[gentoo-dev] Request for enhancement

[gentoo-dev] Request for enhancement

[Bo Majewski]

I think the "Final network configuration" can benefit from adding a 
comment to indicate that 3c59x is a 3Com network module. It was less 
than obvious to me, since I never owned one. The 3c59x looked like a 
magic number, needed by Gentoo. Only after seeing an error message 
during boot sequence I realized my mistake. Documentatin  should caution 
people to include modules they compiled. A good way of doing this would 
be to include some comments in the code. I have RealTek 8139 Ethernet 
card and for such you need to include 8139too module.  Maybe something 
like this:

Code listing 16.5: /etc/modules.autoload:
# If you compiled 3Com Ethernet card as a module uncomment the following
# 3c59x
# If you compiled RealTek 8139 Ethernet card as a module uncomment the 
following
# 8139too

# Etc., for other cards ...

Bo

Re: [gentoo-dev] Request for enhancement

[Stuart Bouyer]

[-- Attachment #1: Type: text/plain, Size: 1465 bytes --]

On 火, 2002-10-08 at 02:49, Bo Majewski wrote:
> I think the "Final network configuration" can benefit from adding a 
> comment to indicate that 3c59x is a 3Com network module. 

This is a good idea, maybe a link to a list of common nics and their
modules on or off site.

It was less 
> than obvious to me, since I never owned one. The 3c59x looked like a 
> magic number, needed by Gentoo. Only after seeing an error message 
> during boot sequence I realized my mistake. Documentatin  should caution 
> people to include modules they compiled. A good way of doing this would 
> be to include some comments in the code. I have RealTek 8139 Ethernet 
> card and for such you need to include 8139too module.  Maybe something 
> like this:
> 
> Code listing 16.5: /etc/modules.autoload:
> # If you compiled 3Com Ethernet card as a module uncomment the following
> # 3c59x
> # If you compiled RealTek 8139 Ethernet card as a module uncomment the 
> following
> # 8139too
> 
> # Etc., for other cards ...
> 
I think this is bad idea!

I for one don't want my modules.autoload full of a list of nic cards
just because other people don't take the time to read the documentation
and find out how to load their modules. 

Yes, I know I could delete the entries, but why should I have to? 

If we go this way, next people will be asking for a list of sound, scsi,
ide and other misc pci cards. Where do we draw the line? 
 
Stu

-- 

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

[gentoo-dev] Re: Request for enhancement

[Andreas Waschbuesch]

Bo Majewski wrote:

> [...]
> # If you compiled RealTek 8139 Ethernet card as a module uncomment the
> # following
> # 8139too

Let me correct this:

# If you got a RealTek 8129/8139 chip based ethernet card throw it away
# and buy a decent nic such as Intel, 3Com, DEC etc. etc.

