* [PATCH 07/17] ath5k: unify rx descriptor error handling
From: Bruno Randolf @ 2010-06-16 10:11 UTC (permalink / raw)
To: linville; +Cc: ath5k-devel, linux-wireless
In-Reply-To: <20100616100809.10067.34787.stgit@tt-desk>
There is no reason for a special handling (return) here, just break like we do
with the checks before.
Signed-off-by: Bruno Randolf <br1@einfach.org>
Acked-by: Bob Copeland <me@bobcopeland.com>
---
drivers/net/wireless/ath/ath5k/base.c | 3 +--
1 files changed, 1 insertions(+), 2 deletions(-)
diff --git a/drivers/net/wireless/ath/ath5k/base.c b/drivers/net/wireless/ath/ath5k/base.c
index 5479f85..d9df114 100644
--- a/drivers/net/wireless/ath/ath5k/base.c
+++ b/drivers/net/wireless/ath/ath5k/base.c
@@ -1937,8 +1937,7 @@ ath5k_tasklet_rx(unsigned long data)
else if (unlikely(ret)) {
ATH5K_ERR(sc, "error in processing rx descriptor\n");
sc->stats.rxerr_proc++;
- spin_unlock(&sc->rxbuflock);
- return;
+ break;
}
sc->stats.rx_all_count++;
^ permalink raw reply related
* [PATCH 08/17] ath5k: split descriptor handling and frame receive
From: Bruno Randolf @ 2010-06-16 10:11 UTC (permalink / raw)
To: linville; +Cc: ath5k-devel, linux-wireless
In-Reply-To: <20100616100809.10067.34787.stgit@tt-desk>
Move frame reception into it's own function to have a clearer separation
between buffer and descriptor handling and things that are done when we
actually receive a frame.
Signed-off-by: Bruno Randolf <br1@einfach.org>
Acked-by: Bob Copeland <me@bobcopeland.com>
---
drivers/net/wireless/ath/ath5k/base.c | 149 +++++++++++++++++----------------
1 files changed, 78 insertions(+), 71 deletions(-)
diff --git a/drivers/net/wireless/ath/ath5k/base.c b/drivers/net/wireless/ath/ath5k/base.c
index d9df114..c54d1fd 100644
--- a/drivers/net/wireless/ath/ath5k/base.c
+++ b/drivers/net/wireless/ath/ath5k/base.c
@@ -357,7 +357,6 @@ static void ath5k_txq_release(struct ath5k_softc *sc);
static int ath5k_rx_start(struct ath5k_softc *sc);
static void ath5k_rx_stop(struct ath5k_softc *sc);
static unsigned int ath5k_rx_decrypted(struct ath5k_softc *sc,
- struct ath5k_desc *ds,
struct sk_buff *skb,
struct ath5k_rx_status *rs);
static void ath5k_tasklet_rx(unsigned long data);
@@ -1732,8 +1731,8 @@ ath5k_rx_stop(struct ath5k_softc *sc)
}
static unsigned int
-ath5k_rx_decrypted(struct ath5k_softc *sc, struct ath5k_desc *ds,
- struct sk_buff *skb, struct ath5k_rx_status *rs)
+ath5k_rx_decrypted(struct ath5k_softc *sc, struct sk_buff *skb,
+ struct ath5k_rx_status *rs)
{
struct ath5k_hw *ah = sc->ah;
struct ath_common *common = ath5k_hw_common(ah);
@@ -1900,9 +1899,83 @@ static int ath5k_remove_padding(struct sk_buff *skb)
}
static void
-ath5k_tasklet_rx(unsigned long data)
+ath5k_receive_frame(struct ath5k_softc *sc, struct sk_buff *skb,
+ struct ath5k_rx_status *rs)
{
struct ieee80211_rx_status *rxs;
+
+ /* The MAC header is padded to have 32-bit boundary if the
+ * packet payload is non-zero. The general calculation for
+ * padsize would take into account odd header lengths:
+ * padsize = (4 - hdrlen % 4) % 4; However, since only
+ * even-length headers are used, padding can only be 0 or 2
+ * bytes and we can optimize this a bit. In addition, we must
+ * not try to remove padding from short control frames that do
+ * not have payload. */
+ ath5k_remove_padding(skb);
+
+ rxs = IEEE80211_SKB_RXCB(skb);
+
+ rxs->flag = 0;
+ if (unlikely(rs->rs_status & AR5K_RXERR_MIC))
+ rxs->flag |= RX_FLAG_MMIC_ERROR;
+
+ /*
+ * always extend the mac timestamp, since this information is
+ * also needed for proper IBSS merging.
+ *
+ * XXX: it might be too late to do it here, since rs_tstamp is
+ * 15bit only. that means TSF extension has to be done within
+ * 32768usec (about 32ms). it might be necessary to move this to
+ * the interrupt handler, like it is done in madwifi.
+ *
+ * Unfortunately we don't know when the hardware takes the rx
+ * timestamp (beginning of phy frame, data frame, end of rx?).
+ * The only thing we know is that it is hardware specific...
+ * On AR5213 it seems the rx timestamp is at the end of the
+ * frame, but i'm not sure.
+ *
+ * NOTE: mac80211 defines mactime at the beginning of the first
+ * data symbol. Since we don't have any time references it's
+ * impossible to comply to that. This affects IBSS merge only
+ * right now, so it's not too bad...
+ */
+ rxs->mactime = ath5k_extend_tsf(sc->ah, rs->rs_tstamp);
+ rxs->flag |= RX_FLAG_TSFT;
+
+ rxs->freq = sc->curchan->center_freq;
+ rxs->band = sc->curband->band;
+
+ rxs->signal = sc->ah->ah_noise_floor + rs->rs_rssi;
+
+ rxs->antenna = rs->rs_antenna;
+
+ if (rs->rs_antenna > 0 && rs->rs_antenna < 5)
+ sc->stats.antenna_rx[rs->rs_antenna]++;
+ else
+ sc->stats.antenna_rx[0]++; /* invalid */
+
+ rxs->rate_idx = ath5k_hw_to_driver_rix(sc, rs->rs_rate);
+ rxs->flag |= ath5k_rx_decrypted(sc, skb, rs);
+
+ if (rxs->rate_idx >= 0 && rs->rs_rate ==
+ sc->curband->bitrates[rxs->rate_idx].hw_value_short)
+ rxs->flag |= RX_FLAG_SHORTPRE;
+
+ ath5k_debug_dump_skb(sc, skb, "RX ", 0);
+
+ ath5k_update_beacon_rssi(sc, skb, rs->rs_rssi);
+
+ /* check beacons in IBSS mode */
+ if (sc->opmode == NL80211_IFTYPE_ADHOC)
+ ath5k_check_ibss_tsf(sc, skb, rxs);
+
+ ieee80211_rx(sc->hw, skb);
+}
+
+static void
+ath5k_tasklet_rx(unsigned long data)
+{
struct ath5k_rx_status rs = {};
struct sk_buff *skb, *next_skb;
dma_addr_t next_skb_addr;
@@ -1912,7 +1985,6 @@ ath5k_tasklet_rx(unsigned long data)
struct ath5k_buf *bf;
struct ath5k_desc *ds;
int ret;
- int rx_flag;
spin_lock(&sc->rxbuflock);
if (list_empty(&sc->rxbuf)) {
@@ -1920,8 +1992,6 @@ ath5k_tasklet_rx(unsigned long data)
goto unlock;
}
do {
- rx_flag = 0;
-
bf = list_first_entry(&sc->rxbuf, struct ath5k_buf, list);
BUG_ON(bf->skb == NULL);
skb = bf->skb;
@@ -1970,7 +2040,6 @@ ath5k_tasklet_rx(unsigned long data)
goto accept;
}
if (rs.rs_status & AR5K_RXERR_MIC) {
- rx_flag |= RX_FLAG_MMIC_ERROR;
sc->stats.rxerr_mic++;
goto accept;
}
@@ -2001,69 +2070,7 @@ accept:
PCI_DMA_FROMDEVICE);
skb_put(skb, rs.rs_datalen);
- /* The MAC header is padded to have 32-bit boundary if the
- * packet payload is non-zero. The general calculation for
- * padsize would take into account odd header lengths:
- * padsize = (4 - hdrlen % 4) % 4; However, since only
- * even-length headers are used, padding can only be 0 or 2
- * bytes and we can optimize this a bit. In addition, we must
- * not try to remove padding from short control frames that do
- * not have payload. */
- ath5k_remove_padding(skb);
-
- rxs = IEEE80211_SKB_RXCB(skb);
-
- /*
- * always extend the mac timestamp, since this information is
- * also needed for proper IBSS merging.
- *
- * XXX: it might be too late to do it here, since rs_tstamp is
- * 15bit only. that means TSF extension has to be done within
- * 32768usec (about 32ms). it might be necessary to move this to
- * the interrupt handler, like it is done in madwifi.
- *
- * Unfortunately we don't know when the hardware takes the rx
- * timestamp (beginning of phy frame, data frame, end of rx?).
- * The only thing we know is that it is hardware specific...
- * On AR5213 it seems the rx timestamp is at the end of the
- * frame, but i'm not sure.
- *
- * NOTE: mac80211 defines mactime at the beginning of the first
- * data symbol. Since we don't have any time references it's
- * impossible to comply to that. This affects IBSS merge only
- * right now, so it's not too bad...
- */
- rxs->mactime = ath5k_extend_tsf(sc->ah, rs.rs_tstamp);
- rxs->flag = rx_flag | RX_FLAG_TSFT;
-
- rxs->freq = sc->curchan->center_freq;
- rxs->band = sc->curband->band;
-
- rxs->signal = sc->ah->ah_noise_floor + rs.rs_rssi;
-
- rxs->antenna = rs.rs_antenna;
-
- if (rs.rs_antenna > 0 && rs.rs_antenna < 5)
- sc->stats.antenna_rx[rs.rs_antenna]++;
- else
- sc->stats.antenna_rx[0]++; /* invalid */
-
- rxs->rate_idx = ath5k_hw_to_driver_rix(sc, rs.rs_rate);
- rxs->flag |= ath5k_rx_decrypted(sc, ds, skb, &rs);
-
- if (rxs->rate_idx >= 0 && rs.rs_rate ==
- sc->curband->bitrates[rxs->rate_idx].hw_value_short)
- rxs->flag |= RX_FLAG_SHORTPRE;
-
- ath5k_debug_dump_skb(sc, skb, "RX ", 0);
-
- ath5k_update_beacon_rssi(sc, skb, rs.rs_rssi);
-
- /* check beacons in IBSS mode */
- if (sc->opmode == NL80211_IFTYPE_ADHOC)
- ath5k_check_ibss_tsf(sc, skb, rxs);
-
- ieee80211_rx(sc->hw, skb);
+ ath5k_receive_frame(sc, skb, &rs);
bf->skb = next_skb;
bf->skbaddr = next_skb_addr;
^ permalink raw reply related
* [PATCH 09/17] ath5k: move checks and stats into new function
From: Bruno Randolf @ 2010-06-16 10:11 UTC (permalink / raw)
To: linville; +Cc: ath5k-devel, linux-wireless
In-Reply-To: <20100616100809.10067.34787.stgit@tt-desk>
Create a new function ath5k_receive_frame_ok() which checks for errors, updates
error statistics and tells us if we want to further "receive" this frame or
not. This way we can avoid a goto and have a cleaner separation between buffer
handling and other things.
Signed-off-by: Bruno Randolf <br1@einfach.org>
---
drivers/net/wireless/ath/ath5k/base.c | 130 ++++++++++++++++++---------------
1 files changed, 70 insertions(+), 60 deletions(-)
diff --git a/drivers/net/wireless/ath/ath5k/base.c b/drivers/net/wireless/ath/ath5k/base.c
index c54d1fd..a4482b5 100644
--- a/drivers/net/wireless/ath/ath5k/base.c
+++ b/drivers/net/wireless/ath/ath5k/base.c
@@ -1973,6 +1973,61 @@ ath5k_receive_frame(struct ath5k_softc *sc, struct sk_buff *skb,
ieee80211_rx(sc->hw, skb);
}
+/** ath5k_frame_receive_ok() - Do we want to receive this frame or not?
+ *
+ * Check if we want to further process this frame or not. Also update
+ * statistics. Return true if we want this frame, false if not.
+ */
+static bool
+ath5k_receive_frame_ok(struct ath5k_softc *sc, struct ath5k_rx_status *rs)
+{
+ sc->stats.rx_all_count++;
+
+ if (unlikely(rs->rs_status)) {
+ if (rs->rs_status & AR5K_RXERR_CRC)
+ sc->stats.rxerr_crc++;
+ if (rs->rs_status & AR5K_RXERR_FIFO)
+ sc->stats.rxerr_fifo++;
+ if (rs->rs_status & AR5K_RXERR_PHY) {
+ sc->stats.rxerr_phy++;
+ if (rs->rs_phyerr > 0 && rs->rs_phyerr < 32)
+ sc->stats.rxerr_phy_code[rs->rs_phyerr]++;
+ return false;
+ }
+ if (rs->rs_status & AR5K_RXERR_DECRYPT) {
+ /*
+ * Decrypt error. If the error occurred
+ * because there was no hardware key, then
+ * let the frame through so the upper layers
+ * can process it. This is necessary for 5210
+ * parts which have no way to setup a ``clear''
+ * key cache entry.
+ *
+ * XXX do key cache faulting
+ */
+ sc->stats.rxerr_decrypt++;
+ if (rs->rs_keyix == AR5K_RXKEYIX_INVALID &&
+ !(rs->rs_status & AR5K_RXERR_CRC))
+ return true;
+ }
+ if (rs->rs_status & AR5K_RXERR_MIC) {
+ sc->stats.rxerr_mic++;
+ return true;
+ }
+
+ /* let crypto-error packets fall through in MNTR */
+ if ((rs->rs_status & ~(AR5K_RXERR_DECRYPT|AR5K_RXERR_MIC)) ||
+ sc->opmode != NL80211_IFTYPE_MONITOR)
+ return false;
+ }
+
+ if (unlikely(rs->rs_more)) {
+ sc->stats.rxerr_jumbo++;
+ return false;
+ }
+ return true;
+}
+
static void
ath5k_tasklet_rx(unsigned long data)
{
@@ -2010,70 +2065,27 @@ ath5k_tasklet_rx(unsigned long data)
break;
}
- sc->stats.rx_all_count++;
-
- if (unlikely(rs.rs_status)) {
- if (rs.rs_status & AR5K_RXERR_CRC)
- sc->stats.rxerr_crc++;
- if (rs.rs_status & AR5K_RXERR_FIFO)
- sc->stats.rxerr_fifo++;
- if (rs.rs_status & AR5K_RXERR_PHY) {
- sc->stats.rxerr_phy++;
- if (rs.rs_phyerr > 0 && rs.rs_phyerr < 32)
- sc->stats.rxerr_phy_code[rs.rs_phyerr]++;
- goto next;
- }
- if (rs.rs_status & AR5K_RXERR_DECRYPT) {
- /*
- * Decrypt error. If the error occurred
- * because there was no hardware key, then
- * let the frame through so the upper layers
- * can process it. This is necessary for 5210
- * parts which have no way to setup a ``clear''
- * key cache entry.
