STAGE1_CFLAGS can be used to accelerate the just-built stage1 compiler
which especially improves its performance on some of the large generated
files during bootstrap. It defaults to nothing (i.e. -O0).
The downside is that if the native compiler is buggy, there's a greater
risk of a failed bootstrap. Those with a modern native compiler, ideally
a recent version of GCC, should be able to use -O1 or -O2 without issue
to get a faster build.
PR rtl-optimization/111619
* doc/install.texi (Building a native compiler): Discuss STAGE1_CFLAGS.
---
Sorry, I'd completely forgot about this and only noticed when trying to reduce
my local branch and get more things ready to submit..
This takes your suggestion as-is with some minor phrasing tweaks. I agree
with your take here and think this works better indeed.
OK?
gcc/doc/install.texi | 7 +++++++
1 file changed, 7 insertions(+)
diff --git a/gcc/doc/install.texi b/gcc/doc/install.texi
index fc771448274a..576b7eead5ec 100644
--- a/gcc/doc/install.texi
+++ b/gcc/doc/install.texi
@@ -3125,6 +3125,13 @@ Again, if the native compiler miscompiles the stage1
compiler, you may
need to work around this by avoiding non-working parts of the stage1
compiler. Use @code{STAGE1_TFLAGS} to this end.
+You can use @code{STAGE1_CFLAGS} to set the flags passed to the host compiler
+when building the stage1 compiler. The default is to pass @option{-g}, but
when
+the host compiler is GCC, this results in a non-optimized build of the stage1
+compiler. You can speed up the bootstrap by using @samp{STAGE1_CFLAGS='-O2'}
+at the increased risk of miscompiling the stage1 compiler when the host
+compiler is buggy.
+
If you used the flag @option{--enable-languages=@dots{}} to restrict
the compilers to be built, only those you've actually enabled will be
built. This will of course only build those runtime libraries, for
--
2.51.0
