Re: bug#327: 23.0.60; make bootstrap fails from CVS

[Glenn Morris]

Ted Zlatanov wrote:

> The XXX Makefile works and compiles calc-aent.el as expected.

Then I'm baffled, since it's doing the same thing as the normal
lisp/Makefile. (VPATH? see below).

> I notice the compilation fails for the first file in a subdirectory.

Yes, I already remarked on that, but I don't know what it means. (VPATH?)

> I am attaching my generated lisp/Makefile

The same as my own, save for the expected differences in
lisp,VPATH,srcdir variables.

> Make considers the right .el file but for some reason gives up on it.

I'm sorry, I don't know what's going on. The only suggestions I have
left:

Try moving your CVS tree to eg /tmp/foo and building there, just in
case there is something about the path to your sources that is
confusing make for reasons I am failing to see.

In the test Makefile, increase ELCFILES to the full list from the
normal makefile, and see if it still works.

Throwing ideas at random, could it be something to do with VPATH? Try
adding the calc dir to VPATH at the top of Makefile, or deleting VPATH
altogether. But then I don't know why it works for everyone (?) else.

If all else fails, please try asking on help-make at gnu.org.

Help, anyone?