UbiquitousOS
/
OpenCloudOS-Kernel
mirror of https://gitee.com/OpenCloudOS/OpenCloudOS-Kernel.git
11
0
Fork 0
Code Issues Projects Releases Wiki Activity
OpenCloudOS-Kernel/drivers/net/wireless/ath/ath5k/desc.c

693 lines
19 KiB
C
Raw Normal View History

ath5k: HW code cleanup * No code changes... * Split hw.c to multiple files for better maintenance and add some documentation on each file code is going to grow soon (eeprom.c for example is going to get much stuff currently developed on ath_info) so it's better this way. * Rename following functions to maintain naming scheme: ah_setup_xtx_desc -> ah_setup_mrr_tx_desc (Because xtx doesn't say much, it's actually a multi-rate-retry tx descriptor) ath5k_hw_put_tx/rx_buf - > ath5k_hw_set_tx/rxdp ath5k_hw_get_tx/rx_buf -> ath5k_hw_get_tx/rxdp (We don't put any "buf" we set descriptor pointers on hw) ath5k_hw_tx_start -> ath5k_hw_start_tx_dma ath5k_hw_start_rx -> ath5k_hw_start_rx_dma ath5k_hw_stop_pcu_recv -> ath5k_hw_stop_rx_pcu (It's easier this way to identify them, we also have ath5k_hw_start_rx_pcu which completes the set) ath5k_hw_set_intr -> ath5k_hw_set_imr (As in get_isr we set imr here, not "intr") * Move ath5k_hw_setup_rx_desc on ah->ah_setup_rx_desc so we can include support for different rx descriptors in the future * Further cleanups so that checkpatch doesn't complain (only some > 80 col warnings for eeprom.h and reg.h as usual due to comments) Tested on 5211 and 5213 cards and works ok. Changes-licensed-under: ISC Signed-off-by: Nick Kossifidis <mickflemm@gmail.com> Acked-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-08-30 03:45:39 +08:00
/*
* Copyright (c) 2004-2008 Reyk Floeter <reyk@openbsd.org>
* Copyright (c) 2006-2008 Nick Kossifidis <mickflemm@gmail.com>
* Copyright (c) 2007-2008 Pavel Roskin <proski@gnu.org>
*
* Permission to use, copy, modify, and 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.
*
*/
/******************************\
Hardware Descriptor Functions
\******************************/
#include "ath5k.h"
#include "reg.h"
#include "debug.h"
#include "base.h"
/*
* TX Descriptors
*/
/*
* Initialize the 2-word tx control descriptor on 5210/5211
*/
static int
ath5k_hw_setup_2word_tx_desc(struct ath5k_hw *ah, struct ath5k_desc *desc,
ath5k: Fix TX/RX padding for all frames Currently, the padding position is based on ieee80211_get_hdrlen_from_skb(). This is not correct since the HW does padding on RX (and expect the same padding to be present on TX) at the following position : - management : 24 + 6 if 4-addr format - control : 24 + 6 if 4-addr format - data : 24 + 6 if 4-addr format + 2 if QoS - invalid : 24 + 6 if 4-addr format whereas ieee80211_get_hdrlen_from_skb() is : - management : 24 - control : 16 except for ACK/CTS where it is 10 - data : 24 + 6 if 4-addr format + 2 if QoS + 2 if QoS & order - invalid : 24 So, correct frames are not affected : management frames do not use 4-addr format, control frames have no body and invalid frames are ... not valid by definition. However, in order to use monitor interface for debugging purpose, one must be able to send/receive any frames, be it correct or not. Such frames are affected by incorrect padding. Moreover, since padding is added on TX, we need to remove it before calling ieee80211_tx_status. This affect TX packets received by monitor interfaces. It has been tested between an ath5k based card (AR5212) and an ar9170usb based card (netgear WNDA3100) using a frame generator and a monitor interface for each card. v2: Added ath5k_add_padding / ath5k_remove_padding Signed-off-by: Benoit Papillault <benoit.papillault@free.fr> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-02-28 06:05:26 +08:00
