This is the mail archive of the gcc@gcc.gnu.org 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]

Re: Major bootstrap time regression on March 30


On Thu, Apr 07, 2005 at 03:08:50PM -0700, James E Wilson wrote:

> Is it possible that something changed on Diego's machine?  Such as a new 
> binutils, or a new kernel, etc?
> 
Yes.  I rebooted the machine into a new kernel on 2005-03-31
(2.6.10-1.770_FC3).  The slowdown coincided with the box being
rebooted into the new kernel:

Date		Bootstrap time (secs)

2005-03-29		5,663
2005-03-30		5,704	<-- 2.6.10-1.760_FC3
2005-03-31		7,026	<-- 2.6.10-1.770_FC3
2005-04-01		7,768
[ ... ]
2005-04-06		8,048
2005-04-07		8,197

Bootstrap times have been slowly creeping up since then.

However, build times for SPEC haven't changed much.  It could be
that the new kernel is not too good with heavy I/O.  I've got
binutils-2.15.92.0.2-5.  Dunno if that has been recently
upgraded.

I'm rebooting the machine into the previous kernel right now to
see if it changes things.  Tomorrow's run will use kernel
2.6.10-1.760_FC3.

Another thing, has our library code base (libjava, libstdc++)
grown significantly lately?


Diego.


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