Age | Commit message (Collapse) | Author |
|
|
|
The loop in the earlier patch would invoke undefined behavior in two
ways: First, it would check whether it was looking at a space before
it checked whether the pointer was in-range. Second, it would let a
pointer reach a position _before_ the start of a string, which is
not allowed.
I've removed the assertion about empty messages: empty messages can
be their own warning IMO.
I've also added tests for this formatting code, to make sure it
actually works.
|
|
|
|
|
|
|
|
This is an automated commit, generated by this command:
./scripts/maint/rename_c_identifier.py \
EXPOSE_CLEAN_BACKTRACE BACKTRACE_PRIVATE \
TOR_CHANNEL_INTERNAL_ CHANNEL_OBJECT_PRIVATE \
CHANNEL_PRIVATE_ CHANNEL_FILE_PRIVATE \
EXPOSE_ROUTERDESC_TOKEN_TABLE ROUTERDESC_TOKEN_TABLE_PRIVATE \
SCHEDULER_PRIVATE_ SCHEDULER_PRIVATE
|
|
|
|
|
|
Coverity can't see that it is not in fact going to read
uninitialized memory here, so we initialize these values
unconditionally.
Bugfix on 0.4.0.1-alpha.
|
|
Part of ticket 29976.
|
|
Part of ticket 29976.
|
|
Also, split the formatting code shared by control.c and
control_events.c into controller_fmt.c.
|
|
|
|
Part of ticket 27617.
|
|
Redefine the set of bootstrap phases to allow display of finer-grained
progress in the early connection stages of connecting to a relay.
This includes adding intermediate phases for proxy and PT connections.
Also add a separate new phase to indicate obtaining enough directory
info to build circuits so we can report that independently of actually
initiating an ORCONN to build the first application circuit.
Previously, we would claim to be connecting to a relay when we had
merely finished obtaining directory info.
Part of ticket 27167.
|
|
Replace a few invocations of control_event_bootstrap() with calls from
the bootstrap tracker subsystem. This mostly leaves behavior
unchanged. The actual behavior changes come in the next commit.
Part of ticket 27167.
|
|
Existing cached directory information can cause misleadingly high
bootstrap percentages. To improve user experience, defer reporting of
directory information progress until at least one connection has
succeeded to a relay or bridge.
Closes ticket 27169.
|
|
I am very glad to have written this script.
|
|
|
|
|
|
|
|
|
|
This option was used for shadow testing previously, but is no longer
used for anything. It interferes with refactoring our token buckets.
|
|
|
|
|
|
Some of them like to munge the global event mask, so it's important
to have that behavior isolated.
|
|
Part of ticket 15431, checks for bugs similar to 13085.
|
|
|
|
Part of fix for 13172
|
|
Because in 95 years, we or our successors will surely care about
enforcing the BSD license terms on this code. Right?
|
|
Before the 11825 fix, these were all silently ignored.
|
|
|
|
|
|
CIDs: 1130994, 1130993, 1130992, 1130991
|
|
Don't cast uint64_t * to const uint64_t * explicitly. The cast is always
safe, so C does it for us. Doing the cast explitictly can hide bugs if
the input is secretly the wrong type.
Suggested by Nick.
|
|
Suggested by nickm.
|
|
Suggested by nickm.
|
|
|