVzb2x2ZSh2bSwgZXhlYywgcHJvbWlzZSwgaW5zdGFuY2UsIG1vZHVsZSwgZW50cmll cyk7CkBAIC0xOTQsOCArMTkyLDcgQEAgc3RhdGljIHZvaWQgY29tcGlsZUFuZEluc3RhbnRpYXRl KFZNJiB2bSwgRXhlY1N0YXRlKiBleGVjLCBKU1Byb21pc2VEZWZlcnJlZCogcHIKICAgICB2bS5w cm9taXNlRGVmZXJyZWRUaW1lci0+YWRkUGVuZGluZ1Byb21pc2UocHJvbWlzZSwgV1RGTW92ZShk ZXBlbmRlbmNpZXMpKTsKIAogICAgIFJlZjxQbGFuPiBwbGFuID0gYWRvcHRSZWYoKm5ldyBQbGFu KHZtLCAqc291cmNlLCBQbGFuOjpWYWxpZGF0aW9uLCBbc291cmNlLCBwcm9taXNlLCBpbXBvcnRP YmplY3QsIGdsb2JhbE9iamVjdF0gKFBsYW4mIHApIG11dGFibGUgewotICAgICAgICBSZWZQdHI8 UGxhbj4gcGxhbiA9IG1ha2VSZWYocCk7Ci0gICAgICAgIHBsYW4tPnZtKCkucHJvbWlzZURlZmVy cmVkVGltZXItPnNjaGVkdWxlV29ya1Nvb24ocHJvbWlzZSwgW3NvdXJjZSwgcHJvbWlzZSwgaW1w b3J0T2JqZWN0LCBnbG9iYWxPYmplY3QsIHBsYW4gPSBXVEZNb3ZlKHBsYW4pXSAoKSBtdXRhYmxl IHsKKyAgICAgICAgcC52bSgpLnByb21pc2VEZWZlcnJlZFRpbWVyLT5zY2hlZHVsZVdvcmtTb29u KHByb21pc2UsIFtzb3VyY2UsIHByb21pc2UsIGltcG9ydE9iamVjdCwgZ2xvYmFsT2JqZWN0LCBw bGFuID0gbWFrZVJlZlB0cihwKV0gKCkgbXV0YWJsZSB7CiAgICAgICAgICAgICBWTSYgdm0gPSBw bGFuLT52bSgpOwogICAgICAgICAgICAgYXV0byBzY29wZSA9IERFQ0xBUkVfQ0FUQ0hfU0NPUEUo dm0pOwogICAgICAgICAgICAgRXhlY1N0YXRlKiBleGVjID0gZ2xvYmFsT2JqZWN0LT5nbG9iYWxF eGVjKCk7Cg==