d/freebsd-nq
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
a0a2aa7916
freebsd-nq/sys/dev/ath/ath_hal/ar5416/ar5416_ani.c

855 lines
26 KiB
C
Raw Normal View History

Update to later code from my repository: o many bug fixes o add new periodic calibration api o break up 5416 periodic calibration code in preparation for 928x o move get noise floor to rf backends o 5416-specific ani (still disabled) o modularize 5210 eeprom format a la other eeprom formats o start cleaning up regdomain code o prepare for proper 1/2 and 1/4 width channel support o bring back 900MHz card support o clean up 5212 rf version handling o add 1/2 and 1/4 width channel support for 5212 parts o split 5212 rfgain handling out o improve ani debugging o add AH_USE_INIPDGAIN compile option o purge a bunch of dead 5212 state o add 1/2 and 1/4 rate modes o remove HAL_CAP_CHAN_HALFRATE and HAL_CAP_CHAN_QUARTERRATE; the same info can now be deduced from the set of supported modes
2008-11-28 00:48:05 +00:00
/*
Overhaul regulatory support: o remove HAL_CHANNEL; convert the hal to use net80211 channels; this mostly involves mechanical changes to variable names and channel attribute macros o gut HAL_CHANNEL_PRIVATE as most of the contents are now redundant with the net80211 channel available o change api for ath_hal_init_channels: no more reglass id's, no more outdoor indication (was a noop), anM contents o add ath_hal_getchannels to have the hal construct a channel list without altering runtime state; this is used to retrieve the calibration list for the device in ath_getradiocaps o add ath_hal_set_channels to take a channel list and regulatory data from above and construct internal state to match (maps frequencies for 900MHz cards, setup for CTL lookups, etc) o compact the private channel table: we keep one private channel per frequency instead of one per HAL_CHANNEL; this gives a big space savings and potentially improves ani and calibration by sharing state (to be seen; didn't see anything in testing); a new config option AH_MAXCHAN controls the table size (default to 96 which was chosen to be ~3x the largest expected size) o shrink ani state and change to mirror private channel table (one entry per frequency indexed by ic_devdata) o move ani state flags to private channel state o remove country codes; use net80211 definitions instead o remove GSM regulatory support; it's no longer needed now that we pass in channel lists from above o consolidate ADHOC_NO_11A attribute with DISALLOW_ADHOC_11A o simplify initial channel list construction based on the EEPROM contents; we preserve country code support for now but may want to just fallback to a WWR sku and dispatch the discovered country code up to user space so the channel list can be constructed using the master regdomain tables o defer to net80211 for max antenna gain o eliminate sorting of internal channel table; now that we use ic_devdata as an index, table lookups are O(1) o remove internal copy of the country code; the public one is sufficient o remove AH_SUPPORT_11D conditional compilation; we always support 11d o remove ath_hal_ispublicsafetysku; not needed any more o remove ath_hal_isgsmsku; no more GSM stuff o move Conformance Test Limit (CTL) state from private channel to a lookup using per-band pointers cached in the private state block o remove regulatory class id support; was unused and belongs in net80211 o fix channel list construction to set IEEE80211_CHAN_NOADHOC, IEEE80211_CHAN_NOHOSTAP, and IEEE80211_CHAN_4MSXMIT o remove private channel flags CHANNEL_DFS and CHANNEL_4MS_LIMIT; these are now set in the constructed net80211 channel o store CHANNEL_NFCREQUIRED (Noise Floor Required) channel attribute in one of the driver-private flag bits of the net80211 channel o move 900MHz frequency mapping into the hal; the mapped frequency is stored in the private channel and used throughout the hal (no more mapping in the driver and/or net80211) o remove ath_hal_mhz2ieee; it's no longer needed as net80211 does the calculation and available in the net80211 channel o change noise floor calibration logic to work with compacted private channel table setup; this may require revisiting as we no longer can distinguish channel attributes (e.g. 11b vs 11g vs turbo) but since the data is used only to calculate status data we can live with it for now o change ah_getChipPowerLimits internal method to operate on a single channel instead of all channels in the private channel table o add ath_hal_gethwchannel to map a net80211 channel to a h/w frequency (always the same except for 900MHz channels) o add HAL_EEBADREG and HAL_EEBADCC status codes to better identify regulatory problems o remove CTRY_DEBUG and CTRY_DEFAULT enum's; these come from net80211 now o change ath_hal_getwirelessmodes to really return wireless modes supported by the hardware (was previously applying regulatory constraints) o return channel interference status with IEEE80211_CHANSTATE_CWINT (should change to a callback so hal api's can take const pointers) o remove some #define's no longer needed with the inclusion of <net80211/_ieee80211.h> Sponsored by: Carlson Wireless
2009-01-28 18:00:22 +00:00
* Copyright (c) 2002-2009 Sam Leffler, Errno Consulting
Update to later code from my repository: o many bug fixes o add new periodic calibration api o break up 5416 periodic calibration code in preparation for 928x o move get noise floor to rf backends o 5416-specific ani (still disabled) o modularize 5210 eeprom format a la other eeprom formats o start cleaning up regdomain code o prepare for proper 1/2 and 1/4 width channel support o bring back 900MHz card support o clean up 5212 rf version handling o add 1/2 and 1/4 width channel support for 5212 parts o split 5212 rfgain handling out o improve ani debugging o add AH_USE_INIPDGAIN compile option o purge a bunch of dead 5212 state o add 1/2 and 1/4 rate modes o remove HAL_CAP_CHAN_HALFRATE and HAL_CAP_CHAN_QUARTERRATE; the same info can now be deduced from the set of supported modes
2008-11-28 00:48:05 +00:00
* Copyright (c) 2002-2008 Atheros Communications, Inc.
*
* Permission to use, copy, modify, and/or distribute this software for any
* purpose with or without fee is hereby granted, provided that the above
* copyright notice and this permission notice appear in all copies.
*
* THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL WARRANTIES
* WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF
* MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR
* ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES
* WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN
* ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF
* OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
*
Overhaul regulatory support: o remove HAL_CHANNEL; convert the hal to use net80211 channels; this mostly involves mechanical changes to variable names and channel attribute macros o gut HAL_CHANNEL_PRIVATE as most of the contents are now redundant with the net80211 channel available o change api for ath_hal_init_channels: no more reglass id's, no more outdoor indication (was a noop), anM contents o add ath_hal_getchannels to have the hal construct a channel list without altering runtime state; this is used to retrieve the calibration list for the device in ath_getradiocaps o add ath_hal_set_channels to take a channel list and regulatory data from above and construct internal state to match (maps frequencies for 900MHz cards, setup for CTL lookups, etc) o compact the private channel table: we keep one private channel per frequency instead of one per HAL_CHANNEL; this gives a big space savings and potentially improves ani and calibration by sharing state (to be seen; didn't see anything in testing); a new config option AH_MAXCHAN controls the table size (default to 96 which was chosen to be ~3x the largest expected size) o shrink ani state and change to mirror private channel table (one entry per frequency indexed by ic_devdata) o move ani state flags to private channel state o remove country codes; use net80211 definitions instead o remove GSM regulatory support; it's no longer needed now that we pass in channel lists from above o consolidate ADHOC_NO_11A attribute with DISALLOW_ADHOC_11A o simplify initial channel list construction based on the EEPROM contents; we preserve country code support for now but may want to just fallback to a WWR sku and dispatch the discovered country code up to user space so the channel list can be constructed using the master regdomain tables o defer to net80211 for max antenna gain o eliminate sorting of internal channel table; now that we use ic_devdata as an index, table lookups are O(1) o remove internal copy of the country code; the public one is sufficient o remove AH_SUPPORT_11D conditional compilation; we always support 11d o remove ath_hal_ispublicsafetysku; not needed any more o remove ath_hal_isgsmsku; no more GSM stuff o move Conformance Test Limit (CTL) state from private channel to a lookup using per-band pointers cached in the private state block o remove regulatory class id support; was unused and belongs in net80211 o fix channel list construction to set IEEE80211_CHAN_NOADHOC, IEEE80211_CHAN_NOHOSTAP, and IEEE80211_CHAN_4MSXMIT o remove private channel flags CHANNEL_DFS and CHANNEL_4MS_LIMIT; these are now set in the constructed net80211 channel o store CHANNEL_NFCREQUIRED (Noise Floor Required) channel attribute in one of the driver-private flag bits of the net80211 channel o move 900MHz frequency mapping into the hal; the mapped frequency is stored in the private channel and used throughout the hal (no more mapping in the driver and/or net80211) o remove ath_hal_mhz2ieee; it's no longer needed as net80211 does the calculation and available in the net80211 channel o change noise floor calibration logic to work with compacted private channel table setup; this may require revisiting as we no longer can distinguish channel attributes (e.g. 11b vs 11g vs turbo) but since the data is used only to calculate status data we can live with it for now o change ah_getChipPowerLimits internal method to operate on a single channel instead of all channels in the private channel table o add ath_hal_gethwchannel to map a net80211 channel to a h/w frequency (always the same except for 900MHz channels) o add HAL_EEBADREG and HAL_EEBADCC status codes to better identify regulatory problems o remove CTRY_DEBUG and CTRY_DEFAULT enum's; these come from net80211 now o change ath_hal_getwirelessmodes to really return wireless modes supported by the hardware (was previously applying regulatory constraints) o return channel interference status with IEEE80211_CHANSTATE_CWINT (should change to a callback so hal api's can take const pointers) o remove some #define's no longer needed with the inclusion of <net80211/_ieee80211.h> Sponsored by: Carlson Wireless
2009-01-28 18:00:22 +00:00
* $FreeBSD$
Update to later code from my repository: o many bug fixes o add new periodic calibration api o break up 5416 periodic calibration code in preparation for 928x o move get noise floor to rf backends o 5416-specific ani (still disabled) o modularize 5210 eeprom format a la other eeprom formats o start cleaning up regdomain code o prepare for proper 1/2 and 1/4 width channel support o bring back 900MHz card support o clean up 5212 rf version handling o add 1/2 and 1/4 width channel support for 5212 parts o split 5212 rfgain handling out o improve ani debugging o add AH_USE_INIPDGAIN compile option o purge a bunch of dead 5212 state o add 1/2 and 1/4 rate modes o remove HAL_CAP_CHAN_HALFRATE and HAL_CAP_CHAN_QUARTERRATE; the same info can now be deduced from the set of supported modes
2008-11-28 00:48:05 +00:00
*/
#include "opt_ah.h"
/*
* XXX this is virtually the same code as for 5212; we reuse
* storage in the 5212 state block; need to refactor.
*/
#include "ah.h"
#include "ah_internal.h"
#include "ah_desc.h"
#include "ar5416/ar5416.h"
#include "ar5416/ar5416reg.h"
#include "ar5416/ar5416phy.h"
/*
* Anti noise immunity support. We track phy errors and react
* to excessive errors by adjusting the noise immunity parameters.
*/
#define HAL_EP_RND(x, mul) \
((((x)%(mul)) >= ((mul)/2)) ? ((x) + ((mul) - 1)) / (mul) : (x)/(mul))
#define BEACON_RSSI(ahp) \
HAL_EP_RND(ahp->ah_stats.ast_nodestats.ns_avgbrssi, \
HAL_RSSI_EP_MULTIPLIER)
/*
* ANI processing tunes radio parameters according to PHY errors
* and related information. This is done for for noise and spur
* immunity in all operating modes if the device indicates it's
* capable at attach time. In addition, when there is a reference
* rssi value (e.g. beacon frames from an ap in station mode)
* further tuning is done.
*
* ANI_ENA indicates whether any ANI processing should be done;
* this is specified at attach time.
*
* ANI_ENA_RSSI indicates whether rssi-based processing should
* done, this is enabled based on operating mode and is meaningful
* only if ANI_ENA is true.
*
* ANI parameters are typically controlled only by the hal. The
* AniControl interface however permits manual tuning through the
* diagnostic api.
