(In the tree where I building the core branch used for the app as
distributed, that is.)
Change-Id: Ice622c79ff9c7f56f4e58f68fe65e5d89696681b
(cherry picked from commit 21dc19f7a28174fbd5eabd0f60b6f1cedcafd89c)
Reviewed-on: https://gerrit.libreoffice.org/79381
Reviewed-by: Tor Lillqvist <tml@collabora.com>
Tested-by: Tor Lillqvist <tml@collabora.com>
It is now possible to run ./configure --with-sanitizer=address
(or any least of valid sanitizers) to build with sanitizers.
When --with-sanitizer is specified, we build with -O1 and
don't omit frame-pointer. We also enable RTTI (which should
be enabled anyway, but just in case).
UBSan (undefined) sanitizer can cause 'typeinfo' errors.
Fixing the source is best, but as a workaround, disabling
vptr sanitizer works. Just pass -fno-sanitize=vptr to CFLAGS
and CXXFLAGS.
Change-Id: I4031aa872b1b1ef779703135394f3733952e5cd1
Reviewed-on: https://gerrit.libreoffice.org/79329
Reviewed-by: Ashod Nakashian <ashnakash@gmail.com>
Tested-by: Ashod Nakashian <ashnakash@gmail.com>
As the comment details, this avoids having C++ objects
in the same frame as setjmp, which may reset their
contents without the dtor getting called.
Change-Id: I851ae8bffb4356d465a25dfc815a1fecb489fa30
Reviewed-on: https://gerrit.libreoffice.org/79338
Reviewed-by: Ashod Nakashian <ashnakash@gmail.com>
Tested-by: Ashod Nakashian <ashnakash@gmail.com>
No longer is it necessary to run with --verbose
in case we need --debugrun.
Also, by default the log level is now warning instead
of error, which should hopefully be more useful in
troubleshooting issues (or at least to force us
better categorize log messages).
Change-Id: Iad4279ecf3bf77780adcd786d4f46d9c964f302d
Reviewed-on: https://gerrit.libreoffice.org/79351
Reviewed-by: Ashod Nakashian <ashnakash@gmail.com>
Tested-by: Ashod Nakashian <ashnakash@gmail.com>
Now the wsd docbroker thread and its peer kit
thread are trivial to match, since they are called
docbroker_xxx and kitbroker_xxx (where xxx is the
instance ID) respectively.
Also, label spare kit instances as kit_spare_xxx
to differentiate from ones with actual documents,
where xxx is a running counter to differentiate
spare instances from one another.
Now we are able to easily see (and count) the number
of spare kit instances, and match wsd and kit threads
handling a given document.
Unit-test logic updated to reflect the new thread
naming scheme.
Change-Id: I154dc8f200fbe0e65f3f5984e6dad2cef1b52e22
Reviewed-on: https://gerrit.libreoffice.org/79328
Reviewed-by: Ashod Nakashian <ashnakash@gmail.com>
Tested-by: Ashod Nakashian <ashnakash@gmail.com>
The thread name helps one track threads in logs.
When renaming threads it's important to log the process
and previous thread name (if any), so grepping is more
fruitful and tracking is easier.
Change-Id: I47a948d77629b387cc1e9fd58fdd88e1ae1168df
Reviewed-on: https://gerrit.libreoffice.org/79327
Reviewed-by: Ashod Nakashian <ashnakash@gmail.com>
Tested-by: Ashod Nakashian <ashnakash@gmail.com>
"Session" is too generic and can be misleading.
This is really a connection ID, or a connection-session ID,
so it's more sensible to call it ConnectionId to
signify that it's a connection-specific ID.