- *
- * XXX do key cache faulting
- */
- sc->stats.rxerr_decrypt++;
- if (rs.rs_keyix == AR5K_RXKEYIX_INVALID &&
- !(rs.rs_status & AR5K_RXERR_CRC))
- goto accept;
- }
- if (rs.rs_status & AR5K_RXERR_MIC) {
- sc->stats.rxerr_mic++;
- goto accept;
- }
+ if (ath5k_receive_frame_ok(sc, &rs)) {
+ next_skb = ath5k_rx_skb_alloc(sc, &next_skb_addr);
- /* let crypto-error packets fall through in MNTR */
- if ((rs.rs_status &
- ~(AR5K_RXERR_DECRYPT|AR5K_RXERR_MIC)) ||
- sc->opmode != NL80211_IFTYPE_MONITOR)
+ /*
+ * If we can't replace bf->skb with a new skb under
+ * memory pressure, just skip this packet
+ */
+ if (!next_skb)
goto next;
- }
-
- if (unlikely(rs.rs_more)) {
- sc->stats.rxerr_jumbo++;
- goto next;
- }
-accept:
- next_skb = ath5k_rx_skb_alloc(sc, &next_skb_addr);
-
- /*
- * If we can't replace bf->skb with a new skb under memory
- * pressure, just skip this packet
- */
- if (!next_skb)
- goto next;
+ pci_unmap_single(sc->pdev, bf->skbaddr,
+ common->rx_bufsize,
+ PCI_DMA_FROMDEVICE);
- pci_unmap_single(sc->pdev, bf->skbaddr, common->rx_bufsize,
- PCI_DMA_FROMDEVICE);
- skb_put(skb, rs.rs_datalen);
+ skb_put(skb, rs.rs_datalen);
- ath5k_receive_frame(sc, skb, &rs);
+ ath5k_receive_frame(sc, skb, &rs);
- bf->skb = next_skb;
- bf->skbaddr = next_skb_addr;
+ bf->skb = next_skb;
+ bf->skbaddr = next_skb_addr;
+ }
next:
list_move_tail(&bf->list, &sc->rxbuf);
} while (ath5k_rxbuf_setup(sc, bf) == 0);
@@ -2082,8 +2094,6 @@ unlock:
}
-
-
/*************\
* TX Handling *
\*************/
^ permalink raw reply related
* [PATCH 10/17] ath5k: use direct function calls for descriptors when possible
From: Bruno Randolf @ 2010-06-16 10:12 UTC (permalink / raw)
To: linville; +Cc: ath5k-devel, linux-wireless
In-Reply-To: <20100616100809.10067.34787.stgit@tt-desk>
Use direct function calls for ath5k_hw_setup_rx_desc() and
ath5k_hw_setup_mrr_tx_desc() instead of a function pointer which always pointed
to the same function in the case of ath5k_hw_setup_rx_desc() and which is
easily unified in the case of ath5k_hw_setup_mrr_tx_desc().
Also simplify the initialization function for the remaining function pointers.
Signed-off-by: Bruno Randolf <br1@einfach.org>
---
drivers/net/wireless/ath/ath5k/ath5k.h | 10 ++++----
drivers/net/wireless/ath/ath5k/base.c | 7 +++--
drivers/net/wireless/ath/ath5k/desc.c | 41 +++++++++-----------------------
3 files changed, 20 insertions(+), 38 deletions(-)
diff --git a/drivers/net/wireless/ath/ath5k/ath5k.h b/drivers/net/wireless/ath/ath5k/ath5k.h
index 387c120..ea6362a 100644
--- a/drivers/net/wireless/ath/ath5k/ath5k.h
+++ b/drivers/net/wireless/ath/ath5k/ath5k.h
@@ -1127,15 +1127,10 @@ struct ath5k_hw {
/*
* Function pointers
*/
- int (*ah_setup_rx_desc)(struct ath5k_hw *ah, struct ath5k_desc *desc,
- u32 size, unsigned int flags);
int (*ah_setup_tx_desc)(struct ath5k_hw *, struct ath5k_desc *,
unsigned int, unsigned int, int, enum ath5k_pkt_type,
unsigned int, unsigned int, unsigned int, unsigned int,
unsigned int, unsigned int, unsigned int, unsigned int);
- int (*ah_setup_mrr_tx_desc)(struct ath5k_hw *, struct ath5k_desc *,
- unsigned int, unsigned int, unsigned int, unsigned int,
- unsigned int, unsigned int);
int (*ah_proc_tx_desc)(struct ath5k_hw *, struct ath5k_desc *,
struct ath5k_tx_status *);
int (*ah_proc_rx_desc)(struct ath5k_hw *, struct ath5k_desc *,
@@ -1236,6 +1231,11 @@ int ath5k_hw_set_slot_time(struct ath5k_hw *ah, unsigned int slot_time);
/* Hardware Descriptor Functions */
int ath5k_hw_init_desc_functions(struct ath5k_hw *ah);
+int ath5k_hw_setup_rx_desc(struct ath5k_hw *ah, struct ath5k_desc *desc,
+ u32 size, unsigned int flags);
+int ath5k_hw_setup_mrr_tx_desc(struct ath5k_hw *ah, struct ath5k_desc *desc,
+ unsigned int tx_rate1, u_int tx_tries1, u_int tx_rate2,
+ u_int tx_tries2, unsigned int tx_rate3, u_int tx_tries3);
/* GPIO Functions */
void ath5k_hw_set_ledstate(struct ath5k_hw *ah, unsigned int state);
diff --git a/drivers/net/wireless/ath/ath5k/base.c b/drivers/net/wireless/ath/ath5k/base.c
index a4482b5..20328bd 100644
--- a/drivers/net/wireless/ath/ath5k/base.c
+++ b/drivers/net/wireless/ath/ath5k/base.c
@@ -769,7 +769,8 @@ ath5k_attach(struct pci_dev *pdev, struct ieee80211_hw *hw)
* return false w/o doing anything. MAC's that do
* support it will return true w/o doing anything.
*/
- ret = ah->ah_setup_mrr_tx_desc(ah, NULL, 0, 0, 0, 0, 0, 0);
+ ret = ath5k_hw_setup_mrr_tx_desc(ah, NULL, 0, 0, 0, 0, 0, 0);
+
if (ret < 0)
goto err;
if (ret > 0)
@@ -1245,7 +1246,7 @@ ath5k_rxbuf_setup(struct ath5k_softc *sc, struct ath5k_buf *bf)
ds = bf->desc;
ds->ds_link = bf->daddr; /* link to self */
ds->ds_data = bf->skbaddr;
- ret = ah->ah_setup_rx_desc(ah, ds, ah->common.rx_bufsize, 0);
+ ret = ath5k_hw_setup_rx_desc(ah, ds, ah->common.rx_bufsize, 0);
if (ret) {
ATH5K_ERR(sc, "%s: could not setup RX desc\n", __func__);
return ret;
@@ -1354,7 +1355,7 @@ ath5k_txbuf_setup(struct ath5k_softc *sc, struct ath5k_buf *bf,
mrr_tries[i] = info->control.rates[i + 1].count;
}
- ah->ah_setup_mrr_tx_desc(ah, ds,
+ ath5k_hw_setup_mrr_tx_desc(ah, ds,
mrr_rate[0], mrr_tries[0],
mrr_rate[1], mrr_tries[1],
mrr_rate[2], mrr_tries[2]);
diff --git a/drivers/net/wireless/ath/ath5k/desc.c b/drivers/net/wireless/ath/ath5k/desc.c
index da5dbb6..b5a5d45 100644
--- a/drivers/net/wireless/ath/ath5k/desc.c
+++ b/drivers/net/wireless/ath/ath5k/desc.c
@@ -277,13 +277,17 @@ static int ath5k_hw_setup_4word_tx_desc(struct ath5k_hw *ah,
/*
* Initialize a 4-word multi rate retry tx control descriptor on 5212
*/
-static int
+int
ath5k_hw_setup_mrr_tx_desc(struct ath5k_hw *ah, struct ath5k_desc *desc,
unsigned int tx_rate1, u_int tx_tries1, u_int tx_rate2,
u_int tx_tries2, unsigned int tx_rate3, u_int tx_tries3)
{
struct ath5k_hw_4w_tx_ctl *tx_ctl;
+ /* no mrr support for cards older than 5212 */
+ if (ah->ah_version < AR5K_AR5212)
+ return 0;
+
/*
* Rates can be 0 as long as the retry count is 0 too.
* A zero rate and nonzero retry count will put the HW into a mode where
@@ -323,15 +327,6 @@ ath5k_hw_setup_mrr_tx_desc(struct ath5k_hw *ah, struct ath5k_desc *desc,
return 0;
}
-/* no mrr support for cards older than 5212 */
-static int
-ath5k_hw_setup_no_mrr(struct ath5k_hw *ah, struct ath5k_desc *desc,
- unsigned int tx_rate1, u_int tx_tries1, u_int tx_rate2,
- u_int tx_tries2, unsigned int tx_rate3, u_int tx_tries3)
-{
- return 0;
-}
-
/*
* Proccess the tx status descriptor on 5210/5211
*/
@@ -480,8 +475,8 @@ static int ath5k_hw_proc_4word_tx_status(struct ath5k_hw *ah,
/*
* Initialize an rx control descriptor
*/
-static int ath5k_hw_setup_rx_desc(struct ath5k_hw *ah, struct ath5k_desc *desc,
- u32 size, unsigned int flags)
+int ath5k_hw_setup_rx_desc(struct ath5k_hw *ah, struct ath5k_desc *desc,
+ u32 size, unsigned int flags)
{
struct ath5k_hw_rx_ctl *rx_ctl;
@@ -658,29 +653,15 @@ static int ath5k_hw_proc_5212_rx_status(struct ath5k_hw *ah,
*/
int ath5k_hw_init_desc_functions(struct ath5k_hw *ah)
{
-
- if (ah->ah_version != AR5K_AR5210 &&
- ah->ah_version != AR5K_AR5211 &&
- ah->ah_version != AR5K_AR5212)
- return -ENOTSUPP;
-
if (ah->ah_version == AR5K_AR5212) {
- ah->ah_setup_rx_desc = ath5k_hw_setup_rx_desc;
ah->ah_setup_tx_desc = ath5k_hw_setup_4word_tx_desc;
- ah->ah_setup_mrr_tx_desc = ath5k_hw_setup_mrr_tx_desc;
ah->ah_proc_tx_desc = ath5k_hw_proc_4word_tx_status;
- } else {
- ah->ah_setup_rx_desc = ath5k_hw_setup_rx_desc;
+ ah->ah_proc_rx_desc = ath5k_hw_proc_5212_rx_status;
+ } else if (ah->ah_version <= AR5K_AR5211) {
ah->ah_setup_tx_desc = ath5k_hw_setup_2word_tx_desc;
- ah->ah_setup_mrr_tx_desc = ath5k_hw_setup_no_mrr;
ah->ah_proc_tx_desc = ath5k_hw_proc_2word_tx_status;
- }
-
- if (ah->ah_version == AR5K_AR5212)
- ah->ah_proc_rx_desc = ath5k_hw_proc_5212_rx_status;
- else if (ah->ah_version <= AR5K_AR5211)
ah->ah_proc_rx_desc = ath5k_hw_proc_5210_rx_status;
-
+ } else
+ return -ENOTSUPP;
return 0;
}
-
^ permalink raw reply related
* [PATCH 11/17] ath5k: cosmetic changes in ath5k_hw_proc_5212_rx_status()
From: Bruno Randolf @ 2010-06-16 10:12 UTC (permalink / raw)
To: linville; +Cc: ath5k-devel, linux-wireless
In-Reply-To: <20100616100809.10067.34787.stgit@tt-desk>
Just whitespace and indentation.
Signed-off-by: Bruno Randolf <br1@einfach.org>
Acked-by: Bob Copeland <me@bobcopeland.com>
---
drivers/net/wireless/ath/ath5k/desc.c | 10 +++++-----
1 files changed, 5 insertions(+), 5 deletions(-)
diff --git a/drivers/net/wireless/ath/ath5k/desc.c b/drivers/net/wireless/ath/ath5k/desc.c
index b5a5d45..50fc931 100644
--- a/drivers/net/wireless/ath/ath5k/desc.c
+++ b/drivers/net/wireless/ath/ath5k/desc.c
@@ -577,7 +577,8 @@ static int ath5k_hw_proc_5210_rx_status(struct ath5k_hw *ah,
* Proccess the rx status descriptor on 5212
*/
static int ath5k_hw_proc_5212_rx_status(struct ath5k_hw *ah,
- struct ath5k_desc *desc, struct ath5k_rx_status *rs)
+ struct ath5k_desc *desc,
+ struct ath5k_rx_status *rs)
{
struct ath5k_hw_rx_status *rx_status;
struct ath5k_hw_rx_error *rx_err;
@@ -589,7 +590,7 @@ static int ath5k_hw_proc_5212_rx_status(struct ath5k_hw *ah,
/* No frame received / not ready */
if (unlikely(!(rx_status->rx_status_1 &
- AR5K_5212_RX_DESC_STATUS1_DONE)))
+ AR5K_5212_RX_DESC_STATUS1_DONE)))
return -EINPROGRESS;
/*
@@ -615,7 +616,7 @@ static int ath5k_hw_proc_5212_rx_status(struct ath5k_hw *ah,
*/
if (rx_status->rx_status_1 & AR5K_5212_RX_DESC_STATUS1_KEY_INDEX_VALID)
rs->rs_keyix = AR5K_REG_MS(rx_status->rx_status_1,
- AR5K_5212_RX_DESC_STATUS1_KEY_INDEX);
+ AR5K_5212_RX_DESC_STATUS1_KEY_INDEX);
else
rs->rs_keyix = AR5K_RXKEYIX_INVALID;
@@ -623,7 +624,7 @@ static int ath5k_hw_proc_5212_rx_status(struct ath5k_hw *ah,
* Receive/descriptor errors
*/
if (!(rx_status->rx_status_1 &
- AR5K_5212_RX_DESC_STATUS1_FRAME_RECEIVE_OK)) {
+ AR5K_5212_RX_DESC_STATUS1_FRAME_RECEIVE_OK)) {
if (rx_status->rx_status_1 &
AR5K_5212_RX_DESC_STATUS1_CRC_ERROR)
rs->rs_status |= AR5K_RXERR_CRC;
@@ -644,7 +645,6 @@ static int ath5k_hw_proc_5212_rx_status(struct ath5k_hw *ah,
AR5K_5212_RX_DESC_STATUS1_MIC_ERROR)
rs->rs_status |= AR5K_RXERR_MIC;
}
-
return 0;
}
^ permalink raw reply related
* [PATCH 12/17] ath5k: remove pointless rx error overlay struct
From: Bruno Randolf @ 2010-06-16 10:12 UTC (permalink / raw)
To: linville; +Cc: ath5k-devel, linux-wireless
In-Reply-To: <20100616100809.10067.34787.stgit@tt-desk>
ath5k_hw_rx_error was only used once, where we could easily just use
ath5k_hw_rx_status as well, so remove it.