*/
#define ANI_ENA(ah) \
(AH5212(ah)->ah_procPhyErr & HAL_ANI_ENA)
#define ANI_ENA_RSSI(ah) \
(AH5212(ah)->ah_procPhyErr & HAL_RSSI_ANI_ENA)
#define ah_mibStats ah_stats.ast_mibstats
static void
enableAniMIBCounters(struct ath_hal *ah, const struct ar5212AniParams *params)
{
struct ath_hal_5212 *ahp = AH5212(ah);
HALDEBUG(ah, HAL_DEBUG_ANI, "%s: Enable mib counters: "
"OfdmPhyErrBase 0x%x cckPhyErrBase 0x%x\n",
__func__, params->ofdmPhyErrBase, params->cckPhyErrBase);
OS_REG_WRITE(ah, AR_FILTOFDM, 0);
OS_REG_WRITE(ah, AR_FILTCCK, 0);
OS_REG_WRITE(ah, AR_PHYCNT1, params->ofdmPhyErrBase);
OS_REG_WRITE(ah, AR_PHYCNT2, params->cckPhyErrBase);
OS_REG_WRITE(ah, AR_PHY_ERR_MASK_1, AR_PHY_ERR_OFDM_TIMING);
OS_REG_WRITE(ah, AR_PHY_ERR_MASK_2, AR_PHY_ERR_CCK_TIMING);
ar5212UpdateMibCounters(ah, &ahp->ah_mibStats); /* save+clear counters*/
ar5212EnableMibCounters(ah); /* enable everything */
}
static void
disableAniMIBCounters(struct ath_hal *ah)
{
struct ath_hal_5212 *ahp = AH5212(ah);
HALDEBUG(ah, HAL_DEBUG_ANI, "Disable MIB counters\n");
ar5212UpdateMibCounters(ah, &ahp->ah_mibStats); /* save stats */
ar5212DisableMibCounters(ah); /* disable everything */
OS_REG_WRITE(ah, AR_PHY_ERR_MASK_1, 0);
OS_REG_WRITE(ah, AR_PHY_ERR_MASK_2, 0);
}
static void
setPhyErrBase(struct ath_hal *ah, struct ar5212AniParams *params)
{
if (params->ofdmTrigHigh >= AR_PHY_COUNTMAX) {
HALDEBUG(ah, HAL_DEBUG_ANY,
"OFDM Trigger %d is too high for hw counters, using max\n",
params->ofdmTrigHigh);
params->ofdmPhyErrBase = 0;
} else
params->ofdmPhyErrBase = AR_PHY_COUNTMAX - params->ofdmTrigHigh;
if (params->cckTrigHigh >= AR_PHY_COUNTMAX) {
HALDEBUG(ah, HAL_DEBUG_ANY,
"CCK Trigger %d is too high for hw counters, using max\n",
params->cckTrigHigh);
params->cckPhyErrBase = 0;
} else
params->cckPhyErrBase = AR_PHY_COUNTMAX - params->cckTrigHigh;
}
/*
* Setup ANI handling. Sets all thresholds and reset the
* channel statistics. Note that ar5416AniReset should be
* called by ar5416Reset before anything else happens and
* that's where we force initial settings.
*/
void
ar5416AniAttach(struct ath_hal *ah, const struct ar5212AniParams *params24,
const struct ar5212AniParams *params5, HAL_BOOL enable)
{
struct ath_hal_5212 *ahp = AH5212(ah);
if (params24 != AH_NULL) {
OS_MEMCPY(&ahp->ah_aniParams24, params24, sizeof(*params24));
setPhyErrBase(ah, &ahp->ah_aniParams24);
}
if (params5 != AH_NULL) {
OS_MEMCPY(&ahp->ah_aniParams5, params5, sizeof(*params5));
setPhyErrBase(ah, &ahp->ah_aniParams5);
}
OS_MEMZERO(ahp->ah_ani, sizeof(ahp->ah_ani));
/* Enable MIB Counters */
enableAniMIBCounters(ah, &ahp->ah_aniParams24 /*XXX*/);
if (enable) { /* Enable ani now */
HALASSERT(params24 != AH_NULL && params5 != AH_NULL);
ahp->ah_procPhyErr |= HAL_ANI_ENA;
} else {
ahp->ah_procPhyErr &= ~HAL_ANI_ENA;
}
}
/*
* Cleanup any ANI state setup.
*/
void
ar5416AniDetach(struct ath_hal *ah)
{
HALDEBUG(ah, HAL_DEBUG_ANI, "Detaching Ani\n");
disableAniMIBCounters(ah);
}
/*
* Control Adaptive Noise Immunity Parameters
*/
HAL_BOOL
ar5416AniControl(struct ath_hal *ah, HAL_ANI_CMD cmd, int param)
{
typedef int TABLE[];
struct ath_hal_5212 *ahp = AH5212(ah);
struct ar5212AniState *aniState = ahp->ah_curani;
const struct ar5212AniParams *params = aniState->params;
OS_MARK(ah, AH_MARK_ANI_CONTROL, cmd);
switch (cmd) {
case HAL_ANI_NOISE_IMMUNITY_LEVEL: {
u_int level = param;
if (level >= params->maxNoiseImmunityLevel) {
HALDEBUG(ah, HAL_DEBUG_ANY,
"%s: level out of range (%u > %u)\n",
__func__, level, params->maxNoiseImmunityLevel);
return AH_FALSE;
}
OS_REG_RMW_FIELD(ah, AR_PHY_DESIRED_SZ,
AR_PHY_DESIRED_SZ_TOT_DES, params->totalSizeDesired[level]);
OS_REG_RMW_FIELD(ah, AR_PHY_AGC_CTL1,
AR_PHY_AGC_CTL1_COARSE_LOW, params->coarseLow[level]);
OS_REG_RMW_FIELD(ah, AR_PHY_AGC_CTL1,
AR_PHY_AGC_CTL1_COARSE_HIGH, params->coarseHigh[level]);
OS_REG_RMW_FIELD(ah, AR_PHY_FIND_SIG,
AR_PHY_FIND_SIG_FIRPWR, params->firpwr[level]);
if (level > aniState->noiseImmunityLevel)
ahp->ah_stats.ast_ani_niup++;
else if (level < aniState->noiseImmunityLevel)
ahp->ah_stats.ast_ani_nidown++;
aniState->noiseImmunityLevel = level;
break;
}
case HAL_ANI_OFDM_WEAK_SIGNAL_DETECTION: {
static const TABLE m1ThreshLow = { 127, 50 };
static const TABLE m2ThreshLow = { 127, 40 };
static const TABLE m1Thresh = { 127, 0x4d };
static const TABLE m2Thresh = { 127, 0x40 };
static const TABLE m2CountThr = { 31, 16 };
static const TABLE m2CountThrLow = { 63, 48 };
u_int on = param ? 1 : 0;
OS_REG_RMW_FIELD(ah, AR_PHY_SFCORR_LOW,
AR_PHY_SFCORR_LOW_M1_THRESH_LOW, m1ThreshLow[on]);
OS_REG_RMW_FIELD(ah, AR_PHY_SFCORR_LOW,
AR_PHY_SFCORR_LOW_M2_THRESH_LOW, m2ThreshLow[on]);
OS_REG_RMW_FIELD(ah, AR_PHY_SFCORR,
AR_PHY_SFCORR_M1_THRESH, m1Thresh[on]);
OS_REG_RMW_FIELD(ah, AR_PHY_SFCORR,
AR_PHY_SFCORR_M2_THRESH, m2Thresh[on]);
OS_REG_RMW_FIELD(ah, AR_PHY_SFCORR,
AR_PHY_SFCORR_M2COUNT_THR, m2CountThr[on]);
OS_REG_RMW_FIELD(ah, AR_PHY_SFCORR_LOW,
AR_PHY_SFCORR_LOW_M2COUNT_THR_LOW, m2CountThrLow[on]);
OS_REG_RMW_FIELD(ah, AR_PHY_SFCORR_EXT,
AR_PHY_SFCORR_EXT_M1_THRESH_LOW, m1ThreshLow[on]);
OS_REG_RMW_FIELD(ah, AR_PHY_SFCORR_EXT,
AR_PHY_SFCORR_EXT_M2_THRESH_LOW, m2ThreshLow[on]);
OS_REG_RMW_FIELD(ah, AR_PHY_SFCORR_EXT,
AR_PHY_SFCORR_EXT_M1_THRESH, m1Thresh[on]);
OS_REG_RMW_FIELD(ah, AR_PHY_SFCORR_EXT,
AR_PHY_SFCORR_EXT_M2_THRESH, m2Thresh[on]);
if (on) {
OS_REG_SET_BIT(ah, AR_PHY_SFCORR_LOW,
AR_PHY_SFCORR_LOW_USE_SELF_CORR_LOW);
} else {
OS_REG_CLR_BIT(ah, AR_PHY_SFCORR_LOW,
AR_PHY_SFCORR_LOW_USE_SELF_CORR_LOW);
}
if (on)
ahp->ah_stats.ast_ani_ofdmon++;
else
ahp->ah_stats.ast_ani_ofdmoff++;
aniState->ofdmWeakSigDetectOff = !on;
break;
}
case HAL_ANI_CCK_WEAK_SIGNAL_THR: {
static const TABLE weakSigThrCck = { 8, 6 };
u_int high = param ? 1 : 0;
OS_REG_RMW_FIELD(ah, AR_PHY_CCK_DETECT,
AR_PHY_CCK_DETECT_WEAK_SIG_THR_CCK, weakSigThrCck[high]);
if (high)
ahp->ah_stats.ast_ani_cckhigh++;
else
ahp->ah_stats.ast_ani_ccklow++;
aniState->cckWeakSigThreshold = high;
break;
}
case HAL_ANI_FIRSTEP_LEVEL: {
u_int level = param;
if (level >= params->maxFirstepLevel) {
HALDEBUG(ah, HAL_DEBUG_ANY,
"%s: level out of range (%u > %u)\n",
__func__, level, params->maxFirstepLevel);
return AH_FALSE;
}
OS_REG_RMW_FIELD(ah, AR_PHY_FIND_SIG,
AR_PHY_FIND_SIG_FIRSTEP, params->firstep[level]);
if (level > aniState->firstepLevel)
ahp->ah_stats.ast_ani_stepup++;
else if (level < aniState->firstepLevel)
ahp->ah_stats.ast_ani_stepdown++;
aniState->firstepLevel = level;
break;
}
case HAL_ANI_SPUR_IMMUNITY_LEVEL: {
u_int level = param;
if (level >= params->maxSpurImmunityLevel) {
HALDEBUG(ah, HAL_DEBUG_ANY,
"%s: level out of range (%u > %u)\n",
__func__, level, params->maxSpurImmunityLevel);
return AH_FALSE;
}
OS_REG_RMW_FIELD(ah, AR_PHY_TIMING5,
AR_PHY_TIMING5_CYCPWR_THR1, params->cycPwrThr1[level]);
if (level > aniState->spurImmunityLevel)
ahp->ah_stats.ast_ani_spurup++;
else if (level < aniState->spurImmunityLevel)
ahp->ah_stats.ast_ani_spurdown++;
aniState->spurImmunityLevel = level;
break;
}
case HAL_ANI_PRESENT:
break;
case HAL_ANI_MODE:
if (param == 0) {
ahp->ah_procPhyErr &= ~HAL_ANI_ENA;
/* Turn off HW counters if we have them */
ar5416AniDetach(ah);
ar5212SetRxFilter(ah,
ar5212GetRxFilter(ah) &~ HAL_RX_FILTER_PHYERR);
} else { /* normal/auto mode */
/* don't mess with state if already enabled */
if (ahp->ah_procPhyErr & HAL_ANI_ENA)
break;
ar5212SetRxFilter(ah,
ar5212GetRxFilter(ah) &~ HAL_RX_FILTER_PHYERR);
/* Enable MIB Counters */
enableAniMIBCounters(ah, ahp->ah_curani != AH_NULL ?
