Date: Sun, 10 Aug 2003 14:11:18 -0400 From: Adam McDougall <mcdouga9@egr.msu.edu> To: David O'Brien <obrien@freebsd.org> Cc: sparc64@freebsd.org Subject: Re: ponderous 'make world' times post GCC 3.3... Message-ID: <20030810181117.GG85229@egr.msu.edu> In-Reply-To: <20030807062536.GA68747@dragon.nuxi.com> References: <20030807062536.GA68747@dragon.nuxi.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On Wed, Aug 06, 2003 at 11:25:36PM -0700, David O'Brien wrote:
Am I the only one that saw 'make world' go from almost 3 hours with GCC
3.2 to:
--------------------------------------------------------------
>>> make world completed on Wed Aug 6 20:49:47 PDT 2003
(started Wed Aug 6 09:49:30 PDT 2003)
--------------------------------------------------------------
11h17.00s real 9h29m42.85s user 1h15m22.05s sys
post GCC 3.3? This is a 500mhz Blade 100 with a GENERIC minus WITNESS*
kernel.
--
-- David (obrien@FreeBSD.org)
I have been rebuilding my dual cpu Ultra 60 occasionally, including
several times after the gcc upgrade, and have not noticed a large
slowdown. Here are some datapoints, maybe it will help discover
where the problem lies if someone can try at least a similar
build configuration. Notables in config and build are NOPROFILE=yes,
no DEBUG=-g, no INVARIENTS*, no WITNESS*, SCHED_4BSD (sometimes
I use ULE, but not in the past month), SMP, /etc/malloc.conf -> aj,
make -j2 MAKE="make -j2" buildworld. I always cvsup before building
if it is a new day. I can return more information or tests if anyone
wants.
Infact, I occasionally compile the samba port as a dirty benchmark,
and the compile time dropped from ~9.5 minutes to 8 minutes between
may and july 18.
bw1.log:Script started on Sun Jun 8 17:16:52 2003
bw1.log:Script done on Sun Jun 8 19:36:52 2003
bw2.log:Script started on Wed Jun 11 22:51:51 2003
bw2.log:Script done on Thu Jun 12 00:54:38 2003
bw3.log:Script started on Fri Jun 13 18:26:45 2003
bw3.log:Script done on Fri Jun 13 20:18:21 2003
bw4.log:Script started on Tue Jun 17 18:45:08 2003
bw4.log:Script done on Tue Jun 17 20:34:12 2003
bw5.log:Script started on Wed Jun 18 17:44:28 2003
bw5.log:Script done on Wed Jun 18 19:33:28 2003
bw6.log:Script started on Tue Jun 24 18:28:54 2003
bw6.log:Script done on Tue Jun 24 20:17:47 2003
bw7.log:Script started on Tue Jul 1 18:25:01 2003
bw7.log:Script done on Tue Jul 1 20:57:22 2003
bw8.log:Script started on Wed Jul 2 19:59:14 2003
bw8.log:Script done on Wed Jul 2 22:46:29 2003
bw9.log:Script started on Fri Jul 4 10:15:08 2003
bw9.log:Script done on Fri Jul 4 13:04:01 2003
bw10.log:Script started on Wed Jul 16 19:24:34 2003
bw10.log:Script done on Wed Jul 16 21:02:10 2003
bw11.log:Script started on Thu Jul 17 17:23:07 2003
bw11.log:Script done on Thu Jul 17 19:00:30 2003
bw12.log:Script started on Thu Jul 17 19:35:37 2003
bw12.log:Script done on Thu Jul 17 21:13:46 2003
bw13.log:Script started on Thu Jul 17 23:53:55 2003
bw13.log:Script done on Fri Jul 18 03:10:21 2003
bw14.log:Script started on Sun Jul 27 12:03:38 2003
bw14.log:Script done on Sun Jul 27 14:03:42 2003
bw15.log:Script started on Sun Aug 10 10:53:46 2003
bw15.log:Script done on Sun Aug 10 12:53:04 2003
build options from make.conf:
CFLAGS=-pipe -O
WITH_LIBMAP=YES
NOPROFILE=yes
Non-commented kernel config options:
machine sparc64
cpu SUN4U
ident U60
maxusers 0
options OFW_NEWPCI
options SCHED_4BSD #4BSD scheduler
options INET #InterNETworking
options FFS #Berkeley Fast Filesystem
options SOFTUPDATES #Enable FFS soft updates support
options UFS_ACL #Support for access control lists
options UFS_DIRHASH #Improve performance on big directories
options NFSCLIENT #Network Filesystem Client
options NFSSERVER #Network Filesystem Server
options CD9660 #ISO 9660 Filesystem
options PROCFS #Process filesystem (requires PSEUDOFS)
options PSEUDOFS #Pseudo-filesystem framework
options COMPAT_43 #Compatible with BSD 4.3 [KEEP THIS!]
options COMPAT_FREEBSD4 #Keep this for a while
options SCSI_DELAY=3000 #Delay (in ms) before probing SCSI
options KTRACE #ktrace(1) syscall trace support
options SYSVSHM #SYSV-style shared memory
options SYSVMSG #SYSV-style message queues
options SYSVSEM #SYSV-style semaphores
options _KPOSIX_PRIORITY_SCHEDULING #Posix P1003_1B real-time extensions
options SMP # Symmetric MultiProcessor Kernel
device apb # Sun APB PCI-PCI bridge
device ebus
device isa
device pci
device sbus
device ata
device atadisk # ATA disk drives
device atapicd # ATAPI CDROM drives
device ahc # AHA2940 and onboard AIC7xxx devices
device isp # Qlogic family
device ispfw # Firmware module for Qlogic host adapters
device sym # NCR/Symbios Logic (newer chipsets + those of `ncr')
device scbus # SCSI bus (required)
device ch # SCSI media changers
device da # Direct Access (disks)
device sa # Sequential Access (tape etc)
device cd # CD
device pass # Passthrough device (direct SCSI access)
device ses # SCSI Environmental Services (and SAF-TE)
device ofw_console # OpenBoot firmware console device
device genclock # Generic clock interface
device eeprom # eeprom (really an ebus driver for the MK48Txx)
device "mk48txx" # Mostek MK48T02, MK48T08, MK48T59 clock
device sab # Siemens SAB82532 based serial ports
device miibus # MII bus support
device gem # Sun GEM/Sun ERI/Apple GMAC
device hme # Sun HME (Happy Meal Ethernet)
device rl # RealTek 8129/8139
device random # Entropy device
device loop # Network loopback
device ether # Ethernet support
device pty # Pseudo-ttys (telnet etc)
device bpf #Berkeley packet filter
options OFW_PCI_DEBUG
Hardware:
real memory = 805306368 (768 MB)
cpu0: Sun Microsystems UltraSparc-II Processor (360.03 MHz CPU)
cpu1: Sun Microsystems UltraSparc-II Processor (360.03 MHz CPU)
da0: <FUJITSU MXF3364LC 0319> Fixed Direct Access SCSI-2 device
da0: 40.000MB/s transfers (20.000MHz, offset 16, 16bit), Tagged Queueing Enabled
da0: 34732MB (71132960 512 byte sectors: 255H 63S/T 4427C)
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20030810181117.GG85229>