Re: -pipe switch in kernel compilation

2000-01-11 Thread George Cox

On 11/01 07:39, Andreas Klemm wrote:

 Where 4 MB isn't sufficient anymore with a GENERIC kernel.  You need at least
 6 MB or so to boot, then compile a custom kernel and then, if you are lucky,
 can perhaps run with 4 MB.

Here are the two constituent process of a compilation spotted earlier today:

  PID USERNAME PRI NICE  SIZERES STATE  C   TIME   WCPUCPU COMMAND  
 1897 root  31   0  4224K  4096K STOP   1   0:04 13.32%  7.91% cc1  
 1898 root  -6   0  1616K  1216K STOP   0   0:00  0.00%  0.00% as   

So it looks like a round 6M is required for a '-pipe' enabled compilation.
Thank heavens for lower memory prices! :-)


 But that information is about 1-2 years old, don't know, if we perhaps
 already need 6-8 MB nowadays...  Though it's zillions better than M$ crap.

Absolutely.  My computer is a free software-only zone. :-)

best;


gjvc

-- 
[gjvc]
4.4BSD 4.ever!


To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message



Re: -pipe switch in kernel compilation

2000-01-11 Thread Wilko Bulte

On Tue, Jan 11, 2000 at 07:39:40AM +0100, Andreas Klemm wrote:
 On Mon, Jan 10, 2000 at 04:55:43PM +, George Cox wrote:
  G'day,
  
  While compiling a kernel today, I noticed that the '-pipe' option to gcc
  was not being used.  Is there any reason for this?
 
 I think this is the (historical) default, so that people with
 only 4-8 MB of RAM don't get into trouble ;-)
 
 Where 4 MB isn't sufficient anymore with a GENERIC kernel.
 You need at least 6 MB or so to boot, then compile a custom
 kernel and then, if you are lucky, can perhaps run with 4 MB.

FWIW: 3.3R ran (crawled?) in 4Mb. I tried it 2 months ago on a 386SX40 with
4Mb.

Compiling a GENERIC kernel was 5 hours or so ;-) That is when I gave up
on my idea to 'make buildworld'.

But still impressive, it was stable.

Wilko
-- 
Wilko Bulte Arnhem, The Netherlands   - The FreeBSD Project 
WWW : http://www.tcja.nl  http://www.freebsd.org


To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message



Re: -pipe switch in kernel compilation

2000-01-11 Thread Warner Losh

In message [EMAIL PROTECTED] Doug 
Russell writes:
: See, I knew there was a reason I hung on to all these 1M 30 pin SIMMs. :)
: Old 386/40s sure make nice little router/modem/whatever boxes.  :)

Until their hard disks go south :-(.  The biggest problems I have with 
them is that they also tend to dislike newer ATA disks.

Then again, I have 512k 72pin SIMMS around for when I want to try to
boot the kernel in 2MB of memory.  Not that I want to do this very
often...

Warner


To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message



Re: -pipe switch in kernel compilation

2000-01-11 Thread Brooks Davis

On Tue, Jan 11, 2000 at 01:53:10PM -0700, Warner Losh wrote:
 In message [EMAIL PROTECTED] Doug 
Russell writes:
 : See, I knew there was a reason I hung on to all these 1M 30 pin SIMMs. :)
 : Old 386/40s sure make nice little router/modem/whatever boxes.  :)
 
 Until their hard disks go south :-(.  The biggest problems I have with 
 them is that they also tend to dislike newer ATA disks.

I think that's why IBM has jumpers on some of their disks that limit
them to 2GB.  I can't see why else you would want to take a perfectly
good 8GB+ disk and use it as a 2GB drive.  Of course, some of them may
not even tolerate that much space.

-- Brooks

-- 
"Those who desire to give up freedom in order to gain security, will not
have, nor do they deserve, either one"
   --Thomas Jefferson 


To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message



Re: -pipe switch in kernel compilation

2000-01-11 Thread Warner Losh

In message [EMAIL PROTECTED] Brooks Davis writes:
: On Tue, Jan 11, 2000 at 01:53:10PM -0700, Warner Losh wrote:
:  In message [EMAIL PROTECTED] Doug 
:Russell writes:
:  : See, I knew there was a reason I hung on to all these 1M 30 pin SIMMs. :)
:  : Old 386/40s sure make nice little router/modem/whatever boxes.  :)
:  
:  Until their hard disks go south :-(.  The biggest problems I have with 
:  them is that they also tend to dislike newer ATA disks.
: 
: I think that's why IBM has jumpers on some of their disks that limit
: them to 2GB.  I can't see why else you would want to take a perfectly
: good 8GB+ disk and use it as a 2GB drive.  Of course, some of them may
: not even tolerate that much space.

