Low overhead looping problem

[Saira Viqar]

We are porting GCC to our own target. We have encountered a problem during
the optimization phase. It is not able to match the
"decrement_and_branch_until_zero" pattern though we have given the rtl
pattern specified in the manual-- "GNU compiler internals" , topic 8.12 .For
further reference please see the attatched md file.

 <<avaz.h>>  <<avaz.md>> 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: avaz.h
Type: application/octet-stream
Size: 36216 bytes
Desc: not available
URL: <https://gcc.gnu.org/pipermail/gcc/attachments/20020312/9ee5fa72/attachment.obj>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: avaz.md
Type: application/octet-stream
Size: 8188 bytes
Desc: not available
URL: <https://gcc.gnu.org/pipermail/gcc/attachments/20020312/9ee5fa72/attachment-0001.obj>