+The time the "make" step takes depends heavily on optimization levels.
+Large amounts of memory (>128MB) will be required by the compiler,
+also depending on optimization. To give you a rough idea of what you
+have to expect the following table may be helpful. It was measured on
+a P-III/500MHz with "enough" memory:
+
+step: | GCC optimization level: | comment:
+ | -O0 | -O1 | -O2 |
+--------------+--------+--------+--------+-------------------------------
+make | ~4m | ~6m | ~10m | building shared and static lib
+make check | ~2m | ~2m | ~7m | mostly due to compilation
+
+
+COMMON PROBLEMS
+===============
+
+Problems with CLN
+-----------------
+
+You should use at least CLN V1.0.3, since during the development of
+GiNaC various bugs have been discovered and fixed in earlier versions.
+To protect you, the "configure" script checks for a feature that was
+added in V1.0.3 so it won't continue with earlier versions anyhow.
+Please install CLN properly on your system before continuing with
+GiNaC.
+
+Problems with building ginsh
+----------------------------
+
+The most common reason why this doesn't succeed is the absence of
+libreadline and/or the corresponding header files. Depending on what
+your system/distribution is, you will have to install a package called
+libreadline and maybe libreadline-dev. If your system's vendor
+doesn't supply such packages, go to <ftp://ftp.gnu.org/gnu/readline/>
+and compile it yourself. After that it should work fine.
+
+If you want to build a completely static ginsh and your compiler still
+complains about unresolved objects try linking with libncurses, too.
+
+
+Working with the Cint C++ interpreter
+=====================================