I've retired perfectly good systems that can't handle having a 600MB
disk connected to them.  Anything over 540MB seems to give it fits,
although I suppose that a BIOS upgrade would help.  I couldn't even
lie to the BIOS and tell it that the drive's geometry was such that
only 540MB would be used.

Warner


To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message



Re: -pipe switch in kernel compilation

2000-01-11 Thread Peter Jeremy

On 2000-Jan-12 07:58:12 +1100, Brooks Davis [EMAIL PROTECTED] wrote:
I think that's why IBM has jumpers on some of their disks that limit
them to 2GB.  I can't see why else you would want to take a perfectly
good 8GB+ disk and use it as a 2GB drive.

The volumes probably mean it's not cost-effective to either continue
to produce smaller older technology drives, or design a new small
drive.

  Of course, some of them may not even tolerate that much space.

AFAIK, the BIOS in my Toshiba T1850 won't allow anything bigger than
130MB.  Getting FreeBSD into my current 80MB isn't fun (and the major
reason it hasn't been updated from 2.2.5 - I can't cut 3.x down far
enough to fit, and -current is even larger).

Peter


To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message



Re: -pipe switch in kernel compilation

2000-01-11 Thread Warner Losh

Before you ask, I've also tried to use the tapr[*] CF - IDE adapter to
see if I could get these systems to boot off a 16MB CF card with no
luck.  I even have a mini486 based system from NEC that I bought
surplus that I thought could use the CF card, but no joy.  Works great 
on all the modern machines I've tried it on.  likely a BIOS issue, but 
I don't wanna write bios for a machine that only cost me $49.99 that
has no docs.

Warner

[*] http://www.tapr.org/, as featured on slashdot a long time ago.


To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message



Re: -pipe switch in kernel compilation

2000-01-11 Thread Wilko Bulte

On Tue, Jan 11, 2000 at 01:32:49PM -0700, Doug Russell wrote:
 
 On Tue, 11 Jan 2000, Wilko Bulte wrote:
 
  FWIW: 3.3R ran (crawled?) in 4Mb. I tried it 2 months ago on a 386SX40 with
  4Mb.
  
  Compiling a GENERIC kernel was 5 hours or so ;-) That is when I gave up
  on my idea to 'make buildworld'.
  
  But still impressive, it was stable.
 
 See, I knew there was a reason I hung on to all these 1M 30 pin SIMMs. :)
 Old 386/40s sure make nice little router/modem/whatever boxes.  :)

I guess with Picobsd it would run better. People wanted to can the board,
brandnew in the wrapper. That is when I rescued it (for no good reason other
than curiosity what FreeBSD would think of it).

[followup to -chat]

-- 
Wilko Bulte Arnhem, The Netherlands   - The FreeBSD Project 
WWW : http://www.tcja.nl  http://www.freebsd.org


To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message



Re: -pipe switch in kernel compilation

2000-01-11 Thread Bernd Walter

On Tue, Jan 11, 2000 at 01:32:49PM -0700, Doug Russell wrote:
 
 On Tue, 11 Jan 2000, Wilko Bulte wrote:
 
  FWIW: 3.3R ran (crawled?) in 4Mb. I tried it 2 months ago on a 386SX40 with
  4Mb.
  
  Compiling a GENERIC kernel was 5 hours or so ;-) That is when I gave up
  on my idea to 'make buildworld'.
  
  But still impressive, it was stable.
 
