Anyone else sees a bootstrap failure (Linux-x86)?

Mon Oct 20 19:03:00 GMT 2003

Ranjit Mathew writes:
 > Andrew Haley wrote:
 > 
 > > > "-pipe" is useful and eliminates a lot of temporary files creation,
 > > > somewhat speeding up the build.
 > > 
 > > Not on any sane platform. Really, have you ever seen a significant
 > > speedup? Or is file creation so horribly slow on Windows that this is
 > > a real issue?
 > 
 > >From the GCC manual:
 > 
 > ----------------------------- 8< -----------------------------
 > -pipe
 > Use pipes rather than temporary files for communication 
 > between the various stages of compilation. This fails to 
 > work on some systems where the assembler is unable to read from 
 > a pipe; but the GNU assembler has no trouble.
 > ----------------------------- 8< -----------------------------
 > 
 > It *looks like* "-pipe" should be at least as fast, if not
 > faster, than using temporary files. However, I have never
 > really properly compared the two alternatives.
If -pipe was such a good idea it would be the default. It isn't.
 > This choice was based more on a gut feeling and a large amount of
 > anecdotal evidence (mainly, Slashdot comments - the ones on the
 > *old* Slashdot, not the weirdness that currently carries the same
 > name).
 > 
 > By "sane", do you mean a system that does a healthy amount of
 > file-system caching to avoid unnecessary writes to the disk, making
 > using temporary files almost as fast as pipes?
Faster. No interlocks, you see. 
Write the whole file, then read it. That's much less work than all
the management of pipes.
 > BTW, I build only on x86-Linux now.
Then -pipe is slowing you down.
Andrew.