ahp->ah_curani->params: &ahp->ah_aniParams24 /*XXX*/);
ahp->ah_procPhyErr |= HAL_ANI_ENA;
}
break;
#ifdef AH_PRIVATE_DIAG
case HAL_ANI_PHYERR_RESET:
ahp->ah_stats.ast_ani_ofdmerrs = 0;
ahp->ah_stats.ast_ani_cckerrs = 0;
break;
#endif /* AH_PRIVATE_DIAG */
default:
HALDEBUG(ah, HAL_DEBUG_ANY, "%s: invalid cmd %u\n",
__func__, cmd);
return AH_FALSE;
}
return AH_TRUE;
}
static void
ar5416AniOfdmErrTrigger(struct ath_hal *ah)
{
struct ath_hal_5212 *ahp = AH5212(ah);
Overhaul regulatory support: o remove HAL_CHANNEL; convert the hal to use net80211 channels; this mostly involves mechanical changes to variable names and channel attribute macros o gut HAL_CHANNEL_PRIVATE as most of the contents are now redundant with the net80211 channel available o change api for ath_hal_init_channels: no more reglass id's, no more outdoor indication (was a noop), anM contents o add ath_hal_getchannels to have the hal construct a channel list without altering runtime state; this is used to retrieve the calibration list for the device in ath_getradiocaps o add ath_hal_set_channels to take a channel list and regulatory data from above and construct internal state to match (maps frequencies for 900MHz cards, setup for CTL lookups, etc) o compact the private channel table: we keep one private channel per frequency instead of one per HAL_CHANNEL; this gives a big space savings and potentially improves ani and calibration by sharing state (to be seen; didn't see anything in testing); a new config option AH_MAXCHAN controls the table size (default to 96 which was chosen to be ~3x the largest expected size) o shrink ani state and change to mirror private channel table (one entry per frequency indexed by ic_devdata) o move ani state flags to private channel state o remove country codes; use net80211 definitions instead o remove GSM regulatory support; it's no longer needed now that we pass in channel lists from above o consolidate ADHOC_NO_11A attribute with DISALLOW_ADHOC_11A o simplify initial channel list construction based on the EEPROM contents; we preserve country code support for now but may want to just fallback to a WWR sku and dispatch the discovered country code up to user space so the channel list can be constructed using the master regdomain tables o defer to net80211 for max antenna gain o eliminate sorting of internal channel table; now that we use ic_devdata as an index, table lookups are O(1) o remove internal copy of the country code; the public one is sufficient o remove AH_SUPPORT_11D conditional compilation; we always support 11d o remove ath_hal_ispublicsafetysku; not needed any more o remove ath_hal_isgsmsku; no more GSM stuff o move Conformance Test Limit (CTL) state from private channel to a lookup using per-band pointers cached in the private state block o remove regulatory class id support; was unused and belongs in net80211 o fix channel list construction to set IEEE80211_CHAN_NOADHOC, IEEE80211_CHAN_NOHOSTAP, and IEEE80211_CHAN_4MSXMIT o remove private channel flags CHANNEL_DFS and CHANNEL_4MS_LIMIT; these are now set in the constructed net80211 channel o store CHANNEL_NFCREQUIRED (Noise Floor Required) channel attribute in one of the driver-private flag bits of the net80211 channel o move 900MHz frequency mapping into the hal; the mapped frequency is stored in the private channel and used throughout the hal (no more mapping in the driver and/or net80211) o remove ath_hal_mhz2ieee; it's no longer needed as net80211 does the calculation and available in the net80211 channel o change noise floor calibration logic to work with compacted private channel table setup; this may require revisiting as we no longer can distinguish channel attributes (e.g. 11b vs 11g vs turbo) but since the data is used only to calculate status data we can live with it for now o change ah_getChipPowerLimits internal method to operate on a single channel instead of all channels in the private channel table o add ath_hal_gethwchannel to map a net80211 channel to a h/w frequency (always the same except for 900MHz channels) o add HAL_EEBADREG and HAL_EEBADCC status codes to better identify regulatory problems o remove CTRY_DEBUG and CTRY_DEFAULT enum's; these come from net80211 now o change ath_hal_getwirelessmodes to really return wireless modes supported by the hardware (was previously applying regulatory constraints) o return channel interference status with IEEE80211_CHANSTATE_CWINT (should change to a callback so hal api's can take const pointers) o remove some #define's no longer needed with the inclusion of <net80211/_ieee80211.h> Sponsored by: Carlson Wireless
2009-01-28 18:00:22 +00:00
const struct ieee80211_channel *chan = AH_PRIVATE(ah)->ah_curchan;
Update to later code from my repository: o many bug fixes o add new periodic calibration api o break up 5416 periodic calibration code in preparation for 928x o move get noise floor to rf backends o 5416-specific ani (still disabled) o modularize 5210 eeprom format a la other eeprom formats o start cleaning up regdomain code o prepare for proper 1/2 and 1/4 width channel support o bring back 900MHz card support o clean up 5212 rf version handling o add 1/2 and 1/4 width channel support for 5212 parts o split 5212 rfgain handling out o improve ani debugging o add AH_USE_INIPDGAIN compile option o purge a bunch of dead 5212 state o add 1/2 and 1/4 rate modes o remove HAL_CAP_CHAN_HALFRATE and HAL_CAP_CHAN_QUARTERRATE; the same info can now be deduced from the set of supported modes
2008-11-28 00:48:05 +00:00
struct ar5212AniState *aniState;
const struct ar5212AniParams *params;
HALASSERT(chan != AH_NULL);
if (!ANI_ENA(ah))
return;
aniState = ahp->ah_curani;
params = aniState->params;
/* First, raise noise immunity level, up to max */
if (aniState->noiseImmunityLevel+1 < params->maxNoiseImmunityLevel) {
ar5416AniControl(ah, HAL_ANI_NOISE_IMMUNITY_LEVEL,
aniState->noiseImmunityLevel + 1);
return;
}
/* then, raise spur immunity level, up to max */
if (aniState->spurImmunityLevel+1 < params->maxSpurImmunityLevel) {
ar5416AniControl(ah, HAL_ANI_SPUR_IMMUNITY_LEVEL,
aniState->spurImmunityLevel + 1);
return;
}
if (ANI_ENA_RSSI(ah)) {
int32_t rssi = BEACON_RSSI(ahp);
if (rssi > params->rssiThrHigh) {
/*
* Beacon rssi is high, can turn off ofdm
* weak sig detect.
*/
if (!aniState->ofdmWeakSigDetectOff) {
ar5416AniControl(ah,
HAL_ANI_OFDM_WEAK_SIGNAL_DETECTION,
AH_FALSE);
ar5416AniControl(ah,
HAL_ANI_SPUR_IMMUNITY_LEVEL, 0);
return;
}
/*
* If weak sig detect is already off, as last resort,
* raise firstep level
*/
if (aniState->firstepLevel+1 < params->maxFirstepLevel) {
ar5416AniControl(ah, HAL_ANI_FIRSTEP_LEVEL,
aniState->firstepLevel + 1);
return;
}
} else if (rssi > params->rssiThrLow) {
/*
* Beacon rssi in mid range, need ofdm weak signal
* detect, but we can raise firststepLevel.
*/
if (aniState->ofdmWeakSigDetectOff)
ar5416AniControl(ah,
HAL_ANI_OFDM_WEAK_SIGNAL_DETECTION,
AH_TRUE);
if (aniState->firstepLevel+1 < params->maxFirstepLevel)
ar5416AniControl(ah, HAL_ANI_FIRSTEP_LEVEL,
aniState->firstepLevel + 1);
return;
} else {
/*
* Beacon rssi is low, if in 11b/g mode, turn off ofdm
* weak signal detection and zero firstepLevel to
* maximize CCK sensitivity
*/
Overhaul regulatory support: o remove HAL_CHANNEL; convert the hal to use net80211 channels; this mostly involves mechanical changes to variable names and channel attribute macros o gut HAL_CHANNEL_PRIVATE as most of the contents are now redundant with the net80211 channel available o change api for ath_hal_init_channels: no more reglass id's, no more outdoor indication (was a noop), anM contents o add ath_hal_getchannels to have the hal construct a channel list without altering runtime state; this is used to retrieve the calibration list for the device in ath_getradiocaps o add ath_hal_set_channels to take a channel list and regulatory data from above and construct internal state to match (maps frequencies for 900MHz cards, setup for CTL lookups, etc) o compact the private channel table: we keep one private channel per frequency instead of one per HAL_CHANNEL; this gives a big space savings and potentially improves ani and calibration by sharing state (to be seen; didn't see anything in testing); a new config option AH_MAXCHAN controls the table size (default to 96 which was chosen to be ~3x the largest expected size) o shrink ani state and change to mirror private channel table (one entry per frequency indexed by ic_devdata) o move ani state flags to private channel state o remove country codes; use net80211 definitions instead o remove GSM regulatory support; it's no longer needed now that we pass in channel lists from above o consolidate ADHOC_NO_11A attribute with DISALLOW_ADHOC_11A o simplify initial channel list construction based on the EEPROM contents; we preserve country code support for now but may want to just fallback to a WWR sku and dispatch the discovered country code up to user space so the channel list can be constructed using the master regdomain tables o defer to net80211 for max antenna gain o eliminate sorting of internal channel table; now that we use ic_devdata as an index, table lookups are O(1) o remove internal copy of the country code; the public one is sufficient o remove AH_SUPPORT_11D conditional compilation; we always support 11d o remove ath_hal_ispublicsafetysku; not needed any more o remove ath_hal_isgsmsku; no more GSM stuff o move Conformance Test Limit (CTL) state from private channel to a lookup using per-band pointers cached in the private state block o remove regulatory class id support; was unused and belongs in net80211 o fix channel list construction to set IEEE80211_CHAN_NOADHOC, IEEE80211_CHAN_NOHOSTAP, and IEEE80211_CHAN_4MSXMIT o remove private channel flags CHANNEL_DFS and CHANNEL_4MS_LIMIT; these are now set in the constructed net80211 channel o store CHANNEL_NFCREQUIRED (Noise Floor Required) channel attribute in one of the driver-private flag bits of the net80211 channel o move 900MHz frequency mapping into the hal; the mapped frequency is stored in the private channel and used throughout the hal (no more mapping in the driver and/or net80211) o remove ath_hal_mhz2ieee; it's no longer needed as net80211 does the calculation and available in the net80211 channel o change noise floor calibration logic to work with compacted private channel table setup; this may require revisiting as we no longer can distinguish channel attributes (e.g. 11b vs 11g vs turbo) but since the data is used only to calculate status data we can live with it for now o change ah_getChipPowerLimits internal method to operate on a single channel instead of all channels in the private channel table o add ath_hal_gethwchannel to map a net80211 channel to a h/w frequency (always the same except for 900MHz channels) o add HAL_EEBADREG and HAL_EEBADCC status codes to better identify regulatory problems o remove CTRY_DEBUG and CTRY_DEFAULT enum's; these come from net80211 now o change ath_hal_getwirelessmodes to really return wireless modes supported by the hardware (was previously applying regulatory constraints) o return channel interference status with IEEE80211_CHANSTATE_CWINT (should change to a callback so hal api's can take const pointers) o remove some #define's no longer needed with the inclusion of <net80211/_ieee80211.h> Sponsored by: Carlson Wireless
2009-01-28 18:00:22 +00:00
if (IEEE80211_IS_CHAN_CCK(chan)) {
Update to later code from my repository: o many bug fixes o add new periodic calibration api o break up 5416 periodic calibration code in preparation for 928x o move get noise floor to rf backends o 5416-specific ani (still disabled) o modularize 5210 eeprom format a la other eeprom formats o start cleaning up regdomain code o prepare for proper 1/2 and 1/4 width channel support o bring back 900MHz card support o clean up 5212 rf version handling o add 1/2 and 1/4 width channel support for 5212 parts o split 5212 rfgain handling out o improve ani debugging o add AH_USE_INIPDGAIN compile option o purge a bunch of dead 5212 state o add 1/2 and 1/4 rate modes o remove HAL_CAP_CHAN_HALFRATE and HAL_CAP_CHAN_QUARTERRATE; the same info can now be deduced from the set of supported modes
2008-11-28 00:48:05 +00:00
if (!aniState->ofdmWeakSigDetectOff)
