[Bug c/32447] New: without-decimal-float needed

[Bug c/32447] New: without-decimal-float needed

[malitzke at metronets dot com]

Recent quip: "I do not care that its legal; not in my back-yard" NYMBY.

If you want to complain about not disabling the internal part for dfp, please
submit a new bug but note it will be marked as an enchement because it is
actually hard to do (I already looked into doing it after you and someone else
noticed it).

I find all these -I references to various parts of libdecnumber as an invasion
of my back yard:

/var/tmp/gcc_r43/build-86/./prev-gcc/xgcc
-B/var/tmp/gcc_r43/build-86/./prev-gcc/ -B/usr/i686-pc-linux-gnu/bin/ -c   -O2
-pipe -DIN_GCC   -W -Wall -Wwrite-strings -Wstrict-prototypes
-Wmissing-prototypes -Wold-style-definition -Wmissing-format-attribute
-pedantic -Wno-long-long -Wno-variadic-macros                                 
-Wno-overlength-strings    -DHAVE_CONFIG_H -DGENERATOR_FILE -I. -Ibuild
-I../../gcc-4.3.0/gcc -I../../gcc-4.3.0/gcc/build
-I../../gcc-4.3.0/gcc/../include -I../../gcc-4.3.0/gcc/../libcpp/include 
-I../../gcc-4.3.0/gcc/../libdecnumber -I../../gcc-4.3.0/gcc/../libdecnumber/dpd
-I../libdecnumber    -o build/genemit.o ../../gcc-4.3.0/gcc/genemit.c
/var/tmp/gcc_r43/build-86/./prev-gcc/xgcc
-B/var/tmp/gcc_r43/build-86/./prev-gcc/ -B/usr/i686-pc-linux-gnu/bin/ -c   -O2
-pipe -DIN_GCC   -W -Wall -Wwrite-strings -Wstrict-prototypes
-Wmissing-prototypes -Wold-style-definition -Wmissing-format-attribute
-pedantic -Wno-long-long -Wno-variadic-macros                                 
-Wno-overlength-strings    -DHAVE_CONFIG_H -DGENERATOR_FILE -I. -Ibuild
-I../../gcc-4.3.0/gcc -I../../gcc-4.3.0/gcc/build
-I../../gcc-4.3.0/gcc/../include -I../../gcc-4.3.0/gcc/../libcpp/include 
-I../../gcc-4.3.0/gcc/../libdecnumber -I../../gcc-4.3.0/gcc/../libdecnumber/dpd
-I../libdecnumber    -o build/gencodes.o ../../gcc-4.3.0/gcc/gencodes.c
/var/tmp/gcc_r43/build-86/./prev-gcc/xgcc
-B/var/tmp/gcc_r43/build-86/./prev-gcc/ -B/usr/i686-pc-linux-gnu/bin/ -c   -O2
-pipe -DIN_GCC   -W -Wall -Wwrite-strings -Wstrict-prototypes
-Wmissing-prototypes -Wold-style-definition -Wmissing-format-attribute
-pedantic -Wno-long-long -Wno-variadic-macros                                 
-Wno-overlength-strings    -DHAVE_CONFIG_H -DGENERATOR_FILE -I. -Ibuild
-I../../gcc-4.3.0/gcc -I../../gcc-4.3.0/gcc/build
-I../../gcc-4.3.0/gcc/../include -I../../gcc-4.3.0/gcc/../libcpp/include 
-I../../gcc-4.3.0/gcc/../libdecnumber -I../../gcc-4.3.0/gcc/../libdecnumber/dpd
-I../libdecnumber    -o build/genextract.o ../../gcc-4.3.0/gcc/genextract.c
/var/tmp/gcc_r43/build-86/./prev-gcc/xgcc
-B/var/tmp/gcc_r43/build-86/./prev-gcc/ -B/usr/i686-pc-linux-gnu/bin/ -c   -O2
-pipe -DIN_GCC   -W -Wall -Wwrite-strings -Wstrict-prototypes
-Wmissing-prototypes -Wold-style-definition -Wmissing-format-attribute
-pedantic -Wno-long-long -Wno-variadic-macros                                 
-Wno-overlength-strings    -DHAVE_CONFIG_H -DGENERATOR_FILE -I. -Ibuild
-I../../gcc-4.3.0/gcc -I../../gcc-4.3.0/gcc/build
-I../../gcc-4.3.0/gcc/../include -I../../gcc-4.3.0/gcc/../libcpp/include 
-I../../gcc-4.3.0/gcc/../libdecnumber -I../../gcc-4.3.0/gcc/../libdecnumber/dpd
-I../libdecnumber    -o build/genopinit.o ../../gcc-4.3.0/gcc/genopinit.c
/var/tmp/gcc_r43/build-86/./prev-gcc/xgcc
-B/var/tmp/gcc_r43/build-86/./prev-gcc/ -B/usr/i686-pc-linux-gnu/bin/ -c   -O2
-pipe -DIN_GCC   -W -Wall -Wwrite-strings -Wstrict-prototypes
-Wmissing-prototypes -Wold-style-definition -Wmissing-format-attribute
-pedantic -Wno-long-long -Wno-variadic-macros                                 
-Wno-overlength-strings    -DHAVE_CONFIG_H -DGENERATOR_FILE -I. -Ibuild
-I../../gcc-4.3.0/gcc -I../../gcc-4.3.0/gcc/build
-I../../gcc-4.3.0/gcc/../include -I../../gcc-4.3.0/gcc/../libcpp/include 
-I../../gcc-4.3.0/gcc/../libdecnumber -I../../gcc-4.3.0/gcc/../libdecnumber/dpd
-I../libdecnumber    -o build/genoutput.o ../../gcc-4.3.0/gcc/genoutput.c
/var/tmp/gcc_r43/build-86/./prev-gcc/xgcc
-B/var/tmp/gcc_r43/build-86/./prev-gcc/ -B/usr/i686-pc-linux-gnu/bin/ -c   -O2
-pipe -DIN_GCC   -W -Wall -Wwrite-strings -Wstrict-prototypes
-Wmissing-prototypes -Wold-style-definition -Wmissing-format-attribute
-pedantic -Wno-long-long -Wno-variadic-macros                                 
-Wno-overlength-strings    -DHAVE_CONFIG_H -DGENERATOR_FILE -I. -Ibuild
-I../../gcc-4.3.0/gcc -I../../gcc-4.3.0/gcc/build
-I../../gcc-4.3.0/gcc/../include -I../../gcc-4.3.0/gcc/../libcpp/include 
-I../../gcc-4.3.0/gcc/../libdecnumber -I../../gcc-4.3.0/gcc/../libdecnumber/dpd
-I../libdecnumber    -o build/genpeep.o ../../gcc-4.3.0/gcc/genpeep.c
/var/tmp/gcc_r43/build-86/./prev-gcc/xgcc
-B/var/tmp/gcc_r43/build-86/./prev-gcc/ -B/usr/i686-pc-linux-gnu/bin/ -c   -O2
-pipe -DIN_GCC   -W -Wall -Wwrite-strings -Wstrict-prototypes
-Wmissing-prototypes -Wold-style-definition -Wmissing-format-attribute
-pedantic -Wno-long-long -Wno-variadic-macros                                 
-Wno-overlength-strings    -DHAVE_CONFIG_H -DGENERATOR_FILE -I. -Ibuild
-I../../gcc-4.3.0/gcc -I../../gcc-4.3.0/gcc/build
-I../../gcc-4.3.0/gcc/../include -I../../gcc-4.3.0/gcc/../libcpp/include 
-I../../gcc-4.3.0/gcc/../libdecnumber -I../../gcc-4.3.0/gcc/../libdecnumber/dpd
-I../libdecnumber    -o build/genrecog.o ../../gcc-4.3.0/gcc/genrecog.c

