Re: number of spills and reloads

[James E Wilson <wilson at specifixinc dot com>]

Qiong Cai wrote:
> I'd like to know the number of spills and reloads generated by
> register allocator in GCC.

You need to define exactly what you mean by "spills" and "reloads" 
before those questions can be answered.  And you may not be able to 
define those terms exactly if you don't know how reload works.

> So I add the counter "num_spills" in function "find_reload_regs" and the counter
> "num_reloads" in "find_reg" in "reload1.c" just after the instructions
> printing out the above information.  Is this correct?  Because the
> reload phase is complicated, I'm quite suspicious what I did.

find_reload_regs gets called once for every instruction that needs a 
reload.  find_reg gets called once for every operand of every 
instruction that needs a spill register allocated to satisfy a reload. 
But we only reach the place where we emit the message if we choose to 
allocate a new spill register.

Counting the instructions that need a reload is a very poor measure of 
the number of spills.  There is n_spills which counts the number of hard 
registers allocated for use as spill registers which is one reasonable 
measure.  n_spills is probably very close to your num_reloads.  The 
different is probably eliminable registers like the frame pointer. 
Another measure might be the number of pseudos that get forced to stack 
slots in order to free up hard registers for use as spill registers. 
This can be computed by counting the number of registers in spilled_pseudos.

For the number of reloads, there is n_reloads computed by find_reloads. 
 This is a per instruction count.  You would have to sum it over all 
instructions to get a count for the entire function.  But there are a 
number of ambiguities here.

We sometimes create what is known as "optional" reloads.  These are 
things that don't need a reload, but might result in better code if they 
do get a reload.  Sometimes we perform them.  Sometimes we don't.  You 
may or not want to count them.

We also have secondary reloads.  Sometimes a single reload needs two 
registers, which we handle by creating two reloads, one of which is a 
secondary reload.  So is that 1 reload or 2?  Depends on how you want to 
count them.

Some reloads need a spill register, some don't.  If a reload can never 
result in a spill, do you still want to count it?

Some reloads involve multi-word values which may require multiple spill 
registers.

Etc.

Maybe what you really want is the number of reloads that require use of 
a spill register, which I think you could get by putting a counter at 
the top of find_reg.
--
Jim Wilson, GNU Tools Support, http://www.SpecifixInc.com