Change-Id: I0f12b1cf80d4853a27535101093ef6b7a3e0d438
Reviewed-on: https://gerrit.libreoffice.org/79326
Reviewed-by: Ashod Nakashian <ashnakash@gmail.com>
Tested-by: Ashod Nakashian <ashnakash@gmail.com>
This used to work fine, but at some stage it had regressed. I did not
bother bisecting to find the culprit. Anyway, now the JS part needs to
send a lang parameter in the load message, too, for the corrent
language to be used in the tunnelled dialogs. (Likely it affects also
other things in core.)
Change-Id: I339cddb28357978414669964106cb6fa472c967f
Reviewed-on: https://gerrit.libreoffice.org/77599
Tested-by: Jenkins CollaboraOffice <jenkinscollaboraoffice@gmail.com>
Reviewed-by: Tor Lillqvist <tml@collabora.com>
Tested-by: Tor Lillqvist <tml@collabora.com>
Reviewed-on: https://gerrit.libreoffice.org/79023
There is now other code that takes care of it.
This reverts commit 3cc714b49b6e29429d45818a281cd77a209dbec2.
Change-Id: I9c777e1592d1c5e0fc931307b24e63743cd4da5b
Reviewed-on: https://gerrit.libreoffice.org/78279
Reviewed-by: Tor Lillqvist <tml@collabora.com>
Tested-by: Tor Lillqvist <tml@collabora.com>
Reviewed-on: https://gerrit.libreoffice.org/79014
Avoids a warning when compiling for iOS: format specifies type
'unsigned long' but the argument has type 'uint64_t' (aka 'unsigned
long long').
Change-Id: I8b5205dd0c3a8ae2f531f1647b3e3bac27ea6065
Reviewed-on: https://gerrit.libreoffice.org/78985
Reviewed-by: Tor Lillqvist <tml@collabora.com>
Tested-by: Tor Lillqvist <tml@collabora.com>
We never exit an app process intentionally, so Log::isShutdownCalled()
can always return false.
Change-Id: I6b3cce85ccac5e3ffbb7b29ca552cb9e4441df2a
Reviewed-on: https://gerrit.libreoffice.org/78979
Reviewed-by: Tor Lillqvist <tml@collabora.com>
Tested-by: Tor Lillqvist <tml@collabora.com>
And also others, pointed out by IWYU.
Finally, it's not needed to force-enable HTTPS after all, it seems.
Change-Id: I4d9984d2485b1519bd5a9cc44834735e09a87058
This was originally added in commit
c2c7982e5e (Add TIFF load testcase,
2019-07-04), and still fails without the core.git fix.
Change-Id: I798b82ea8212dfeafc162e4b73f6fff707ac3de2
==18134==ERROR: LeakSanitizer: detected memory leaks
Direct leak of 8 byte(s) in 1 object(s) allocated from:
#0 0x514db8 in operator new(unsigned long) /git/libreoffice/lode/packages/llvm-472c6ef8b0f53061b049039f9775ab127beafbe4.src/compiler-rt/lib/asan/asan_new_delete.cc:105
#1 0x9f836c in runClientTests(bool, bool) /git/libreoffice/online-san/test/test.cpp:113:28
#2 0x9f7975 in main /git/libreoffice/online-san/test/test.cpp:77:12
Change-Id: Ib93625e2a77e37733b9bb052dd73ec3333bc2a71
net/Socket.hpp:405:9: runtime error: member call on address 0x6070007a2210 which does not point to an object of type 'Poco::Logger'
0x6070007a2210: note: object has invalid vptr
05 00 80 0e 7c 01 80 6f 3c 7f 00 00 c8 b8 0f 01 3c 7f 00 00 00 00 00 00 be be be be 38 22 7a 00
^~~~~~~~~~~~~~~~~~~~~~~
invalid vptr
SUMMARY: UndefinedBehaviorSanitizer: undefined-behavior net/Socket.hpp:405:9 in
Which is use-after-free: once Poco::Logger::shutdown() is called, we
need to tolerate LOG_DBG() and other similar calls, even if they don't
go anywhere.
Change-Id: Ic76433743177dd2b604ff34e340309c506d83350