unsigned int pkt_len, unsigned int hdr_len, int padsize,
enum ath5k_pkt_type type,
ath5k: HW code cleanup * No code changes... * Split hw.c to multiple files for better maintenance and add some documentation on each file code is going to grow soon (eeprom.c for example is going to get much stuff currently developed on ath_info) so it's better this way. * Rename following functions to maintain naming scheme: ah_setup_xtx_desc -> ah_setup_mrr_tx_desc (Because xtx doesn't say much, it's actually a multi-rate-retry tx descriptor) ath5k_hw_put_tx/rx_buf - > ath5k_hw_set_tx/rxdp ath5k_hw_get_tx/rx_buf -> ath5k_hw_get_tx/rxdp (We don't put any "buf" we set descriptor pointers on hw) ath5k_hw_tx_start -> ath5k_hw_start_tx_dma ath5k_hw_start_rx -> ath5k_hw_start_rx_dma ath5k_hw_stop_pcu_recv -> ath5k_hw_stop_rx_pcu (It's easier this way to identify them, we also have ath5k_hw_start_rx_pcu which completes the set) ath5k_hw_set_intr -> ath5k_hw_set_imr (As in get_isr we set imr here, not "intr") * Move ath5k_hw_setup_rx_desc on ah->ah_setup_rx_desc so we can include support for different rx descriptors in the future * Further cleanups so that checkpatch doesn't complain (only some > 80 col warnings for eeprom.h and reg.h as usual due to comments) Tested on 5211 and 5213 cards and works ok. Changes-licensed-under: ISC Signed-off-by: Nick Kossifidis <mickflemm@gmail.com> Acked-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-08-30 03:45:39 +08:00
unsigned int tx_power, unsigned int tx_rate0, unsigned int tx_tries0,
unsigned int key_index, unsigned int antenna_mode, unsigned int flags,
unsigned int rtscts_rate, unsigned int rtscts_duration)
{
u32 frame_type;
struct ath5k_hw_2w_tx_ctl *tx_ctl;
unsigned int frame_len;
tx_ctl = &desc->ud.ds_tx5210.tx_ctl;
/*
* Validate input
* - Zero retries don't make sense.
* - A zero rate will put the HW into a mode where it continously sends
* noise on the channel, so it is important to avoid this.
*/
if (unlikely(tx_tries0 == 0)) {
ATH5K_ERR(ah->ah_sc, "zero retries\n");
WARN_ON(1);
return -EINVAL;
}
if (unlikely(tx_rate0 == 0)) {
ATH5K_ERR(ah->ah_sc, "zero rate\n");
WARN_ON(1);
return -EINVAL;
}
/* Clear descriptor */
memset(&desc->ud.ds_tx5210, 0, sizeof(struct ath5k_hw_5210_tx_desc));
/* Setup control descriptor */
/* Verify and set frame length */
/* remove padding we might have added before */
ath5k: Fix TX/RX padding for all frames Currently, the padding position is based on ieee80211_get_hdrlen_from_skb(). This is not correct since the HW does padding on RX (and expect the same padding to be present on TX) at the following position : - management : 24 + 6 if 4-addr format - control : 24 + 6 if 4-addr format - data : 24 + 6 if 4-addr format + 2 if QoS - invalid : 24 + 6 if 4-addr format whereas ieee80211_get_hdrlen_from_skb() is : - management : 24 - control : 16 except for ACK/CTS where it is 10 - data : 24 + 6 if 4-addr format + 2 if QoS + 2 if QoS & order - invalid : 24 So, correct frames are not affected : management frames do not use 4-addr format, control frames have no body and invalid frames are ... not valid by definition. However, in order to use monitor interface for debugging purpose, one must be able to send/receive any frames, be it correct or not. Such frames are affected by incorrect padding. Moreover, since padding is added on TX, we need to remove it before calling ieee80211_tx_status. This affect TX packets received by monitor interfaces. It has been tested between an ath5k based card (AR5212) and an ar9170usb based card (netgear WNDA3100) using a frame generator and a monitor interface for each card. v2: Added ath5k_add_padding / ath5k_remove_padding Signed-off-by: Benoit Papillault <benoit.papillault@free.fr> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-02-28 06:05:26 +08:00
