public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: Wakko Warner <wakko@animx.eu.org>
To: "Randy.Dunlap" <rddunlap@osdl.org>
Cc: linux-kernel@vger.kernel.org
Subject: Re: zImage on 2.6?
Date: Tue, 3 May 2005 19:05:32 -0400	[thread overview]
Message-ID: <20050503230532.GA13063@animx.eu.org> (raw)
In-Reply-To: <20050503153737.5e3627ad.rddunlap@osdl.org>

Randy.Dunlap wrote:
> On Tue, 3 May 2005 18:19:22 -0400 Wakko Warner wrote:
> 
> | Randy.Dunlap wrote:
> | > On Tue, 3 May 2005 12:33:43 -0400 Wakko Warner wrote:
> | > | Yes, I do recall it says "System is 724k".  zImage failes.  bzImage says
> | > | 724k as well and succeeds.
> | > 
> | > The image size needs to be <= 0x7f000 (520192 bytes, 508 KB).
> | > 
> | > (No, I don't know why, just that this is what is being
> | > enforced.)
> | > 
> | > Just cut more out of the kernel image...
> | 
> | Any suggestions?  All that I know of that is modularizable is a module,
> | except for keyboard, ext2, unix sockets, ramdisk+initrd.  Some of the options I need since
> | this is supposed to support all relevent hardware that we use where I work.
> | (all scsi,ide,sata,raid cards,ethernet cards that we have)
> 
> Are those (mostly) modular?

Yes.

> Is this supposed to be a kernel that is used on a regular basis,
> not just a temporary quick boot thing?

I'd say it would be used frequently, but short lived.

> You want one kernel that handles many configurations but fits
> in < 510 KB?  That is aggressive.  :)

=)  hehe, if I can.

> I think of this as being a custom kernel, but it sounds
> like you want a very general-purpose (small) one.

Yes, very custom.

> | Hmm, I did find one I forgot.  CONFIG_MII.
> | 
> | EXPERIMENTAL=y
> | CLEAN_COMPILE=y
> | BROKEN_ON_SMP=y
> 
> Do you need some drivers etc. that are EXPERIMENTAL or BROKEN?
> Just curious.

Actually, not sure, I've always enabled that.

> | ACPI=y
> | ACPI_BOOT=y
> | ACPI_INTERPRETER=y
> | ACPI_BUS=y
> | ACPI_EC=y
> | ACPI_POWER=y
> | ACPI_PCI=y
> | ACPI_SYSTEM=y
> so some systems may need ACPI ?

I wanted to beable to query battery life on laptops that I use this on.  I
can remove it I think, all ACPI that can be a module is.  I don't think it
works that way though.  I'll have to check it on a laptop.

> | ISA=y
> | CARDBUS=y
> | PCMCIA_PROBE=y
> some systems have ISA?  some are PCMCIA/CardBus?

Yes, however, pcmcia is modular.

> | PNP=y
> | ISAPNP=y
> and ISA PNP?  :(

We have some old ISA cards that support PNP (3c509 comes to mind)

> | SCSI_PROC_FS=y
> drop this one
> 
> | AIC7XXX_DEBUG_ENABLE=y
> | AIC7XXX_REG_PRETTY_PRINT=y
> | AIC79XX_DEBUG_ENABLE=y
> | AIC79XX_REG_PRETTY_PRINT=y
> drop these.

All SCSI is a module, these I assume are just capabilites in the modules.

> | SOUND_GAMEPORT=y
> drop.

Couldn't find a configure option to disable this.  It caught my eye too.  I
don't have gameport in input configured.

> | VT=y
> | VT_CONSOLE=y
> | HW_CONSOLE=y
> | VGA_CONSOLE=y
> | DUMMY_CONSOLE=y
> drop that ^^^^^

All of those or DUMMY_CONSOLE?

> | USB_ARCH_HAS_HCD=y
> | USB_ARCH_HAS_OHCI=y
> | USB_DEVICEFS=y
> | USB_OHCI_LITTLE_ENDIAN=y
> | USB_STORAGE_FREECOM=y
> | USB_STORAGE_ISD200=y
> | USB_STORAGE_DPCM=y
> | USB_STORAGE_USBAT=y
> | USB_STORAGE_SDDR09=y
> | USB_STORAGE_SDDR55=y
> | USB_STORAGE_JUMPSHOT=y
> | USB_HIDINPUT=y
> | USB_ALI_M5632=y
> | USB_AN2720=y
> | USB_BELKIN=y
> | USB_GENESYS=y
> | USB_NET1080=y
> | USB_PL2301=y
> | USB_KC2190=y
> | USB_ARMLINUX=y
> | USB_EPSON2888=y
> | USB_ZAURUS=y
> | USB_CDCETHER=y
> | USB_AX8817X=y
> | EXT2_FS=y
> | JOLIET=y
> | UDF_NLS=y
> | NTFS_RW=y
> drop NTFS (you didn't list it as a requirement :)

All USB are modules, NTFS is a module (yes I need it)

> | PROC_FS=y
> | SYSFS=y
> | TMPFS=y
> | RAMFS=y
> | NFS_V3=y
> | LOCKD_V4=y
> | MSDOS_PARTITION=y
> | DEBUG_KERNEL=y
> | MAGIC_SYSRQ=y
> | CRYPTO=y
> drop ^^^^^

Crypto is required by one of the wireless drivers (MADWIFI I believe)

> so what's the problem with using a bzImage kernel?

Nothing, I wanted to see if zImage would be smaller.  As others have stated,
it's not.  However, I did not know this.

I do appriate your time in helping me.  I'm real close to getting this on a
single floppy (if I formatted to 1.7mb, it would be easy.  I know some of
our systems that I would need this on can't use these).  One goal is to use
the same kernel for all booting.

-- 
 Lab tests show that use of micro$oft causes cancer in lab animals

  reply	other threads:[~2005-05-03 23:06 UTC|newest]

Thread overview: 28+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2005-05-03  1:29 zImage on 2.6? Wakko Warner
2005-05-03  2:35 ` Randy.Dunlap
2005-05-03 10:45   ` Wakko Warner
2005-05-03 14:26     ` Randy.Dunlap
2005-05-03 16:33       ` Wakko Warner
2005-05-03 16:59         ` Randy.Dunlap
2005-05-03 22:19           ` Wakko Warner
2005-05-03 22:27             ` segin
2005-05-03 22:31               ` Wakko Warner
2005-05-03 23:15                 ` segin
2005-05-04  0:28                 ` shogunx
2005-05-04  1:29                   ` Wakko Warner
2005-07-12  4:55                   ` H. Peter Anvin
2005-05-03 22:37             ` Randy.Dunlap
2005-05-03 23:05               ` Wakko Warner [this message]
2005-05-03 23:16                 ` Randy.Dunlap
2005-07-12  4:54           ` H. Peter Anvin
2005-05-03 18:28     ` Krzysztof Halasa
2005-07-12  4:58       ` H. Peter Anvin
2005-05-03 16:06 ` Rick Warner
2005-05-03 16:40   ` Wakko Warner
2005-05-03 18:01     ` Lennart Sorensen
2005-05-03 18:44     ` Brian Gerst
2005-05-03 22:20       ` Wakko Warner
2005-05-03 21:42     ` Rick Warner
2005-05-03 22:23       ` Wakko Warner
2005-07-12  5:00         ` H. Peter Anvin
2005-05-04 13:50 ` Wakko Warner

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20050503230532.GA13063@animx.eu.org \
    --to=wakko@animx.eu.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=rddunlap@osdl.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox