This patch is the most scary of all proposals I've been mailing to you
this week until now.
It consists of 3 parts:
1. a patch
2. a perl script (acpolish)
3. a shell script wrapper to invoke the perl-script.
The perl-script reads in each Makefile.in and modifies them
("polishes/beautifies" them :-).
These modifications are not easy to describe:
Basically, it hard-codes some automake Makefile-variables and rules into
RTEMS autoconf-Makefile.ins (Note: autoconf vs. automake!!) and converts
some settings/variables to configure scripts' requirements (Yes,
plural).
E.g. it adds the automake standard variables $top_builddir and $subdir,
adds dependency rules for automatic re-generation of Makefiles from
Makefile.in, adds support variables for relative paths to multiple
configure scripts etc.
The patch is a one-line patch to enable the support of the new features
added by acpolish.
The shell script is a wrapper which pokes around inside of the source
tree for Makefile.ins and invokes acpolish on all autoconf-Makefile.ins.
acpolish is designed to be able to run several times on the same
Makefile.in and may once become a more general tool to convert RTEMS
Makefile.in to automake. Therefore, I'd like to keep it inside of source
tree. (e.g. as contrib/acpolish or c/update-tools/acpolish). However, it
doesn't make sense to export it outside of RTEMS.
To apply this:
cd <source-tree>
patch -p1 -E < <path-to-patch>/rtems-rc-19990318-1.diff
tar xzvf <path-to>/rtems-rc-polish.tar.gz
./rtems-polish.sh
./autogen
Note: The path contrib/acpolish is hard-coded into rtems-polish.sh, if
you decide to put it in an alternative place, please modify
rtems-polish.sh to reflect this change.
Later:
cvs rm make/rtems.cfg (It isn't used anymore)
cvs add contrib
cvs add contrib/acpolish
cvs commit
I've tested this intensively, but naturally I can't exclude bugs.
Ralf.
PS.: Most probably, this is the last "Towards automake" patch. The next
one probably will be a real automake patch.
This one once again changes the scheme to preinstall bsp_specs.
It moves generating PROJECT_ROOT/lib/bsp_specs to
libbsp/<cpu>/<bsp>/wrapup/Makefile.in.
I.e. it decentralizes generation of bsp_specs to a bsp-dependent
directory, because preinstalling bsp_specs in a centralized Makefile
like it has been done until now does not harmonize well with spliting
the toplevel configure script in cpu and bsp-dependent configure scripts
and automake.
First apply the patch (rtems-rc-19990318-0.diff) below, then run the
reorg-bsp_specs.sh script.
IMO, this one is comparatively harmless and eases automake support
significantly.
This is the next step towards automake:
* Two scripts for the toplevel directory:
a) "autogen" (Idea borrowed from libtool and gnome) A helper script to
recursively regenerate autoconf/automake/aclocal generated files
(Still not perfect but sufficient).
b) "missing" (from automake-cvs archive). This file normally is
automatically generated by automake, but we have to manually add
it until we add automake support to the toplevel configure script.
"chmod 755 missing autogen" after applying the patch.
* Changing the toplevel installation directory [ I can hear you
falling off the chair ;-]
Until now rtems installed itself to $(prefix)/rtems. This is in
contradiction to automake and GNU/FSF/Cygnus conventions.
With this patch applied, rtems installs into $(prefix).
To achieve the old behaviour simply configure with
--prefix=<install-dir>/rtems instead of --prefix=<install-dir>
This is a widely visible change and I can understand if you don't
like it at the present point. It enables us to use automake's
default installation paths instead of having to set up installation
paths manually. At the moment this doesn't help much, but in the not
so far future this would enable us to mix cpu-only dependent libraries
into the host's cross-compiler library and header files into
newlib's include directories, tools into the toolchain directories etc.
I would recommend to change the main installation directory, however it's
up to you to draw the final design decision.
OK, I 've made up my mind to cut a big chunk of my automake-patches (:-).