ar5416AniControl(ah,
HAL_ANI_OFDM_WEAK_SIGNAL_DETECTION,
AH_FALSE);
if (aniState->firstepLevel > 0)
ar5416AniControl(ah,
HAL_ANI_FIRSTEP_LEVEL, 0);
return;
}
}
}
}
static void
ar5416AniCckErrTrigger(struct ath_hal *ah)
{
struct ath_hal_5212 *ahp = AH5212(ah);
Overhaul regulatory support: o remove HAL_CHANNEL; convert the hal to use net80211 channels; this mostly involves mechanical changes to variable names and channel attribute macros o gut HAL_CHANNEL_PRIVATE as most of the contents are now redundant with the net80211 channel available o change api for ath_hal_init_channels: no more reglass id's, no more outdoor indication (was a noop), anM contents o add ath_hal_getchannels to have the hal construct a channel list without altering runtime state; this is used to retrieve the calibration list for the device in ath_getradiocaps o add ath_hal_set_channels to take a channel list and regulatory data from above and construct internal state to match (maps frequencies for 900MHz cards, setup for CTL lookups, etc) o compact the private channel table: we keep one private channel per frequency instead of one per HAL_CHANNEL; this gives a big space savings and potentially improves ani and calibration by sharing state (to be seen; didn't see anything in testing); a new config option AH_MAXCHAN controls the table size (default to 96 which was chosen to be ~3x the largest expected size) o shrink ani state and change to mirror private channel table (one entry per frequency indexed by ic_devdata) o move ani state flags to private channel state o remove country codes; use net80211 definitions instead o remove GSM regulatory support; it's no longer needed now that we pass in channel lists from above o consolidate ADHOC_NO_11A attribute with DISALLOW_ADHOC_11A o simplify initial channel list construction based on the EEPROM contents; we preserve country code support for now but may want to just fallback to a WWR sku and dispatch the discovered country code up to user space so the channel list can be constructed using the master regdomain tables o defer to net80211 for max antenna gain o eliminate sorting of internal channel table; now that we use ic_devdata as an index, table lookups are O(1) o remove internal copy of the country code; the public one is sufficient o remove AH_SUPPORT_11D conditional compilation; we always support 11d o remove ath_hal_ispublicsafetysku; not needed any more o remove ath_hal_isgsmsku; no more GSM stuff o move Conformance Test Limit (CTL) state from private channel to a lookup using per-band pointers cached in the private state block o remove regulatory class id support; was unused and belongs in net80211 o fix channel list construction to set IEEE80211_CHAN_NOADHOC, IEEE80211_CHAN_NOHOSTAP, and IEEE80211_CHAN_4MSXMIT o remove private channel flags CHANNEL_DFS and CHANNEL_4MS_LIMIT; these are now set in the constructed net80211 channel o store CHANNEL_NFCREQUIRED (Noise Floor Required) channel attribute in one of the driver-private flag bits of the net80211 channel o move 900MHz frequency mapping into the hal; the mapped frequency is stored in the private channel and used throughout the hal (no more mapping in the driver and/or net80211) o remove ath_hal_mhz2ieee; it's no longer needed as net80211 does the calculation and available in the net80211 channel o change noise floor calibration logic to work with compacted private channel table setup; this may require revisiting as we no longer can distinguish channel attributes (e.g. 11b vs 11g vs turbo) but since the data is used only to calculate status data we can live with it for now o change ah_getChipPowerLimits internal method to operate on a single channel instead of all channels in the private channel table o add ath_hal_gethwchannel to map a net80211 channel to a h/w frequency (always the same except for 900MHz channels) o add HAL_EEBADREG and HAL_EEBADCC status codes to better identify regulatory problems o remove CTRY_DEBUG and CTRY_DEFAULT enum's; these come from net80211 now o change ath_hal_getwirelessmodes to really return wireless modes supported by the hardware (was previously applying regulatory constraints) o return channel interference status with IEEE80211_CHANSTATE_CWINT (should change to a callback so hal api's can take const pointers) o remove some #define's no longer needed with the inclusion of <net80211/_ieee80211.h> Sponsored by: Carlson Wireless
2009-01-28 18:00:22 +00:00
const struct ieee80211_channel *chan = AH_PRIVATE(ah)->ah_curchan;
Update to later code from my repository: o many bug fixes o add new periodic calibration api o break up 5416 periodic calibration code in preparation for 928x o move get noise floor to rf backends o 5416-specific ani (still disabled) o modularize 5210 eeprom format a la other eeprom formats o start cleaning up regdomain code o prepare for proper 1/2 and 1/4 width channel support o bring back 900MHz card support o clean up 5212 rf version handling o add 1/2 and 1/4 width channel support for 5212 parts o split 5212 rfgain handling out o improve ani debugging o add AH_USE_INIPDGAIN compile option o purge a bunch of dead 5212 state o add 1/2 and 1/4 rate modes o remove HAL_CAP_CHAN_HALFRATE and HAL_CAP_CHAN_QUARTERRATE; the same info can now be deduced from the set of supported modes
2008-11-28 00:48:05 +00:00
struct ar5212AniState *aniState;
const struct ar5212AniParams *params;
HALASSERT(chan != AH_NULL);
if (!ANI_ENA(ah))
return;
/* first, raise noise immunity level, up to max */
aniState = ahp->ah_curani;
params = aniState->params;
if (aniState->noiseImmunityLevel+1 < params->maxNoiseImmunityLevel) {
ar5416AniControl(ah, HAL_ANI_NOISE_IMMUNITY_LEVEL,
aniState->noiseImmunityLevel + 1);
return;
}
if (ANI_ENA_RSSI(ah)) {
int32_t rssi = BEACON_RSSI(ahp);
if (rssi > params->rssiThrLow) {
/*
* Beacon signal in mid and high range,
* raise firstep level.
*/
if (aniState->firstepLevel+1 < params->maxFirstepLevel)
ar5416AniControl(ah, HAL_ANI_FIRSTEP_LEVEL,
aniState->firstepLevel + 1);
} else {
/*
* Beacon rssi is low, zero firstep level to maximize
* CCK sensitivity in 11b/g mode.
*/
Overhaul regulatory support: o remove HAL_CHANNEL; convert the hal to use net80211 channels; this mostly involves mechanical changes to variable names and channel attribute macros o gut HAL_CHANNEL_PRIVATE as most of the contents are now redundant with the net80211 channel available o change api for ath_hal_init_channels: no more reglass id's, no more outdoor indication (was a noop), anM contents o add ath_hal_getchannels to have the hal construct a channel list without altering runtime state; this is used to retrieve the calibration list for the device in ath_getradiocaps o add ath_hal_set_channels to take a channel list and regulatory data from above and construct internal state to match (maps frequencies for 900MHz cards, setup for CTL lookups, etc) o compact the private channel table: we keep one private channel per frequency instead of one per HAL_CHANNEL; this gives a big space savings and potentially improves ani and calibration by sharing state (to be seen; didn't see anything in testing); a new config option AH_MAXCHAN controls the table size (default to 96 which was chosen to be ~3x the largest expected size) o shrink ani state and change to mirror private channel table (one entry per frequency indexed by ic_devdata) o move ani state flags to private channel state o remove country codes; use net80211 definitions instead o remove GSM regulatory support; it's no longer needed now that we pass in channel lists from above o consolidate ADHOC_NO_11A attribute with DISALLOW_ADHOC_11A o simplify initial channel list construction based on the EEPROM contents; we preserve country code support for now but may want to just fallback to a WWR sku and dispatch the discovered country code up to user space so the channel list can be constructed using the master regdomain tables o defer to net80211 for max antenna gain o eliminate sorting of internal channel table; now that we use ic_devdata as an index, table lookups are O(1) o remove internal copy of the country code; the public one is sufficient o remove AH_SUPPORT_11D conditional compilation; we always support 11d o remove ath_hal_ispublicsafetysku; not needed any more o remove ath_hal_isgsmsku; no more GSM stuff o move Conformance Test Limit (CTL) state from private channel to a lookup using per-band pointers cached in the private state block o remove regulatory class id support; was unused and belongs in net80211 o fix channel list construction to set IEEE80211_CHAN_NOADHOC, IEEE80211_CHAN_NOHOSTAP, and IEEE80211_CHAN_4MSXMIT o remove private channel flags CHANNEL_DFS and CHANNEL_4MS_LIMIT; these are now set in the constructed net80211 channel o store CHANNEL_NFCREQUIRED (Noise Floor Required) channel attribute in one of the driver-private flag bits of the net80211 channel o move 900MHz frequency mapping into the hal; the mapped frequency is stored in the private channel and used throughout the hal (no more mapping in the driver and/or net80211) o remove ath_hal_mhz2ieee; it's no longer needed as net80211 does the calculation and available in the net80211 channel o change noise floor calibration logic to work with compacted private channel table setup; this may require revisiting as we no longer can distinguish channel attributes (e.g. 11b vs 11g vs turbo) but since the data is used only to calculate status data we can live with it for now o change ah_getChipPowerLimits internal method to operate on a single channel instead of all channels in the private channel table o add ath_hal_gethwchannel to map a net80211 channel to a h/w frequency (always the same except for 900MHz channels) o add HAL_EEBADREG and HAL_EEBADCC status codes to better identify regulatory problems o remove CTRY_DEBUG and CTRY_DEFAULT enum's; these come from net80211 now o change ath_hal_getwirelessmodes to really return wireless modes supported by the hardware (was previously applying regulatory constraints) o return channel interference status with IEEE80211_CHANSTATE_CWINT (should change to a callback so hal api's can take const pointers) o remove some #define's no longer needed with the inclusion of <net80211/_ieee80211.h> Sponsored by: Carlson Wireless
2009-01-28 18:00:22 +00:00
if (IEEE80211_IS_CHAN_CCK(chan)) {
Update to later code from my repository: o many bug fixes o add new periodic calibration api o break up 5416 periodic calibration code in preparation for 928x o move get noise floor to rf backends o 5416-specific ani (still disabled) o modularize 5210 eeprom format a la other eeprom formats o start cleaning up regdomain code o prepare for proper 1/2 and 1/4 width channel support o bring back 900MHz card support o clean up 5212 rf version handling o add 1/2 and 1/4 width channel support for 5212 parts o split 5212 rfgain handling out o improve ani debugging o add AH_USE_INIPDGAIN compile option o purge a bunch of dead 5212 state o add 1/2 and 1/4 rate modes o remove HAL_CAP_CHAN_HALFRATE and HAL_CAP_CHAN_QUARTERRATE; the same info can now be deduced from the set of supported modes
2008-11-28 00:48:05 +00:00
if (aniState->firstepLevel > 0)
ar5416AniControl(ah,
HAL_ANI_FIRSTEP_LEVEL, 0);
}
}
}
}
static void
ar5416AniRestart(struct ath_hal *ah, struct ar5212AniState *aniState)
{
struct ath_hal_5212 *ahp = AH5212(ah);
const struct ar5212AniParams *params = aniState->params;
aniState->listenTime = 0;
/*
* NB: these are written on reset based on the
* ini so we must re-write them!
*/
HALDEBUG(ah, HAL_DEBUG_ANI,
"%s: Writing ofdmbase=%u cckbase=%u\n", __func__,
params->ofdmPhyErrBase, params->cckPhyErrBase);
OS_REG_WRITE(ah, AR_PHY_ERR_1, params->ofdmPhyErrBase);
OS_REG_WRITE(ah, AR_PHY_ERR_2, params->cckPhyErrBase);
OS_REG_WRITE(ah, AR_PHY_ERR_MASK_1, AR_PHY_ERR_OFDM_TIMING);
OS_REG_WRITE(ah, AR_PHY_ERR_MASK_1, AR_PHY_ERR_CCK_TIMING);
/* Clear the mib counters and save them in the stats */
ar5212UpdateMibCounters(ah, &ahp->ah_mibStats);
aniState->ofdmPhyErrCount = 0;
aniState->cckPhyErrCount = 0;
}
/*
* Restore/reset the ANI parameters and reset the statistics.
* This routine must be called for every channel change.
*
* NOTE: This is where ah_curani is set; other ani code assumes
* it is setup to reflect the current channel.