Signed-off-by: Bruno Randolf <br1@einfach.org>
Acked-by: Bob Copeland <me@bobcopeland.com>
---
drivers/net/wireless/ath/ath5k/debug.c | 2 +-
drivers/net/wireless/ath/ath5k/desc.c | 12 ++++--------
drivers/net/wireless/ath/ath5k/desc.h | 24 ++++--------------------
3 files changed, 9 insertions(+), 29 deletions(-)
diff --git a/drivers/net/wireless/ath/ath5k/debug.c b/drivers/net/wireless/ath/ath5k/debug.c
index 02db66e..8c63886 100644
--- a/drivers/net/wireless/ath/ath5k/debug.c
+++ b/drivers/net/wireless/ath/ath5k/debug.c
@@ -925,7 +925,7 @@ ath5k_debug_printrxbuf(struct ath5k_buf *bf, int done,
ds, (unsigned long long)bf->daddr,
ds->ds_link, ds->ds_data,
rd->rx_ctl.rx_control_0, rd->rx_ctl.rx_control_1,
- rd->u.rx_stat.rx_status_0, rd->u.rx_stat.rx_status_1,
+ rd->rx_stat.rx_status_0, rd->rx_stat.rx_status_1,
!done ? ' ' : (rs->rs_status == 0) ? '*' : '!');
}
diff --git a/drivers/net/wireless/ath/ath5k/desc.c b/drivers/net/wireless/ath/ath5k/desc.c
index 50fc931..eb1427c 100644
--- a/drivers/net/wireless/ath/ath5k/desc.c
+++ b/drivers/net/wireless/ath/ath5k/desc.c
@@ -510,7 +510,7 @@ static int ath5k_hw_proc_5210_rx_status(struct ath5k_hw *ah,
{
struct ath5k_hw_rx_status *rx_status;
- rx_status = &desc->ud.ds_rx.u.rx_stat;
+ rx_status = &desc->ud.ds_rx.rx_stat;
/* No frame received / not ready */
if (unlikely(!(rx_status->rx_status_1 &
@@ -581,12 +581,8 @@ static int ath5k_hw_proc_5212_rx_status(struct ath5k_hw *ah,
struct ath5k_rx_status *rs)
{
struct ath5k_hw_rx_status *rx_status;
- struct ath5k_hw_rx_error *rx_err;
- rx_status = &desc->ud.ds_rx.u.rx_stat;
-
- /* Overlay on error */
- rx_err = &desc->ud.ds_rx.u.rx_err;
+ rx_status = &desc->ud.ds_rx.rx_stat;
/* No frame received / not ready */
if (unlikely(!(rx_status->rx_status_1 &
@@ -632,8 +628,8 @@ static int ath5k_hw_proc_5212_rx_status(struct ath5k_hw *ah,
if (rx_status->rx_status_1 &
AR5K_5212_RX_DESC_STATUS1_PHY_ERROR) {
rs->rs_status |= AR5K_RXERR_PHY;
- rs->rs_phyerr |= AR5K_REG_MS(rx_err->rx_error_1,
- AR5K_RX_DESC_ERROR1_PHY_ERROR_CODE);
+ rs->rs_phyerr |= AR5K_REG_MS(rx_status->rx_status_1,
+ AR5K_5212_RX_DESC_STATUS1_PHY_ERROR_CODE);
ath5k_ani_phy_error_report(ah, rs->rs_phyerr);
}
diff --git a/drivers/net/wireless/ath/ath5k/desc.h b/drivers/net/wireless/ath/ath5k/desc.h
index 64538fb..45f2644 100644
--- a/drivers/net/wireless/ath/ath5k/desc.h
+++ b/drivers/net/wireless/ath/ath5k/desc.h
@@ -96,21 +96,8 @@ struct ath5k_hw_rx_status {
#define AR5K_5212_RX_DESC_STATUS1_RECEIVE_TIMESTAMP 0x7fff0000
#define AR5K_5212_RX_DESC_STATUS1_RECEIVE_TIMESTAMP_S 16
#define AR5K_5212_RX_DESC_STATUS1_KEY_CACHE_MISS 0x80000000
-
-/*
- * common hardware RX error descriptor
- */
-struct ath5k_hw_rx_error {
- u32 rx_error_0; /* RX status word 0 */
- u32 rx_error_1; /* RX status word 1 */
-} __packed;
-
-/* RX error word 0 fields/flags */
-#define AR5K_RX_DESC_ERROR0 0x00000000
-
-/* RX error word 1 fields/flags */
-#define AR5K_RX_DESC_ERROR1_PHY_ERROR_CODE 0x0000ff00
-#define AR5K_RX_DESC_ERROR1_PHY_ERROR_CODE_S 8
+#define AR5K_5212_RX_DESC_STATUS1_PHY_ERROR_CODE 0x0000ff00
+#define AR5K_5212_RX_DESC_STATUS1_PHY_ERROR_CODE_S 8
/**
* enum ath5k_phy_error_code - PHY Error codes
@@ -316,11 +303,8 @@ struct ath5k_hw_5212_tx_desc {
* common hardware RX descriptor
*/
struct ath5k_hw_all_rx_desc {
- struct ath5k_hw_rx_ctl rx_ctl;
- union {
- struct ath5k_hw_rx_status rx_stat;
- struct ath5k_hw_rx_error rx_err;
- } u;
+ struct ath5k_hw_rx_ctl rx_ctl;
+ struct ath5k_hw_rx_status rx_stat;
} __packed;
/*
^ permalink raw reply related
* [PATCH 13/17] ath5k: review and add comments for descriptors
From: Bruno Randolf @ 2010-06-16 10:12 UTC (permalink / raw)
To: linville; +Cc: ath5k-devel, linux-wireless
In-Reply-To: <20100616100809.10067.34787.stgit@tt-desk>
I carefully reviewed desh.h against the HAL sources. Added comments and made
differences between 5210, 5211 and 5212 more clear by adding _521x to the
defines which are specific to that chipset. Renamed some defines. No functional
changes.
Signed-off-by: Bruno Randolf <br1@einfach.org>
Acked-by: Bob Copeland <me@bobcopeland.com>
---
drivers/net/wireless/ath/ath5k/desc.c | 22 +--
drivers/net/wireless/ath/ath5k/desc.h | 276 ++++++++++++++++-----------------
2 files changed, 144 insertions(+), 154 deletions(-)
diff --git a/drivers/net/wireless/ath/ath5k/desc.c b/drivers/net/wireless/ath/ath5k/desc.c
index eb1427c..41a490e 100644
--- a/drivers/net/wireless/ath/ath5k/desc.c
+++ b/drivers/net/wireless/ath/ath5k/desc.c
@@ -95,10 +95,10 @@ ath5k_hw_setup_2word_tx_desc(struct ath5k_hw *ah, struct ath5k_desc *desc,
* XXX: I only found that on 5210 code, does it work on 5211 ?
*/
if (ah->ah_version == AR5K_AR5210) {
- if (hdr_len & ~AR5K_2W_TX_DESC_CTL0_HEADER_LEN)
+ if (hdr_len & ~AR5K_2W_TX_DESC_CTL0_HEADER_LEN_5210)
return -EINVAL;
tx_ctl->tx_control_0 |=
- AR5K_REG_SM(hdr_len, AR5K_2W_TX_DESC_CTL0_HEADER_LEN);
+ AR5K_REG_SM(hdr_len, AR5K_2W_TX_DESC_CTL0_HEADER_LEN_5210);
}
/*Differences between 5210-5211*/
@@ -114,7 +114,7 @@ ath5k_hw_setup_2word_tx_desc(struct ath5k_hw *ah, struct ath5k_desc *desc,
}
tx_ctl->tx_control_0 |=
- AR5K_REG_SM(frame_type, AR5K_2W_TX_DESC_CTL0_FRAME_TYPE) |
+ AR5K_REG_SM(frame_type, AR5K_2W_TX_DESC_CTL0_FRAME_TYPE_5210) |
AR5K_REG_SM(tx_rate0, AR5K_2W_TX_DESC_CTL0_XMIT_RATE);
} else {
@@ -123,7 +123,7 @@ ath5k_hw_setup_2word_tx_desc(struct ath5k_hw *ah, struct ath5k_desc *desc,
AR5K_REG_SM(antenna_mode,
AR5K_2W_TX_DESC_CTL0_ANT_MODE_XMIT);
tx_ctl->tx_control_1 |=
- AR5K_REG_SM(type, AR5K_2W_TX_DESC_CTL1_FRAME_TYPE);
+ AR5K_REG_SM(type, AR5K_2W_TX_DESC_CTL1_FRAME_TYPE_5211);
}
#define _TX_FLAGS(_c, _flag) \
if (flags & AR5K_TXDESC_##_flag) { \
@@ -147,7 +147,7 @@ ath5k_hw_setup_2word_tx_desc(struct ath5k_hw *ah, struct ath5k_desc *desc,
AR5K_2W_TX_DESC_CTL0_ENCRYPT_KEY_VALID;
tx_ctl->tx_control_1 |=
AR5K_REG_SM(key_index,
- AR5K_2W_TX_DESC_CTL1_ENCRYPT_KEY_INDEX);
+ AR5K_2W_TX_DESC_CTL1_ENC_KEY_IDX);
}
/*
@@ -156,7 +156,7 @@ ath5k_hw_setup_2word_tx_desc(struct ath5k_hw *ah, struct ath5k_desc *desc,
if ((ah->ah_version == AR5K_AR5210) &&
(flags & (AR5K_TXDESC_RTSENA | AR5K_TXDESC_CTSENA)))
tx_ctl->tx_control_1 |= rtscts_duration &
- AR5K_2W_TX_DESC_CTL1_RTS_DURATION;
+ AR5K_2W_TX_DESC_CTL1_RTS_DURATION_5210;
return 0;
}
@@ -255,7 +255,7 @@ static int ath5k_hw_setup_4word_tx_desc(struct ath5k_hw *ah,
if (key_index != AR5K_TXKEYIX_INVALID) {
tx_ctl->tx_control_0 |= AR5K_4W_TX_DESC_CTL0_ENCRYPT_KEY_VALID;
tx_ctl->tx_control_1 |= AR5K_REG_SM(key_index,
- AR5K_4W_TX_DESC_CTL1_ENCRYPT_KEY_INDEX);
+ AR5K_4W_TX_DESC_CTL1_ENCRYPT_KEY_IDX);
}
/*
@@ -409,11 +409,11 @@ static int ath5k_hw_proc_4word_tx_status(struct ath5k_hw *ah,
ts->ts_rssi = AR5K_REG_MS(tx_status->tx_status_1,
AR5K_DESC_TX_STATUS1_ACK_SIG_STRENGTH);
ts->ts_antenna = (tx_status->tx_status_1 &
- AR5K_DESC_TX_STATUS1_XMIT_ANTENNA) ? 2 : 1;
+ AR5K_DESC_TX_STATUS1_XMIT_ANTENNA_5212) ? 2 : 1;
ts->ts_status = 0;
ts->ts_final_idx = AR5K_REG_MS(tx_status->tx_status_1,
- AR5K_DESC_TX_STATUS1_FINAL_TS_INDEX);
+ AR5K_DESC_TX_STATUS1_FINAL_TS_IX_5212);
/* The longretry counter has the number of un-acked retries
* for the final rate. To get the total number of retries
@@ -527,7 +527,7 @@ static int ath5k_hw_proc_5210_rx_status(struct ath5k_hw *ah,
rs->rs_rate = AR5K_REG_MS(rx_status->rx_status_0,
AR5K_5210_RX_DESC_STATUS0_RECEIVE_RATE);
rs->rs_antenna = AR5K_REG_MS(rx_status->rx_status_0,
- AR5K_5210_RX_DESC_STATUS0_RECEIVE_ANTENNA);
+ AR5K_5210_RX_DESC_STATUS0_RECEIVE_ANT_5211);
rs->rs_more = !!(rx_status->rx_status_0 &
AR5K_5210_RX_DESC_STATUS0_MORE);
/* TODO: this timestamp is 13 bit, later on we assume 15 bit */
@@ -555,7 +555,7 @@ static int ath5k_hw_proc_5210_rx_status(struct ath5k_hw *ah,
rs->rs_status |= AR5K_RXERR_CRC;
if (rx_status->rx_status_1 &
- AR5K_5210_RX_DESC_STATUS1_FIFO_OVERRUN)
+ AR5K_5210_RX_DESC_STATUS1_FIFO_OVERRUN_5210)
rs->rs_status |= AR5K_RXERR_FIFO;
if (rx_status->rx_status_1 &
diff --git a/drivers/net/wireless/ath/ath5k/desc.h b/drivers/net/wireless/ath/ath5k/desc.h
index 45f2644..50f9a12 100644
--- a/drivers/net/wireless/ath/ath5k/desc.h
+++ b/drivers/net/wireless/ath/ath5k/desc.h
@@ -17,28 +17,24 @@
*/
/*
- * Internal RX/TX descriptor structures
- * (rX: reserved fields possibily used by future versions of the ar5k chipset)
+ * RX/TX descriptor structures
*/
/*
- * common hardware RX control descriptor
+ * Common hardware RX control descriptor
*/
struct ath5k_hw_rx_ctl {
u32 rx_control_0; /* RX control word 0 */
u32 rx_control_1; /* RX control word 1 */
} __packed;
-/* RX control word 0 field/sflags */
-#define AR5K_DESC_RX_CTL0 0x00000000
-
/* RX control word 1 fields/flags */
-#define AR5K_DESC_RX_CTL1_BUF_LEN 0x00000fff
-#define AR5K_DESC_RX_CTL1_INTREQ 0x00002000
+#define AR5K_DESC_RX_CTL1_BUF_LEN 0x00000fff /* data buffer length */
+#define AR5K_DESC_RX_CTL1_INTREQ 0x00002000 /* RX interrupt request */
/*
- * common hardware RX status descriptor
- * 5210/11 and 5212 differ only in the flags defined below
+ * Common hardware RX status descriptor
+ * 5210, 5211 and 5212 differ only in the fields and flags defined below
*/
struct ath5k_hw_rx_status {
u32 rx_status_0; /* RX status word 0 */
@@ -47,68 +43,69 @@ struct ath5k_hw_rx_status {
/* 5210/5211 */
/* RX status word 0 fields/flags */
-#define AR5K_5210_RX_DESC_STATUS0_DATA_LEN 0x00000fff
-#define AR5K_5210_RX_DESC_STATUS0_MORE 0x00001000
-#define AR5K_5210_RX_DESC_STATUS0_RECEIVE_RATE 0x00078000
+#define AR5K_5210_RX_DESC_STATUS0_DATA_LEN 0x00000fff /* RX data length */
+#define AR5K_5210_RX_DESC_STATUS0_MORE 0x00001000 /* more desc for this frame */
+#define AR5K_5210_RX_DESC_STATUS0_RECEIVE_ANT_5210 0x00004000 /* [5210] receive on ant 1 TODO */
+#define AR5K_5210_RX_DESC_STATUS0_RECEIVE_RATE 0x00078000 /* reception rate */
#define AR5K_5210_RX_DESC_STATUS0_RECEIVE_RATE_S 15
-#define AR5K_5210_RX_DESC_STATUS0_RECEIVE_SIGNAL 0x07f80000
+#define AR5K_5210_RX_DESC_STATUS0_RECEIVE_SIGNAL 0x07f80000 /* rssi */
#define AR5K_5210_RX_DESC_STATUS0_RECEIVE_SIGNAL_S 19
-#define AR5K_5210_RX_DESC_STATUS0_RECEIVE_ANTENNA 0x38000000
-#define AR5K_5210_RX_DESC_STATUS0_RECEIVE_ANTENNA_S 27
+#define AR5K_5210_RX_DESC_STATUS0_RECEIVE_ANT_5211 0x38000000 /* [5211] receive antenna */
