[svlug] Expectations for Windows XP and Linux
Jim Thompson
jim at netgate.com
Thu Jul 27 19:24:32 PDT 2006
On Jul 27, 2006, at 3:37 PM, Rick Moen wrote:
> Quoting Jim Thompson (gonzopancho at mac.com):
>
> [lots of stuff:]
>
>> Stating that I "harangued" you when what I was trying to do is
>> privately correct your misinformation is unjust and unfriendly.
>
> Folks, this is Jim. He harangues people. ;-> As you've just seen,
> he's a good guy, but he'll chew up an ungodly number of hours of your
> time attempting to quibble over trivial that makes no _earthly_
> difference, such as whether the firm he worked for was an Atheros
> "business partner" or "licensee", whether and when he's an employee at
> all, and so on. After he's driven you nearly to distraction in
> private
> mail (but you'll still think he's a good guy), he'll adopt the
> always-odd combination of extremely aggressive rhetoric with
> (simultaneous) hypersensitivity about personal offence.
1) I got trained (with a switch) by a couple lawyers who didn't like
to use the word "partnership" where such a legal entity wasn't intended.
(Reason: you could get stuck with your "partner"'s problems, debts,
judgments, etc.)
2) I really like *not* being an employee. :-)
The second of these may be hypersensitivity about personal offense,
the first is just good corporate hygiene.
> Stick with him. He's worth the trouble...
>
>> I've got six years of looking at this particular subject, from
>> nearly
>> every angle I can imagine. Do you?
>
> ...even though, at odd intervals, he'll switch from talking to you, to
> attempting completely irrelevant appeals to authority -- e.g., he's
> well
> aware you cannot decide legal/regulatory and technical issues by
> counting how many years each person has dealt with a subject. But if
> you can get past that and other cranky rhetorical twitches, Jim _is_
> knowledgeable and gives you his level-best, straight assessment.
It was just a way of staying, "I've looked at a way through this for
six years now, and I can't find the requisite eye of the needle."
> Jim, one of the reasons I attempted to get you to talk in public,
> rather
> than just at me privately, is my _hope_ that other people will
> converse
> with you, for the public benefit.
My only reason(s) for taking it private were:
1) I didn't really want to subscribe to svlug (I'm not in Silicon(e)
Valley, and have no intent of leaving Hawaii to move there. I'm
involved with what passes for the local LUG here, though HOSEF has
different goals than most LUGs.
2) The message to svlug would bounce until such a time as I subscribed.
3) I wanted to give you a chance to understand the 'why' without
seeming to attack you in public.
> My time is, alas, limited -- as is
> (ultimately) my interest in long research projects concerning
> hardware I
> don't even very much care about in the first place -- and I'm stealing
> time away from ongoing work to post this. (I have no stake in this
> matter, and don't even own any of the affected hardware. I'm
> interested
> only in understanding and documenting the matter.)
Sure. I'm taking time away too, and I've got some hard deadlines
tomorrow as in 'the plane leaves @ 6:15pm and I must be on it".
> I _hope_ that others with an interest in the subject will join in, and
> discuss the matter. If not, at least your post is now searchable
> on the
> Web, which is a f___load more useful pro bono publico than just
> barraging me with private mail.
>
> (Just to stress, I also consider what Jim says to be reliably his
> absolutely honest assessment, to the best of his considerable
> knowledge in this field.)
I try.
>
>> On Jul 27, 2006, at 10:25 AM, Rick Moen wrote:
>>
>>> However, before that happened, Intersil cooperated with the open
>>> source community to the extent of sharing information (_without_
>>> requiring an NDA) on how to write drivers for the hardware.
>>
>> Untrue, you had to sign an NDA with Harris/Intersil to get the
>> programmers manual back in the Prism2/2.5/3.0 days.
>
> Like many of the other things I said, this was paraphrased directly
> from
> the people at the Prism54 project. You'd have to go argue with them.
>
>
> [skipping a lot]
>
>> and detailed which parts apply. You've only quoted one.
>
> If memory serves, the other parts were procural in the sense of saying
> (something like) what categories things fall under. I was looking for
> _relevant_ requirements decreed for manufacturers of Part 15 devices,
> and that was the only clause, of the several, that seemed on-point.
> If you feel otherwise, you're invited to quote, and explain relevance.
>
>> You're not going to crash a commercial jet with your wifi gear (at
>> least, not from the ground).
>
> It was, of course, a way to dramatise the broader point that spectrum
> regulation is a serious matter. Whose slice is adjacent to whose is
> interesting, but rather beside the point.