*/
void
Overhaul regulatory support: o remove HAL_CHANNEL; convert the hal to use net80211 channels; this mostly involves mechanical changes to variable names and channel attribute macros o gut HAL_CHANNEL_PRIVATE as most of the contents are now redundant with the net80211 channel available o change api for ath_hal_init_channels: no more reglass id's, no more outdoor indication (was a noop), anM contents o add ath_hal_getchannels to have the hal construct a channel list without altering runtime state; this is used to retrieve the calibration list for the device in ath_getradiocaps o add ath_hal_set_channels to take a channel list and regulatory data from above and construct internal state to match (maps frequencies for 900MHz cards, setup for CTL lookups, etc) o compact the private channel table: we keep one private channel per frequency instead of one per HAL_CHANNEL; this gives a big space savings and potentially improves ani and calibration by sharing state (to be seen; didn't see anything in testing); a new config option AH_MAXCHAN controls the table size (default to 96 which was chosen to be ~3x the largest expected size) o shrink ani state and change to mirror private channel table (one entry per frequency indexed by ic_devdata) o move ani state flags to private channel state o remove country codes; use net80211 definitions instead o remove GSM regulatory support; it's no longer needed now that we pass in channel lists from above o consolidate ADHOC_NO_11A attribute with DISALLOW_ADHOC_11A o simplify initial channel list construction based on the EEPROM contents; we preserve country code support for now but may want to just fallback to a WWR sku and dispatch the discovered country code up to user space so the channel list can be constructed using the master regdomain tables o defer to net80211 for max antenna gain o eliminate sorting of internal channel table; now that we use ic_devdata as an index, table lookups are O(1) o remove internal copy of the country code; the public one is sufficient o remove AH_SUPPORT_11D conditional compilation; we always support 11d o remove ath_hal_ispublicsafetysku; not needed any more o remove ath_hal_isgsmsku; no more GSM stuff o move Conformance Test Limit (CTL) state from private channel to a lookup using per-band pointers cached in the private state block o remove regulatory class id support; was unused and belongs in net80211 o fix channel list construction to set IEEE80211_CHAN_NOADHOC, IEEE80211_CHAN_NOHOSTAP, and IEEE80211_CHAN_4MSXMIT o remove private channel flags CHANNEL_DFS and CHANNEL_4MS_LIMIT; these are now set in the constructed net80211 channel o store CHANNEL_NFCREQUIRED (Noise Floor Required) channel attribute in one of the driver-private flag bits of the net80211 channel o move 900MHz frequency mapping into the hal; the mapped frequency is stored in the private channel and used throughout the hal (no more mapping in the driver and/or net80211) o remove ath_hal_mhz2ieee; it's no longer needed as net80211 does the calculation and available in the net80211 channel o change noise floor calibration logic to work with compacted private channel table setup; this may require revisiting as we no longer can distinguish channel attributes (e.g. 11b vs 11g vs turbo) but since the data is used only to calculate status data we can live with it for now o change ah_getChipPowerLimits internal method to operate on a single channel instead of all channels in the private channel table o add ath_hal_gethwchannel to map a net80211 channel to a h/w frequency (always the same except for 900MHz channels) o add HAL_EEBADREG and HAL_EEBADCC status codes to better identify regulatory problems o remove CTRY_DEBUG and CTRY_DEFAULT enum's; these come from net80211 now o change ath_hal_getwirelessmodes to really return wireless modes supported by the hardware (was previously applying regulatory constraints) o return channel interference status with IEEE80211_CHANSTATE_CWINT (should change to a callback so hal api's can take const pointers) o remove some #define's no longer needed with the inclusion of <net80211/_ieee80211.h> Sponsored by: Carlson Wireless
2009-01-28 18:00:22 +00:00
ar5416AniReset(struct ath_hal *ah, const struct ieee80211_channel *chan,
Update to later code from my repository: o many bug fixes o add new periodic calibration api o break up 5416 periodic calibration code in preparation for 928x o move get noise floor to rf backends o 5416-specific ani (still disabled) o modularize 5210 eeprom format a la other eeprom formats o start cleaning up regdomain code o prepare for proper 1/2 and 1/4 width channel support o bring back 900MHz card support o clean up 5212 rf version handling o add 1/2 and 1/4 width channel support for 5212 parts o split 5212 rfgain handling out o improve ani debugging o add AH_USE_INIPDGAIN compile option o purge a bunch of dead 5212 state o add 1/2 and 1/4 rate modes o remove HAL_CAP_CHAN_HALFRATE and HAL_CAP_CHAN_QUARTERRATE; the same info can now be deduced from the set of supported modes
2008-11-28 00:48:05 +00:00
HAL_OPMODE opmode, int restore)
{
struct ath_hal_5212 *ahp = AH5212(ah);
Overhaul regulatory support: o remove HAL_CHANNEL; convert the hal to use net80211 channels; this mostly involves mechanical changes to variable names and channel attribute macros o gut HAL_CHANNEL_PRIVATE as most of the contents are now redundant with the net80211 channel available o change api for ath_hal_init_channels: no more reglass id's, no more outdoor indication (was a noop), anM contents o add ath_hal_getchannels to have the hal construct a channel list without altering runtime state; this is used to retrieve the calibration list for the device in ath_getradiocaps o add ath_hal_set_channels to take a channel list and regulatory data from above and construct internal state to match (maps frequencies for 900MHz cards, setup for CTL lookups, etc) o compact the private channel table: we keep one private channel per frequency instead of one per HAL_CHANNEL; this gives a big space savings and potentially improves ani and calibration by sharing state (to be seen; didn't see anything in testing); a new config option AH_MAXCHAN controls the table size (default to 96 which was chosen to be ~3x the largest expected size) o shrink ani state and change to mirror private channel table (one entry per frequency indexed by ic_devdata) o move ani state flags to private channel state o remove country codes; use net80211 definitions instead o remove GSM regulatory support; it's no longer needed now that we pass in channel lists from above o consolidate ADHOC_NO_11A attribute with DISALLOW_ADHOC_11A o simplify initial channel list construction based on the EEPROM contents; we preserve country code support for now but may want to just fallback to a WWR sku and dispatch the discovered country code up to user space so the channel list can be constructed using the master regdomain tables o defer to net80211 for max antenna gain o eliminate sorting of internal channel table; now that we use ic_devdata as an index, table lookups are O(1) o remove internal copy of the country code; the public one is sufficient o remove AH_SUPPORT_11D conditional compilation; we always support 11d o remove ath_hal_ispublicsafetysku; not needed any more o remove ath_hal_isgsmsku; no more GSM stuff o move Conformance Test Limit (CTL) state from private channel to a lookup using per-band pointers cached in the private state block o remove regulatory class id support; was unused and belongs in net80211 o fix channel list construction to set IEEE80211_CHAN_NOADHOC, IEEE80211_CHAN_NOHOSTAP, and IEEE80211_CHAN_4MSXMIT o remove private channel flags CHANNEL_DFS and CHANNEL_4MS_LIMIT; these are now set in the constructed net80211 channel o store CHANNEL_NFCREQUIRED (Noise Floor Required) channel attribute in one of the driver-private flag bits of the net80211 channel o move 900MHz frequency mapping into the hal; the mapped frequency is stored in the private channel and used throughout the hal (no more mapping in the driver and/or net80211) o remove ath_hal_mhz2ieee; it's no longer needed as net80211 does the calculation and available in the net80211 channel o change noise floor calibration logic to work with compacted private channel table setup; this may require revisiting as we no longer can distinguish channel attributes (e.g. 11b vs 11g vs turbo) but since the data is used only to calculate status data we can live with it for now o change ah_getChipPowerLimits internal method to operate on a single channel instead of all channels in the private channel table o add ath_hal_gethwchannel to map a net80211 channel to a h/w frequency (always the same except for 900MHz channels) o add HAL_EEBADREG and HAL_EEBADCC status codes to better identify regulatory problems o remove CTRY_DEBUG and CTRY_DEFAULT enum's; these come from net80211 now o change ath_hal_getwirelessmodes to really return wireless modes supported by the hardware (was previously applying regulatory constraints) o return channel interference status with IEEE80211_CHANSTATE_CWINT (should change to a callback so hal api's can take const pointers) o remove some #define's no longer needed with the inclusion of <net80211/_ieee80211.h> Sponsored by: Carlson Wireless
2009-01-28 18:00:22 +00:00
HAL_CHANNEL_INTERNAL *ichan = ath_hal_checkchannel(ah, chan);
/* XXX bounds check ic_devdata */
struct ar5212AniState *aniState = &ahp->ah_ani[chan->ic_devdata];
Update to later code from my repository: o many bug fixes o add new periodic calibration api o break up 5416 periodic calibration code in preparation for 928x o move get noise floor to rf backends o 5416-specific ani (still disabled) o modularize 5210 eeprom format a la other eeprom formats o start cleaning up regdomain code o prepare for proper 1/2 and 1/4 width channel support o bring back 900MHz card support o clean up 5212 rf version handling o add 1/2 and 1/4 width channel support for 5212 parts o split 5212 rfgain handling out o improve ani debugging o add AH_USE_INIPDGAIN compile option o purge a bunch of dead 5212 state o add 1/2 and 1/4 rate modes o remove HAL_CAP_CHAN_HALFRATE and HAL_CAP_CHAN_QUARTERRATE; the same info can now be deduced from the set of supported modes
2008-11-28 00:48:05 +00:00
uint32_t rxfilter;
Overhaul regulatory support: o remove HAL_CHANNEL; convert the hal to use net80211 channels; this mostly involves mechanical changes to variable names and channel attribute macros o gut HAL_CHANNEL_PRIVATE as most of the contents are now redundant with the net80211 channel available o change api for ath_hal_init_channels: no more reglass id's, no more outdoor indication (was a noop), anM contents o add ath_hal_getchannels to have the hal construct a channel list without altering runtime state; this is used to retrieve the calibration list for the device in ath_getradiocaps o add ath_hal_set_channels to take a channel list and regulatory data from above and construct internal state to match (maps frequencies for 900MHz cards, setup for CTL lookups, etc) o compact the private channel table: we keep one private channel per frequency instead of one per HAL_CHANNEL; this gives a big space savings and potentially improves ani and calibration by sharing state (to be seen; didn't see anything in testing); a new config option AH_MAXCHAN controls the table size (default to 96 which was chosen to be ~3x the largest expected size) o shrink ani state and change to mirror private channel table (one entry per frequency indexed by ic_devdata) o move ani state flags to private channel state o remove country codes; use net80211 definitions instead o remove GSM regulatory support; it's no longer needed now that we pass in channel lists from above o consolidate ADHOC_NO_11A attribute with DISALLOW_ADHOC_11A o simplify initial channel list construction based on the EEPROM contents; we preserve country code support for now but may want to just fallback to a WWR sku and dispatch the discovered country code up to user space so the channel list can be constructed using the master regdomain tables o defer to net80211 for max antenna gain o eliminate sorting of internal channel table; now that we use ic_devdata as an index, table lookups are O(1) o remove internal copy of the country code; the public one is sufficient o remove AH_SUPPORT_11D conditional compilation; we always support 11d o remove ath_hal_ispublicsafetysku; not needed any more o remove ath_hal_isgsmsku; no more GSM stuff o move Conformance Test Limit (CTL) state from private channel to a lookup using per-band pointers cached in the private state block o remove regulatory class id support; was unused and belongs in net80211 o fix channel list construction to set IEEE80211_CHAN_NOADHOC, IEEE80211_CHAN_NOHOSTAP, and IEEE80211_CHAN_4MSXMIT o remove private channel flags CHANNEL_DFS and CHANNEL_4MS_LIMIT; these are now set in the constructed net80211 channel o store CHANNEL_NFCREQUIRED (Noise Floor Required) channel attribute in one of the driver-private flag bits of the net80211 channel o move 900MHz frequency mapping into the hal; the mapped frequency is stored in the private channel and used throughout the hal (no more mapping in the driver and/or net80211) o remove ath_hal_mhz2ieee; it's no longer needed as net80211 does the calculation and available in the net80211 channel o change noise floor calibration logic to work with compacted private channel table setup; this may require revisiting as we no longer can distinguish channel attributes (e.g. 11b vs 11g vs turbo) but since the data is used only to calculate status data we can live with it for now o change ah_getChipPowerLimits internal method to operate on a single channel instead of all channels in the private channel table o add ath_hal_gethwchannel to map a net80211 channel to a h/w frequency (always the same except for 900MHz channels) o add HAL_EEBADREG and HAL_EEBADCC status codes to better identify regulatory problems o remove CTRY_DEBUG and CTRY_DEFAULT enum's; these come from net80211 now o change ath_hal_getwirelessmodes to really return wireless modes supported by the hardware (was previously applying regulatory constraints) o return channel interference status with IEEE80211_CHANSTATE_CWINT (should change to a callback so hal api's can take const pointers) o remove some #define's no longer needed with the inclusion of <net80211/_ieee80211.h> Sponsored by: Carlson Wireless
2009-01-28 18:00:22 +00:00
if ((ichan->privFlags & CHANNEL_ANI_INIT) == 0) {
OS_MEMZERO(aniState, sizeof(*aniState));
if (IEEE80211_IS_CHAN_2GHZ(chan))
aniState->params = &ahp->ah_aniParams24;
else
aniState->params = &ahp->ah_aniParams5;
ichan->privFlags |= CHANNEL_ANI_INIT;
HALASSERT((ichan->privFlags & CHANNEL_ANI_SETUP) == 0);
}
Update to later code from my repository: o many bug fixes o add new periodic calibration api o break up 5416 periodic calibration code in preparation for 928x o move get noise floor to rf backends o 5416-specific ani (still disabled) o modularize 5210 eeprom format a la other eeprom formats o start cleaning up regdomain code o prepare for proper 1/2 and 1/4 width channel support o bring back 900MHz card support o clean up 5212 rf version handling o add 1/2 and 1/4 width channel support for 5212 parts o split 5212 rfgain handling out o improve ani debugging o add AH_USE_INIPDGAIN compile option o purge a bunch of dead 5212 state o add 1/2 and 1/4 rate modes o remove HAL_CAP_CHAN_HALFRATE and HAL_CAP_CHAN_QUARTERRATE; the same info can now be deduced from the set of supported modes
2008-11-28 00:48:05 +00:00
ahp->ah_curani = aniState;
#if 0
Overhaul regulatory support: o remove HAL_CHANNEL; convert the hal to use net80211 channels; this mostly involves mechanical changes to variable names and channel attribute macros o gut HAL_CHANNEL_PRIVATE as most of the contents are now redundant with the net80211 channel available o change api for ath_hal_init_channels: no more reglass id's, no more outdoor indication (was a noop), anM contents o add ath_hal_getchannels to have the hal construct a channel list without altering runtime state; this is used to retrieve the calibration list for the device in ath_getradiocaps o add ath_hal_set_channels to take a channel list and regulatory data from above and construct internal state to match (maps frequencies for 900MHz cards, setup for CTL lookups, etc) o compact the private channel table: we keep one private channel per frequency instead of one per HAL_CHANNEL; this gives a big space savings and potentially improves ani and calibration by sharing state (to be seen; didn't see anything in testing); a new config option AH_MAXCHAN controls the table size (default to 96 which was chosen to be ~3x the largest expected size) o shrink ani state and change to mirror private channel table (one entry per frequency indexed by ic_devdata) o move ani state flags to private channel state o remove country codes; use net80211 definitions instead o remove GSM regulatory support; it's no longer needed now that we pass in channel lists from above o consolidate ADHOC_NO_11A attribute with DISALLOW_ADHOC_11A o simplify initial channel list construction based on the EEPROM contents; we preserve country code support for now but may want to just fallback to a WWR sku and dispatch the discovered country code up to user space so the channel list can be constructed using the master regdomain tables o defer to net80211 for max antenna gain o eliminate sorting of internal channel table; now that we use ic_devdata as an index, table lookups are O(1) o remove internal copy of the country code; the public one is sufficient o remove AH_SUPPORT_11D conditional compilation; we always support 11d o remove ath_hal_ispublicsafetysku; not needed any more o remove ath_hal_isgsmsku; no more GSM stuff o move Conformance Test Limit (CTL) state from private channel to a lookup using per-band pointers cached in the private state block o remove regulatory class id support; was unused and belongs in net80211 o fix channel list construction to set IEEE80211_CHAN_NOADHOC, IEEE80211_CHAN_NOHOSTAP, and IEEE80211_CHAN_4MSXMIT o remove private channel flags CHANNEL_DFS and CHANNEL_4MS_LIMIT; these are now set in the constructed net80211 channel o store CHANNEL_NFCREQUIRED (Noise Floor Required) channel attribute in one of the driver-private flag bits of the net80211 channel o move 900MHz frequency mapping into the hal; the mapped frequency is stored in the private channel and used throughout the hal (no more mapping in the driver and/or net80211) o remove ath_hal_mhz2ieee; it's no longer needed as net80211 does the calculation and available in the net80211 channel o change noise floor calibration logic to work with compacted private channel table setup; this may require revisiting as we no longer can distinguish channel attributes (e.g. 11b vs 11g vs turbo) but since the data is used only to calculate status data we can live with it for now o change ah_getChipPowerLimits internal method to operate on a single channel instead of all channels in the private channel table o add ath_hal_gethwchannel to map a net80211 channel to a h/w frequency (always the same except for 900MHz channels) o add HAL_EEBADREG and HAL_EEBADCC status codes to better identify regulatory problems o remove CTRY_DEBUG and CTRY_DEFAULT enum's; these come from net80211 now o change ath_hal_getwirelessmodes to really return wireless modes supported by the hardware (was previously applying regulatory constraints) o return channel interference status with IEEE80211_CHANSTATE_CWINT (should change to a callback so hal api's can take const pointers) o remove some #define's no longer needed with the inclusion of <net80211/_ieee80211.h> Sponsored by: Carlson Wireless
2009-01-28 18:00:22 +00:00
ath_hal_printf(ah,"%s: chan %u/0x%x restore %d opmode %u%s\n",
__func__, chan->ic_freq, chan->ic_flags, restore, opmode,
ichan->privFlags & CHANNEL_ANI_SETUP ? " setup" : "");
Update to later code from my repository: o many bug fixes o add new periodic calibration api o break up 5416 periodic calibration code in preparation for 928x o move get noise floor to rf backends o 5416-specific ani (still disabled) o modularize 5210 eeprom format a la other eeprom formats o start cleaning up regdomain code o prepare for proper 1/2 and 1/4 width channel support o bring back 900MHz card support o clean up 5212 rf version handling o add 1/2 and 1/4 width channel support for 5212 parts o split 5212 rfgain handling out o improve ani debugging o add AH_USE_INIPDGAIN compile option o purge a bunch of dead 5212 state o add 1/2 and 1/4 rate modes o remove HAL_CAP_CHAN_HALFRATE and HAL_CAP_CHAN_QUARTERRATE; the same info can now be deduced from the set of supported modes
2008-11-28 00:48:05 +00:00
#else
Overhaul regulatory support: o remove HAL_CHANNEL; convert the hal to use net80211 channels; this mostly involves mechanical changes to variable names and channel attribute macros o gut HAL_CHANNEL_PRIVATE as most of the contents are now redundant with the net80211 channel available o change api for ath_hal_init_channels: no more reglass id's, no more outdoor indication (was a noop), anM contents o add ath_hal_getchannels to have the hal construct a channel list without altering runtime state; this is used to retrieve the calibration list for the device in ath_getradiocaps o add ath_hal_set_channels to take a channel list and regulatory data from above and construct internal state to match (maps frequencies for 900MHz cards, setup for CTL lookups, etc) o compact the private channel table: we keep one private channel per frequency instead of one per HAL_CHANNEL; this gives a big space savings and potentially improves ani and calibration by sharing state (to be seen; didn't see anything in testing); a new config option AH_MAXCHAN controls the table size (default to 96 which was chosen to be ~3x the largest expected size) o shrink ani state and change to mirror private channel table (one entry per frequency indexed by ic_devdata) o move ani state flags to private channel state o remove country codes; use net80211 definitions instead o remove GSM regulatory support; it's no longer needed now that we pass in channel lists from above o consolidate ADHOC_NO_11A attribute with DISALLOW_ADHOC_11A o simplify initial channel list construction based on the EEPROM contents; we preserve country code support for now but may want to just fallback to a WWR sku and dispatch the discovered country code up to user space so the channel list can be constructed using the master regdomain tables o defer to net80211 for max antenna gain o eliminate sorting of internal channel table; now that we use ic_devdata as an index, table lookups are O(1) o remove internal copy of the country code; the public one is sufficient o remove AH_SUPPORT_11D conditional compilation; we always support 11d o remove ath_hal_ispublicsafetysku; not needed any more o remove ath_hal_isgsmsku; no more GSM stuff o move Conformance Test Limit (CTL) state from private channel to a lookup using per-band pointers cached in the private state block o remove regulatory class id support; was unused and belongs in net80211 o fix channel list construction to set IEEE80211_CHAN_NOADHOC, IEEE80211_CHAN_NOHOSTAP, and IEEE80211_CHAN_4MSXMIT o remove private channel flags CHANNEL_DFS and CHANNEL_4MS_LIMIT; these are now set in the constructed net80211 channel o store CHANNEL_NFCREQUIRED (Noise Floor Required) channel attribute in one of the driver-private flag bits of the net80211 channel o move 900MHz frequency mapping into the hal; the mapped frequency is stored in the private channel and used throughout the hal (no more mapping in the driver and/or net80211) o remove ath_hal_mhz2ieee; it's no longer needed as net80211 does the calculation and available in the net80211 channel o change noise floor calibration logic to work with compacted private channel table setup; this may require revisiting as we no longer can distinguish channel attributes (e.g. 11b vs 11g vs turbo) but since the data is used only to calculate status data we can live with it for now o change ah_getChipPowerLimits internal method to operate on a single channel instead of all channels in the private channel table o add ath_hal_gethwchannel to map a net80211 channel to a h/w frequency (always the same except for 900MHz channels) o add HAL_EEBADREG and HAL_EEBADCC status codes to better identify regulatory problems o remove CTRY_DEBUG and CTRY_DEFAULT enum's; these come from net80211 now o change ath_hal_getwirelessmodes to really return wireless modes supported by the hardware (was previously applying regulatory constraints) o return channel interference status with IEEE80211_CHANSTATE_CWINT (should change to a callback so hal api's can take const pointers) o remove some #define's no longer needed with the inclusion of <net80211/_ieee80211.h> Sponsored by: Carlson Wireless
2009-01-28 18:00:22 +00:00
HALDEBUG(ah, HAL_DEBUG_ANI, "%s: chan %u/0x%x restore %d opmode %u%s\n",
__func__, chan->ic_freq, chan->ic_flags, restore, opmode,
ichan->privFlags & CHANNEL_ANI_SETUP ? " setup" : "");
Update to later code from my repository: o many bug fixes o add new periodic calibration api o break up 5416 periodic calibration code in preparation for 928x o move get noise floor to rf backends o 5416-specific ani (still disabled) o modularize 5210 eeprom format a la other eeprom formats o start cleaning up regdomain code o prepare for proper 1/2 and 1/4 width channel support o bring back 900MHz card support o clean up 5212 rf version handling o add 1/2 and 1/4 width channel support for 5212 parts o split 5212 rfgain handling out o improve ani debugging o add AH_USE_INIPDGAIN compile option o purge a bunch of dead 5212 state o add 1/2 and 1/4 rate modes o remove HAL_CAP_CHAN_HALFRATE and HAL_CAP_CHAN_QUARTERRATE; the same info can now be deduced from the set of supported modes
2008-11-28 00:48:05 +00:00
#endif
OS_MARK(ah, AH_MARK_ANI_RESET, opmode);
/*
* Turn off PHY error frame delivery while we futz with settings.
*/
rxfilter = ar5212GetRxFilter(ah);
ar5212SetRxFilter(ah, rxfilter &~ HAL_RX_FILTER_PHYERR);
/*
* Automatic processing is done only in station mode right now.
*/
if (opmode == HAL_M_STA)
ahp->ah_procPhyErr |= HAL_RSSI_ANI_ENA;
else
ahp->ah_procPhyErr &= ~HAL_RSSI_ANI_ENA;
/*
* Set all ani parameters. We either set them to initial
* values or restore the previous ones for the channel.
