This is the mail archive of the mailing list for the GCC project.

Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

[RFC] Thoughts on reordering the source tree


The GCC source tree is getting really big.  We currently have in gcc/:

- 337 .c files
- 171 .h files

Personally, I think the source tree is quite a mess, the way it is
now.  A long time ago (I can't find the threads involved) there was
some discussion about re-ordering the source tree a bit.

Now that GCC uses subversion, we can move files around without
destroying the file revision history, right? And we are in stage1, the
perfect time for Big Changes like re-ordering the source tree.

Thoughts on what the source tree could look like (crude, eventual
re-ordering is of course a pre-file decision):
- c-* go to gcc/c (including c-common.*, 29 files)
- *.h go to gcc/include
- ipa-* go to gcc/ipa-opt/ipa
- tree-* go to gcc/gimple-opt/
- RTL optimizations (fwprop, see, gcse, loop-*, etc.) go to gcc/rtl-opt/
- CFG related stuff (cfg*, dominance.c, etc.) go to gcc/cfg
- plugin* go to gcc/plugin/
- ira*, sched-*, reload*, final.c, etc. (low level stuff) goes to
gcc/cg (code generation)
- driver goes to gcc/driver
- basic intermediate language support stays in gcc/ or goes to gcc/ir,
or gcc/tree & gcc/rtl, etc.

Thoughts&comments?  Are there any basic problems/reasons that make
this kind of change impossible/undesirable at this time?


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]