Many C-C TB xpcshell tests fail, Hit MOZ_CRASH(assertion `left != right` failed: src and dst must not alias ) at mozilla/netwerk/base/idna_glue/src/lib.rs:66
Categories
(MailNews Core :: Networking, defect)
Tracking
(Not tracked)
People
(Reporter: ishikawa, Assigned: ishikawa)
References
(Regression)
Details
(Keywords: regression)
Attachments
(1 file, 1 obsolete file)
Starting very recently in the last 24 hours after the local tree update, many C-C TB tests fail.
One of the causes is reported here. (There seem to be other causes.)
The following is a failure from|mach xpcshell-test comm/mail/base/test/unit/test_viewWrapper_imapFolder.js| that shows this MOZ_CRASH error.
00:37.50 pid:415846 [Parent 415846, Main Thread] WARNING: NS_ENSURE_SUCCESS(rv, rv) failed with result 0x804B000A (NS_ERROR_MALFORMED_URI): file /NEW-SSD/NREF-COMM-CENTRAL/mozilla/comm/mailnews/base/src/nsMsgAccountManager.cpp:1855
00:37.50 pid:415846 [Parent 415846, Main Thread] WARNING: (debug) findServerInternal("nobody"_ns, "Local Folders"_ns, "none"_ns, 0, aServer)=rv=0x804b000a aServer=7ffeb8c5c5a0
00:37.50 pid:415846 : file /NEW-SSD/NREF-COMM-CENTRAL/mozilla/comm/mailnews/base/src/nsMsgAccountManager.cpp:2122
00:37.50 pid:415846 [415846] Hit MOZ_CRASH(assertion `left != right` failed: src and dst must not alias
00:37.50 pid:415846 left: 0x7ffeb8c5c2b0
00:37.50 pid:415846 right: 0x7ffeb8c5c2b0) at /NEW-SSD/NREF-COMM-CENTRAL/mozilla/netwerk/base/idna_glue/src/lib.rs:66
00:37.50 pid:415846 #01: RustMozCrash (/NEW-SSD/NREF-COMM-CENTRAL/mozilla/mozglue/static/rust/wrappers.cpp:18)
00:37.50 pid:415846 #02: ??? (/NEW-SSD/moz-obj-dir/objdir-tb3/dist/bin/libxul.so + 0xbc552b4)
00:37.50 pid:415846 #03: std::panicking::rust_panic_with_hook (library/alloc/src/boxed.rs:0)
00:37.50 pid:415846 #04: std::panicking::begin_panic_handler::{{closure}} (library/std/src/panicking.rs:0)
00:37.50 pid:415846 #05: ??? (/NEW-SSD/moz-obj-dir/objdir-tb3/dist/bin/libxul.so + 0xc776b76)
00:37.50 pid:415846 #06: <std::panicking::begin_panic_handler::FormatStringPayload as core::panic::PanicPayload>::take_box (library/std/src/panicking.rs:618)
00:37.50 pid:415846 #07: ??? (/NEW-SSD/moz-obj-dir/objdir-tb3/dist/bin/libxul.so + 0xc7a24c5)
00:37.50 pid:415846 #08: ??? (/NEW-SSD/moz-obj-dir/objdir-tb3/dist/bin/libxul.so + 0xc7a2a7f)
00:37.50 pid:415846 #09: ??? (/NEW-SSD/moz-obj-dir/objdir-tb3/dist/bin/libxul.so + 0xbc74394)
00:37.50 pid:415846 #10: mozilla_net_domain_to_display_impl (src/lib.rs:0)
00:37.50 pid:415846 #11: nsMsgAccountManager::findServerInternal(nsTSubstring<char> const&, nsTSubstring<char> const&, nsTSubstring<char> const&, int, nsIMsgIncomingServer**) [clone .cold] (/NEW-SSD/NREF-COMM-CENTRAL/mozilla/comm/mailnews/base/src/nsMsgAccountManager.cpp:1887)
00:37.50 pid:415846 #12: nsMsgAccountManager::GetLocalFoldersServer(nsIMsgIncomingServer**) [clone .cold] (/NEW-SSD/NREF-COMM-CENTRAL/mozilla/comm/mailnews/base/src/nsMsgAccountManager.cpp:2132)
00:37.50 pid:415846 #13: nsMsgAccountManager::createKeyedAccount(nsTString<char> const&, bool, nsIMsgAccount**) [clone .cold] (/NEW-SSD/NREF-COMM-CENTRAL/mozilla/comm/mailnews/base/src/nsMsgAccountManager.cpp:1624)
...
The line numbers may be slightly off due to my local dump statements.
EDIT: Fuller log is separately attached.
Assignee | ||
Comment 1•4 months ago
•
|
||
try-comm-central is closed due to this and other causes.
EDIT: maybe only comm-central?
Assignee | ||
Comment 2•4 months ago
|
||
The MOZ_CRASH failures during xpcshell test are all from the same location.
And as far as I can tell, in all the cases, the left pointer value and the right pointer value seem to be identical.
Wait the line that calls ConvertToDisplayIDN() certainly pass the same ( pointers to) nsCStrings.
https://searchfox.org/comm-central/source/mailnews/base/src/nsMsgAccountManager.cpp#1885
Updated•4 months ago
|
Assignee | ||
Comment 4•4 months ago
|
||
Updated•4 months ago
|
Assignee | ||
Comment 5•4 months ago
|
||
I am reopening this because the MOZ_CRASH I am seeing is gone with the patch I post here.
Comment 6•4 months ago
|
||
(In reply to ISHIKAWA, Chiaki from comment #5)
I am reopening this because the MOZ_CRASH I am seeing is gone with the patch I post here.
Thanks for the patch! It looks like Geoff's WIP patch from bug 1906992 would already fix it, but let's decouple it and tackle this in here so it's not blocked behind fixing the bigger issue.
Comment 7•4 months ago
|
||
Moving this to MailNews Core as a regression of the Core bug, since this failure is specific to comm-central.
Updated•4 months ago
|
Comment 8•4 months ago
|
||
Fixed as part of bug 1906992
Description
•