+#define AR5K_5210_RX_DESC_STATUS0_RECEIVE_ANT_5211_S 27
/* RX status word 1 fields/flags */
-#define AR5K_5210_RX_DESC_STATUS1_DONE 0x00000001
-#define AR5K_5210_RX_DESC_STATUS1_FRAME_RECEIVE_OK 0x00000002
-#define AR5K_5210_RX_DESC_STATUS1_CRC_ERROR 0x00000004
-#define AR5K_5210_RX_DESC_STATUS1_FIFO_OVERRUN 0x00000008
-#define AR5K_5210_RX_DESC_STATUS1_DECRYPT_CRC_ERROR 0x00000010
-#define AR5K_5210_RX_DESC_STATUS1_PHY_ERROR 0x000000e0
+#define AR5K_5210_RX_DESC_STATUS1_DONE 0x00000001 /* descriptor complete */
+#define AR5K_5210_RX_DESC_STATUS1_FRAME_RECEIVE_OK 0x00000002 /* reception success */
+#define AR5K_5210_RX_DESC_STATUS1_CRC_ERROR 0x00000004 /* CRC error */
+#define AR5K_5210_RX_DESC_STATUS1_FIFO_OVERRUN_5210 0x00000008 /* [5210] FIFO overrun */
+#define AR5K_5210_RX_DESC_STATUS1_DECRYPT_CRC_ERROR 0x00000010 /* decyption CRC failure */
+#define AR5K_5210_RX_DESC_STATUS1_PHY_ERROR 0x000000e0 /* PHY error */
#define AR5K_5210_RX_DESC_STATUS1_PHY_ERROR_S 5
-#define AR5K_5210_RX_DESC_STATUS1_KEY_INDEX_VALID 0x00000100
-#define AR5K_5210_RX_DESC_STATUS1_KEY_INDEX 0x00007e00
+#define AR5K_5210_RX_DESC_STATUS1_KEY_INDEX_VALID 0x00000100 /* key index valid */
+#define AR5K_5210_RX_DESC_STATUS1_KEY_INDEX 0x00007e00 /* decyption key index */
#define AR5K_5210_RX_DESC_STATUS1_KEY_INDEX_S 9
-#define AR5K_5210_RX_DESC_STATUS1_RECEIVE_TIMESTAMP 0x0fff8000
+#define AR5K_5210_RX_DESC_STATUS1_RECEIVE_TIMESTAMP 0x0fff8000 /* 13 bit of TSF */
#define AR5K_5210_RX_DESC_STATUS1_RECEIVE_TIMESTAMP_S 15
-#define AR5K_5210_RX_DESC_STATUS1_KEY_CACHE_MISS 0x10000000
+#define AR5K_5210_RX_DESC_STATUS1_KEY_CACHE_MISS 0x10000000 /* key cache miss */
/* 5212 */
/* RX status word 0 fields/flags */
-#define AR5K_5212_RX_DESC_STATUS0_DATA_LEN 0x00000fff
-#define AR5K_5212_RX_DESC_STATUS0_MORE 0x00001000
-#define AR5K_5212_RX_DESC_STATUS0_DECOMP_CRC_ERROR 0x00002000
-#define AR5K_5212_RX_DESC_STATUS0_RECEIVE_RATE 0x000f8000
+#define AR5K_5212_RX_DESC_STATUS0_DATA_LEN 0x00000fff /* RX data length */
+#define AR5K_5212_RX_DESC_STATUS0_MORE 0x00001000 /* more desc for this frame */
+#define AR5K_5212_RX_DESC_STATUS0_DECOMP_CRC_ERROR 0x00002000 /* decompression CRC error */
+#define AR5K_5212_RX_DESC_STATUS0_RECEIVE_RATE 0x000f8000 /* reception rate */
#define AR5K_5212_RX_DESC_STATUS0_RECEIVE_RATE_S 15
-#define AR5K_5212_RX_DESC_STATUS0_RECEIVE_SIGNAL 0x0ff00000
+#define AR5K_5212_RX_DESC_STATUS0_RECEIVE_SIGNAL 0x0ff00000 /* rssi */
#define AR5K_5212_RX_DESC_STATUS0_RECEIVE_SIGNAL_S 20
-#define AR5K_5212_RX_DESC_STATUS0_RECEIVE_ANTENNA 0xf0000000
+#define AR5K_5212_RX_DESC_STATUS0_RECEIVE_ANTENNA 0xf0000000 /* receive antenna */
#define AR5K_5212_RX_DESC_STATUS0_RECEIVE_ANTENNA_S 28
/* RX status word 1 fields/flags */
-#define AR5K_5212_RX_DESC_STATUS1_DONE 0x00000001
-#define AR5K_5212_RX_DESC_STATUS1_FRAME_RECEIVE_OK 0x00000002
-#define AR5K_5212_RX_DESC_STATUS1_CRC_ERROR 0x00000004
-#define AR5K_5212_RX_DESC_STATUS1_DECRYPT_CRC_ERROR 0x00000008
-#define AR5K_5212_RX_DESC_STATUS1_PHY_ERROR 0x00000010
-#define AR5K_5212_RX_DESC_STATUS1_MIC_ERROR 0x00000020
-#define AR5K_5212_RX_DESC_STATUS1_KEY_INDEX_VALID 0x00000100
-#define AR5K_5212_RX_DESC_STATUS1_KEY_INDEX 0x0000fe00
+#define AR5K_5212_RX_DESC_STATUS1_DONE 0x00000001 /* descriptor complete */
+#define AR5K_5212_RX_DESC_STATUS1_FRAME_RECEIVE_OK 0x00000002 /* frame reception success */
+#define AR5K_5212_RX_DESC_STATUS1_CRC_ERROR 0x00000004 /* CRC error */
+#define AR5K_5212_RX_DESC_STATUS1_DECRYPT_CRC_ERROR 0x00000008 /* decryption CRC failure */
+#define AR5K_5212_RX_DESC_STATUS1_PHY_ERROR 0x00000010 /* PHY error */
+#define AR5K_5212_RX_DESC_STATUS1_MIC_ERROR 0x00000020 /* MIC decrypt error */
+#define AR5K_5212_RX_DESC_STATUS1_KEY_INDEX_VALID 0x00000100 /* key index valid */
+#define AR5K_5212_RX_DESC_STATUS1_KEY_INDEX 0x0000fe00 /* decryption key index */
#define AR5K_5212_RX_DESC_STATUS1_KEY_INDEX_S 9
-#define AR5K_5212_RX_DESC_STATUS1_RECEIVE_TIMESTAMP 0x7fff0000
+#define AR5K_5212_RX_DESC_STATUS1_RECEIVE_TIMESTAMP 0x7fff0000 /* first 15bit of the TSF */
#define AR5K_5212_RX_DESC_STATUS1_RECEIVE_TIMESTAMP_S 16
-#define AR5K_5212_RX_DESC_STATUS1_KEY_CACHE_MISS 0x80000000
-#define AR5K_5212_RX_DESC_STATUS1_PHY_ERROR_CODE 0x0000ff00
+#define AR5K_5212_RX_DESC_STATUS1_KEY_CACHE_MISS 0x80000000 /* key cache miss */
+#define AR5K_5212_RX_DESC_STATUS1_PHY_ERROR_CODE 0x0000ff00 /* phy error code overlays key index and valid fields */
#define AR5K_5212_RX_DESC_STATUS1_PHY_ERROR_CODE_S 8
/**
* enum ath5k_phy_error_code - PHY Error codes
*/
enum ath5k_phy_error_code {
- AR5K_RX_PHY_ERROR_UNDERRUN = 0, /* Transmit underrun */
+ AR5K_RX_PHY_ERROR_UNDERRUN = 0, /* Transmit underrun, [5210] No error */
AR5K_RX_PHY_ERROR_TIMING = 1, /* Timing error */
AR5K_RX_PHY_ERROR_PARITY = 2, /* Illegal parity */
AR5K_RX_PHY_ERROR_RATE = 3, /* Illegal rate */
AR5K_RX_PHY_ERROR_LENGTH = 4, /* Illegal length */
- AR5K_RX_PHY_ERROR_RADAR = 5, /* Radar detect */
+ AR5K_RX_PHY_ERROR_RADAR = 5, /* Radar detect, [5210] 64 QAM rate */
AR5K_RX_PHY_ERROR_SERVICE = 6, /* Illegal service */
AR5K_RX_PHY_ERROR_TOR = 7, /* Transmit override receive */
/* these are specific to the 5212 */
@@ -135,45 +132,41 @@ struct ath5k_hw_2w_tx_ctl {
} __packed;
/* TX control word 0 fields/flags */
-#define AR5K_2W_TX_DESC_CTL0_FRAME_LEN 0x00000fff
-#define AR5K_2W_TX_DESC_CTL0_HEADER_LEN 0x0003f000 /*[5210 ?]*/
-#define AR5K_2W_TX_DESC_CTL0_HEADER_LEN_S 12
-#define AR5K_2W_TX_DESC_CTL0_XMIT_RATE 0x003c0000
+#define AR5K_2W_TX_DESC_CTL0_FRAME_LEN 0x00000fff /* frame length */
+#define AR5K_2W_TX_DESC_CTL0_HEADER_LEN_5210 0x0003f000 /* [5210] header length */
+#define AR5K_2W_TX_DESC_CTL0_HEADER_LEN_5210_S 12
+#define AR5K_2W_TX_DESC_CTL0_XMIT_RATE 0x003c0000 /* tx rate */
#define AR5K_2W_TX_DESC_CTL0_XMIT_RATE_S 18
-#define AR5K_2W_TX_DESC_CTL0_RTSENA 0x00400000
-#define AR5K_2W_TX_DESC_CTL0_CLRDMASK 0x01000000
-#define AR5K_2W_TX_DESC_CTL0_LONG_PACKET 0x00800000 /*[5210]*/
-#define AR5K_2W_TX_DESC_CTL0_VEOL 0x00800000 /*[5211]*/
-#define AR5K_2W_TX_DESC_CTL0_FRAME_TYPE 0x1c000000 /*[5210]*/
-#define AR5K_2W_TX_DESC_CTL0_FRAME_TYPE_S 26
-#define AR5K_2W_TX_DESC_CTL0_ANT_MODE_XMIT_5210 0x02000000
-#define AR5K_2W_TX_DESC_CTL0_ANT_MODE_XMIT_5211 0x1e000000
-
+#define AR5K_2W_TX_DESC_CTL0_RTSENA 0x00400000 /* RTS/CTS enable */
+#define AR5K_2W_TX_DESC_CTL0_LONG_PACKET_5210 0x00800000 /* [5210] long packet */
+#define AR5K_2W_TX_DESC_CTL0_VEOL 0x00800000 /* [5211] virtual end-of-list TODO */
+#define AR5K_2W_TX_DESC_CTL0_CLRDMASK 0x01000000 /* clear destination mask */
+#define AR5K_2W_TX_DESC_CTL0_ANT_MODE_XMIT_5210 0x02000000 /* [5210] antenna selection */
+#define AR5K_2W_TX_DESC_CTL0_ANT_MODE_XMIT_5211 0x1e000000 /* [5211] antenna selection */
#define AR5K_2W_TX_DESC_CTL0_ANT_MODE_XMIT \
(ah->ah_version == AR5K_AR5210 ? \
AR5K_2W_TX_DESC_CTL0_ANT_MODE_XMIT_5210 : \
AR5K_2W_TX_DESC_CTL0_ANT_MODE_XMIT_5211)
-
#define AR5K_2W_TX_DESC_CTL0_ANT_MODE_XMIT_S 25
-#define AR5K_2W_TX_DESC_CTL0_INTREQ 0x20000000
-#define AR5K_2W_TX_DESC_CTL0_ENCRYPT_KEY_VALID 0x40000000
+#define AR5K_2W_TX_DESC_CTL0_FRAME_TYPE_5210 0x1c000000 /* [5210] frame type */
+#define AR5K_2W_TX_DESC_CTL0_FRAME_TYPE_5210_S 26
+#define AR5K_2W_TX_DESC_CTL0_INTREQ 0x20000000 /* TX interrupt request */
+#define AR5K_2W_TX_DESC_CTL0_ENCRYPT_KEY_VALID 0x40000000 /* key is valid */
/* TX control word 1 fields/flags */
-#define AR5K_2W_TX_DESC_CTL1_BUF_LEN 0x00000fff
-#define AR5K_2W_TX_DESC_CTL1_MORE 0x00001000
-#define AR5K_2W_TX_DESC_CTL1_ENCRYPT_KEY_INDEX_5210 0x0007e000
-#define AR5K_2W_TX_DESC_CTL1_ENCRYPT_KEY_INDEX_5211 0x000fe000
-
-#define AR5K_2W_TX_DESC_CTL1_ENCRYPT_KEY_INDEX \
+#define AR5K_2W_TX_DESC_CTL1_BUF_LEN 0x00000fff /* data buffer length */
+#define AR5K_2W_TX_DESC_CTL1_MORE 0x00001000 /* more desc for this frame */
+#define AR5K_2W_TX_DESC_CTL1_ENC_KEY_IDX_5210 0x0007e000 /* [5210] key table index */
+#define AR5K_2W_TX_DESC_CTL1_ENC_KEY_IDX_5211 0x000fe000 /* [5211] key table index */
+#define AR5K_2W_TX_DESC_CTL1_ENC_KEY_IDX \
(ah->ah_version == AR5K_AR5210 ? \
- AR5K_2W_TX_DESC_CTL1_ENCRYPT_KEY_INDEX_5210 : \
- AR5K_2W_TX_DESC_CTL1_ENCRYPT_KEY_INDEX_5211)
-
-#define AR5K_2W_TX_DESC_CTL1_ENCRYPT_KEY_INDEX_S 13
-#define AR5K_2W_TX_DESC_CTL1_FRAME_TYPE 0x00700000 /*[5211]*/
-#define AR5K_2W_TX_DESC_CTL1_FRAME_TYPE_S 20
-#define AR5K_2W_TX_DESC_CTL1_NOACK 0x00800000 /*[5211]*/
-#define AR5K_2W_TX_DESC_CTL1_RTS_DURATION 0xfff80000 /*[5210 ?]*/
+ AR5K_2W_TX_DESC_CTL1_ENC_KEY_IDX_5210 : \
+ AR5K_2W_TX_DESC_CTL1_ENC_KEY_IDX_5211)
+#define AR5K_2W_TX_DESC_CTL1_ENC_KEY_IDX_S 13
+#define AR5K_2W_TX_DESC_CTL1_FRAME_TYPE_5211 0x00700000 /* [5211] frame type */
+#define AR5K_2W_TX_DESC_CTL1_FRAME_TYPE_5211_S 20
+#define AR5K_2W_TX_DESC_CTL1_NOACK 0x00800000 /* [5211] no ACK TODO */
+#define AR5K_2W_TX_DESC_CTL1_RTS_DURATION_5210 0xfff80000 /* [5210] lower 13 bit of duration */
/* Frame types */
#define AR5K_AR5210_TX_DESC_FRAME_TYPE_NORMAL 0x00
@@ -187,60 +180,61 @@ struct ath5k_hw_2w_tx_ctl {
*/
struct ath5k_hw_4w_tx_ctl {
u32 tx_control_0; /* TX control word 0 */
+ u32 tx_control_1; /* TX control word 1 */
+ u32 tx_control_2; /* TX control word 2 */
+ u32 tx_control_3; /* TX control word 3 */
+} __packed;
-#define AR5K_4W_TX_DESC_CTL0_FRAME_LEN 0x00000fff
-#define AR5K_4W_TX_DESC_CTL0_XMIT_POWER 0x003f0000
+/* TX control word 0 fields/flags */
+#define AR5K_4W_TX_DESC_CTL0_FRAME_LEN 0x00000fff /* frame length */
+#define AR5K_4W_TX_DESC_CTL0_XMIT_POWER 0x003f0000 /* transmit power */
#define AR5K_4W_TX_DESC_CTL0_XMIT_POWER_S 16
-#define AR5K_4W_TX_DESC_CTL0_RTSENA 0x00400000
-#define AR5K_4W_TX_DESC_CTL0_VEOL 0x00800000
-#define AR5K_4W_TX_DESC_CTL0_CLRDMASK 0x01000000
-#define AR5K_4W_TX_DESC_CTL0_ANT_MODE_XMIT 0x1e000000
+#define AR5K_4W_TX_DESC_CTL0_RTSENA 0x00400000 /* RTS/CTS enable */
+#define AR5K_4W_TX_DESC_CTL0_VEOL 0x00800000 /* virtual end-of-list */
+#define AR5K_4W_TX_DESC_CTL0_CLRDMASK 0x01000000 /* clear destination mask */
+#define AR5K_4W_TX_DESC_CTL0_ANT_MODE_XMIT 0x1e000000 /* TX antenna selection */
#define AR5K_4W_TX_DESC_CTL0_ANT_MODE_XMIT_S 25
-#define AR5K_4W_TX_DESC_CTL0_INTREQ 0x20000000
-#define AR5K_4W_TX_DESC_CTL0_ENCRYPT_KEY_VALID 0x40000000
-#define AR5K_4W_TX_DESC_CTL0_CTSENA 0x80000000
+#define AR5K_4W_TX_DESC_CTL0_INTREQ 0x20000000 /* TX interrupt request */
+#define AR5K_4W_TX_DESC_CTL0_ENCRYPT_KEY_VALID 0x40000000 /* destination index valid */
+#define AR5K_4W_TX_DESC_CTL0_CTSENA 0x80000000 /* precede frame with CTS */
- u32 tx_control_1; /* TX control word 1 */
-
-#define AR5K_4W_TX_DESC_CTL1_BUF_LEN 0x00000fff
-#define AR5K_4W_TX_DESC_CTL1_MORE 0x00001000
-#define AR5K_4W_TX_DESC_CTL1_ENCRYPT_KEY_INDEX 0x000fe000
-#define AR5K_4W_TX_DESC_CTL1_ENCRYPT_KEY_INDEX_S 13
-#define AR5K_4W_TX_DESC_CTL1_FRAME_TYPE 0x00f00000
+/* TX control word 1 fields/flags */
+#define AR5K_4W_TX_DESC_CTL1_BUF_LEN 0x00000fff /* data buffer length */
+#define AR5K_4W_TX_DESC_CTL1_MORE 0x00001000 /* more desc for this frame */
+#define AR5K_4W_TX_DESC_CTL1_ENCRYPT_KEY_IDX 0x000fe000 /* destination table index */
+#define AR5K_4W_TX_DESC_CTL1_ENCRYPT_KEY_IDX_S 13
+#define AR5K_4W_TX_DESC_CTL1_FRAME_TYPE 0x00f00000 /* frame type */
#define AR5K_4W_TX_DESC_CTL1_FRAME_TYPE_S 20
-#define AR5K_4W_TX_DESC_CTL1_NOACK 0x01000000
-#define AR5K_4W_TX_DESC_CTL1_COMP_PROC 0x06000000
+#define AR5K_4W_TX_DESC_CTL1_NOACK 0x01000000 /* no ACK */
+#define AR5K_4W_TX_DESC_CTL1_COMP_PROC 0x06000000 /* compression processing */
#define AR5K_4W_TX_DESC_CTL1_COMP_PROC_S 25
-#define AR5K_4W_TX_DESC_CTL1_COMP_IV_LEN 0x18000000
+#define AR5K_4W_TX_DESC_CTL1_COMP_IV_LEN 0x18000000 /* length of frame IV */
#define AR5K_4W_TX_DESC_CTL1_COMP_IV_LEN_S 27
-#define AR5K_4W_TX_DESC_CTL1_COMP_ICV_LEN 0x60000000
+#define AR5K_4W_TX_DESC_CTL1_COMP_ICV_LEN 0x60000000 /* length of frame ICV */
#define AR5K_4W_TX_DESC_CTL1_COMP_ICV_LEN_S 29
- u32 tx_control_2; /* TX control word 2 */
-
-#define AR5K_4W_TX_DESC_CTL2_RTS_DURATION 0x00007fff
-#define AR5K_4W_TX_DESC_CTL2_DURATION_UPDATE_ENABLE 0x00008000
-#define AR5K_4W_TX_DESC_CTL2_XMIT_TRIES0 0x000f0000
-#define AR5K_4W_TX_DESC_CTL2_XMIT_TRIES0_S 16
-#define AR5K_4W_TX_DESC_CTL2_XMIT_TRIES1 0x00f00000
-#define AR5K_4W_TX_DESC_CTL2_XMIT_TRIES1_S 20
-#define AR5K_4W_TX_DESC_CTL2_XMIT_TRIES2 0x0f000000
-#define AR5K_4W_TX_DESC_CTL2_XMIT_TRIES2_S 24
-#define AR5K_4W_TX_DESC_CTL2_XMIT_TRIES3 0xf0000000
-#define AR5K_4W_TX_DESC_CTL2_XMIT_TRIES3_S 28
-
- u32 tx_control_3; /* TX control word 3 */
-
-#define AR5K_4W_TX_DESC_CTL3_XMIT_RATE0 0x0000001f
-#define AR5K_4W_TX_DESC_CTL3_XMIT_RATE1 0x000003e0
+/* TX control word 2 fields/flags */
+#define AR5K_4W_TX_DESC_CTL2_RTS_DURATION 0x00007fff /* RTS/CTS duration */
+#define AR5K_4W_TX_DESC_CTL2_DURATION_UPD_EN 0x00008000 /* frame duration update */
+#define AR5K_4W_TX_DESC_CTL2_XMIT_TRIES0 0x000f0000 /* series 0 max attempts */
+#define AR5K_4W_TX_DESC_CTL2_XMIT_TRIES0_S 16
+#define AR5K_4W_TX_DESC_CTL2_XMIT_TRIES1 0x00f00000 /* series 1 max attempts */
+#define AR5K_4W_TX_DESC_CTL2_XMIT_TRIES1_S 20
+#define AR5K_4W_TX_DESC_CTL2_XMIT_TRIES2 0x0f000000 /* series 2 max attempts */
+#define AR5K_4W_TX_DESC_CTL2_XMIT_TRIES2_S 24
+#define AR5K_4W_TX_DESC_CTL2_XMIT_TRIES3 0xf0000000 /* series 3 max attempts */
+#define AR5K_4W_TX_DESC_CTL2_XMIT_TRIES3_S 28
+
+/* TX control word 3 fields/flags */
+#define AR5K_4W_TX_DESC_CTL3_XMIT_RATE0 0x0000001f /* series 0 tx rate */
+#define AR5K_4W_TX_DESC_CTL3_XMIT_RATE1 0x000003e0 /* series 1 tx rate */
#define AR5K_4W_TX_DESC_CTL3_XMIT_RATE1_S 5
-#define AR5K_4W_TX_DESC_CTL3_XMIT_RATE2 0x00007c00
+#define AR5K_4W_TX_DESC_CTL3_XMIT_RATE2 0x00007c00 /* series 2 tx rate */
#define AR5K_4W_TX_DESC_CTL3_XMIT_RATE2_S 10
-#define AR5K_4W_TX_DESC_CTL3_XMIT_RATE3 0x000f8000
+#define AR5K_4W_TX_DESC_CTL3_XMIT_RATE3 0x000f8000 /* series 3 tx rate */
#define AR5K_4W_TX_DESC_CTL3_XMIT_RATE3_S 15
-#define AR5K_4W_TX_DESC_CTL3_RTS_CTS_RATE 0x01f00000
+#define AR5K_4W_TX_DESC_CTL3_RTS_CTS_RATE 0x01f00000 /* RTS or CTS rate */
#define AR5K_4W_TX_DESC_CTL3_RTS_CTS_RATE_S 20
-} __packed;
/*
* Common TX status descriptor
@@ -251,37 +245,34 @@ struct ath5k_hw_tx_status {
} __packed;
/* TX status word 0 fields/flags */
-#define AR5K_DESC_TX_STATUS0_FRAME_XMIT_OK 0x00000001
-#define AR5K_DESC_TX_STATUS0_EXCESSIVE_RETRIES 0x00000002
-#define AR5K_DESC_TX_STATUS0_FIFO_UNDERRUN 0x00000004
-#define AR5K_DESC_TX_STATUS0_FILTERED 0x00000008
-/*???
-#define AR5K_DESC_TX_STATUS0_RTS_FAIL_COUNT 0x000000f0
-#define AR5K_DESC_TX_STATUS0_RTS_FAIL_COUNT_S 4
-*/
-#define AR5K_DESC_TX_STATUS0_SHORT_RETRY_COUNT 0x000000f0
+#define AR5K_DESC_TX_STATUS0_FRAME_XMIT_OK 0x00000001 /* TX success */
+#define AR5K_DESC_TX_STATUS0_EXCESSIVE_RETRIES 0x00000002 /* excessive retries */
+#define AR5K_DESC_TX_STATUS0_FIFO_UNDERRUN 0x00000004 /* FIFO underrun */
+#define AR5K_DESC_TX_STATUS0_FILTERED 0x00000008 /* TX filter indication */
+/* according to the HAL sources the spec has short/long retry counts reversed.
+ * we have it reversed to the HAL sources as well, for 5210 and 5211.
+ * For 5212 these fields are defined as RTS_FAIL_COUNT and DATA_FAIL_COUNT,
+ * but used respectively as SHORT and LONG retry count in the code later. This
+ * is consistent with the definitions here... TODO: check */
+#define AR5K_DESC_TX_STATUS0_SHORT_RETRY_COUNT 0x000000f0 /* short retry count */
#define AR5K_DESC_TX_STATUS0_SHORT_RETRY_COUNT_S 4
-/*???
-#define AR5K_DESC_TX_STATUS0_DATA_FAIL_COUNT 0x00000f00
-#define AR5K_DESC_TX_STATUS0_DATA_FAIL_COUNT_S 8
-*/
-#define AR5K_DESC_TX_STATUS0_LONG_RETRY_COUNT 0x00000f00
+#define AR5K_DESC_TX_STATUS0_LONG_RETRY_COUNT 0x00000f00 /* long retry count */
#define AR5K_DESC_TX_STATUS0_LONG_RETRY_COUNT_S 8
-#define AR5K_DESC_TX_STATUS0_VIRT_COLL_COUNT 0x0000f000
-#define AR5K_DESC_TX_STATUS0_VIRT_COLL_COUNT_S 12
-#define AR5K_DESC_TX_STATUS0_SEND_TIMESTAMP 0xffff0000
+#define AR5K_DESC_TX_STATUS0_VIRTCOLL_CT_5211 0x0000f000 /* [5211+] virtual collision count */
+#define AR5K_DESC_TX_STATUS0_VIRTCOLL_CT_5212_S 12
+#define AR5K_DESC_TX_STATUS0_SEND_TIMESTAMP 0xffff0000 /* TX timestamp */
#define AR5K_DESC_TX_STATUS0_SEND_TIMESTAMP_S 16
/* TX status word 1 fields/flags */
-#define AR5K_DESC_TX_STATUS1_DONE 0x00000001
-#define AR5K_DESC_TX_STATUS1_SEQ_NUM 0x00001ffe
+#define AR5K_DESC_TX_STATUS1_DONE 0x00000001 /* descriptor complete */
+#define AR5K_DESC_TX_STATUS1_SEQ_NUM 0x00001ffe /* TX sequence number */
#define AR5K_DESC_TX_STATUS1_SEQ_NUM_S 1
-#define AR5K_DESC_TX_STATUS1_ACK_SIG_STRENGTH 0x001fe000
+#define AR5K_DESC_TX_STATUS1_ACK_SIG_STRENGTH 0x001fe000 /* signal strength of ACK */
#define AR5K_DESC_TX_STATUS1_ACK_SIG_STRENGTH_S 13
-#define AR5K_DESC_TX_STATUS1_FINAL_TS_INDEX 0x00600000
-#define AR5K_DESC_TX_STATUS1_FINAL_TS_INDEX_S 21
-#define AR5K_DESC_TX_STATUS1_COMP_SUCCESS 0x00800000
-#define AR5K_DESC_TX_STATUS1_XMIT_ANTENNA 0x01000000
+#define AR5K_DESC_TX_STATUS1_FINAL_TS_IX_5212 0x00600000 /* [5212] final TX attempt series ix */
+#define AR5K_DESC_TX_STATUS1_FINAL_TS_IX_5212_S 21
+#define AR5K_DESC_TX_STATUS1_COMP_SUCCESS_5212 0x00800000 /* [5212] compression status */
+#define AR5K_DESC_TX_STATUS1_XMIT_ANTENNA_5212 0x01000000 /* [5212] transmit antenna */
/*
* 5210/5211 hardware TX descriptor
@@ -300,7 +291,7 @@ struct ath5k_hw_5212_tx_desc {
} __packed;
/*
- * common hardware RX descriptor
+ * Common hardware RX descriptor
*/
struct ath5k_hw_all_rx_desc {
struct ath5k_hw_rx_ctl rx_ctl;
@@ -308,7 +299,7 @@ struct ath5k_hw_all_rx_desc {
} __packed;
/*
- * Atheros hardware descriptor
+ * Atheros hardware DMA descriptor
* This is read and written to by the hardware
*/
struct ath5k_desc {
@@ -330,4 +321,3 @@ struct ath5k_desc {
#define AR5K_TXDESC_CTSENA 0x0008
#define AR5K_TXDESC_INTREQ 0x0010
#define AR5K_TXDESC_VEOL 0x0020 /*[5211+]*/
-
^ permalink raw reply related
* [PATCH 14/17] ath5k: update 5210/5211 frame types
From: Bruno Randolf @ 2010-06-16 10:12 UTC (permalink / raw)
To: linville; +Cc: ath5k-devel, linux-wireless
In-Reply-To: <20100616100809.10067.34787.stgit@tt-desk>
Update 5210 frame types to match the HAL. We have to apply the same bitshift to
the constants as we use later.
Add 5211 specific frame types.
Signed-off-by: Bruno Randolf <br1@einfach.org>
---
drivers/net/wireless/ath/ath5k/desc.c | 2 +-
drivers/net/wireless/ath/ath5k/desc.h | 12 +++++++-----
2 files changed, 8 insertions(+), 6 deletions(-)
diff --git a/drivers/net/wireless/ath/ath5k/desc.c b/drivers/net/wireless/ath/ath5k/desc.c
index 41a490e..f1f1a22 100644
--- a/drivers/net/wireless/ath/ath5k/desc.c
+++ b/drivers/net/wireless/ath/ath5k/desc.c
@@ -110,7 +110,7 @@ ath5k_hw_setup_2word_tx_desc(struct ath5k_hw *ah, struct ath5k_desc *desc,
case AR5K_PKT_TYPE_PIFS:
frame_type = AR5K_AR5210_TX_DESC_FRAME_TYPE_PIFS;
default:
- frame_type = type /*<< 2 ?*/;
+ frame_type = type;
}
tx_ctl->tx_control_0 |=
diff --git a/drivers/net/wireless/ath/ath5k/desc.h b/drivers/net/wireless/ath/ath5k/desc.h
index 50f9a12..1aca4af 100644
--- a/drivers/net/wireless/ath/ath5k/desc.h
+++ b/drivers/net/wireless/ath/ath5k/desc.h
@@ -169,11 +169,13 @@ struct ath5k_hw_2w_tx_ctl {
#define AR5K_2W_TX_DESC_CTL1_RTS_DURATION_5210 0xfff80000 /* [5210] lower 13 bit of duration */
/* Frame types */
-#define AR5K_AR5210_TX_DESC_FRAME_TYPE_NORMAL 0x00
-#define AR5K_AR5210_TX_DESC_FRAME_TYPE_ATIM 0x04
-#define AR5K_AR5210_TX_DESC_FRAME_TYPE_PSPOLL 0x08
-#define AR5K_AR5210_TX_DESC_FRAME_TYPE_NO_DELAY 0x0c
-#define AR5K_AR5210_TX_DESC_FRAME_TYPE_PIFS 0x10
+#define AR5K_AR5210_TX_DESC_FRAME_TYPE_NORMAL 0
+#define AR5K_AR5210_TX_DESC_FRAME_TYPE_ATIM 1
+#define AR5K_AR5210_TX_DESC_FRAME_TYPE_PSPOLL 2
+#define AR5K_AR5210_TX_DESC_FRAME_TYPE_NO_DELAY 3
+#define AR5K_AR5211_TX_DESC_FRAME_TYPE_BEACON 3
+#define AR5K_AR5210_TX_DESC_FRAME_TYPE_PIFS 4
+#define AR5K_AR5211_TX_DESC_FRAME_TYPE_PRESP 4
/*
* 5212 hardware 4-word TX control descriptor
^ permalink raw reply related
* [PATCH 15/17] ath5k: take descriptor differences between 5210 and 5211 into account
From: Bruno Randolf @ 2010-06-16 10:12 UTC (permalink / raw)
To: linville; +Cc: ath5k-devel, linux-wireless
In-Reply-To: <20100616100809.10067.34787.stgit@tt-desk>
There are some differences between 5210 and 5211 descriptors which we did not
take into account before.
Signed-off-by: Bruno Randolf <br1@einfach.org>
---
drivers/net/wireless/ath/ath5k/desc.c | 29 ++++++++++++++++++++++-------
drivers/net/wireless/ath/ath5k/desc.h | 6 +++---
2 files changed, 25 insertions(+), 10 deletions(-)
diff --git a/drivers/net/wireless/ath/ath5k/desc.c b/drivers/net/wireless/ath/ath5k/desc.c
index f1f1a22..019525d 100644
--- a/drivers/net/wireless/ath/ath5k/desc.c
+++ b/drivers/net/wireless/ath/ath5k/desc.c
@@ -91,8 +91,7 @@ ath5k_hw_setup_2word_tx_desc(struct ath5k_hw *ah, struct ath5k_desc *desc,
tx_ctl->tx_control_1 = pkt_len & AR5K_2W_TX_DESC_CTL1_BUF_LEN;
/*
- * Verify and set header length
- * XXX: I only found that on 5210 code, does it work on 5211 ?
+ * Verify and set header length (only 5210)
*/
if (ah->ah_version == AR5K_AR5210) {
if (hdr_len & ~AR5K_2W_TX_DESC_CTL0_HEADER_LEN_5210)
@@ -125,19 +124,28 @@ ath5k_hw_setup_2word_tx_desc(struct ath5k_hw *ah, struct ath5k_desc *desc,
tx_ctl->tx_control_1 |=
AR5K_REG_SM(type, AR5K_2W_TX_DESC_CTL1_FRAME_TYPE_5211);
}
+
#define _TX_FLAGS(_c, _flag) \
if (flags & AR5K_TXDESC_##_flag) { \
tx_ctl->tx_control_##_c |= \
AR5K_2W_TX_DESC_CTL##_c##_##_flag; \
}
-
+#define _TX_FLAGS_5211(_c, _flag) \
+ if (flags & AR5K_TXDESC_##_flag) { \
+ tx_ctl->tx_control_##_c |= \
+ AR5K_2W_TX_DESC_CTL##_c##_##_flag##_5211; \
+ }
_TX_FLAGS(0, CLRDMASK);
- _TX_FLAGS(0, VEOL);
_TX_FLAGS(0, INTREQ);
_TX_FLAGS(0, RTSENA);
- _TX_FLAGS(1, NOACK);
+
+ if (ah->ah_version == AR5K_AR5211) {
+ _TX_FLAGS_5211(0, VEOL);
+ _TX_FLAGS_5211(1, NOACK);
+ }
#undef _TX_FLAGS
+#undef _TX_FLAGS_5211
/*
* WEP crap
@@ -526,13 +534,20 @@ static int ath5k_hw_proc_5210_rx_status(struct ath5k_hw *ah,
AR5K_5210_RX_DESC_STATUS0_RECEIVE_SIGNAL);
rs->rs_rate = AR5K_REG_MS(rx_status->rx_status_0,
AR5K_5210_RX_DESC_STATUS0_RECEIVE_RATE);
- rs->rs_antenna = AR5K_REG_MS(rx_status->rx_status_0,
- AR5K_5210_RX_DESC_STATUS0_RECEIVE_ANT_5211);
rs->rs_more = !!(rx_status->rx_status_0 &
AR5K_5210_RX_DESC_STATUS0_MORE);
/* TODO: this timestamp is 13 bit, later on we assume 15 bit */
rs->rs_tstamp = AR5K_REG_MS(rx_status->rx_status_1,
AR5K_5210_RX_DESC_STATUS1_RECEIVE_TIMESTAMP);
+
+ if (ah->ah_version == AR5K_AR5211)
+ rs->rs_antenna = AR5K_REG_MS(rx_status->rx_status_0,
+ AR5K_5210_RX_DESC_STATUS0_RECEIVE_ANT_5211);
+ else
+ rs->rs_antenna = (rx_status->rx_status_0 &
+ AR5K_5210_RX_DESC_STATUS0_RECEIVE_ANT_5210)
+ ? 2 : 1;
+
rs->rs_status = 0;
rs->rs_phyerr = 0;
diff --git a/drivers/net/wireless/ath/ath5k/desc.h b/drivers/net/wireless/ath/ath5k/desc.h
index 1aca4af..b2adb2a 100644
--- a/drivers/net/wireless/ath/ath5k/desc.h
+++ b/drivers/net/wireless/ath/ath5k/desc.h
@@ -45,7 +45,7 @@ struct ath5k_hw_rx_status {
/* RX status word 0 fields/flags */
#define AR5K_5210_RX_DESC_STATUS0_DATA_LEN 0x00000fff /* RX data length */
#define AR5K_5210_RX_DESC_STATUS0_MORE 0x00001000 /* more desc for this frame */
-#define AR5K_5210_RX_DESC_STATUS0_RECEIVE_ANT_5210 0x00004000 /* [5210] receive on ant 1 TODO */
+#define AR5K_5210_RX_DESC_STATUS0_RECEIVE_ANT_5210 0x00004000 /* [5210] receive on ant 1 */
#define AR5K_5210_RX_DESC_STATUS0_RECEIVE_RATE 0x00078000 /* reception rate */
#define AR5K_5210_RX_DESC_STATUS0_RECEIVE_RATE_S 15
#define AR5K_5210_RX_DESC_STATUS0_RECEIVE_SIGNAL 0x07f80000 /* rssi */
@@ -139,7 +139,7 @@ struct ath5k_hw_2w_tx_ctl {
#define AR5K_2W_TX_DESC_CTL0_XMIT_RATE_S 18
#define AR5K_2W_TX_DESC_CTL0_RTSENA 0x00400000 /* RTS/CTS enable */
#define AR5K_2W_TX_DESC_CTL0_LONG_PACKET_5210 0x00800000 /* [5210] long packet */
-#define AR5K_2W_TX_DESC_CTL0_VEOL 0x00800000 /* [5211] virtual end-of-list TODO */
+#define AR5K_2W_TX_DESC_CTL0_VEOL_5211 0x00800000 /* [5211] virtual end-of-list */
#define AR5K_2W_TX_DESC_CTL0_CLRDMASK 0x01000000 /* clear destination mask */
#define AR5K_2W_TX_DESC_CTL0_ANT_MODE_XMIT_5210 0x02000000 /* [5210] antenna selection */
#define AR5K_2W_TX_DESC_CTL0_ANT_MODE_XMIT_5211 0x1e000000 /* [5211] antenna selection */
@@ -165,7 +165,7 @@ struct ath5k_hw_2w_tx_ctl {
#define AR5K_2W_TX_DESC_CTL1_ENC_KEY_IDX_S 13
#define AR5K_2W_TX_DESC_CTL1_FRAME_TYPE_5211 0x00700000 /* [5211] frame type */
#define AR5K_2W_TX_DESC_CTL1_FRAME_TYPE_5211_S 20
-#define AR5K_2W_TX_DESC_CTL1_NOACK 0x00800000 /* [5211] no ACK TODO */
+#define AR5K_2W_TX_DESC_CTL1_NOACK_5211 0x00800000 /* [5211] no ACK */
#define AR5K_2W_TX_DESC_CTL1_RTS_DURATION_5210 0xfff80000 /* [5210] lower 13 bit of duration */
/* Frame types */
^ permalink raw reply related
* [PATCH 16/17] ath5k: review RX descriptor functions
From: Bruno Randolf @ 2010-06-16 10:12 UTC (permalink / raw)
To: linville; +Cc: ath5k-devel, linux-wireless
In-Reply-To: <20100616100809.10067.34787.stgit@tt-desk>
Reviewed RX descriptor functions against the HAL sources. Some minor changes:
- check size before making changes to the descriptor
- whitespace
- add comments about 5210 timestamps. this needs to be adressed later!
- FIFO overrun error only available on 5210
- rs_phyerr should not be OR'ed
- clear the whole ath5k_rx_status structure before using, instead of
zeroing specific fields.
Signed-off-by: Bruno Randolf <br1@einfach.org>
---
drivers/net/wireless/ath/ath5k/desc.c | 34 +++++++++++++++++++--------------
1 files changed, 20 insertions(+), 14 deletions(-)
diff --git a/drivers/net/wireless/ath/ath5k/desc.c b/drivers/net/wireless/ath/ath5k/desc.c
index 019525d..c0037b6 100644
--- a/drivers/net/wireless/ath/ath5k/desc.c
+++ b/drivers/net/wireless/ath/ath5k/desc.c
@@ -499,10 +499,11 @@ int ath5k_hw_setup_rx_desc(struct ath5k_hw *ah, struct ath5k_desc *desc,
*/
memset(&desc->ud.ds_rx, 0, sizeof(struct ath5k_hw_all_rx_desc));
+ if (unlikely(size & ~AR5K_DESC_RX_CTL1_BUF_LEN))
+ return -EINVAL;
+
/* Setup descriptor */
rx_ctl->rx_control_1 = size & AR5K_DESC_RX_CTL1_BUF_LEN;
- if (unlikely(rx_ctl->rx_control_1 != size))
- return -EINVAL;
if (flags & AR5K_RXDESC_INTREQ)
rx_ctl->rx_control_1 |= AR5K_DESC_RX_CTL1_INTREQ;
@@ -522,9 +523,11 @@ static int ath5k_hw_proc_5210_rx_status(struct ath5k_hw *ah,
/* No frame received / not ready */
if (unlikely(!(rx_status->rx_status_1 &
- AR5K_5210_RX_DESC_STATUS1_DONE)))
+ AR5K_5210_RX_DESC_STATUS1_DONE)))
return -EINPROGRESS;
+ memset(rs, 0, sizeof(struct ath5k_rx_status));
+
/*
* Frame receive status
*/
@@ -536,7 +539,11 @@ static int ath5k_hw_proc_5210_rx_status(struct ath5k_hw *ah,
AR5K_5210_RX_DESC_STATUS0_RECEIVE_RATE);
rs->rs_more = !!(rx_status->rx_status_0 &
AR5K_5210_RX_DESC_STATUS0_MORE);
- /* TODO: this timestamp is 13 bit, later on we assume 15 bit */
+ /* TODO: this timestamp is 13 bit, later on we assume 15 bit!
+ * also the HAL code for 5210 says the timestamp is bits [10..22] of the
+ * TSF, and extends the timestamp here to 15 bit.
+ * we need to check on 5210...
+ */
rs->rs_tstamp = AR5K_REG_MS(rx_status->rx_status_1,
AR5K_5210_RX_DESC_STATUS1_RECEIVE_TIMESTAMP);
@@ -548,9 +555,6 @@ static int ath5k_hw_proc_5210_rx_status(struct ath5k_hw *ah,
AR5K_5210_RX_DESC_STATUS0_RECEIVE_ANT_5210)
? 2 : 1;
- rs->rs_status = 0;
- rs->rs_phyerr = 0;
-
/*
* Key table status
*/
@@ -564,19 +568,21 @@ static int ath5k_hw_proc_5210_rx_status(struct ath5k_hw *ah,
* Receive/descriptor errors
*/
if (!(rx_status->rx_status_1 &
- AR5K_5210_RX_DESC_STATUS1_FRAME_RECEIVE_OK)) {
+ AR5K_5210_RX_DESC_STATUS1_FRAME_RECEIVE_OK)) {
if (rx_status->rx_status_1 &
AR5K_5210_RX_DESC_STATUS1_CRC_ERROR)
rs->rs_status |= AR5K_RXERR_CRC;
- if (rx_status->rx_status_1 &
- AR5K_5210_RX_DESC_STATUS1_FIFO_OVERRUN_5210)
+ /* only on 5210 */
+ if ((ah->ah_version == AR5K_AR5210) &&
+ (rx_status->rx_status_1 &
+ AR5K_5210_RX_DESC_STATUS1_FIFO_OVERRUN_5210))
rs->rs_status |= AR5K_RXERR_FIFO;
if (rx_status->rx_status_1 &
AR5K_5210_RX_DESC_STATUS1_PHY_ERROR) {
rs->rs_status |= AR5K_RXERR_PHY;
- rs->rs_phyerr |= AR5K_REG_MS(rx_status->rx_status_1,
+ rs->rs_phyerr = AR5K_REG_MS(rx_status->rx_status_1,
AR5K_5210_RX_DESC_STATUS1_PHY_ERROR);
}
@@ -604,6 +610,8 @@ static int ath5k_hw_proc_5212_rx_status(struct ath5k_hw *ah,
AR5K_5212_RX_DESC_STATUS1_DONE)))
return -EINPROGRESS;
+ memset(rs, 0, sizeof(struct ath5k_rx_status));
+
/*
* Frame receive status
*/
@@ -619,8 +627,6 @@ static int ath5k_hw_proc_5212_rx_status(struct ath5k_hw *ah,
AR5K_5212_RX_DESC_STATUS0_MORE);
rs->rs_tstamp = AR5K_REG_MS(rx_status->rx_status_1,
AR5K_5212_RX_DESC_STATUS1_RECEIVE_TIMESTAMP);
- rs->rs_status = 0;
- rs->rs_phyerr = 0;
/*
* Key table status
@@ -643,7 +649,7 @@ static int ath5k_hw_proc_5212_rx_status(struct ath5k_hw *ah,
if (rx_status->rx_status_1 &
AR5K_5212_RX_DESC_STATUS1_PHY_ERROR) {
rs->rs_status |= AR5K_RXERR_PHY;
- rs->rs_phyerr |= AR5K_REG_MS(rx_status->rx_status_1,
+ rs->rs_phyerr = AR5K_REG_MS(rx_status->rx_status_1,
AR5K_5212_RX_DESC_STATUS1_PHY_ERROR_CODE);
ath5k_ani_phy_error_report(ah, rs->rs_phyerr);
}
^ permalink raw reply related
* [PATCH 17/17] ath5k: report PHY error frames only for chips which need it
From: Bruno Randolf @ 2010-06-16 10:12 UTC (permalink / raw)
To: linville; +Cc: ath5k-devel, linux-wireless
In-Reply-To: <20100616100809.10067.34787.stgit@tt-desk>
Only report PHY error frames for ANI on chipsets which do not have PHY error
counters in hardware.
Signed-off-by: Bruno Randolf <br1@einfach.org>
Acked-by: Bob Copeland <me@bobcopeland.com>
---
drivers/net/wireless/ath/ath5k/desc.c | 3 ++-
1 files changed, 2 insertions(+), 1 deletions(-)
diff --git a/drivers/net/wireless/ath/ath5k/desc.c b/drivers/net/wireless/ath/ath5k/desc.c
index c0037b6..4324438 100644
--- a/drivers/net/wireless/ath/ath5k/desc.c
+++ b/drivers/net/wireless/ath/ath5k/desc.c
@@ -651,7 +651,8 @@ static int ath5k_hw_proc_5212_rx_status(struct ath5k_hw *ah,
rs->rs_status |= AR5K_RXERR_PHY;
rs->rs_phyerr = AR5K_REG_MS(rx_status->rx_status_1,
AR5K_5212_RX_DESC_STATUS1_PHY_ERROR_CODE);
- ath5k_ani_phy_error_report(ah, rs->rs_phyerr);
+ if (!ah->ah_capabilities.cap_has_phyerr_counters)
+ ath5k_ani_phy_error_report(ah, rs->rs_phyerr);
}
if (rx_status->rx_status_1 &
^ permalink raw reply related
* Re: [PATCH 0/5] Fix set but unused variable warnings
From: Matthew Wilcox @ 2010-06-16 11:09 UTC (permalink / raw)
To: Julian Calaby
Cc: Justin P. Mattock, linux-kernel, linux-wireless, linux-pci,
linux-scsi
In-Reply-To: <AANLkTikwKHITQJDaSf9BnHQ9FLiQ-KYlCWNiH438otsV@mail.gmail.com>
On Wed, Jun 16, 2010 at 03:52:58PM +1000, Julian Calaby wrote:
> Given that patches 3, 4 and 5 seem to be a cases of missing error
> handling, (3 and 4 in particular seem to be breaking things rather
> than fixing them) in my humble opinion, I think this set needs some
> work and discussion.
>
> Justin, maybe you'd be better off posting the actual error messages
> (split up by subsystem) and letting the lists discuss them, rather
> than posting patches which are obviously wrong. (like the ones I've
> pointed out)
My first impression of patch 5 was "that can't be right", but upon review,
I concluded that was the best solution. If we return the PTR_ERR, this
is going to confuse every user of scsi_host_alloc who are currently only
checking for NULL.
--
Matthew Wilcox Intel Open Source Technology Centre
"Bill, look, we understand that you're interested in selling us this
operating system, but compare it to ours. We can't possibly take such
a retrograde step."
^ permalink raw reply
* Re: [PATCH 0/5] Fix set but unused variable warnings
From: Julian Calaby @ 2010-06-16 11:30 UTC (permalink / raw)
To: Matthew Wilcox
Cc: Justin P. Mattock, linux-kernel, linux-wireless, linux-pci,
linux-scsi
In-Reply-To: <20100616110914.GF9298@parisc-linux.org>
On Wed, Jun 16, 2010 at 21:09, Matthew Wilcox <matthew@wil.cx> wrote:
> On Wed, Jun 16, 2010 at 03:52:58PM +1000, Julian Calaby wrote:
>> Given that patches 3, 4 and 5 seem to be a cases of missing error
>> handling, (3 and 4 in particular seem to be breaking things rather
>> than fixing them) in my humble opinion, I think this set needs some
>> work and discussion.
>>
>> Justin, maybe you'd be better off posting the actual error messages
>> (split up by subsystem) and letting the lists discuss them, rather
>> than posting patches which are obviously wrong. (like the ones I've
>> pointed out)
>
> My first impression of patch 5 was "that can't be right", but upon review,
> I concluded that was the best solution. If we return the PTR_ERR, this
> is going to confuse every user of scsi_host_alloc who are currently only
> checking for NULL.
I did a double take at patch 4 when skimming over them, then took a
proper look at the lot of them. Whilst I agree that 5 isn't doing
anything particularly bad, it does seem wrong, especially in the
context of 3 and 4.
Thanks,
--
Julian Calaby
Email: julian.calaby@gmail.com
.Plan: http://sites.google.com/site/juliancalaby/
^ permalink raw reply
* Re: [PATCH] zd1211rw: ignore unknown regulatory domain.
From: Kouhei Sutou @ 2010-06-16 12:55 UTC (permalink / raw)
To: mcgrof; +Cc: Stephen.Chen, David.Quan, linux-wireless, Michael.Green
In-Reply-To: <AANLkTin9bG4qDEKwgOXmlR3k9wZ-JBSIlh3JzYjzVDeY@mail.gmail.com>
[-- Attachment #1: Type: Text/Plain, Size: 1898 bytes --]
Hi,
In <AANLkTin9bG4qDEKwgOXmlR3k9wZ-JBSIlh3JzYjzVDeY@mail.gmail.com>
"Re: [PATCH] zd1211rw: ignore unknown regulatory domain." on Tue, 15 Jun 2010 08:22:09 -0700,
"Luis R. Rodriguez" <mcgrof@gmail.com> wrote:
> On Tue, Jun 15, 2010 at 7:25 AM, Kouhei Sutou <kou@clear-code.com> wrote:
>> Hi,
>>
>> In <AANLkTinEMAQu33Tlv81-Cdx9fLZK-a-7uJlbAo7te2_w@mail.gmail.com>
>> "Re: [PATCH] zd1211rw: ignore unknown regulatory domain." on Sun, 13 Jun 2010 13:23:20 -0700,
>> "Luis R. Rodriguez" <mcgrof@gmail.com> wrote:
>>
>>>> I'm using PLANEX GW-US54GXS (2019:5303) and it works with
>>>> the attached patch. Could you consider to merge the attached
>>>> patch?
>>>>
>>>> Problem: GW-US54GXS uses zd1211rw but zd1211rw doesn't have
>>>> a regulatory domain reported by GW-US54GXS (0x49).
>>>>
>>>> Solutions:
>>>>
>>>> (1) add a new regulatory domain (0x49). Here is a patch to
>>>> use the approach:
>>>
>>> Stephen, David, does 0x49 map to JP for zd1211 ? Are there other ones?
>>> Here is our list so far:
>> ...
>>> Kouhei, if no regulatory domain is found, instead we should world
>>> roam, we cannot allow letting the user change regulatory domains at
>>> their whim. We can, however let them choose one to help compliance,
>>> but you can only help compliance once you know your actual regulatory
>>> domain.
>>
>> Luis, thanks for your input.
>> It seems that we can't use GW-US54GXS until 0x49 regulatory
>> domain is registered to zd1211rw. Is it right?
>>
>> If it is right, what I can do for GW-US54GXS? Should I wait
>> a response from Stephen and/or David?
>
> We spoke and 0x49 - > JP is ok, can you send a patch for that?
OK.
I'll attach a patch for that.
I used 'ZD_REGDOMAIN_JAPAN_GW_US54GXS' as a macro name for
0x49 because 0x49 is used by GW-US54GXS. Is it OK? Or should
I use other name for it?
Thanks,
--
kou
[-- Attachment #2: 0001-zd1211rw-add-0x49-JP-regulatory-domain-map.patch --]
[-- Type: Text/X-Patch, Size: 1440 bytes --]
>From a530e66e6e995d8e8bc00ece2a9b3540ef88688f Mon Sep 17 00:00:00 2001
From: Kouhei Sutou <kou@clear-code.com>
Date: Wed, 16 Jun 2010 21:53:59 +0900
Subject: [PATCH] zd1211rw: add 0x49 -> JP regulatory domain map
0x49 is used by PLANEX GW-US54GXS (2019:5303).
Signed-off-by: Kouhei Sutou <kou@clear-code.com>
---
drivers/net/wireless/zd1211rw/zd_mac.c | 1 +
drivers/net/wireless/zd1211rw/zd_mac.h | 1 +
2 files changed, 2 insertions(+), 0 deletions(-)
diff --git a/drivers/net/wireless/zd1211rw/zd_mac.c b/drivers/net/wireless/zd1211rw/zd_mac.c
index 163a8a0..faf45f1 100644
--- a/drivers/net/wireless/zd1211rw/zd_mac.c
+++ b/drivers/net/wireless/zd1211rw/zd_mac.c
@@ -43,6 +43,7 @@ static struct zd_reg_alpha2_map reg_alpha2_map[] = {
{ ZD_REGDOMAIN_ETSI, "DE" }, /* Generic ETSI, use most restrictive */
{ ZD_REGDOMAIN_JAPAN, "JP" },
{ ZD_REGDOMAIN_JAPAN_ADD, "JP" },
+ { ZD_REGDOMAIN_JAPAN_GW_US54GXS, "JP" },
{ ZD_REGDOMAIN_SPAIN, "ES" },
{ ZD_REGDOMAIN_FRANCE, "FR" },
};
diff --git a/drivers/net/wireless/zd1211rw/zd_mac.h b/drivers/net/wireless/zd1211rw/zd_mac.h
index 630c298..7ab19d5 100644
--- a/drivers/net/wireless/zd1211rw/zd_mac.h
+++ b/drivers/net/wireless/zd1211rw/zd_mac.h
@@ -214,6 +214,7 @@ struct zd_mac {
#define ZD_REGDOMAIN_FRANCE 0x32
#define ZD_REGDOMAIN_JAPAN_ADD 0x40
#define ZD_REGDOMAIN_JAPAN 0x41
+#define ZD_REGDOMAIN_JAPAN_GW_US54GXS 0x49
enum {
MIN_CHANNEL24 = 1,
--
1.7.1
^ permalink raw reply related
* Re: [PATCH 2.6.32.y] ath9k: re-enable ps by default for new single chip families
From: Justin P. Mattock @ 2010-06-16 13:33 UTC (permalink / raw)
To: Kristoffer Ericson
Cc: Luis R. Rodriguez, stable, greg, linux-wireless, Peter Stuge,
John W. Linville
In-Reply-To: <20100616081227.GA1118@boggieman>
On 06/16/2010 01:12 AM, Kristoffer Ericson wrote:
> On Tue, Jun 15, 2010 at 06:19:19PM -0400, Luis R. Rodriguez wrote:
>> commit 14acdde6e527950f66c084dbf19bad6fbfcaeedc upstream.
>>
>> The newer single chip hardware family of chipsets have not been
>> experiencing issues with power saving set by default with recent
>> fixes merged (even into stable). The remaining issues are only
>> reported with AR5416 and since enabling PS by default can increase
>> power savings considerably best to take advantage of that feature
>> as this has been tested properly.
>
> I havent had any issues lately. But that said Ive moved on to
> 2.6.33, 2.6.34 which seems to work fine. Believe you set
> to disabled by default on those?
>
> /Kristoffer
>
just tried to add this patch to the latest HEAD, but realized
ath_set_hw_capab is missing(or I cant seem to grep it)
if you have a patch for the latest tree let me know I can see if I'm
hitting anything like before.
Justin P. Mattock
^ permalink raw reply
* Re: [PATCH] zd1211rw: ignore unknown regulatory domain.
From: John W. Linville @ 2010-06-16 13:44 UTC (permalink / raw)
To: Kouhei Sutou
Cc: mcgrof, Stephen.Chen, David.Quan, linux-wireless, Michael.Green
In-Reply-To: <20100616.215523.1718341995174283889.kou@clear-code.com>
On Wed, Jun 16, 2010 at 09:55:23PM +0900, Kouhei Sutou wrote:
> I used 'ZD_REGDOMAIN_JAPAN_GW_US54GXS' as a macro name for
> 0x49 because 0x49 is used by GW-US54GXS. Is it OK? Or should
> I use other name for it?
Hopefully someone can suggest a better name. Anyone know what inspired
"ZD_REGDOMAIN_JAPAN_ADD"?
John
--
John W. Linville Someday the world will need a hero, and you
linville@tuxdriver.com might be all we have. Be ready.
^ permalink raw reply
* Re: [PATCH] ath9k: Modify LED blinking pattern during wifi activity.
From: John W. Linville @ 2010-06-16 13:38 UTC (permalink / raw)
To: Vivek Natarajan; +Cc: linux-wireless
In-Reply-To: <AANLkTimprPdviZPX26qcH-uM8FyNY09JEH3kRvqwDi-S@mail.gmail.com>
On Wed, Jun 16, 2010 at 11:21:27AM +0530, Vivek Natarajan wrote:
> On Wed, Jun 16, 2010 at 1:21 AM, John W. Linville
> <linville@tuxdriver.com> wrote:
> > On Tue, Jun 15, 2010 at 10:50:17AM +0530, Vivek Natarajan wrote:
> >> Some vendors require the LED to be ON always irrespective of any
> >> radio activity. Introducing a module parameter to enable this,
> >> so that one can choose between always on or led blink during
> >> activity.
> >>
> >> Signed-off-by: Vivek Natarajan <vnatarajan@atheros.com>
> >
> > Any particular reason the always-on behaviour is the default?
>
> There is no specific reason for setting it as default. It is only that
> some customers preferred the led to be always on instead of blinking.
Well it is just that as I read the patch, you are changing the
behaviour for everyone rather than simply giving a new option for
those that don't like it that way it already is.
Look, I don't particular care about the behaviour. Blinking w/
activity makes sense to me. But I also get to read the seemingly
endless, whining complaints about how the blinking is distracting for
laptop users. I just would rather not add another endless stream of
complaints that say "my wifi LED used to blink w/ activity, now it
doesn't -- fix it!"
Perhaps we could standardize this somehow? Anyone care to make
a proposal?
John
--
John W. Linville Someday the world will need a hero, and you
linville@tuxdriver.com might be all we have. Be ready.
^ permalink raw reply
* "ressource temporarily unavailable" on 2.4 GHz, not on 5GHz
From: Dennis Borgmann @ 2010-06-16 14:05 UTC (permalink / raw)
To: linux-wireless, hostap
[-- Attachment #1: Type: text/plain, Size: 717 bytes --]
Hello wireless-list!
Hello hostapd-list!
I am using ath5k with hostapd-0.6.9 on a debian running kernel 2.6.34
and I am observing a "ressource temporarily unavailable" - error code 11
if I pump out multicast-packets quite fast. I wrote a small
test-program, that handles this test and as soon as I go down to a
8ms-pause or below in betweens the packets, the error is thrown quite
soon after starting the program. Funny thing is, that this error only
occurs in g-band (2.4 GHz), not in a-band (5GHz).
What could be the problem? Is this due to backoff-times in 2.4 GHz, that
don't occur on 5 GHz? Or could this be a bug?
Find attached my small piece of code to test and reproduce this effect.