cf. FreeBSD source: /src/sys/pci/if_rl.c

,----[ FreeBSD source: /src/sys/pci/if_rl.c ]
| $FreeBSD: src/sys/pci/if_rl.c,v 1.47 2000/10/15 14:18:59 phk Exp $
| 
| 
| RealTek 8129/8139 PCI NIC driver
| 
| Supports several extremely cheap PCI 10/100 adapters based on
| the RealTek chipset. Datasheets can be obtained from
| www.realtek.com.tw.
| 
| Written by Bill Paul <wpaul@ctr.columbia.edu>
| Electrical Engineering Department
| Columbia University, New York City
| 
| 
| 
| The RealTek 8139 PCI NIC redefines the meaning of 'low end.' This is
| probably the worst PCI ethernet controller ever made, with the
| possible exception of the FEAST chip made by SMC. The 8139 supports
| bus-master DMA, but it has a terrible interface that nullifies any
| performance gains that bus-master DMA usually offers.
| 
| For transmission, the chip offers a series of four TX descriptor
| registers. Each transmit frame must be in a contiguous buffer, aligned
| on a longword (32-bit) boundary. This means we almost always have to
| do mbuf copies in order to transmit a frame, except in the unlikely
| case where a) the packet fits into a single mbuf, and b) the packet
| is 32-bit aligned within the mbuf's data area. The presence of only
| four descriptor registers means that we can never have more than four
| packets queued for transmission at any one time.
| 
| Reception is not much better. The driver has to allocate a single
| large buffer area (up to 64K in size) into which the chip will DMA
| received frames. Because we don't know where within this region
| received packets will begin or end, we have no choice but to copy data
| from the buffer area into mbufs in order to pass the packets up to the
| higher protocol levels.
| 
| It's impossible given this rotten design to really achieve decent
| performance at 100Mbps, unless you happen to have a 400Mhz PII or
| some equally overmuscled CPU to drive it.
| 
| On the bright side, the 8139 does have a built-in PHY, although
| rather than using an MDIO serial interface like most other NICs, the
| PHY registers are directly accessible through the 8139's register
| space. The 8139 supports autonegotiation, as well as a 64-bit
| multicast filter.
| 
| The 8129 chip is an older version of the 8139 that uses an external
| PHY chip. 
| The 8129 has a serial MDIO interface for accessing the MII where
| the 8139 lets you directly access the on-board PHY registers. We need
| to select which interface to use depending on the chip type.
| 
`----

Re: [gentoo-dev] Re: Request for enhancement

[Matthew Walker]

> | The RealTek 8139 PCI NIC redefines the meaning of 'low end.' This is |
> probably the worst PCI ethernet controller ever made, with the
> | possible exception of the FEAST chip made by SMC. The 8139 supports |
> bus-master DMA, but it has a terrible interface that nullifies any |
> performance gains that bus-master DMA usually offers.

Joy. Just what I wanted to read about my ethernet controller. ;)

Re: [gentoo-dev] Re: Request for enhancement

[Jon Nelson]

On Tue, 08 Oct 2002 00:14:48 +0200
Andreas Waschbuesch <awaschb@gwdg.de> wrote:

> Bo Majewski wrote:
> 
> > [...]
> > # If you compiled RealTek 8139 Ethernet card as a module uncomment the
> > # following
> > # 8139too
> 
> Let me correct this:
> 
> # If you got a RealTek 8129/8139 chip based ethernet card throw it away
> # and buy a decent nic such as Intel, 3Com, DEC etc. etc.

Let's keep this crap off the lists, huh?  I have worked with literally tens
of thousands of ethernet cards in my lifetime, most in the last 5 years,
and statistically, I've had more problems with 3com that with anything
else (except Winbond).  I recommend and use RealTek 8139's,
because they work, they work well, and I have yet to see any *actual*
evidence that on modern machines they perform any less well than
just about anything else.  Plus, they are 7 bucks a pop.

-- 
Pound for pound, the amoeba is the most vicious animal on earth.

Jon Nelson <jnelson@jamponi.net>
C and Python Code Gardener

[gentoo-dev] Re: Re: Request for enhancement

[Andreas Waschbuesch]

Jon Nelson wrote:

> [...]
> I recommend and use RealTek 8139's,
> because they work, they work well, and I have yet to see any *actual*
> evidence that on modern machines they perform any less well than
> just about anything else.  Plus, they are 7 bucks a pop.

You get what You pay for. 

BTW: "Re-Al-Tek" is an Altaic (vulgo: Korean) word and means: "Now that 
we got Your a** we're gonna get Your bases too - long nosed man!"

Re: [gentoo-dev] Request for enhancement

[Karsten Schulz]

On Montag, 7. Oktober 2002 19:49, Bo Majewski wrote:
> I think the "Final network configuration" can benefit from adding a
> comment to indicate that 3c59x is a 3Com network module.

there is no need for this. As for every other module, you can read the 
information in /usr/src/linux/Documentation and/or run "modinfo <modulename>"
The /etc/modules.autoload file is a configuration file, not a documentation 
file.

just my 0.02¢

have fun,
Karsten


-- 
Fachbegriffe aus der Computerwelt:
EDV  - Einige Daten Verschwinden