frame_len = pkt_len - padsize + FCS_LEN;
ath5k: HW code cleanup * No code changes... * Split hw.c to multiple files for better maintenance and add some documentation on each file code is going to grow soon (eeprom.c for example is going to get much stuff currently developed on ath_info) so it's better this way. * Rename following functions to maintain naming scheme: ah_setup_xtx_desc -> ah_setup_mrr_tx_desc (Because xtx doesn't say much, it's actually a multi-rate-retry tx descriptor) ath5k_hw_put_tx/rx_buf - > ath5k_hw_set_tx/rxdp ath5k_hw_get_tx/rx_buf -> ath5k_hw_get_tx/rxdp (We don't put any "buf" we set descriptor pointers on hw) ath5k_hw_tx_start -> ath5k_hw_start_tx_dma ath5k_hw_start_rx -> ath5k_hw_start_rx_dma ath5k_hw_stop_pcu_recv -> ath5k_hw_stop_rx_pcu (It's easier this way to identify them, we also have ath5k_hw_start_rx_pcu which completes the set) ath5k_hw_set_intr -> ath5k_hw_set_imr (As in get_isr we set imr here, not "intr") * Move ath5k_hw_setup_rx_desc on ah->ah_setup_rx_desc so we can include support for different rx descriptors in the future * Further cleanups so that checkpatch doesn't complain (only some > 80 col warnings for eeprom.h and reg.h as usual due to comments) Tested on 5211 and 5213 cards and works ok. Changes-licensed-under: ISC Signed-off-by: Nick Kossifidis <mickflemm@gmail.com> Acked-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-08-30 03:45:39 +08:00
if (frame_len & ~AR5K_2W_TX_DESC_CTL0_FRAME_LEN)
return -EINVAL;
tx_ctl->tx_control_0 = frame_len & AR5K_2W_TX_DESC_CTL0_FRAME_LEN;
/* Verify and set buffer length */
/* NB: beacon's BufLen must be a multiple of 4 bytes */
if (type == AR5K_PKT_TYPE_BEACON)
pkt_len = roundup(pkt_len, 4);
if (pkt_len & ~AR5K_2W_TX_DESC_CTL1_BUF_LEN)
return -EINVAL;
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 ?
*/
if (ah->ah_version == AR5K_AR5210) {
if (hdr_len & ~AR5K_2W_TX_DESC_CTL0_HEADER_LEN)
return -EINVAL;
tx_ctl->tx_control_0 |=
AR5K_REG_SM(hdr_len, AR5K_2W_TX_DESC_CTL0_HEADER_LEN);
}
ath5k: Fix TX/RX padding for all frames Currently, the padding position is based on ieee80211_get_hdrlen_from_skb(). This is not correct since the HW does padding on RX (and expect the same padding to be present on TX) at the following position : - management : 24 + 6 if 4-addr format - control : 24 + 6 if 4-addr format - data : 24 + 6 if 4-addr format + 2 if QoS - invalid : 24 + 6 if 4-addr format whereas ieee80211_get_hdrlen_from_skb() is : - management : 24 - control : 16 except for ACK/CTS where it is 10 - data : 24 + 6 if 4-addr format + 2 if QoS + 2 if QoS & order - invalid : 24 So, correct frames are not affected : management frames do not use 4-addr format, control frames have no body and invalid frames are ... not valid by definition. However, in order to use monitor interface for debugging purpose, one must be able to send/receive any frames, be it correct or not. Such frames are affected by incorrect padding. Moreover, since padding is added on TX, we need to remove it before calling ieee80211_tx_status. This affect TX packets received by monitor interfaces. It has been tested between an ath5k based card (AR5212) and an ar9170usb based card (netgear WNDA3100) using a frame generator and a monitor interface for each card. v2: Added ath5k_add_padding / ath5k_remove_padding Signed-off-by: Benoit Papillault <benoit.papillault@free.fr> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-02-28 06:05:26 +08:00
/*Differences between 5210-5211*/