Kind regards,
Dennis
[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #2: connection_test_multicast.c --]
[-- Type: text/x-csrc; name="connection_test_multicast.c", Size: 2300 bytes --]
#include <sys/types.h> /* See NOTES */
#include <sys/socket.h>
#include <stdio.h>
#include <netinet/in.h>
#include <arpa/inet.h>
#include <errno.h>
#include <linux/ip.h>
#include <string.h>
#include <stdlib.h>
#define MULTICAST_IP "224.0.0.1"
#define PORT 12345
struct sockaddr_in server_address;
int fd_socket;
int init_connection_client( void ) {
int sock, opt;
errno = 0;
//create a socket
sock = socket( PF_INET, SOCK_DGRAM, 0 );
if( sock == -1 ) {
printf("%s:%u socket() error (%s), error-code: %d\n", __FILE__, __LINE__, strerror(errno), sock);
exit(EXIT_FAILURE);
}
server_address.sin_family = PF_INET;
server_address.sin_addr.s_addr = inet_addr( MULTICAST_IP );
server_address.sin_port = htons( PORT );
if( bind( sock, (struct sockaddr*)&server_address, sizeof(struct sockaddr)) == -1 ) {
printf("%s:%u bind error (%s), error-code: %d\n", __FILE__, __LINE__, strerror(errno), errno);
exit(EXIT_FAILURE);
}
//set socket options
//IPTOS = type of service and precedence
//IPTOS_LOWDELAY = minimize delay
//IPTOS_PREC_PRIORITY =
opt = IPTOS_PREC_PRIORITY | IPTOS_LOWDELAY;
if( setsockopt( sock, IPPROTO_IP, IP_TOS, &opt, sizeof(opt)) == -1 ) {
printf("%s:%u setsockopt(IP_TOS) error (%s), error-code: %d\n", __FILE__, __LINE__, strerror(errno), errno);
exit(EXIT_FAILURE);
}
//give these packets the maximum priority
//SOL_SOCKET = manipulate at socket level
opt = 7;
if( setsockopt( sock, SOL_SOCKET, SO_PRIORITY, &opt, sizeof(opt)) == -1 ) {
printf("%s:%u setsockopt(SO_PRIORITY) error (%s), error-code: %d\n", __FILE__, __LINE__, strerror(errno), errno);
exit(EXIT_FAILURE);
}
return sock;
}
int send_data( unsigned char *network_data, size_t len ) {
if( sendto( fd_socket, network_data, len, MSG_DONTWAIT, \
(struct sockaddr*)&server_address, sizeof(struct sockaddr_in) ) == -1 ) {
printf("%s:%u sendto() error (%s), error-code: %d\n", __FILE__, __LINE__, strerror(errno), errno);
return errno;
}
return 0;
}
int main(int argc, char **argv) {
unsigned char *data_for_transmission;
int sleeper=atoi(argv[1]);
printf("%d\n",sleeper);
data_for_transmission = malloc( 1068 );
fd_socket=init_connection_client();
while(1) {
send_data( data_for_transmission, 1068 );
usleep(sleeper);
}
return 0;
}
^ permalink raw reply
* Re: "ressource temporarily unavailable" on 2.4 GHz, not on 5GHz
From: John W. Linville @ 2010-06-16 14:23 UTC (permalink / raw)
To: Dennis Borgmann; +Cc: linux-wireless, hostap
In-Reply-To: <4C18DA3D.1030303@googlemail.com>
On Wed, Jun 16, 2010 at 04:05:49PM +0200, Dennis Borgmann wrote:
> Hello wireless-list!
> Hello hostapd-list!
>
> I am using ath5k with hostapd-0.6.9 on a debian running kernel 2.6.34
> and I am observing a "ressource temporarily unavailable" - error code 11
> if I pump out multicast-packets quite fast. I wrote a small
> test-program, that handles this test and as soon as I go down to a
> 8ms-pause or below in betweens the packets, the error is thrown quite
> soon after starting the program. Funny thing is, that this error only
> occurs in g-band (2.4 GHz), not in a-band (5GHz).
>
> What could be the problem? Is this due to backoff-times in 2.4 GHz, that
> don't occur on 5 GHz? Or could this be a bug?
My only initial thought is that the 2.5GHz band is usually more
utilized. So there is more likely to be retransmissions, etc.
> Find attached my small piece of code to test and reproduce this effect.
Perhaps you could include actual dmesg output?
John
--
John W. Linville Someday the world will need a hero, and you
linville@tuxdriver.com might be all we have. Be ready.
^ permalink raw reply
* [PATCH] mac80211: fix warn, enum may be used uninitialized
From: Christoph Fritz @ 2010-06-16 14:37 UTC (permalink / raw)
To: John W. Linville, Johannes Berg
Cc: kernel-janitors, linux-wireless, David S. Miller, Kalle Valo
regression introduced by b8d92c9c141ee3dc9b3537b1f0ffb4a54ea8d9b2
In function ‘ieee80211_work_rx_queued_mgmt’:
warning: ‘rma’ may be used uninitialized in this function
this re-adds default value WORK_ACT_NONE back to rma
Signed-off-by: Christoph Fritz <chf.fritz@googlemail.com>
---
net/mac80211/work.c | 2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/net/mac80211/work.c b/net/mac80211/work.c
index be3d4a6..b025dc7 100644
--- a/net/mac80211/work.c
+++ b/net/mac80211/work.c
@@ -715,7 +715,7 @@ static void ieee80211_work_rx_queued_mgmt(struct ieee80211_local *local,
struct ieee80211_rx_status *rx_status;
struct ieee80211_mgmt *mgmt;
struct ieee80211_work *wk;
- enum work_action rma;
+ enum work_action rma = WORK_ACT_NONE;
u16 fc;
rx_status = (struct ieee80211_rx_status *) skb->cb;
--
1.7.1
^ permalink raw reply related
* Re: "ressource temporarily unavailable" on 2.4 GHz, not on 5GHz
From: Bob Copeland @ 2010-06-16 14:38 UTC (permalink / raw)
To: Dennis Borgmann; +Cc: linux-wireless, hostap
In-Reply-To: <4C18DA3D.1030303@googlemail.com>
On Wed, Jun 16, 2010 at 10:05 AM, Dennis Borgmann
<dennis.borgmann@googlemail.com> wrote:
> Hello wireless-list!
> Hello hostapd-list!
>
> I am using ath5k with hostapd-0.6.9 on a debian running kernel 2.6.34
> and I am observing a "ressource temporarily unavailable" - error code 11
[...]
> What could be the problem? Is this due to backoff-times in 2.4 GHz, that
> don't occur on 5 GHz? Or could this be a bug?
Does dmesg report any error (e.g. no available txbuf)?
My guess is you have more frequent successful delivery in 5 GHz than in
ISM band so you aren't running out of send buffers somewhere in the stack.
Of course, your program should expect and handle EAGAIN if you are
using MSG_DONTWAIT.
--
Bob Copeland %% www.bobcopeland.com
^ permalink raw reply
* Re: "ressource temporarily unavailable" on 2.4 GHz, not on 5GHz
From: John W. Linville @ 2010-06-16 14:21 UTC (permalink / raw)
To: Dennis Borgmann; +Cc: linux-wireless, hostap
In-Reply-To: <4C18DA3D.1030303@googlemail.com>
On Wed, Jun 16, 2010 at 04:05:49PM +0200, Dennis Borgmann wrote:
> Hello wireless-list!
> Hello hostapd-list!
>
> I am using ath5k with hostapd-0.6.9 on a debian running kernel 2.6.34
> and I am observing a "ressource temporarily unavailable" - error code 11
> if I pump out multicast-packets quite fast. I wrote a small
> test-program, that handles this test and as soon as I go down to a
> 8ms-pause or below in betweens the packets, the error is thrown quite
> soon after starting the program. Funny thing is, that this error only
> occurs in g-band (2.4 GHz), not in a-band (5GHz).
>
> What could be the problem? Is this due to backoff-times in 2.4 GHz, that
> don't occur on 5 GHz? Or could this be a bug?
My only initial thought is that the 2.5GHz band is usually more
utilized. So there is more likely to be retransmissions, etc.
> Find attached my small piece of code to test and reproduce this effect.
Perhaps you could include actual dmesg output?
John
--
John W. Linville Someday the world will need a hero, and you
linville@tuxdriver.com might be all we have. Be ready.
^ permalink raw reply
* Compat-wireless build on 2.6.34
From: Ben Gamari @ 2010-06-16 14:49 UTC (permalink / raw)
To: linux-wireless
Is the latest compat-wireless package known to build against 2.6.34? I
have tried both June 15 and June 14 and both fail with errors similar
to the following. Thanks for your help!
- Ben
make -C /lib/modules/2.6.34-ben/build
M=/home/ben/compat-wireless-2010-06-15 modules
make[1]: Entering directory `/home/ben/linux-stable'
CC [M]
/home/ben/compat-wireless-2010-06-15/drivers/net/wireless/ipw2x00/ipw2100.o
CC [M]
/home/ben/compat-wireless-2010-06-15/drivers/net/wireless/ipw2x00/ipw2200.o
CC [M]
/home/ben/compat-wireless-2010-06-15/drivers/net/wireless/ipw2x00/libipw_module.o
CC [M]
/home/ben/compat-wireless-2010-06-15/drivers/net/wireless/ipw2x00/libipw_tx.o
/home/ben/compat-wireless-2010-06-15/drivers/net/wireless/ipw2x00/ipw2100.c:
In function ‘ipw2100_alloc_device’:
/home/ben/compat-wireless-2010-06-15/drivers/net/wireless/ipw2x00/ipw2100.c:6138:
error: ‘struct net_device’ has no member named ‘wireless_handlers’
/home/ben/compat-wireless-2010-06-15/drivers/net/wireless/ipw2x00/ipw2100.c:6144:
error: ‘struct net_device’ has no member named ‘wireless_data’
/home/ben/compat-wireless-2010-06-15/drivers/net/wireless/ipw2x00/ipw2100.c:
At top level:
/home/ben/compat-wireless-2010-06-15/drivers/net/wireless/ipw2x00/ipw2100.c:8406:
error: unknown field ‘num_private’ specified in initializer
/home/ben/compat-wireless-2010-06-15/drivers/net/wireless/ipw2x00/ipw2100.c:8406:
warning: initialization makes pointer from integer without a cast
/home/ben/compat-wireless-2010-06-15/drivers/net/wireless/ipw2x00/ipw2100.c:8407:
error: unknown field ‘num_private_args’ specified in initializer
/home/ben/compat-wireless-2010-06-15/drivers/net/wireless/ipw2x00/ipw2100.c:8407:
warning: excess elements in struct initializer
/home/ben/compat-wireless-2010-06-15/drivers/net/wireless/ipw2x00/ipw2100.c:8407:
warning: (near initialization for ‘ipw2100_wx_handler_def’)
/home/ben/compat-wireless-2010-06-15/drivers/net/wireless/ipw2x00/ipw2100.c:8408:
error: unknown field ‘private’ specified in initializer
/home/ben/compat-wireless-2010-06-15/drivers/net/wireless/ipw2x00/ipw2100.c:8408:
warning: excess elements in struct initializer
/home/ben/compat-wireless-2010-06-15/drivers/net/wireless/ipw2x00/ipw2100.c:8408:
warning: (near initialization for ‘ipw2100_wx_handler_def’)
/home/ben/compat-wireless-2010-06-15/drivers/net/wireless/ipw2x00/ipw2100.c:8409:
error: unknown field ‘private_args’ specified in initializer
/home/ben/compat-wireless-2010-06-15/drivers/net/wireless/ipw2x00/ipw2100.c:8409:
warning: excess elements in struct initializer
/home/ben/compat-wireless-2010-06-15/drivers/net/wireless/ipw2x00/ipw2100.c:8409:
warning: (near initialization for ‘ipw2100_wx_handler_def’)
/home/ben/compat-wireless-2010-06-15/drivers/net/wireless/ipw2x00/ipw2200.c:10118:
error: unknown field ‘num_private’ specified in initializer
/home/ben/compat-wireless-2010-06-15/drivers/net/wireless/ipw2x00/ipw2200.c:10118:
warning: initialization makes pointer from integer without a cast
/home/ben/compat-wireless-2010-06-15/drivers/net/wireless/ipw2x00/ipw2200.c:10119:
error: unknown field ‘num_private_args’ specified in initializer
/home/ben/compat-wireless-2010-06-15/drivers/net/wireless/ipw2x00/ipw2200.c:10119:
warning: excess elements in struct initializer
/home/ben/compat-wireless-2010-06-15/drivers/net/wireless/ipw2x00/ipw2200.c:10119:
warning: (near initialization for ‘ipw_wx_handler_def’)
/home/ben/compat-wireless-2010-06-15/drivers/net/wireless/ipw2x00/ipw2200.c:10120:
error: unknown field ‘private’ specified in initializer
/home/ben/compat-wireless-2010-06-15/drivers/net/wireless/ipw2x00/ipw2200.c:10120:
warning: excess elements in struct initializer
/home/ben/compat-wireless-2010-06-15/drivers/net/wireless/ipw2x00/ipw2200.c:10120:
warning: (near initialization for ‘ipw_wx_handler_def’)
/home/ben/compat-wireless-2010-06-15/drivers/net/wireless/ipw2x00/ipw2200.c:10121:
error: unknown field ‘private_args’ specified in initializer
/home/ben/compat-wireless-2010-06-15/drivers/net/wireless/ipw2x00/ipw2200.c:10121:
warning: excess elements in struct initializer
/home/ben/c
^ permalink raw reply
* Re: [PATCH] zd1211rw: ignore unknown regulatory domain.
From: Kouhei Sutou @ 2010-06-16 15:02 UTC (permalink / raw)
To: linville; +Cc: mcgrof, Stephen.Chen, David.Quan, linux-wireless, Michael.Green
In-Reply-To: <20100616134421.GB2404@tuxdriver.com>
Hi,
In <20100616134421.GB2404@tuxdriver.com>
"Re: [PATCH] zd1211rw: ignore unknown regulatory domain." on Wed, 16 Jun 2010 09:44:22 -0400,
"John W. Linville" <linville@tuxdriver.com> wrote:
> On Wed, Jun 16, 2010 at 09:55:23PM +0900, Kouhei Sutou wrote:
>
>> I used 'ZD_REGDOMAIN_JAPAN_GW_US54GXS' as a macro name for
>> 0x49 because 0x49 is used by GW-US54GXS. Is it OK? Or should
>> I use other name for it?
>
> Hopefully someone can suggest a better name. Anyone know what inspired
> "ZD_REGDOMAIN_JAPAN_ADD"?
'ZD_REGDOMAIN_JAPAN_3' was used in another patch:
https://patchwork.kernel.org/patch/47220/
But I don't know why '3' was used...
Thanks,
--
kou
^ permalink raw reply
* Re: [PATCH] zd1211rw: ignore unknown regulatory domain.
From: John W. Linville @ 2010-06-16 15:29 UTC (permalink / raw)
To: Kouhei Sutou
Cc: mcgrof, Stephen.Chen, David.Quan, linux-wireless, Michael.Green
In-Reply-To: <20100617.000223.656204364469558896.kou@clear-code.com>
On Thu, Jun 17, 2010 at 12:02:23AM +0900, Kouhei Sutou wrote:
> Hi,
>
> In <20100616134421.GB2404@tuxdriver.com>
> "Re: [PATCH] zd1211rw: ignore unknown regulatory domain." on Wed, 16 Jun 2010 09:44:22 -0400,
> "John W. Linville" <linville@tuxdriver.com> wrote:
>
> > On Wed, Jun 16, 2010 at 09:55:23PM +0900, Kouhei Sutou wrote:
> >
> >> I used 'ZD_REGDOMAIN_JAPAN_GW_US54GXS' as a macro name for
> >> 0x49 because 0x49 is used by GW-US54GXS. Is it OK? Or should
> >> I use other name for it?
> >
> > Hopefully someone can suggest a better name. Anyone know what inspired
> > "ZD_REGDOMAIN_JAPAN_ADD"?
>
> 'ZD_REGDOMAIN_JAPAN_3' was used in another patch:
> https://patchwork.kernel.org/patch/47220/
>
> But I don't know why '3' was used...
'3' makes sense to me. The better question is why not change 'ADD' to '2'? :-)
John
--
John W. Linville Someday the world will need a hero, and you
linville@tuxdriver.com might be all we have. Be ready.
^ permalink raw reply
page: next (older) | prev (newer) | latest
- recent:[subjects (threaded)|topics (new)|topics (active)]
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox