Devs who look at ICS Source. Look here!

Search This thread

hasoon2000

Senior Member
Feb 22, 2008
3,574
9,705
Orland Park, Illinois
www.hasoon2000.info
So we have a portal writer that has written multiple articles on the stuff I made and the stuff we've been doing. He asked us for more info on the Amaze source so he can write about it and help us get more recognition. Don't believe me? Look for yourself.

https://twitter.com/PoorCollegeGuy/status/222516466711396352

So can you say what new changes have been made. Did HTC make it harder on us to hack it? Did they not release anything we need? anything so we can get this published. Write it all here and I'll pass the link to this thread to him when we get responses.

Please, only respond here with information on source. NOTHING ELSE! I want to keep it nice and clean here.
 
S

sportsstar89

Guest
Well my understanding is the wi-fi drivers are there they just commented out what we need to build it and a compat-wireless modules that they didn't include but i found another way to get it working just fine
 

hasoon2000

Senior Member
Feb 22, 2008
3,574
9,705
Orland Park, Illinois
www.hasoon2000.info
Well my understanding is the wi-fi drivers are there they just commented out what we need to build it and a compat-wireless modules that they didn't include but i found another way to get it working just fine

That is one long run on sentence o_O lol. Any other info you guys found out. We need to add more so he can write more about it!
 
S

sportsstar89

Guest
That is one long run on sentence o_O lol. Any other info you guys found out. We need to add more so he can write more about it!

Don't criticize my grammar and writing ability I go to school for that lol but my teachers tell me I'm the best writer in the class

Sent from my HTC_Amaze_4G using xda premium
 

oryan_dunn

Senior Member
Dec 23, 2010
679
183
Well my understanding is the wi-fi drivers are there they just commented out what we need to build it and a compat-wireless modules that they didn't include but i found another way to get it working just fine

Wifi is working using the existing modules that HTC provided. Currently we are still looking for a way to get the modules that are compiled by the kernel to work.

Sent from my HTC One X using xda premium

Yeah, it appears that they commented out the building of the wl12xx drivers, but the stock ICS does have wl12xx modules. The evidence (commentd out makefile, presence of compat.ko, wl12xx.ko, wl12xx_sdio.ko, etc.) points to HTC using a version of compat-wireless drivers, but which one, I don't know. I also don't know if HTC made any modifications to the wireless drivers, but if they did, they'd be obligated by the GPL to release the updates.

Since the tweaks sportsstar used to get wifi work in his kernel worked, that means that we should have everything we need on the kernel side of things, we just need to get and build the proper version of compat wireless. I also wonder if we need to match the version of compat-wireless with the binary firmware blob from TI. In all the work building different versions of compat-wireless, it could be we also need the proper firmware.
 

kozmikkick

Senior Member
Nov 26, 2010
1,278
1,215
Seattle, WA
Yeah, it appears that they commented out the building of the wl12xx drivers, but the stock ICS does have wl12xx modules. The evidence (commentd out makefile, presence of compat.ko, wl12xx.ko, wl12xx_sdio.ko, etc.) points to HTC using a version of compat-wireless drivers, but which one, I don't know. I also don't know if HTC made any modifications to the wireless drivers, but if they did, they'd be obligated by the GPL to release the updates.

Since the tweaks sportsstar used to get wifi work in his kernel worked, that means that we should have everything we need on the kernel side of things, we just need to get and build the proper version of compat wireless. I also wonder if we need to match the version of compat-wireless with the binary firmware blob from TI. In all the work building different versions of compat-wireless, it could be we also need the proper firmware.

The only problem with the current tweak to use existing wifi modules is that I think it may be breaking other functions. Not sure about this yet though.

When I get my amaze this week my plan is to try a few different firmware from ti. I think you're right about needing to swap out the blobs. If this is the case then it should make things pretty easy.

Sent from my HTC One X using xda premium
 

oryan_dunn

Senior Member
Dec 23, 2010
679
183
Ok, since this is a generic thread about ICS source, I'm moving all of my kernel ramblings here, instead of taking up space in Kozmik's kernel thread.

