Les Mikesell wrote:
max wrote:
That's a question you'd have to direct to the author of that file, who
said that they were. He ought to know.
You based your argument around something you haven't confirmed? That
hardly seems wise but I can't say I have never done the same, lets all
try to avoid this in the future shall we, in such manner are asses
made;^). The author is also a programmer yes? not a lawyer? I am not
sure the author is qualified in the legal sense to make this
determination. The rapid evolution of technologies makes these things
especially difficult. Where can I find his quotes? Are they in the
documentation? I ask because *if* I forward this along I'll get asked
for all this info, it seems quite relevant depending on the basis for
this judgement but IANAL (i have always found this combination of
letters amusing because I think a good lawyer is anal retentive or I
guess any thorough person really, and that's certainly a quality I'd
want in my lawyer :^)
The place to start should probably be with CPU microcode that the kernel
contains and installs. There's a pretty good argument that most of the
other device firmware doesn't become part of 'the Program' and is a
separate thing even when delivered by the kernel. That argument might
be a little harder on the thing actually executing 'the Program', but if
it works there, it should apply to everything else.
My take on it is that it should be the same as downloading a song to an
ipod or other player. Regardless of how you store or encode the bits of
that song on the way to the device, it is still a separate, independent
work and it's just someone else's data bits to the rest of the process
even if that process has the GPL restrictions on its own components.
But they've put the firmware itself under the GPL:
+++ linux-2.6.7.patch/drivers/net/bcm/5701rls.h 2004-06-22 16:00:00.000000000 -0700
@@ -0,0 +1,198 @@
+/******************************************************************************/
+/* */
+/* Broadcom BCM5700 Linux Network Driver, Copyright (c) 2000 - 2003 Broadcom */
+/* Corporation. */
+/* All rights reserved. */
+/* */
+/* This program is free software; you can redistribute it and/or modify */
+/* it under the terms of the GNU General Public License as published by */
+/* the Free Software Foundation, located in the file LICENSE. */
+/* */
+/* History: */
+/******************************************************************************/
+
+typedef unsigned long U32;
+int t3FwReleaseMajor = 0x0;
+int t3FwReleaseMinor = 0x0;
+int t3FwReleaseFix = 0x0;
+U32 t3FwStartAddr = 0x08000000;
+U32 t3FwTextAddr = 0x08000000;
+int t3FwTextLen = 0x9c0;
+U32 t3FwRodataAddr = 0x080009c0;
+int t3FwRodataLen = 0x60;
+U32 t3FwDataAddr = 0x08000a40;
+int t3FwDataLen = 0x20;
+U32 t3FwSbssAddr = 0x08000a60;
+int t3FwSbssLen = 0xc;
+U32 t3FwBssAddr = 0x08000a70;
+int t3FwBssLen = 0x10;
+U32 t3FwText[(0x9c0/4) + 1] = {
+0x0,
+0x10000003, 0x0, 0xd, 0xd,
+0x3c1d0800, 0x37bd3ffc, 0x3a0f021, 0x3c100800,
+0x26100000, 0xe000018, 0x0, 0xd,
+0x3c1d0800, 0x37bd3ffc, 0x3a0f021, 0x3c100800,
+0x26100034, 0xe00021c, 0x0, 0xd,
+0x0, 0x0, 0x0, 0x27bdffe0,
+0x3c1cc000, 0xafbf0018, 0xaf80680c, 0xe00004c,
+0x241b2105, 0x97850000, 0x97870002, 0x9782002c,
+0x9783002e, 0x3c040800, 0x248409c0, 0xafa00014,
+0x21400, 0x621825, 0x52c00, 0xafa30010,
+0x8f860010, 0xe52825, 0xe000060, 0x24070102,
+0x3c02ac00, 0x34420100, 0x3c03ac01, 0x34630100,
+0xaf820490, 0x3c02ffff, 0xaf820494, 0xaf830498,
+0xaf82049c, 0x24020001, 0xaf825ce0, 0xe00003f,
+0xaf825d00, 0xe000140, 0x0, 0x8fbf0018,
... more pages of hex
I personally downloaded this file believing that I was getting a free GPL
driver. Broadcom says so in the patch itself, in the included LICENSE file, and
their website when you click to download it. Red Hat is shipping it as GPLv2.
So either they have to provide the source (if they have it), stop shipping it,
or *at least* stop saying they are shipping a GPLv2 kernel if they are
unwilling/unable to provide the source.
-Jeff
--
fedora-devel-list mailing list
fedora-devel-list@xxxxxxxxxx
https://www.redhat.com/mailman/listinfo/fedora-devel-list