I asked for a quote of that new IBM P 570 computer to put this matter into
perspective.

In my opinion decimal float should have been handled like libiconv (GNU charset
conversion library for libc which doesn't implement it).

Further info in PR32314.

bugzilla will not accept bje@au.ibm.com.


-- 
           Summary: without-decimal-float needed
           Product: gcc
           Version: 4.3.0
            Status: UNCONFIRMED
          Severity: enhancement
          Priority: P3
         Component: c
        AssignedTo: unassigned at gcc dot gnu dot org
        ReportedBy: malitzke at metronets dot com
GCC target triplet: any architecture


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=32447

[Bug c/32447] without-decimal-float needed

[malitzke at metronets dot com]

------- Comment #1 from malitzke at metronets dot com  2007-06-21 08:08 -------
Disclosure:

I am not an IBM hater and never was. My first significant program ran on an IBM
1410 with a Tape Operating System (TOS) on the day President Kennedy was
assassinated. I then used an IBM 1620 II extensively.; migrated to an IBM 1800
and to an IBM 1130. All worked well for these early systems, but, I had to open
the door fairly often late at night for an IBM customer service man. At another
firm I worked with the then largest and newest IBM 360's. Overall, I had twice
teams come from the Almaden Labs to resolve issues with DASD's and the FORTRAN
H compiler (early seventies).

