OpenBSD Journal

Heads up! New iwn firmware released

Contributed by johan on from the stick-it-to-the-bad-guys dept.

Damien Bergamini (damien@) has committed an update to the man page for the iwn(4) wireless device. The update describes a firmware update by Intel.

As you know, Intel refuses to grant distribution rights without contractual obligations, so OpenBSD cannot include the firmware file and the users have to download it on their own.

For further reading please see this article here on Undeadly which explains the matter in depth.

(Comments are closed)

Comments

  1. By Anonymous Coward (82.101.210.49) on

    What exactly are the contractual obligations that Intel requires for distribution?

    Comments
    1. Comments

      1. By Anonymous Coward (82.101.210.49) on

        In other words, you don't know what they are either?

        Comments

        1. By Anonymous Coward (89.27.61.5) on

          > In other words, you don't know what they are either?

          I was reading that link and I think this part is not ok

          "No reverse engineering, decompilation, or disassembly of this software is permitted."

          Comments

          1. By Mike Swanson (76.121.21.10) on

            > > In other words, you don't know what they are either?
            >
            > I was reading that link and I think this part is not ok
            >
            > "No reverse engineering, decompilation, or disassembly of this software is permitted."

            None of which is actually pushed by the OpenBSD developers, they just want redistribution rights.

            Which from that file, seems like it should be possible, but I'd rather look at the original source rather than some "Intell Wifi on Linux" site.

            Comments

            1. By Anonymous Coward (71.126.39.55) on

              > Which from that file, seems like it should be possible, but I'd rather look at the original source rather than some "Intell Wifi on Linux" site.

              The file linked above IS the readme associated with the original source. If you look up the Whois information on intellinuxwireless.org, you will see that it's owned by the Intel Corportation. It's also where Damien obtained the firmware (see his commit message linked in the article).

            2. By Anonymous Coward (89.27.61.5) on

              > > > In other words, you don't know what they are either?
              > >
              > > I was reading that link and I think this part is not ok
              > >
              > > "No reverse engineering, decompilation, or disassembly of this software is permitted."
              >
              > None of which is actually pushed by the OpenBSD developers, they just want redistribution rights.
              >
              > Which from that file, seems like it should be possible, but I'd rather look at the original source rather than some "Intell Wifi on Linux" site.

              It is not possible to read just one part of a license. If they distribute it, they also agree to not reverse engineer it.

              Atleast in some drivers there have been efforts to reverse engineer.

            3. By tedu (udet) on

              > > > In other words, you don't know what they are either?
              > >
              > > I was reading that link and I think this part is not ok
              > >
              > > "No reverse engineering, decompilation, or disassembly of this software is permitted."
              >
              > None of which is actually pushed by the OpenBSD developers, they just want redistribution rights.

              What we want is free software. Those restrictions are not free.

              Comments

              1. By Anonymous Coward (2a01:198:25d:0:20a:e4ff:fe32:17b2) on

                > What we want is free software. Those restrictions are not free.

                It's not software. It's firmware. Redistribution rights should be enough.

                Comments

                1. By Anonymous Coward (128.237.231.106) on

                  > It's not software. It's firmware. Redistribution rights should be enough.

                  The license doesn't only grant redistribution rights. It also contains a "limited patent license". And according to the FAQ at 

                   http://intellinuxwireless.org/?n=FAQ&s=license#bsd_case
                  Intel plans to sue you for patent infringement if you use the firmware in a proprietary commercial product.

                  Comments

                  1. By Anonymous Coward (2a01:198:25d:0:20a:e4ff:fe32:17b2) on

                    > The license doesn't only grant redistribution rights. It also
                    > contains a "limited patent license". And according to the FAQ at
                    >
                    > Intel plans to sue you for patent infringement if you use the firmware
                    > in a proprietary commercial product.

                    This is funny. I wonder if they do that willingly... I mean, saving
                    the few cents for the EEPROM. If the firmware came with/on the
                    hardware, they couldn't do that...

                    They're just crazy.

                  2. By Anonymous Coward (71.126.39.55) on

                    > Intel plans to sue you for patent infringement if you use the firmware in a proprietary commercial product.

