Some svn numbers

[Mike Stump]

On Oct 11, 2005, at 7:42 PM, Daniel Berlin wrote:
> So, Mike Stump said his update times with svn from toolchain.org  
> were a
> little slow.

> Let me know if it's still slow for you.

Ok, had a chance to upgrade my svn to 1.2.3, and time from gcc.gnu.org:

$ cd gcc
$ time svn update *.c
[ lots o output that is annoying ]
real    3m28.134s
user    0m7.665s
sys     0m5.403s

 From top:
Networks:     16981 ipkts/2674K          10405 opkts /1370K
Disks:          106 reads/781K             345 writes/4755K

$ time cvs update *.c
real    0m1.329s
user    0m0.452s
sys     0m0.076s

 From top:
Networks:       160 ipkts/43K               54 opkts /11K
Disks:            0 reads/0K                 0 writes/0K
VM:               0 pageins                  0 pageouts

svn is only 156x slower than cvs.  :-(  It used to be in the 500x  
slower range (or so) with 1.1.3, so, great progress, but...  62x  
worse on the network reads, 124x worse on network writes, infinitely  
worse on the disk reads and writes.

Yes, I really do run updates this way on occasion, usually during the  
last stages of a targeted reghunt, but, not often.  I don't see this  
as a transition issue, just, my fingers will fat finger it from time  
to time and they are gonna have to learn to not do that anymore.

Now, just so people understand the normal update command, here is  
what I see:

mrs $ time cvs update
real    0m16.373s
user    0m0.715s
sys     0m2.533s

 From top:
Networks:       689 ipkts/133K             191 opkts /151K
Disks:         3189 reads/13252K           834 writes/4130K

mrs $ time svn update
real    0m4.833s
user    0m0.577s
sys     0m0.778s
 From top:
Networks:       176 ipkts/20K               35 opkts /7K
Disks:            6 reads/48K              864 writes/12894K

All operations were hot nops.