Many people don't understand what is going on "around" WiFi. They
think spectrum regulation is a stupid idea, that the air is 'free',
and they should be able to put any speech they want into it.
In a perfect world (with perfect receiver architectures), they would
be correct.
> [skipping a lot]
>
>> Nothing but the (closed source) HAL stands between you, or any
>> 'disgruntled' person emitting strong RF in these bands using an
>> ordinary Atheros card.
>
> To reiterate, nobody with any sense wants that ability on a wifi card.
> The control for _that_ can be in a binary file, or a ROM, or
> microcode,
> for all I care. (If it's a binary file, it would be nice for the
> public
> to have the right to redistribute it, which is not currently the case
> for, e.g., some Conexant/Intersil and Intel "firmware images".)
But is for the madwifi HAL.
> However, since the Atheros "HAL" apparently includes a great deal more
> functionality than that, it is not OS-neutral, resulting in a
> situation
> -- correct me if I'm wrong -- where "Madwifi" will run only on
> officially ported platforms, and is fully maintainable in perpetuity
> only by people with access to the proprietary source under NDA.
Its actually a license agreement, rather than an NDA, but it does
include NDA-like provisions, and I quibble.
> To repeat the earlier comparison, I wouldn't particularly have
> minded if
> my Lucent Orinoco and Cisco Aironet 802.11b cards had used "binary
> image" files instead of ROMs, as long as -- like ROMs -- they would
> Just
> Work in perpetuity until the device burned out, and be OS-neutral.
You might have minded if you wanted your XXX card to do something new
and different (say, WPA?) and couldn't make it do that.
You might have minded if you wanted to experiment with alternative
MACs, and couldn't.
>> The FCC doesn't want the general public
>> enabled in this manner. I think they're wise.
>
> Apologies for not being able to oblige you by disagreeing.
>
>
>>> "User-accessible controls" do not tend to include, e.g., solder
>>> joints on CB radios. Or mail-order copies of schematic diagrams.
>>> Or, I would wager, C source code.
>>
>> A wager? Are you offering to buy-out Atheros (current market cap:
>> USD 830.5M), removing the risk for its officers and shareholders,
>> then "open source" the HAL source code and see what the FCC (never
>> mind various other government agencies around the world) will do?
>
> I had in mind more like a pint of the beer of your choosing. ;->
>
> (C'mon, Jim. Cut the bullshit rhetoric melodrama.)
I'm OK betting you a beer (or three, though I recently stopped
drinking for no good reason), but you're asking Atheros to
(literally) "bet the farm" by
requiring them to "open source" the HAL. It wasn't melodrama, it
was what will have to happen before
I've previously discussed (on ESR's blog and elsewhere) that there is
probably a way to "open source" the HAL but have any resulting
binaries have to be "signed"
by Atheros (or one of its partners), with the chipset able to protect
itself by looking at its (public) key and verifying that the "HAL"
was signed with a private key.
(Implementation of same left as an exercise for the reader.)
>
>> You may wish to read the NPRM on "software defined radios": http://
>> www.fcc.gov/Bureaus/Engineering_Technology/Notices/2000/fcc00430.pdf
>> please pay particular attention to the rules about firmware.
>>
>> Quoting: --- A software defined radio is a radio that includes a
>> transmitter in which the operating parameters of the transmitter,
>> including the frequency range, modulation type or maximum
>> radiated or
>> conducted output power can be altered by making a change in software
>> without making any hardware changes. ---
>>
>> As discussed above, the Atheros radios fit this description, and thus
>> fall prey to it. Full stop.
>
> By your logic, Atheros is already in violation, since it has
> reportedly
> taken _no_ steps as required on "software radios" per section 2.932,
> the section immediately following the above:
>
> (e) Manufacturers must take steps to ensure that only software
> that has
> been approved by the FCC or a TCB can be loaded into a transmitter.
> The software must not allow the user to operate the transmitter
> with frequencies, output power, modulation types or other
> parameters
> outside of those that were approved. Manufacturers may use
> authentication codes or any other means to meet these requirements,
> and must describe the methods in their application for equipment
> authorization.
Atheros took such steps by limiting access to the HAL code. You
seem to object to such steps.
> That is, there is now a third-party "HAL" equivalent (which you
> claimed
> on linux-elitists to have been written illegally, but whose author has
> not yet been sued or enjoined), and reportedly also a separate
> clean-room rewriting effort.
a) If you can view the HAL source and this "third-party HAL", you can
quite clearly see that one is a derivative work of the other.
b) Atheros is the copyright holder, not me. They get to make the
decision.
c) It won't drive modern chipsets, and the market for ar5212 (and
earlier) devices is drying up. (market forces, mostly)
d) (minor quibble) it doesn't do a very good job with 5212s, either.