Anyways, I ran modinfo on all the modules in the T-Mobile ICS update. Here are the results:
Code:
filename:       bluetooth.ko
alias:          net-pf-31
license:        GPL
version:        2.16
description:    Bluetooth Core ver 2.16
author:         Marcel Holtmann <marcel@holtmann.org>
srcversion:     7F6941AC162E74AA0B26DAF
depends:        
vermagic:       3.0.16-ga4fb83c SMP preempt mod_unload ARMv7 
parm:           disable_esco:Disable eSCO connection creation (bool)
parm:           disable_ertm:Disable enhanced retransmission mode (bool)
parm:           enable_hs:Enable High Speed (bool)
parm:           enable_mgmt:Enable Management interface (bool)
parm:           enable_le:Enable LE support (bool)
parm:           force_acl_master:Always try to switch to master role on ACL link (int)


filename:       bnep.ko
alias:          bt-proto-4
license:        GPL
version:        1.3
description:    Bluetooth BNEP ver 1.3
author:         Marcel Holtmann <marcel@holtmann.org>
srcversion:     77A9BE70DF012F020FC7798
depends:        bluetooth
vermagic:       3.0.16-ga4fb83c SMP preempt mod_unload ARMv7 
parm:           compress_src:Compress sources headers (bool)
parm:           compress_dst:Compress destination headers (bool)


filename:       btwilink.ko
license:        GPL
version:        1.0
description:    Bluetooth Driver for TI Shared Transport1.0
author:         Raja Mani <raja_mani@ti.com>
srcversion:     269FE2F6F705EA176E7CD4D
depends:        bluetooth
vermagic:       3.0.16-ga4fb83c SMP preempt mod_unload ARMv7 


filename:       cfg80211.ko
description:    wireless configuration support
license:        GPL
author:         Johannes Berg
depends:        
vermagic:       3.0.16-ga4fb83c SMP preempt mod_unload ARMv7 
parm:           ieee80211_regdom:IEEE 802.11 regulatory domain code (charp)
parm:           cfg80211_disable_40mhz_24ghz:Disable 40MHz support in the 2.4GHz band (bool)


filename:       compat.ko
license:        GPL
description:    Kernel compatibility module
author:         Luis R. Rodriguez
license:        Dual BSD/GPL
author:         Broadcom Corporation
description:    Cordic functions
license:        Dual BSD/GPL
author:         Broadcom Corporation
description:    CRC8 (by Williams, Ross N.) function
depends:        
vermagic:       3.0.16-ga4fb83c SMP preempt mod_unload ARMv7 
parm:           compat_base_tree:The upstream tree used as base for this backport (charp)
parm:           compat_base_tree_version:The git-describe of the upstream base tree (charp)
parm:           compat_version:Version of the kernel compat backport work (charp)


filename:       fm_drv.ko
license:        GPL
depends:        
staging:        Y
vermagic:       3.0.16-ga4fb83c SMP preempt mod_unload ARMv7 


filename:       hid-magicmouse.ko
license:        GPL
alias:          hid:b0005v000005ACp0000030E
alias:          hid:b0005v000005ACp0000030D
depends:        
vermagic:       3.0.16-ga4fb83c SMP preempt mod_unload ARMv7 
parm:           emulate_3button:Emulate a middle button (bool)
parm:           emulate_scroll_wheel:Emulate a scroll wheel (bool)
parm:           scroll_speed:Scroll speed, value from 0 (slow) to 63 (fast)
parm:           scroll_acceleration:Accelerate sequential scroll events (bool)
parm:           report_touches:Emit touch records (otherwise, only use them for emulation) (bool)
parm:           report_undeciphered:Report undeciphered multi-touch state field using a MSC_RAW event (bool)


filename:       hidp.ko
alias:          bt-proto-6
license:        GPL
version:        1.2
description:    Bluetooth HIDP ver 1.2
author:         Marcel Holtmann <marcel@holtmann.org>
srcversion:     AC3509329E41D74FA70B7C0
depends:        bluetooth
vermagic:       3.0.16-ga4fb83c SMP preempt mod_unload ARMv7 