                    A recent US Supreme Court ruling might be relevant: 

                     http://www.eff.org/deeplinks/2008/06/supreme-court-victory-patent-first-sale-doctrine
                    Basically, since Intel has licensed the firmware to be used in any open source operating system, the ruling suggests that Intel may have implicitly licensed the firmware to be used by any vendor who incorporates that open source operating system into a proprietary commercial product.

                    So, if Intel did sue you for patent infringement, the odds are pretty good that you could win (assuming that you could afford to defend yourself in court).

                    Comments

                    1. By Anonymous Coward (82.101.210.49) on

                      > So, if Intel did sue you for patent infringement, the odds are pretty good that you could win (assuming that you could afford to defend yourself in court).

                      That theory is at least dubious, particularly because the post you linked to reads:

                      "Unfortunately, the Court did not take the opportunity to issue a broad ruling on whether other sorts of labels, or licenses, or contracts might be enough to defeat the patent exhaustion doctrine."

                    2. By tedu (udet) on

                      > > Intel plans to sue you for patent infringement if you use the firmware in a proprietary commercial product.
                      >
                      >
                      > A recent US Supreme Court ruling might be relevant:
                      >
                      > http://www.eff.org/deeplinks/2008/06/supreme-court-victory-patent-first-sale-doctrine
                      >
                      > Basically, since Intel has licensed the firmware to be used in any open source operating system, the ruling suggests that Intel may have implicitly licensed the firmware to be used by any vendor who incorporates that open source operating system into a proprietary commercial product.
                      >
                      >
                      > So, if Intel did sue you for patent infringement, the odds are pretty good that you could win (assuming that you could afford to defend yourself in court).

                      It would be a whole lot simpler if Intel simply released the firmware with a reasonable license. Until then, it stays out of OpenBSD.

                2. By tedu (udet) on

                  > > What we want is free software. Those restrictions are not free.
                  >
                  > It's not software. It's firmware. Redistribution rights should be enough.

                  they are not.

          2. By Brynet (Brynet) on

            > > In other words, you don't know what they are either?
            >
            > I was reading that link and I think this part is not ok
            >
            > "No reverse engineering, decompilation, or disassembly of this software is permitted."

            They didn't reverse engineer the firmware, or decompile/disassemble it.. it's loaded onto the device as-is.

            It's simply redistributed by Damien Bergamini.. who I presume lives in France.

            Comments

            1. By hugin (212.198.62.141) on

              > It's simply redistributed by Damien Bergamini.. who I presume lives in France.

              Yes, and what !
              what's the connection between citizenship and firmware ???

              Comments

              1. By Colin D. (cdidier) on http://cybione.org/

                The clause "No reverse engineering, decompilation, or disassembly of this software is permitted." does not apply in France.

                Comments

                1. By Anonymous Coward (82.101.210.49) on

                  > The clause "No reverse engineering, decompilation, or disassembly of this software is permitted." does not apply in France.

                  But that statement only holds when such acts are necessary to achieve "interoperability".

                2. By hugin (212.198.62.141) on

                  > The clause "No reverse engineering, decompilation, or disassembly of >this software is permitted." does not apply in France.

                  I don't know what Damien have done within and I'm not a layer, but I'm pretty sure, he's not out of the french law.

                  The law in the US differs from Europe, for instance pattents on softwares...but anyway for a better understanding, you may contact him.

                  Assuming Intel has registered their product/pattent in the US, therefore that does prevail only in US territory, so as to europeen laws aren't a concern in the US, etc.

                  Micro$oft has been nailled in Europe, never in North America!(to my knowledge, Intel is scheduled for the hammer as well !).

                  Comments

                  1. By Brynet (Brynet) on

                    > > The clause "No reverse engineering, decompilation, or disassembly of >this software is permitted." does not apply in France.
                    >
                    > I don't know what Damien have done within and I'm not a layer, but I'm pretty sure, he's not out of the french law.
                    >
                    > The law in the US differs from Europe, for instance pattents on softwares...but anyway for a better understanding, you may contact him.
                    >
                    > Assuming Intel has registered their product/pattent in the US, therefore that does prevail only in US territory, so as to europeen laws aren't a concern in the US, etc.
                    >
                    > Micro$oft has been nailled in Europe, never in North America!(to my knowledge, Intel is scheduled for the hammer as well !).
                    >

                    How many times does it have to be repeated? the firmware was NOT decompiled.. it was NOT reverse engineered.

