Commits on Source 22
-
Thomas Lake authored
-
Andrew Price authored
Experiment: success Conclusion: no
-
Andrew Price authored
Various cleanups and simplifications of the makefiles, plus: - an additional 'make RELEASE_BUILD=1' mode which switches on optimisation and -D_FORTIFY_SOURCE=2, and switches off -Werror - Use RELEASE_BUILD=1 in rpm and deb packaging - Build with PIE and full RELRO by default - Add CFLAGS_APPEND to append options instead of overriding CFLAGS
-
Andrew Price authored
-
Andrew Price authored
And fix some issues it complains about. Also tweak one issue found by adding -Wmissing-field-initializers temporarily (that warning is too fussy to enable by default).
-
-
Andrew Price authored
It's a bit stream-of-consciousness at the moment but I hope it will be useful to help people get started.
-
Andrew Price authored
-
Andrew Price authored
Use -MF to make gcc output the autogenerated rules into dotfiles so that they don't clutter up listings.
-
Andrew Price authored
I don't believe this stuff is used any more and it's confusing to have a web/ directory and a webclient/ directory.
-
Andrew Price authored
When ipc_check() attempts to reconnect it passes ipc_parent as the target to connect to. ipc_connect() then reconnects and frees ipc_parent and sets it to strdup(target). When target is equal to ipc_parent this means we're strduping a freed pointer and reconnecting a second time (or using ipc_parent anywhere subsequently) will fail. Add a check to make sure they're not pointing to the same thing.
-
Andrew Price authored
-
Andrew Price authored
js.c:17:32: error: option ‘-Winvalid-offsetof’ is valid for C++/ObjC++ but not for C [-Werror=pragmas] #pragma GCC diagnostic ignored "-Winvalid-offsetof" ^~~~~~~~~~~~~~~~~~~~ js.c:21:32: error: option ‘-Winvalid-offsetof’ is valid for C++/ObjC++ but not for C [-Werror=pragmas] #pragma GCC diagnostic warning "-Winvalid-offsetof" ^~~~~~~~~~~~~~~~~~~~ cc1: all warnings being treated as errors ../../Makefile.common:104: recipe for target 'js.o' failed
-
Andrew Price authored
New in gcc 6: comms.c: In function ‘json_escape’: comms.c:337:8: error: this ‘else’ clause does not guard... [-Werror=misleading-indentation] }else ^~~~ comms.c:338:32: note: ...this statement, but the latter is misleadingly indented as if it is guarded by the ‘else’ mws_add(line, "%c", *in); break; ^~~~~ cc1: all warnings being treated as errors
-
Justin Mitchell authored
with -fpie. Update the credits a little
-
Andrew Price authored
-
Andrew Price authored
-
Justin Mitchell authored
-
Justin Mitchell authored
-
Justin Mitchell authored
Needed the server to send an error response instead of immediately closing, and wait for it to be sent. also needed certain parts of the client shutdown proceedure tonot cleanup/free things that hadnt yet been initialised.
-
Andrew Price authored
Makefile.common uses it to find the mw.rev file