Below you can find a drop-in replacement of the aclocal directory. It
contains a lot of new macro files, most of them are directly cut from rtems
top-level configure script, some are new some are identical to former
versions. The motivation behind these files is to reuse parts from rtems
current top-level configure.in script in up-coming subdirectory configure.in
scripts.
I'd like to ask you to untar the archive ontop of the source tree and to
add/commit these files to CVS. Adding these files should not have any
influence on RTEMS momentary configuration (except of you are required to
run aclocal -I aclocal && autoconf afterwards), because most of them
currently are not used at all.
---------
BTW: Please upgrade to autoconf-2.13 and automake-2.4, if you havn't done
this already (egcs/CVS require them, too). My upcoming automake files
require automake-2.4 which requires autoconf-2.13 or later.
Erik Ivanenko pointed out a problem in the ne2000.c driver I
submitted: it did not work correctly with bootp. Here is a patch,
based on a patch he sent me.
First, the unlimited patch. I have compiled the unlmited patch for the
Linux posix BSP only and it seems to work cleanly. I would like a really
major application run on this change before commiting as the changes are
very core and significant. I am currently building all the tests to run.
I have no targets suitable to test on at the moment.
I have tested the patch for inline functions and macros.
Turning macros on has found some core bugs. I have fixed these but have
not run all the tests. Please review the patch for these changes. They
are:
1) The conditional compilation for MP support broke the core messages
code. You cannot embed a conditional macro in another macro. The Send
and Urgent Send calls are macros.
2) User extensions handler initialisation now has two parameters. I have
updated the macros to support the extra parameter.
The patch also contains the gcc-target-default.cfg fix required to build
the kernel. More of a by product than a fix for you.
problems that prevented the 19990302 snapshot from running on
the efi332.
I'm happy to report that rtems-19990302 is running on the efi332
board. I have enclosed a few minor patches below to the efi332 bsp. All
patches are within that library but one. make/custom/efi332.cfg has a
patch to select the right CPU_CFLAGS (at one time -m68332 was a
problem... -mcpu32 or -m68332 work fine now).
would work. At the same time, the initial implementation of F_SETFL
was added. A support routine was added to convert internal libio
flags back to the POSIX style. Eventually the internal representation
should be eliminated in the interest of simplicity and code reduction.
This problem was reported by Jake Janovetz <janovetz@tempest.ece.uiuc.edu>.
Comments follow:
Please find attached, the updated network driver. I have verified
that it is working as expected, by timestamping the error messages
generated from the ISR.
If you've taken a look inside, the network driver has a reset thread
in addition to the RX and TX threads. It is possible to avoid the
additional reset thread by allowing the TX driver to time out and then
checking status bits set by the ISR. However, this approach demands
that a transmission is necessary for the NIC to be reset.
Due to Eric V's ISR handling, I suppose that the reset routine could
be called from the "ISR" itself, due to the 8259 interrupt mode, and
that the interrupt is acknowledged prior to running the "ISR".
(Providing that no NIC interrupts are generated during reset -- I
worry about re-entrancy. )
This would be a minor improvement, but you know, I don't want to make
this driver my lifes work.
----------------------------------------------------------------------
----------------------------------------------------------------------
> 5) rtems-rc-19990202-1.diff/reorg-install.sh
>
> reorg-install.sh fixes a Makefile variable name clash of RTEMS
> configuration files and automake/autoconf standards.
> Until now, RTEMS used $(INSTALL) for install-if-change. Automake and
> autoconf use $(INSTALL) for a bsd-compatible install. As
> install-if-change and bsd-install are not compatible, I renamed all
> references to install-if-changed to $(INSTALL_CHANGED) and used
> $(INSTALL) for bsd-install (==automake/autoconf standard). When
> automake will be introduced install-if-change will probably be replaced
> by $(INSTALL) and therefore will slowly vanish. For the moment, this
> patch fixes a very nasty problem which prevents adding any automake file
> until now (There are still more).
rather then NAME_MAX. NAME_MAX is 255 and that lets IMFS chew up memory
too fast. Perhaps in the future, the places in IMFS that put a maximum
length name string on the stack and the jnode structure does not include
a maximu length name string can be fixed so this is not a problem.