filename:       mac80211.ko
license:        GPL
description:    IEEE 802.11 subsystem
depends:        cfg80211
vermagic:       3.0.16-ga4fb83c SMP preempt mod_unload ARMv7 
parm:           ieee80211_default_rc_algo:Default rate control algorithm for mac80211 to use (charp)
parm:           max_nullfunc_tries:Maximum nullfunc tx tries before disconnecting (reason 4). (int)
parm:           max_probe_tries:Maximum probe tries before disconnecting (reason 4). (int)
parm:           probe_wait_ms:Maximum time(ms) to wait for probe response before disconnecting (reason 4). (int)


filename:       rfcomm.ko
alias:          bt-proto-3
license:        GPL
version:        1.11
description:    Bluetooth RFCOMM ver 1.11
author:         Marcel Holtmann <marcel@holtmann.org>
srcversion:     436C8832D80BAD759212601
depends:        bluetooth
vermagic:       3.0.16-ga4fb83c SMP preempt mod_unload ARMv7 
parm:           disable_cfc:Disable credit based flow control (bool)
parm:           channel_mtu:Default MTU for the RFCOMM channel (int)
parm:           l2cap_mtu:Default MTU for the L2CAP connection (uint)
parm:           l2cap_ertm:Use L2CAP ERTM mode for connection (bool)


filename:       wl12xx.ko
author:         Juuso Oikarinen <juuso.oikarinen@nokia.com>
author:         Luciano Coelho <coelho@ti.com>
license:        GPL
alias:          platform:wl12xx
depends:        cfg80211,mac80211
vermagic:       3.0.16-ga4fb83c SMP preempt mod_unload ARMv7 
parm:           debug_level:wl12xx debugging level (uint)
parm:           fwlog:charp
parm:           keymap:FW logger options: continuous, ondemand, dbgpins or disable
parm:           bug_on_recovery:BUG() on fw recovery (bool)
parm:           fref:FREF clock: 19.2, 26, 26x, 38.4, 38.4x, 52 (charp)
parm:           tcxo:TCXO clock: 19.2, 26, 38.4, 52, 16.368, 32.736, 16.8, 33.6 (charp)


filename:       wl12xx_sdio.ko
firmware:       ti-connectivity/wl128x-fw-4-mr.bin
firmware:       ti-connectivity/wl128x-fw-4-sr.bin
firmware:       ti-connectivity/wl127x-fw-4-mr.bin
firmware:       ti-connectivity/wl127x-fw-4-sr.bin
author:         Juuso Oikarinen <juuso.oikarinen@nokia.com>
author:         Luciano Coelho <coelho@ti.com>
license:        GPL
alias:          sdio:c*v0097d4076*
depends:        wl12xx
vermagic:       3.0.16-ga4fb83c SMP preempt mod_unload ARMv7

They are all GPL with the exception of compat.ko being dual licensed. I don't think that should be an issue (ie. doubt HTC had modifications) as I think that modules function is to interface with the other drivers (wl12xx, etc.) to the current kernel version. The rest are GPL and therefore HTC must release the source for it.
 

brainbone

Senior Member
Jun 8, 2010
89
34
They are all GPL with the exception of compat.ko being dual licensed.

I'm not perfectly clear on dual licenses, so it should be posed to FSF, but from what I understand, it allows the code to be compatible with GPL and BSD projects. However, if used in a GPL project, I believe the GPL portion of the license is what takes effect, since the BSD portion is not compatible with the GPL project.

A quote from Linus Torvalds:

"This is no different from the fact that we have some drivers that are GPLv2/BSD licensed. Within the kernel, they are GPLv2. But on their own, you can choose to use them under the BSD license, make your changes to them, and release them commercially. And correct - I cannot (and neither can anybody else) then accept those *non*GPLv2 changes back."

I'm having trouble with "on their own" and "within the kernel".
 
Last edited:
  • Like
Reactions: fcpelayo

johnride

Senior Member
Feb 16, 2012
115
13
I guess this simply means that if the modules are released within a kernel like a single entity then it has to be GPLv2 but if the module's source is not released as "part of the Linux kernel" but just like "a module to make that piece of hardware working with the Linux kernel" then it can be released under BSD license.