>
>
>>> In any event, it's obvious from the fact that it's _not_ OS-neutral
>>> that Atheros chose to obscure in its binary "HAL" blob way, way, way
>>> more functionality than, e.g., Intersil did.
>>
>> This statement is false. There is a lot of functionality in the
>> firmware for the Prism/PrismGT/Duette chipsets that you have in your
>> hands as 'C' code via net80211.
>
> Read what I said more carefully. Note the question of OS-neutrality.
Prism & Hermes (and even the Broadcom and Intel 'Centrino') MACs all
have computational resources on board. They *run* the firmware. You
seem to be under the (false) impression that they simply push some
values into the correct registers and get out of the way.
The Atheros (and rtl818x, as well as others) MACs are all hardware
state machines. They have no CPU resources to exploit. (They have
AES/TKIP/WEP cores which can
operate on packets in external memory, and they are capable of
generating some control frames that have tight timing requirements
(CTS, ACKs), but they are a fundamentally
different architecture from the chipsets you endorse.
Are you taking the position that the hardware is broken (or unusable)
if you don't have source code with which to operate it?
Moreover, the actual "HAL" (not the OS-specific wrapper) is OS-
agnostic, though not CPU agnostic. The same "HAL core" for x86 is
used by linux, freebsd, and netbsd (and others, there is a Windows
driver that uses the "madwifi/freebsd" HAL.) If you change
architectures (to, say, xscale) the same situation applies. You can
literally take the compiled instruction stream for the HAL under
linux on xscale and reuse it under netbsd (and soon FreeBSD, but
thats changing the subject again.)
Point in fact, the core of the HAL is OS-neutral. (there are
wrappers for linux, freebsd, etc.)
>
>> From time to time I debate making someone who is a) trustworthy and
>> b) trusted in the community subject to a "netgate" NDA and showing
>> them the HAL source code, likely acting as a contractor to the
>> company, just so yet another person could say, "Yep, there isn't
>> anything in there to fear", but then, you don't want to believe me
>> when I say that, and I have no real reason to toe the Atheros party
>> line (other than it happens to be the truth), so why would adding
>> one
>> more person help quell the discussion?
>
> I'd be glad to answer your (rather bizarre, aside from being
> pointlessly
> aggressive) question if you can calm down, and then tell me what in
> Sam
> Hill it has to do with the preceding discussion. I can't for the life
> of me imagine why you've asked it.
Other smart, honest people have looked at this and honestly see no
way around it.
>
>> Or are you simply continuing to jump up and down about the subject?
>
> {pause for irony}
>
>
>> Atheros is well-aware that many more people would like to have the
>> source code to the HAL.
>
> Stop. Breathe. Listen:
>
> I have no objection to binary-only code in my option ROMs, or in
> ROM-equivalent "firmware images". I live with it every day, as does
> everyone else on this mailing list: The server I'm posting this from
> has binary-only motherboard firmware, binary-only SCSI firmware, and
> probably binary-only ethernet firmware, not to mention binary-only
> microcode in various places.
Few of these would survive if you moved them to a PowerPC machine, or
even a 64-bit x86 machine. (I'm assuming you're on something 32-bit
x86 based.)
yes, the SCSI firmware would, but that is, quite likely, because your
SCSI controller has on-die computational resources.
The "core" of the HAL is no different. Its a collection of entry
points with known names and associated instruction streams.
> That code will work until the equipment it's embedded in dies. I
> might
> reflash it to use different binary-only code, or I might not. It
> would
> be all the same to me if it were embodied in software blobs that get
> hurled into the appropriate memory space in some well-documented
> fashion
> at initialisation time. Why? Because that would be functionally
> equivalent -- and support full hardware functionality on any OS
> without
> exception or reservation, past, present, or future.
>
> My understanding is that Madwifi does not manage that; that
> proprietary
> software coders alone have the ability to maintain it and "port" it to
> additional hardware platforms. Correct?
If by "hardware platforms" you mean CPU architectures, then yes, its
a straight-forward recompile after building a cross-compiler or
hosting a compiler on
The actual "madwifi" project doesn't have access to the HAL source to
do this. Atheros does, Sam Leffler does, I do, and around 20 other
people that I could name do.
Any one of these parties could run the compiler for you, should you
develop or encounter a new CPU architecture that needed wireless.
The situation is not ideal, but its very likely as good as it can get.
Jim
More information about the svlug
mailing list