ath5k: HW code cleanup * No code changes... * Split hw.c to multiple files for better maintenance and add some documentation on each file code is going to grow soon (eeprom.c for example is going to get much stuff currently developed on ath_info) so it's better this way. * Rename following functions to maintain naming scheme: ah_setup_xtx_desc -> ah_setup_mrr_tx_desc (Because xtx doesn't say much, it's actually a multi-rate-retry tx descriptor) ath5k_hw_put_tx/rx_buf - > ath5k_hw_set_tx/rxdp ath5k_hw_get_tx/rx_buf -> ath5k_hw_get_tx/rxdp (We don't put any "buf" we set descriptor pointers on hw) ath5k_hw_tx_start -> ath5k_hw_start_tx_dma ath5k_hw_start_rx -> ath5k_hw_start_rx_dma ath5k_hw_stop_pcu_recv -> ath5k_hw_stop_rx_pcu (It's easier this way to identify them, we also have ath5k_hw_start_rx_pcu which completes the set) ath5k_hw_set_intr -> ath5k_hw_set_imr (As in get_isr we set imr here, not "intr") * Move ath5k_hw_setup_rx_desc on ah->ah_setup_rx_desc so we can include support for different rx descriptors in the future * Further cleanups so that checkpatch doesn't complain (only some > 80 col warnings for eeprom.h and reg.h as usual due to comments) Tested on 5211 and 5213 cards and works ok. Changes-licensed-under: ISC Signed-off-by: Nick Kossifidis <mickflemm@gmail.com> Acked-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-08-30 03:45:39 +08:00
if (ah->ah_version == AR5K_AR5210) {
switch (type) {
case AR5K_PKT_TYPE_BEACON:
case AR5K_PKT_TYPE_PROBE_RESP:
frame_type = AR5K_AR5210_TX_DESC_FRAME_TYPE_NO_DELAY;
case AR5K_PKT_TYPE_PIFS:
frame_type = AR5K_AR5210_TX_DESC_FRAME_TYPE_PIFS;
default:
frame_type = type /*<< 2 ?*/;
}
tx_ctl->tx_control_0 |=
AR5K_REG_SM(frame_type, AR5K_2W_TX_DESC_CTL0_FRAME_TYPE) |
AR5K_REG_SM(tx_rate0, AR5K_2W_TX_DESC_CTL0_XMIT_RATE);
} else {
tx_ctl->tx_control_0 |=
AR5K_REG_SM(tx_rate0, AR5K_2W_TX_DESC_CTL0_XMIT_RATE) |
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);
}
#define _TX_FLAGS(_c, _flag) \
if (flags & AR5K_TXDESC_##_flag) { \
tx_ctl->tx_control_##_c |= \
AR5K_2W_TX_DESC_CTL##_c##_##_flag; \
}
_TX_FLAGS(0, CLRDMASK);
_TX_FLAGS(0, VEOL);
_TX_FLAGS(0, INTREQ);
_TX_FLAGS(0, RTSENA);
_TX_FLAGS(1, NOACK);
#undef _TX_FLAGS
/*
* WEP crap
*/
if (key_index != AR5K_TXKEYIX_INVALID) {
tx_ctl->tx_control_0 |=
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);
}
/*
* RTS/CTS Duration [5210 ?]
*/
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;
return 0;
}
/*
* Initialize the 4-word tx control descriptor on 5212
*/
static int ath5k_hw_setup_4word_tx_desc(struct ath5k_hw *ah,
struct ath5k_desc *desc, unsigned int pkt_len, unsigned int hdr_len,
ath5k: Fix TX/RX padding for all frames Currently, the padding position is based on ieee80211_get_hdrlen_from_skb(). This is not correct since the HW does padding on RX (and expect the same padding to be present on TX) at the following position : - management : 24 + 6 if 4-addr format - control : 24 + 6 if 4-addr format - data : 24 + 6 if 4-addr format + 2 if QoS - invalid : 24 + 6 if 4-addr format whereas ieee80211_get_hdrlen_from_skb() is : - management : 24 - control : 16 except for ACK/CTS where it is 10 - data : 24 + 6 if 4-addr format + 2 if QoS + 2 if QoS & order - invalid : 24 So, correct frames are not affected : management frames do not use 4-addr format, control frames have no body and invalid frames are ... not valid by definition. However, in order to use monitor interface for debugging purpose, one must be able to send/receive any frames, be it correct or not. Such frames are affected by incorrect padding. Moreover, since padding is added on TX, we need to remove it before calling ieee80211_tx_status. This affect TX packets received by monitor interfaces. It has been tested between an ath5k based card (AR5212) and an ar9170usb based card (netgear WNDA3100) using a frame generator and a monitor interface for each card. v2: Added ath5k_add_padding / ath5k_remove_padding Signed-off-by: Benoit Papillault <benoit.papillault@free.fr> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-02-28 06:05:26 +08:00