* XXX if ANI follows hardware, we don't care what mode we're
* XXX in, we should keep the ani parameters
*/
Overhaul regulatory support: o remove HAL_CHANNEL; convert the hal to use net80211 channels; this mostly involves mechanical changes to variable names and channel attribute macros o gut HAL_CHANNEL_PRIVATE as most of the contents are now redundant with the net80211 channel available o change api for ath_hal_init_channels: no more reglass id's, no more outdoor indication (was a noop), anM contents o add ath_hal_getchannels to have the hal construct a channel list without altering runtime state; this is used to retrieve the calibration list for the device in ath_getradiocaps o add ath_hal_set_channels to take a channel list and regulatory data from above and construct internal state to match (maps frequencies for 900MHz cards, setup for CTL lookups, etc) o compact the private channel table: we keep one private channel per frequency instead of one per HAL_CHANNEL; this gives a big space savings and potentially improves ani and calibration by sharing state (to be seen; didn't see anything in testing); a new config option AH_MAXCHAN controls the table size (default to 96 which was chosen to be ~3x the largest expected size) o shrink ani state and change to mirror private channel table (one entry per frequency indexed by ic_devdata) o move ani state flags to private channel state o remove country codes; use net80211 definitions instead o remove GSM regulatory support; it's no longer needed now that we pass in channel lists from above o consolidate ADHOC_NO_11A attribute with DISALLOW_ADHOC_11A o simplify initial channel list construction based on the EEPROM contents; we preserve country code support for now but may want to just fallback to a WWR sku and dispatch the discovered country code up to user space so the channel list can be constructed using the master regdomain tables o defer to net80211 for max antenna gain o eliminate sorting of internal channel table; now that we use ic_devdata as an index, table lookups are O(1) o remove internal copy of the country code; the public one is sufficient o remove AH_SUPPORT_11D conditional compilation; we always support 11d o remove ath_hal_ispublicsafetysku; not needed any more o remove ath_hal_isgsmsku; no more GSM stuff o move Conformance Test Limit (CTL) state from private channel to a lookup using per-band pointers cached in the private state block o remove regulatory class id support; was unused and belongs in net80211 o fix channel list construction to set IEEE80211_CHAN_NOADHOC, IEEE80211_CHAN_NOHOSTAP, and IEEE80211_CHAN_4MSXMIT o remove private channel flags CHANNEL_DFS and CHANNEL_4MS_LIMIT; these are now set in the constructed net80211 channel o store CHANNEL_NFCREQUIRED (Noise Floor Required) channel attribute in one of the driver-private flag bits of the net80211 channel o move 900MHz frequency mapping into the hal; the mapped frequency is stored in the private channel and used throughout the hal (no more mapping in the driver and/or net80211) o remove ath_hal_mhz2ieee; it's no longer needed as net80211 does the calculation and available in the net80211 channel o change noise floor calibration logic to work with compacted private channel table setup; this may require revisiting as we no longer can distinguish channel attributes (e.g. 11b vs 11g vs turbo) but since the data is used only to calculate status data we can live with it for now o change ah_getChipPowerLimits internal method to operate on a single channel instead of all channels in the private channel table o add ath_hal_gethwchannel to map a net80211 channel to a h/w frequency (always the same except for 900MHz channels) o add HAL_EEBADREG and HAL_EEBADCC status codes to better identify regulatory problems o remove CTRY_DEBUG and CTRY_DEFAULT enum's; these come from net80211 now o change ath_hal_getwirelessmodes to really return wireless modes supported by the hardware (was previously applying regulatory constraints) o return channel interference status with IEEE80211_CHANSTATE_CWINT (should change to a callback so hal api's can take const pointers) o remove some #define's no longer needed with the inclusion of <net80211/_ieee80211.h> Sponsored by: Carlson Wireless
2009-01-28 18:00:22 +00:00
if (restore && (ichan->privFlags & CHANNEL_ANI_SETUP)) {
Update to later code from my repository: o many bug fixes o add new periodic calibration api o break up 5416 periodic calibration code in preparation for 928x o move get noise floor to rf backends o 5416-specific ani (still disabled) o modularize 5210 eeprom format a la other eeprom formats o start cleaning up regdomain code o prepare for proper 1/2 and 1/4 width channel support o bring back 900MHz card support o clean up 5212 rf version handling o add 1/2 and 1/4 width channel support for 5212 parts o split 5212 rfgain handling out o improve ani debugging o add AH_USE_INIPDGAIN compile option o purge a bunch of dead 5212 state o add 1/2 and 1/4 rate modes o remove HAL_CAP_CHAN_HALFRATE and HAL_CAP_CHAN_QUARTERRATE; the same info can now be deduced from the set of supported modes
2008-11-28 00:48:05 +00:00
ar5416AniControl(ah, HAL_ANI_NOISE_IMMUNITY_LEVEL,
aniState->noiseImmunityLevel);
ar5416AniControl(ah, HAL_ANI_SPUR_IMMUNITY_LEVEL,
aniState->spurImmunityLevel);
ar5416AniControl(ah, HAL_ANI_OFDM_WEAK_SIGNAL_DETECTION,
!aniState->ofdmWeakSigDetectOff);
ar5416AniControl(ah, HAL_ANI_CCK_WEAK_SIGNAL_THR,
aniState->cckWeakSigThreshold);
ar5416AniControl(ah, HAL_ANI_FIRSTEP_LEVEL,
aniState->firstepLevel);
} else {
ar5416AniControl(ah, HAL_ANI_NOISE_IMMUNITY_LEVEL, 0);
ar5416AniControl(ah, HAL_ANI_SPUR_IMMUNITY_LEVEL, 0);
ar5416AniControl(ah, HAL_ANI_OFDM_WEAK_SIGNAL_DETECTION,
AH_TRUE);
ar5416AniControl(ah, HAL_ANI_CCK_WEAK_SIGNAL_THR, AH_FALSE);
ar5416AniControl(ah, HAL_ANI_FIRSTEP_LEVEL, 0);
Overhaul regulatory support: o remove HAL_CHANNEL; convert the hal to use net80211 channels; this mostly involves mechanical changes to variable names and channel attribute macros o gut HAL_CHANNEL_PRIVATE as most of the contents are now redundant with the net80211 channel available o change api for ath_hal_init_channels: no more reglass id's, no more outdoor indication (was a noop), anM contents o add ath_hal_getchannels to have the hal construct a channel list without altering runtime state; this is used to retrieve the calibration list for the device in ath_getradiocaps o add ath_hal_set_channels to take a channel list and regulatory data from above and construct internal state to match (maps frequencies for 900MHz cards, setup for CTL lookups, etc) o compact the private channel table: we keep one private channel per frequency instead of one per HAL_CHANNEL; this gives a big space savings and potentially improves ani and calibration by sharing state (to be seen; didn't see anything in testing); a new config option AH_MAXCHAN controls the table size (default to 96 which was chosen to be ~3x the largest expected size) o shrink ani state and change to mirror private channel table (one entry per frequency indexed by ic_devdata) o move ani state flags to private channel state o remove country codes; use net80211 definitions instead o remove GSM regulatory support; it's no longer needed now that we pass in channel lists from above o consolidate ADHOC_NO_11A attribute with DISALLOW_ADHOC_11A o simplify initial channel list construction based on the EEPROM contents; we preserve country code support for now but may want to just fallback to a WWR sku and dispatch the discovered country code up to user space so the channel list can be constructed using the master regdomain tables o defer to net80211 for max antenna gain o eliminate sorting of internal channel table; now that we use ic_devdata as an index, table lookups are O(1) o remove internal copy of the country code; the public one is sufficient o remove AH_SUPPORT_11D conditional compilation; we always support 11d o remove ath_hal_ispublicsafetysku; not needed any more o remove ath_hal_isgsmsku; no more GSM stuff o move Conformance Test Limit (CTL) state from private channel to a lookup using per-band pointers cached in the private state block o remove regulatory class id support; was unused and belongs in net80211 o fix channel list construction to set IEEE80211_CHAN_NOADHOC, IEEE80211_CHAN_NOHOSTAP, and IEEE80211_CHAN_4MSXMIT o remove private channel flags CHANNEL_DFS and CHANNEL_4MS_LIMIT; these are now set in the constructed net80211 channel o store CHANNEL_NFCREQUIRED (Noise Floor Required) channel attribute in one of the driver-private flag bits of the net80211 channel o move 900MHz frequency mapping into the hal; the mapped frequency is stored in the private channel and used throughout the hal (no more mapping in the driver and/or net80211) o remove ath_hal_mhz2ieee; it's no longer needed as net80211 does the calculation and available in the net80211 channel o change noise floor calibration logic to work with compacted private channel table setup; this may require revisiting as we no longer can distinguish channel attributes (e.g. 11b vs 11g vs turbo) but since the data is used only to calculate status data we can live with it for now o change ah_getChipPowerLimits internal method to operate on a single channel instead of all channels in the private channel table o add ath_hal_gethwchannel to map a net80211 channel to a h/w frequency (always the same except for 900MHz channels) o add HAL_EEBADREG and HAL_EEBADCC status codes to better identify regulatory problems o remove CTRY_DEBUG and CTRY_DEFAULT enum's; these come from net80211 now o change ath_hal_getwirelessmodes to really return wireless modes supported by the hardware (was previously applying regulatory constraints) o return channel interference status with IEEE80211_CHANSTATE_CWINT (should change to a callback so hal api's can take const pointers) o remove some #define's no longer needed with the inclusion of <net80211/_ieee80211.h> Sponsored by: Carlson Wireless
2009-01-28 18:00:22 +00:00
ichan->privFlags |= CHANNEL_ANI_SETUP;
Update to later code from my repository: o many bug fixes o add new periodic calibration api o break up 5416 periodic calibration code in preparation for 928x o move get noise floor to rf backends o 5416-specific ani (still disabled) o modularize 5210 eeprom format a la other eeprom formats o start cleaning up regdomain code o prepare for proper 1/2 and 1/4 width channel support o bring back 900MHz card support o clean up 5212 rf version handling o add 1/2 and 1/4 width channel support for 5212 parts o split 5212 rfgain handling out o improve ani debugging o add AH_USE_INIPDGAIN compile option o purge a bunch of dead 5212 state o add 1/2 and 1/4 rate modes o remove HAL_CAP_CHAN_HALFRATE and HAL_CAP_CHAN_QUARTERRATE; the same info can now be deduced from the set of supported modes
2008-11-28 00:48:05 +00:00
}
ar5416AniRestart(ah, aniState);
/* restore RX filter mask */
ar5212SetRxFilter(ah, rxfilter);
}
/*
* Process a MIB interrupt. We may potentially be invoked because
* any of the MIB counters overflow/trigger so don't assume we're
* here because a PHY error counter triggered.
*/
void
ar5416ProcessMibIntr(struct ath_hal *ah, const HAL_NODE_STATS *stats)
{
struct ath_hal_5212 *ahp = AH5212(ah);
uint32_t phyCnt1, phyCnt2;
HALDEBUG(ah, HAL_DEBUG_ANI, "%s: mibc 0x%x phyCnt1 0x%x phyCnt2 0x%x "
"filtofdm 0x%x filtcck 0x%x\n",
__func__, OS_REG_READ(ah, AR_MIBC),
OS_REG_READ(ah, AR_PHYCNT1), OS_REG_READ(ah, AR_PHYCNT2),
OS_REG_READ(ah, AR_FILTOFDM), OS_REG_READ(ah, AR_FILTCCK));
/*
* First order of business is to clear whatever caused
* the interrupt so we don't keep getting interrupted.
* We have the usual mib counters that are reset-on-read
* and the additional counters that appeared starting in
* Hainan. We collect the mib counters and explicitly
* zero additional counters we are not using. Anything
* else is reset only if it caused the interrupt.
*/
/* NB: these are not reset-on-read */
phyCnt1 = OS_REG_READ(ah, AR_PHY_ERR_1);
phyCnt2 = OS_REG_READ(ah, AR_PHY_ERR_2);
/* not used, always reset them in case they are the cause */
OS_REG_WRITE(ah, AR_FILTOFDM, 0);
OS_REG_WRITE(ah, AR_FILTCCK, 0);
if ((OS_REG_READ(ah, AR_SLP_MIB_CTRL) & AR_SLP_MIB_PENDING) == 0)
OS_REG_WRITE(ah, AR_SLP_MIB_CTRL, AR_SLP_MIB_CLEAR);
/* Clear the mib counters and save them in the stats */
ar5212UpdateMibCounters(ah, &ahp->ah_mibStats);
ahp->ah_stats.ast_nodestats = *stats;
/*
* Check for an ani stat hitting the trigger threshold.
* When this happens we get a MIB interrupt and the top
* 2 bits of the counter register will be 0b11, hence
* the mask check of phyCnt?.
*/
if (((phyCnt1 & AR_MIBCNT_INTRMASK) == AR_MIBCNT_INTRMASK) ||
((phyCnt2 & AR_MIBCNT_INTRMASK) == AR_MIBCNT_INTRMASK)) {
struct ar5212AniState *aniState = ahp->ah_curani;
const struct ar5212AniParams *params = aniState->params;
uint32_t ofdmPhyErrCnt, cckPhyErrCnt;
ofdmPhyErrCnt = phyCnt1 - params->ofdmPhyErrBase;
ahp->ah_stats.ast_ani_ofdmerrs +=
ofdmPhyErrCnt - aniState->ofdmPhyErrCount;
aniState->ofdmPhyErrCount = ofdmPhyErrCnt;
cckPhyErrCnt = phyCnt2 - params->cckPhyErrBase;
ahp->ah_stats.ast_ani_cckerrs +=
cckPhyErrCnt - aniState->cckPhyErrCount;
aniState->cckPhyErrCount = cckPhyErrCnt;
/*
* NB: figure out which counter triggered. If both
* trigger we'll only deal with one as the processing
* clobbers the error counter so the trigger threshold
* check will never be true.
