Github user rvs commented on a diff in the pull request:
https://github.com/apache/guacamole-server/pull/206#discussion_r238508851
--- Diff: configure.ac ---
@@ -140,6 +140,46 @@ AC_SUBST([COMMON_SSH_INCLUDE],
'-I$(top_srcdir)/src/common-ssh')
AC_SUBST([TERMINAL_LTLIB],
'$(top_builddir)/src/terminal/libguac_terminal.la')
AC_SUBST([TERMINAL_INCLUDE], '-I$(top_srcdir)/src/terminal $(PANGO_CFLAGS)
$(PANGOCAIRO_CFLAGS) $(COMMON_INCLUDE)')
+# pthread stack size
+AC_ARG_WITH(pthread_stack,
+ [AS_HELP_STRING([--with-pthread_stack=<size in bytes>],
+ [explicitly set pthread stack size (8MB is
recommended)])
+ ],pthread_stack_size=$withval
+ AC_DEFINE_UNQUOTED([PTHREAD_STACK_SIZE],
[$pthread_stack_size], [pthread stack size (8MB is recommended)])
+)
+if test "x$ac_cv_lib_pthread_pthread_create" = xyes; then
+ AC_MSG_CHECKING([whether default pthread stack is larger than 8MB])
+ ac_save_libs="$LIBS"
+ LIBS="$LIBS -lpthread"
+ AC_RUN_IFELSE([AC_LANG_SOURCE([[
--- End diff --
I honestly don't know if there is. I couldn't come up with one. That said
-- this is basically a runtime check for a warning message. As such I can
remove it (while leaving the rest of the patch). It would, however, be *very*
unfortunate to do so since in my personal example it cost me a day to figure
out that it was the stack size that was killing guacd on alpine. Anybody
building guacd in the future on a similar platform will surely appreciate
getting this type of a message early.
ON THE OTHER HAND -- if you're willing to make it a firm guarantee -- not
an option that this size of a stack is alway requested -- perhaps we can
simplify this patch quite a bit. Is this something that would be better
discussed on the mailing list?
---