 See, I knew there was a reason I hung on to all these 1M 30 pin SIMMs. :)
 Old 386/40s sure make nice little router/modem/whatever boxes.  :)
 
Yes and a recent current still works :)
Including the IPv6 parts that are already in current.

cicely4# dmesg
Copyright (c) 1992-2000 The FreeBSD Project.
Copyright (c) 1982, 1986, 1989, 1991, 1993
The Regents of the University of California. All rights reserved.
FreeBSD 4.0-CURRENT #0: Sun Jan  9 16:48:52 CET 2000
[EMAIL PROTECTED]:/var/d0/src-2000-01-07/src/sys/compile/CICELY4
Timecounter "i8254"  frequency 1193182 Hz
CPU: i386DX (386-class CPU)
real memory  = 8388608 (8192K bytes)
avail memory = 5668864 (5536K bytes)
Preloaded elf kernel "kernel" at 0xc02ba000.
Preloaded userconfig_script "/boot/kernel.conf" at 0xc02ba09c.
npx0: math processor on motherboard
npx0: using IRQ 13 interface
isa0: ISA bus on motherboard
fdc0: NEC 72065B or clone at port 0x3f0-0x3f7 irq 6 drq 2 on isa0
fdc0: FIFO enabled, 8 bytes threshold
fd0: 1440-KB 3.5" drive on fdc0 drive 0
fd1: 1200-KB 5.25" drive on fdc0 drive 1
WARNING: "fd" is usurping "fd"'s cdevsw[]
WARNING: "fd" is usurping "fd"'s bmaj
aha0 at port 0x334-0x337 irq 11 drq 5 on isa0
aha0: AHA-1542CF FW Rev. E.0 (ID=45) SCSI Host Adapter, SCSI ID 7, 16 CCBs
atkbdc0: keyboard controller (i8042) at port 0x60-0x6f on isa0
atkbd0: AT Keyboard irq 1 on atkbdc0
vga0: Generic ISA VGA at port 0x3b0-0x3df iomem 0xa-0xb on isa0
sc0: System console on isa0
sc0: VGA 16 virtual consoles, flags=0x200
sio0 at port 0x3f8-0x3ff irq 4 flags 0x10 on isa0
sio0: type 16550A
sio1 at port 0x2f8-0x2ff irq 3 on isa0
sio1: type 16550A
sio2 at port 0x3e8-0x3ef irq 5 on isa0
sio2: type 16550A
sio3 at port 0x2e8-0x2ef irq 7 on isa0
sio3: type 16550A
ed0 at port 0x300-0x31f iomem 0xcc000-0xc irq 10 on isa0
ed0: supplying EUI64: 00:00:c0:ff:fe:27:94:9f
ed0: address 00:00:c0:27:94:9f, type WD8013WC (16 bit) 
Waiting 15 seconds for SCSI devices to settle
Mounting root from ufs:/dev/da0s1a
da0 at aha0 bus 0 target 0 lun 0
da0: IBM 0663L12 s 62 Fixed Direct Access SCSI-2 device 
da0: 5.000MB/s transfers (5.000MHz, offset 8)
da0: 958MB (1962030 512 byte sectors: 64H 32S/T 958C)

-- 
B.Walter  COSMO-Project  http://www.cosmo-project.de
[EMAIL PROTECTED] Usergroup[EMAIL PROTECTED]



To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message



-pipe switch in kernel compilation

2000-01-10 Thread George Cox

G'day,

While compiling a kernel today, I noticed that the '-pipe' option to gcc
was not being used.  Is there any reason for this?

best;


gjvc

-- 
[gjvc]
4.4BSD 4.ever!


To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message



Re: -pipe switch in kernel compilation

2000-01-10 Thread Andreas Klemm

On Mon, Jan 10, 2000 at 04:55:43PM +, George Cox wrote:
 G'day,
 
 While compiling a kernel today, I noticed that the '-pipe' option to gcc
 was not being used.  Is there any reason for this?

I think this is the (historical) default, so that people with
only 4-8 MB of RAM don't get into trouble ;-)

Where 4 MB isn't sufficient anymore with a GENERIC kernel.
You need at least 6 MB or so to boot, then compile a custom
kernel and then, if you are lucky, can perhaps run with 4 MB.

But that information is about 1-2 years old, don't know, if
we perhaps already need 6-8 MB nowadays...
Though it's zillions better than M$ crap.

-- 
Andreas Klemm  http://www.FreeBSD.ORG/~andreas
 http://www.freebsd.org/~fsmp/SMP/SMP.html
   powered by Symmetric MultiProcessor FreeBSD
Get new songs from our band: http://www.freebsd.org/~andreas/64bits/index.html


To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message