                    The firmware image is distributed as-in.

                    Linux ignorance is confusing the difference between a driver and firmware again.

          3. By Paul 'WEiRD' de Weerd (weerd) on http://www.weirdnet.nl/openbsd/

            > > In other words, you don't know what they are either?
            >
            > I was reading that link and I think this part is not ok
            >
            > "No reverse engineering, decompilation, or disassembly of this software is permitted."

            I think it's actually this bit :

            Redistribution. Redistribution and use in binary form, without modification, are permitted provided that the following conditions are met:

            The file as it is redistributed now is fine (in binary form, without modification). Have a look at /sys/dev/microcode, where all such firmwares should end up in the end. Those are obviously not the original binaries without modification.

            But, IANAL

    2. By Anonymous Coward (82.101.210.49) on

      > What exactly are the contractual obligations that Intel requires for distribution?

      I'm not surprised Intel has not cooperated with the OpenBSD community, as the community apparently is unable to adequately explain its demands.

      Comments

      1. By Paul 'WEiRD' de Weerd (weerd) on http://www.weirdnet.nl/openbsd/

        > > What exactly are the contractual obligations that Intel requires for distribution?
        >
        > I'm not surprised Intel has not cooperated with the OpenBSD community, as the community apparently is unable to adequately explain its demands.

        Intel should be cooperating with the OpenBSD developers. The community is larger than just the developers, not everyone in the comunity fully understands the intricacies of the issues at hand. Some random comments on undeadly (including my own) do not represent the requirements OpenBSD has for firmware distribution.

        Intel knows what OpenBSD demands (eg. the firmware for fxp(4) NICs is correctly licensed), but in the case of these wireless firmwares they chose to be incompatible with what OpenBSD requires for inclusion.

      2. By Otto Moerbeek (otto) on http://www.drijf.net

        > > What exactly are the contractual obligations that Intel requires for distribution?
        >
        > I'm not surprised Intel has not cooperated with the OpenBSD community, as the community apparently is unable to adequately explain its demands.

        Beware: I never looked at this license in detail.

        OpenBSD only want to redistribute things in base that have no strings attached: users should be ablt to do with the software what they want. In this case, there are strings attached for the users.

  2. By Dean (63.228.83.29) on

    What is the preferred way to update the firmware?
    I tried pkg_add -ui http://<link to file.tgz> and it didn't work.
    I then did a pkg_delete iwn-firmware-5.0 ;success
    and a pkg_add http://<link to file.tgz> ;success

    Is that the best way, or did I miss something?
    And thanks to Damien and all the rest of the wireless and OBSD developers!

    Comments

    1. By Anonymous Coward (98.127.110.254) on

      I am definitely not a pkg_add expert, but I'm willing to bet that the error message was along the lines of "conflicts with earlier version".

      Including an error message would've been extremely helpful here.

    2. By Anonymous Coward (2a01:348:108:100:230:18ff:fea0:6af6) on

      > What is the preferred way to update the firmware?
      > I tried pkg_add -ui http://<link to file.tgz> and it didn't work.

      -u scans ${PKG_PATH} and works out which of the packages you have installed can be updated to new ones, and calls pkg_add -r to update them.

      > I then did a pkg_delete iwn-firmware-5.0 ;success
      > and a pkg_add http://<link to file.tgz> ;success
      >
      > Is that the best way, or did I miss something?
      > And thanks to Damien and all the rest of the wireless and OBSD developers!

      you want pkg_add -r to update here.

Latest Articles

Credits

Copyright © - Daniel Hartmeier. All rights reserved. Articles and comments are copyright their respective authors, submission implies license to publish on this web site. Contents of the archive prior to as well as images and HTML templates were copied from the fabulous original deadly.org with Jose's and Jim's kind permission. This journal runs as CGI with httpd(8) on OpenBSD, the source code is BSD licensed. undeadly \Un*dead"ly\, a. Not subject to death; immortal. [Obs.]