*/
if (aniState->ofdmPhyErrCount > params->ofdmTrigHigh)
ar5416AniOfdmErrTrigger(ah);
if (aniState->cckPhyErrCount > params->cckTrigHigh)
ar5416AniCckErrTrigger(ah);
/* NB: always restart to insure the h/w counters are reset */
ar5416AniRestart(ah, aniState);
}
}
static void
ar5416AniLowerImmunity(struct ath_hal *ah)
{
struct ath_hal_5212 *ahp = AH5212(ah);
struct ar5212AniState *aniState;
const struct ar5212AniParams *params;
HALASSERT(ANI_ENA(ah));
aniState = ahp->ah_curani;
params = aniState->params;
if (ANI_ENA_RSSI(ah)) {
int32_t rssi = BEACON_RSSI(ahp);
if (rssi > params->rssiThrHigh) {
/*
* Beacon signal is high, leave ofdm weak signal
* detection off or it may oscillate. Let it fall
* through.
*/
} else if (rssi > params->rssiThrLow) {
/*
* Beacon rssi in mid range, turn on ofdm weak signal
* detection or lower firstep level.
*/
if (aniState->ofdmWeakSigDetectOff) {
ar5416AniControl(ah,
HAL_ANI_OFDM_WEAK_SIGNAL_DETECTION,
AH_TRUE);
return;
}
if (aniState->firstepLevel > 0) {
ar5416AniControl(ah, HAL_ANI_FIRSTEP_LEVEL,
aniState->firstepLevel - 1);
return;
}
} else {
/*
* Beacon rssi is low, reduce firstep level.
*/
if (aniState->firstepLevel > 0) {
ar5416AniControl(ah, HAL_ANI_FIRSTEP_LEVEL,
aniState->firstepLevel - 1);
return;
}
}
}
/* then lower spur immunity level, down to zero */
if (aniState->spurImmunityLevel > 0) {
ar5416AniControl(ah, HAL_ANI_SPUR_IMMUNITY_LEVEL,
aniState->spurImmunityLevel - 1);
return;
}
/*
* if all else fails, lower noise immunity level down to a min value
* zero for now
*/
if (aniState->noiseImmunityLevel > 0) {
ar5416AniControl(ah, HAL_ANI_NOISE_IMMUNITY_LEVEL,
aniState->noiseImmunityLevel - 1);
return;
}
}
#define CLOCK_RATE 44000 /* XXX use mac_usec or similar */
/* convert HW counter values to ms using 11g clock rate, goo9d enough
for 11a and Turbo */
/*
* Return an approximation of the time spent ``listening'' by
* deducting the cycles spent tx'ing and rx'ing from the total
* cycle count since our last call. A return value <0 indicates
* an invalid/inconsistent time.
*/
static int32_t
ar5416AniGetListenTime(struct ath_hal *ah)
{
struct ath_hal_5212 *ahp = AH5212(ah);
struct ar5212AniState *aniState;
uint32_t txFrameCount, rxFrameCount, cycleCount;
int32_t listenTime;
txFrameCount = OS_REG_READ(ah, AR_TFCNT);
rxFrameCount = OS_REG_READ(ah, AR_RFCNT);
cycleCount = OS_REG_READ(ah, AR_CCCNT);
aniState = ahp->ah_curani;
if (aniState->cycleCount == 0 || aniState->cycleCount > cycleCount) {
/*
* Cycle counter wrap (or initial call); it's not possible
* to accurately calculate a value because the registers
* right shift rather than wrap--so punt and return 0.
*/
listenTime = 0;
ahp->ah_stats.ast_ani_lzero++;
} else {
int32_t ccdelta = cycleCount - aniState->cycleCount;
int32_t rfdelta = rxFrameCount - aniState->rxFrameCount;
int32_t tfdelta = txFrameCount - aniState->txFrameCount;
listenTime = (ccdelta - rfdelta - tfdelta) / CLOCK_RATE;
}
aniState->cycleCount = cycleCount;
aniState->txFrameCount = txFrameCount;
aniState->rxFrameCount = rxFrameCount;
return listenTime;
}
/*
* Update ani stats in preparation for listen time processing.
*/
static void
updateMIBStats(struct ath_hal *ah, struct ar5212AniState *aniState)
{
struct ath_hal_5212 *ahp = AH5212(ah);
const struct ar5212AniParams *params = aniState->params;
uint32_t phyCnt1, phyCnt2;
int32_t ofdmPhyErrCnt, cckPhyErrCnt;
/* Clear the mib counters and save them in the stats */
ar5212UpdateMibCounters(ah, &ahp->ah_mibStats);
/* NB: these are not reset-on-read */
phyCnt1 = OS_REG_READ(ah, AR_PHY_ERR_1);
phyCnt2 = OS_REG_READ(ah, AR_PHY_ERR_2);
/* NB: these are spec'd to never roll-over */
ofdmPhyErrCnt = phyCnt1 - params->ofdmPhyErrBase;
if (ofdmPhyErrCnt < 0) {
HALDEBUG(ah, HAL_DEBUG_ANI, "OFDM phyErrCnt %d phyCnt1 0x%x\n",
ofdmPhyErrCnt, phyCnt1);
ofdmPhyErrCnt = AR_PHY_COUNTMAX;
}
ahp->ah_stats.ast_ani_ofdmerrs +=
ofdmPhyErrCnt - aniState->ofdmPhyErrCount;
aniState->ofdmPhyErrCount = ofdmPhyErrCnt;
cckPhyErrCnt = phyCnt2 - params->cckPhyErrBase;
if (cckPhyErrCnt < 0) {
HALDEBUG(ah, HAL_DEBUG_ANI, "CCK phyErrCnt %d phyCnt2 0x%x\n",
cckPhyErrCnt, phyCnt2);
cckPhyErrCnt = AR_PHY_COUNTMAX;
}
ahp->ah_stats.ast_ani_cckerrs +=
cckPhyErrCnt - aniState->cckPhyErrCount;
aniState->cckPhyErrCount = cckPhyErrCnt;
}
/*
* Do periodic processing. This routine is called from the
* driver's rx interrupt handler after processing frames.
*/
void
ar5416AniPoll(struct ath_hal *ah, const HAL_NODE_STATS *stats,
Overhaul regulatory support: o remove HAL_CHANNEL; convert the hal to use net80211 channels; this mostly involves mechanical changes to variable names and channel attribute macros o gut HAL_CHANNEL_PRIVATE as most of the contents are now redundant with the net80211 channel available o change api for ath_hal_init_channels: no more reglass id's, no more outdoor indication (was a noop), anM contents o add ath_hal_getchannels to have the hal construct a channel list without altering runtime state; this is used to retrieve the calibration list for the device in ath_getradiocaps o add ath_hal_set_channels to take a channel list and regulatory data from above and construct internal state to match (maps frequencies for 900MHz cards, setup for CTL lookups, etc) o compact the private channel table: we keep one private channel per frequency instead of one per HAL_CHANNEL; this gives a big space savings and potentially improves ani and calibration by sharing state (to be seen; didn't see anything in testing); a new config option AH_MAXCHAN controls the table size (default to 96 which was chosen to be ~3x the largest expected size) o shrink ani state and change to mirror private channel table (one entry per frequency indexed by ic_devdata) o move ani state flags to private channel state o remove country codes; use net80211 definitions instead o remove GSM regulatory support; it's no longer needed now that we pass in channel lists from above o consolidate ADHOC_NO_11A attribute with DISALLOW_ADHOC_11A o simplify initial channel list construction based on the EEPROM contents; we preserve country code support for now but may want to just fallback to a WWR sku and dispatch the discovered country code up to user space so the channel list can be constructed using the master regdomain tables o defer to net80211 for max antenna gain o eliminate sorting of internal channel table; now that we use ic_devdata as an index, table lookups are O(1) o remove internal copy of the country code; the public one is sufficient o remove AH_SUPPORT_11D conditional compilation; we always support 11d o remove ath_hal_ispublicsafetysku; not needed any more o remove ath_hal_isgsmsku; no more GSM stuff o move Conformance Test Limit (CTL) state from private channel to a lookup using per-band pointers cached in the private state block o remove regulatory class id support; was unused and belongs in net80211 o fix channel list construction to set IEEE80211_CHAN_NOADHOC, IEEE80211_CHAN_NOHOSTAP, and IEEE80211_CHAN_4MSXMIT o remove private channel flags CHANNEL_DFS and CHANNEL_4MS_LIMIT; these are now set in the constructed net80211 channel o store CHANNEL_NFCREQUIRED (Noise Floor Required) channel attribute in one of the driver-private flag bits of the net80211 channel o move 900MHz frequency mapping into the hal; the mapped frequency is stored in the private channel and used throughout the hal (no more mapping in the driver and/or net80211) o remove ath_hal_mhz2ieee; it's no longer needed as net80211 does the calculation and available in the net80211 channel o change noise floor calibration logic to work with compacted private channel table setup; this may require revisiting as we no longer can distinguish channel attributes (e.g. 11b vs 11g vs turbo) but since the data is used only to calculate status data we can live with it for now o change ah_getChipPowerLimits internal method to operate on a single channel instead of all channels in the private channel table o add ath_hal_gethwchannel to map a net80211 channel to a h/w frequency (always the same except for 900MHz channels) o add HAL_EEBADREG and HAL_EEBADCC status codes to better identify regulatory problems o remove CTRY_DEBUG and CTRY_DEFAULT enum's; these come from net80211 now o change ath_hal_getwirelessmodes to really return wireless modes supported by the hardware (was previously applying regulatory constraints) o return channel interference status with IEEE80211_CHANSTATE_CWINT (should change to a callback so hal api's can take const pointers) o remove some #define's no longer needed with the inclusion of <net80211/_ieee80211.h> Sponsored by: Carlson Wireless
2009-01-28 18:00:22 +00:00
const struct ieee80211_channel *chan)
Update to later code from my repository: o many bug fixes o add new periodic calibration api o break up 5416 periodic calibration code in preparation for 928x o move get noise floor to rf backends o 5416-specific ani (still disabled) o modularize 5210 eeprom format a la other eeprom formats o start cleaning up regdomain code o prepare for proper 1/2 and 1/4 width channel support o bring back 900MHz card support o clean up 5212 rf version handling o add 1/2 and 1/4 width channel support for 5212 parts o split 5212 rfgain handling out o improve ani debugging o add AH_USE_INIPDGAIN compile option o purge a bunch of dead 5212 state o add 1/2 and 1/4 rate modes o remove HAL_CAP_CHAN_HALFRATE and HAL_CAP_CHAN_QUARTERRATE; the same info can now be deduced from the set of supported modes
2008-11-28 00:48:05 +00:00
{
struct ath_hal_5212 *ahp = AH5212(ah);
struct ar5212AniState *aniState = ahp->ah_curani;
const struct ar5212AniParams *params;
int32_t listenTime;
ahp->ah_stats.ast_nodestats.ns_avgbrssi = stats->ns_avgbrssi;
/* XXX can aniState be null? */
if (aniState == AH_NULL)
return;
if (!ANI_ENA(ah))
return;
listenTime = ar5416AniGetListenTime(ah);
if (listenTime < 0) {
ahp->ah_stats.ast_ani_lneg++;
/* restart ANI period if listenTime is invalid */
ar5416AniRestart(ah, aniState);
}
/* XXX beware of overflow? */
aniState->listenTime += listenTime;
OS_MARK(ah, AH_MARK_ANI_POLL, aniState->listenTime);
params = aniState->params;
if (aniState->listenTime > 5*params->period) {
/*
* Check to see if need to lower immunity if
* 5 aniPeriods have passed
*/
updateMIBStats(ah, aniState);
if (aniState->ofdmPhyErrCount <= aniState->listenTime *
params->ofdmTrigLow/1000 &&
aniState->cckPhyErrCount <= aniState->listenTime *
params->cckTrigLow/1000)
ar5416AniLowerImmunity(ah);
ar5416AniRestart(ah, aniState);
} else if (aniState->listenTime > params->period) {
updateMIBStats(ah, aniState);
/* check to see if need to raise immunity */
if (aniState->ofdmPhyErrCount > aniState->listenTime *
params->ofdmTrigHigh / 1000) {
ar5416AniOfdmErrTrigger(ah);
ar5416AniRestart(ah, aniState);
} else if (aniState->cckPhyErrCount > aniState->listenTime *
params->cckTrigHigh / 1000) {
ar5416AniCckErrTrigger(ah);
ar5416AniRestart(ah, aniState);
}
}
}
Reference in New Issue Copy Permalink