Sent from my HTC_Amaze
 

oryan_dunn

Senior Member
Dec 23, 2010
679
183
I guess this simply means that if the modules are released within a kernel like a single entity then it has to be GPLv2 but if the module's source is not released as "part of the Linux kernel" but just like "a module to make that piece of hardware working with the Linux kernel" then it can be released under BSD license.

Sent from my HTC_Amaze

The only module that's dual licensed is the compat.ko module, not the actual driver files, which are all still GPL. I assume that if HTC did make modifications to compat.ko, that they'd have to update the license to say BSD and not BSD/GPL. I believe all the code in compat-wireless is from the kernel itself, just a newer version, that you compile for an older kernel.

On another note, I am supposed to get a call from HTCdev escalation on the GPL driver issue. I wonder if the person who will call me will actually know what's going on, or if it will just be the generic customer support escalation team.
 
  • Like
Reactions: Ashcunak

brainbone

Senior Member
Jun 8, 2010
89
34
I assume that if HTC did make modifications to compat.ko, that they'd have to update the license to say BSD and not BSD/GPL.

I'm still fuzzy on dual licensing, but seeing that HTC released the module as dual licensed (GPL/BSD), doesn't that allow the user to decide the license? So, if you decide that it is GPL, shouldn't they release the source for it?

On another note, I am supposed to get a call from HTCdev escalation on the GPL driver issue. I wonder if the person who will call me will actually know what's going on, or if it will just be the generic customer support escalation team.

Either way, keep up the good fight.
 
Last edited:

oryan_dunn

Senior Member
Dec 23, 2010
679
183
I ended up talking to HTC this weekend. It was as I expected, some guy who couldn't really do anything, but he listened and wasn't really all that hostile. He said he'd pass it on to the dept responsible. I told him that HTC was oblicated to release the source for all GPL code and that they need to release it the day the executable code is in the hands of customers. We'll see if it leads anywhere.
 

eggydrums

Senior Member
Dec 13, 2010
2,071
306
I ended up talking to HTC this weekend. It was as I expected, some guy who couldn't really do anything, but he listened and wasn't really all that hostile. He said he'd pass it on to the dept responsible. I told him that HTC was oblicated to release the source for all GPL code and that they need to release it the day the executable code is in the hands of customers. We'll see if it leads anywhere.

Im hoping the tmo kernel source might include what the devs need, but has htc released stuff like this by itself?


Sent from my iPad using Tapatalk HD
 
S

sportsstar89

Guest
I ended up talking to HTC this weekend. It was as I expected, some guy who couldn't really do anything, but he listened and wasn't really all that hostile. He said he'd pass it on to the dept responsible. I told him that HTC was oblicated to release the source for all GPL code and that they need to release it the day the executable code is in the hands of customers. We'll see if it leads anywhere.

Well this is what i got from the gb drivers
filename: /home/steve/modules/kineto_gan.ko
license: GPL
version: 1.0
alias: gannet
description: Kineto GAN Virtual Ethernet Device
srcversion: FA8E773097534C9326F7196
depends:
vermagic: 2.6.35.13-g52e85af SMP preempt mod_unload ARMv7
filename: /home/steve/modules/st_drv.ko
license: GPL
description: Shared Transport Driver for TI BT/FM/GPS combo chips
author: Pavan Savoy <pavan_savoy@ti.com>
depends:
staging: Y
vermagic: 2.6.35.13-g52e85af SMP preempt mod_unload ARMv7
filename: /home/steve/modules/tiap_drv.ko
license: Dual BSD/GPL
description: TI WLAN Embedded Station Driver
author: Ohad Ben-Cohen <ohad@wizery.com>
alias: TIWLAN_SDIO
license: GPL
description: TI WLAN 1283 SDIO interface
alias: sdio:c07v*d*
depends:
vermagic: 2.6.35.13-g52e85af SMP preempt mod_unload ARMv7
parm: driver_status:int
parm: g_sdio_debug_level:TIWLAN SDIO debug level (int)
filename: /home/steve/modules/ti_hci_drv.ko
license: GPL
depends: st_drv
staging: Y
vermagic: 2.6.35.13-g52e85af SMP preempt mod_unload ARMv7
filename: /home/steve/modules/tiwlan_drv.ko
author: Ohad Ben-Cohen <ohad@wizery.com>
alias: TIWLAN_SDIO
license: GPL
description: TI WLAN 1283 SDIO interface
license: Dual BSD/GPL
description: TI WLAN Embedded Station Driver
alias: sdio:c07v*d*
depends:
vermagic: 2.6.35.13-g52e85af SMP preempt mod_unload ARMv7

But tiwlan_drv.ko and tiap_drv.ko doesn't get built while compiling the kernel
 

kozmikkick

Senior Member
Nov 26, 2010
1,278
1,215
Seattle, WA
My response from HTC was that they were using proprietary drivers.

Even still since the source is available from Texas it should be possible to use their code if we swap out the firmware. May have to do some calibrations and build our own WPA_supplicant though.

Sent from my HTC One X using xda premium
 

Top Liked Posts

  • There are no posts matching your filters.
  • 6
    So we have a portal writer that has written multiple articles on the stuff I made and the stuff we've been doing. He asked us for more info on the Amaze source so he can write about it and help us get more recognition. Don't believe me? Look for yourself.

    https://twitter.com/PoorCollegeGuy/status/222516466711396352

    So can you say what new changes have been made. Did HTC make it harder on us to hack it? Did they not release anything we need? anything so we can get this published. Write it all here and I'll pass the link to this thread to him when we get responses.

    Please, only respond here with information on source. NOTHING ELSE! I want to keep it nice and clean here.
    5
    Yeah, it appears that they commented out the building of the wl12xx drivers, but the stock ICS does have wl12xx modules. The evidence (commentd out makefile, presence of compat.ko, wl12xx.ko, wl12xx_sdio.ko, etc.) points to HTC using a version of compat-wireless drivers, but which one, I don't know. I also don't know if HTC made any modifications to the wireless drivers, but if they did, they'd be obligated by the GPL to release the updates.

    Since the tweaks sportsstar used to get wifi work in his kernel worked, that means that we should have everything we need on the kernel side of things, we just need to get and build the proper version of compat wireless. I also wonder if we need to match the version of compat-wireless with the binary firmware blob from TI. In all the work building different versions of compat-wireless, it could be we also need the proper firmware.

    The only problem with the current tweak to use existing wifi modules is that I think it may be breaking other functions. Not sure about this yet though.

    When I get my amaze this week my plan is to try a few different firmware from ti. I think you're right about needing to swap out the blobs. If this is the case then it should make things pretty easy.

    Sent from my HTC One X using xda premium
    4
    Just a heads up, we are going to have an article written on this soon. Our portal writer is going fix up some small kinks and then we will hopefully have more exposure and support.
    3
    Wifi is working using the existing modules that HTC provided. Currently we are still looking for a way to get the modules that are compiled by the kernel to work.

    Sent from my HTC One X using xda premium
    3
    Well my understanding is the wi-fi drivers are there they just commented out what we need to build it and a compat-wireless modules that they didn't include but i found another way to get it working just fine

    Wifi is working using the existing modules that HTC provided. Currently we are still looking for a way to get the modules that are compiled by the kernel to work.

    Sent from my HTC One X using xda premium

    Yeah, it appears that they commented out the building of the wl12xx drivers, but the stock ICS does have wl12xx modules. The evidence (commentd out makefile, presence of compat.ko, wl12xx.ko, wl12xx_sdio.ko, etc.) points to HTC using a version of compat-wireless drivers, but which one, I don't know. I also don't know if HTC made any modifications to the wireless drivers, but if they did, they'd be obligated by the GPL to release the updates.

    Since the tweaks sportsstar used to get wifi work in his kernel worked, that means that we should have everything we need on the kernel side of things, we just need to get and build the proper version of compat wireless. I also wonder if we need to match the version of compat-wireless with the binary firmware blob from TI. In all the work building different versions of compat-wireless, it could be we also need the proper firmware.