FP issues on this target:
The default variants of libc, libm and libgcc assume that a 68881
coprocessor is present. Without the FPSP, any floating point operation,
including printf() with a "%f" format specifier, is likely to cause an
unimplemented instruction exception.
The FPSP works with the default variants of libc, libm and libgcc. It does not
work in conjunction with the msoft-float variants. The paranoia test goes into
an infinite loop at milestone 40. I am guessing that floor() is returning an
incorrect value.
The msoft-float variants of libc, libm and libgcc appear to do floating point
I/O properly. They only failed in paranoia. Offhand, I can't think of why they
would conflict with the FPSP, so I think that there is something wrong with the
msoft-float code. It might be my installation.
Given my experiences, I decided to install the FPSP in bsp_start(), and to link
against the default variants of libc, libm and libgcc. This causes the
executables to increase in size by about 60 KB. The README file and the
mvme167.cfg specify how to remove the FPSP, and how to link against the
msoft-float variants of the libraries. This is not what Eric Norum had done: on
my host, his gen68360_040 port links RTEMS code with the msoft-float variants
of libc and libm, and the default variant of libgcc. In this configuration, the
output of printf() with "%f" is garbage on my target.
FP issues on this target:
The default variants of libc, libm and libgcc assume that a 68881
coprocessor is present. Without the FPSP, any floating point operation,
including printf() with a "%f" format specifier, is likely to cause an
unimplemented instruction exception.
The FPSP works with the default variants of libc, libm and libgcc. It does not
work in conjunction with the msoft-float variants. The paranoia test goes into
an infinite loop at milestone 40. I am guessing that floor() is returning an
incorrect value.
The msoft-float variants of libc, libm and libgcc appear to do floating point
I/O properly. They only failed in paranoia. Offhand, I can't think of why they
would conflict with the FPSP, so I think that there is something wrong with the
msoft-float code. It might be my installation.
Given my experiences, I decided to install the FPSP in bsp_start(), and to link
against the default variants of libc, libm and libgcc. This causes the
executables to increase in size by about 60 KB. The README file and the
mvme167.cfg specify how to remove the FPSP, and how to link against the
msoft-float variants of the libraries. This is not what Eric Norum had done: on
my host, his gen68360_040 port links RTEMS code with the msoft-float variants
of libc and libm, and the default variant of libgcc. In this configuration, the
output of printf() with "%f" is garbage on my target.
is long but I hate to lose the information so I am including it here.
> I am still fixing and recompiling but this is the issue that was not the
> result of another patch. This is a fundamental build issue that I value
> your opinion on.
This is difficult issue (I.e. I have no destinct solution for it)
Background:
(gnu-) make's implicit rules apply CFLAGS, CPPFLAGS, CXXFLAGS, ASFLAGS and
LDFLAGS (cf. make.info/Implicit Rules/Catalogue of Rules), only.
In brief:
CPPFLAGS .. passed to the c-preprocessor
CFLAGS ... passed to the c-compiler
CXXFLAGS ... equivalent to CFLAGS but passed to the c++ compiler
(Attention: CFLAGS is not passed to the c++ compiler)
ASFLAGS .. equivalent to CFLAGS, but passed to the assembler
LDFLAGS .. equivalent to CFLAGS, but passed to the linker
A bit oversimplifying, these make rules are as follows
.c.o:
$(CC) $(CPPFLAGS) $(CFLAGS) -c
.cc.o:
$(CXX) $(CPPFLAGS) $(CXXFLAGS) -c
.S.s:
$(CPP) $(CPPFLAGS)
.s.o:
$(AS) $(ASFLAGS)
My reading of the documentation (make.info) is that {AS|AR|CC|CXX|CPP}FLAGS
are ment to be passed to the related tools directly, however examinating
the rule set of gmake (gmake -p -f /dev/null") shows that many rules use
$(CC) instead of the related tools (eg. linker rules) etc.
