Bug 632 - iwconfig ethN txpower auto: should be an error
: iwconfig ethN txpower auto: should be an error
Status: RESOLVED DUPLICATE of bug 572
: IPW2200
Wireless Tools
: 1.0.3
: All All
: P2 normal
Assigned To:
:
:
:
:
:
  Show dependency treegraph
 
Reported: 2005-04-13 11:33 by
Modified: 2005-05-19 07:41 (History)


Attachments
Return an error on attempts to set txpower to auto (307 bytes, patch)
2005-04-13 11:36, Kai Groner
Details | Diff


Note

You need to log in before you can comment on or make changes to this bug.


Description From 2005-04-13 11:33:53
Calls to set txpower ignore iwreq_data.txpower.fixed.  When this bit is clear
iwreq_data.txpower.value should be ignored, and the call should fail (or
transmit power should become "auto").

FYI, the wireless networking scripts I use attempts to reset various parameters
such as txpower before configuring the device.  This results in txpower being
set to -1 dBm (iwconfig shows 255 dBm) which is often not enough.

I have a patch to correct this.  It seems to work fine on my computer (2.6.11,
wireless-tools 27, gentoo).
------- Comment #1 From 2005-04-13 11:36:02 -------
Created an attachment (id=343) [details]
Return an error on attempts to set txpower to auto
------- Comment #2 From 2005-04-14 08:46:41 -------
Thanks.  Applied for 1.0.4.
------- Comment #3 From 2005-05-19 04:12:21 -------
From the bug description, it should disable user to set txpower value when 
txpower is set to auto mode. But the patch disabled "txpower auto" setting. Is 
it correct?

From iwconfig manual, when setting txpower to auto, the power control is 
enabled in the driver (it may be not supported), and user can't set txpower 
anymore. If setting txpower to fixed, user will control the power. 

So user should be able to switch from those two mode (auto, fixed), right? Of 
course, if the driver doesn't support power control, disable "txpower auto" 
setting is also reasonable. But is it the driver want?




------- Comment #4 From 2005-05-19 07:37:06 -------

*** This bug has been marked as a duplicate of 572 ***
------- Comment #5 From 2005-05-19 07:41:24 -------
Notes from scrub:

chuyee	so logics, is it your intention to disable "txpower auto"?
roote	was it ok to classify it as blocker?
logics_sbux	chuyee: the intention was if 'txpower auto' is supplied, the
reported power isn't -1
logics_sbux	for some reason i thought the patch came from you originally; i
didn't look at it much when i merged it
crystal_	logics: zhuyi's patch is in #572
chuyee	i c. mine is to set it to 20dbm
crystal_	And seems the patch is not merged correctly. Now the lowest value is
set to -IPW_TX_POWER_MAX.
crystal_	which is -20, but it was -12 before...
logics_sbux	ok; i've changed -IPW_TX_POWER_MAX to IPW_TX_POWER_MIN and yanked
the !wrqu->power.fixed from the -EINVAL path
logics_sbux	let's stick 632 as a dup of 572
logics_sbux	and i'll add a patch against 1.0.4 that "fixes" it to what chuyee
actually wrote
logics_sbux	vs. what i somehow introduced to the build...
brix	heh
crystal_	seems it is different...
crystal_	I mean bug 632 is for another issue...
crystal_	Calls to set txpower ignore iwreq_data.txpower.fixed... Yi's patch
didn't fix this, right?
logics_sbux	he did. if .fixed isn't set it sets the power to IPW_TX_POWER_DEFAULT
logics_sbux	(which is essentially 'auto')