On the other hand I really got to hate PL/I, as I could interpret ABEND dumps
and  colleagues came to my office with 1-2 feet high dumps that mostly by the
PL/I compiler going off on a tangent. This interfered with my work and I turned
to satellite telecommunications. 

I only returned to programming as a hobby and as a retirement activity. 


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=32447

[Bug middle-end/32447] without-decimal-float needed

[pinskia at gcc dot gnu dot org]

------- Comment #2 from pinskia at gcc dot gnu dot org  2007-06-21 09:11 -------
You are misrepresenting the problem again.  So the issue is that the size of
cc1/cc1plus is increased with the decimal float back-end support.  Now
Disabling this support in general is a hard problem.  I would dare you to do
#ifdef SUPPORT_DFP around all the uses and see if that works.  I doubt it
because the DFP is now embedded inside GCC sources.  This is true of vector
support.


-- 

pinskia at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
          Component|c                           |middle-end


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=32447

[Bug middle-end/32447] without-decimal-float needed

[malitzke at metronets dot com]

------- Comment #3 from malitzke at metronets dot com  2007-06-21 14:33 -------
Thanks for helping out again. Enjoy Japan. I was there quite often, dealing
with NEC and Mitsubishi, but as a buyer representative for for multi-million $
projects. At that level it was pleasure to do business, even enduring sitting
cross-legged on the floor; washing down raw fish with japanese whiskey on the
rocks. But to matters at hand.

What is called a compiler is actually a crime against traditional usage of the
term.
"compilare to plunder; 1: to collect into a volume 2: to compose out of
materials from other documents" this more accurately describes the function of
a traditional loader. A, so-called, compiler is really a translator. In the
case of GCC the Italian saying "tradutore; traditore" translator; traitor seems
appropriate. I know how hard it is to do technical translation, but that was my
first well-paid career at about 15 years of age.

My third career was being a real-time assembly language programmer. I received
a letter of commendation as a super-programmer for committing the following
programming crimes 1) jumping into the middle of instructions; 2) writing
self-modifying code. You do what you have to do to put food on the table. I
actually jeopardized a three month engagement by showing with the help of
simple queuing theory that the project (in-spite of the above tricks) was
doomed to fail. The COBOL manager assured me that I was doing a marvelous job
squeezing code into 128 byte overlays and not to worry about over-all design.
They bought back that contract.

In my fourth career I was termed by the company president to-be "our secret
weapon" for writing specifications, contracts, and inter-national standards
that with-stood the test of legality but were blatantly uni-sided in favor of
my employer (very good money).

Now, being retired and considering the C and FORTRAN compilers as quite
important tools in my UNIX tool-chest; I am trying to prevent GCC being
destroyed by mis-interpretation and mis-use of standards legal shenanigans,
etc.
My loyalty is to "my" tools and not to the people involved with GCC or even GCC
as now interpreted. See PR32347.

Incidentally, two quotes from K&R "Again because the language reflects the
capabilities of current computers, C programs tend to be efficient enough that
there is no compulsion to write assembly language instead" (Intro 1st ed).
'(ANSI) established a committee whose goal was to produce "and unambiguous and
machine-independent definition of the language C" while still _retaining_ its
_spirit_'. Preface 2nd Ed, emphasis added. Ritchie got a "Turing Award" and the
current crop of people should respect the creator's wishes. If they want to
make changes against that original design and call it something else; just as
Ritchie acknowledges BCPL.  


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=32447

[Bug middle-end/32447] without-decimal-float needed

[manu at gcc dot gnu dot org]

------- Comment #4 from manu at gcc dot gnu dot org  2010-02-21 00:02 -------
This is not going to happen because no one working on GCC is interested on
pursuing it. Your only chance is 

1) to do it yourself to show GCC devs that can be done and it is beneficial.
2) to pay someone to do it

Both of those options have the risk that GCC devs may not be convinced and
decide to not apply your patch.

Sorry, it is nothing personal, just explaining what the real situation is.


-- 

manu at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |manu at gcc dot gnu dot org
             Status|UNCONFIRMED                 |RESOLVED
         Resolution|                            |INVALID


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=32447