I.e. these flags should not rely on being passed through cpp or gcc. With
gcc being the common frontend for all of these tools of a gnu-toolchain the
situation becomes difficult (Which option is passed to whom and which tool
really uses it?), because these variable can also contain the toolchain's
frontend (eg. AS=gcc, LD=gcc, CPP=gcc -E).
For some commonly used options the situation is quite clear:
* -g -> CFLAGS
* -OX -> CFLAGS
* -D -> CPPFLAGS
* -A -> CPPFLAGS
But where to add -m, -B, -specs, -qrtems_XXX ?
* -B, -specs, -qrtems_XXX are gcc-frontend options
* -m is a combinations of flags to go to different destinations, in many
(all?) cases, the following is valid
-m is expanded by gcc into a set of -D and -A options
-m is interpreted by cc1 as a machine flag to generate a specific
instruction set.
-m is interpreted by gcc as an implicit linker search path for multilibs to
set up calls to LD.
>From my point of view this indicates we can either destingush between these
different usages (= separately add -m to CFLAGS, LDFLAGS etc) or to add it
to CPPFLAGS and use gcc (the frontend) instead of calling each tool
directly (less error prone) -- I vote for CPPFLAGS, but I am not sure.
-----------------
Now, where to add CPU_CFLAGS?
AFAIS, in probably all cases CPU_CFLAGS contain -D -A, and -m options,
only.
* -D and -A are supposed to go to CPPFLAGS
* -mXXX options can have multiple meanings (It can be gcc, collect2/ld and
cc1/cc1plus option simultaneously)
Here, I made a mistake - I destinguished between CPU_DEFINES to be added to
CPPFLAGS and CPU_CFLAGS to be added to CFLAGS and CXXFLAGS (cf.
gcc-target-default.cfg), generally assuming CPU_CFLAGS are CFLAGS.
This breaks preprocessing *.S into *.i files because CPU_CFLAGS flags were
not added to CPPFLAGS. Hence *all* *.S were compiled without taking
-mXX-flags into account. The i960/cvme BSP was the only one which
explicitly checked for a specific -m flag (-mca) and refused to compile
without it -- all other CPUs/BSPs silently swallowed this.
IMO, we can either
1) add CPU_CFLAGS and CPU_DEFINES to CPPFLAGS, thus silently convert
CPU_CFLAGS's meaning into CPU_DEFINES (Alternative solution: rename
CPU_CFLAGS to CPU_DEFINES and merge CPU_FLAGS with CPU_DEFINES).
or
2) destinguish between CPU_DEFINES and CPU_CFLAGS. In this case we would
need to check the contents of each CPU_CFLAGS in custom/*.cfg and move the
some parts of the contents to CPU_DEFINES and keep other parts in
CPU_CFLAGS (CFLAGS must contain options for the c/c++-compiler only!).
Though Solution 2) is the clearer one, I implemented 1) which is the
simplier one (the patch below).
ATTENTION: This patch is small in size, but affects almost everything.
------------
Additional complications araise with linking:
Some BSPs call LD and AS directly (esp. gcc-2.7 make-exe rules). If LD=gcc
then LDFLAGS are supposed to be gcc-options, but if LD=ld then LDFLAGS is
supposed to contain ld-options.
An analog thought is valid for AS, but luckily enough ASFLAGS is not used
of inside the whole source tree.
Most RTEMS' custom/*.cfg use $(CC) $(CFLAGS) to link with gcc-2.8 make-exe
rules. With the patch below (CPU_CFLAGS added to CPPFLAGS) this means
CPU_CFLAGS will not be passed to the linker, which is incorrect for
multilibbed CPU's.
gmake's default rule set contains a variety of rules for linking, all
ending up in calling $(CC) $(LDFLAGS) for linking at their very end.
IMO, this means we should use something like
LINK.o = $(CC) $(LDFLAGS) in gcc-target-default.cfg
+ modify all gcc-2.8 make-exe rules to use
$(LINK.o) .......
+ setup LDFLAGS according to the requirements of the above.
I.e. we should use $(CC) for linking instead of calling the linker (LD)
directly and set LDFLAGS = $(CPPFLAGS) $(CFLAGS) or similar.