int padsize,
ath5k: HW code cleanup * No code changes... * Split hw.c to multiple files for better maintenance and add some documentation on each file code is going to grow soon (eeprom.c for example is going to get much stuff currently developed on ath_info) so it's better this way. * Rename following functions to maintain naming scheme: ah_setup_xtx_desc -> ah_setup_mrr_tx_desc (Because xtx doesn't say much, it's actually a multi-rate-retry tx descriptor) ath5k_hw_put_tx/rx_buf - > ath5k_hw_set_tx/rxdp ath5k_hw_get_tx/rx_buf -> ath5k_hw_get_tx/rxdp (We don't put any "buf" we set descriptor pointers on hw) ath5k_hw_tx_start -> ath5k_hw_start_tx_dma ath5k_hw_start_rx -> ath5k_hw_start_rx_dma ath5k_hw_stop_pcu_recv -> ath5k_hw_stop_rx_pcu (It's easier this way to identify them, we also have ath5k_hw_start_rx_pcu which completes the set) ath5k_hw_set_intr -> ath5k_hw_set_imr (As in get_isr we set imr here, not "intr") * Move ath5k_hw_setup_rx_desc on ah->ah_setup_rx_desc so we can include support for different rx descriptors in the future * Further cleanups so that checkpatch doesn't complain (only some > 80 col warnings for eeprom.h and reg.h as usual due to comments) Tested on 5211 and 5213 cards and works ok. Changes-licensed-under: ISC Signed-off-by: Nick Kossifidis <mickflemm@gmail.com> Acked-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-08-30 03:45:39 +08:00
enum ath5k_pkt_type type, unsigned int tx_power, unsigned int tx_rate0,
unsigned int tx_tries0, unsigned int key_index,
unsigned int antenna_mode, unsigned int flags,
unsigned int rtscts_rate,
unsigned int rtscts_duration)
{
struct ath5k_hw_4w_tx_ctl *tx_ctl;
unsigned int frame_len;
ATH5K_TRACE(ah->ah_sc);
tx_ctl = &desc->ud.ds_tx5212.tx_ctl;
/*
* Validate input
* - Zero retries don't make sense.
* - A zero rate will put the HW into a mode where it continously sends
* noise on the channel, so it is important to avoid this.
*/
if (unlikely(tx_tries0 == 0)) {
ATH5K_ERR(ah->ah_sc, "zero retries\n");
WARN_ON(1);
return -EINVAL;
}
if (unlikely(tx_rate0 == 0)) {
ATH5K_ERR(ah->ah_sc, "zero rate\n");
WARN_ON(1);
return -EINVAL;
}
ath5k: Add tx power calibration support * Add tx power calibration support * Add a few tx power limits * Hardcode default power to 12.5dB * Disable TPC for now v2: Address Jiri's comments Signed-off-by: Nick Kossifidis <mickflemm@gmail.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-03-16 04:20:35 +08:00
tx_power += ah->ah_txpower.txp_offset;
if (tx_power > AR5K_TUNE_MAX_TXPOWER)
tx_power = AR5K_TUNE_MAX_TXPOWER;
ath5k: HW code cleanup * No code changes... * Split hw.c to multiple files for better maintenance and add some documentation on each file code is going to grow soon (eeprom.c for example is going to get much stuff currently developed on ath_info) so it's better this way. * Rename following functions to maintain naming scheme: ah_setup_xtx_desc -> ah_setup_mrr_tx_desc (Because xtx doesn't say much, it's actually a multi-rate-retry tx descriptor) ath5k_hw_put_tx/rx_buf - > ath5k_hw_set_tx/rxdp ath5k_hw_get_tx/rx_buf -> ath5k_hw_get_tx/rxdp (We don't put any "buf" we set descriptor pointers on hw) ath5k_hw_tx_start -> ath5k_hw_start_tx_dma ath5k_hw_start_rx -> ath5k_hw_start_rx_dma ath5k_hw_stop_pcu_recv -> ath5k_hw_stop_rx_pcu (It's easier this way to identify them, we also have ath5k_hw_start_rx_pcu which completes the set) ath5k_hw_set_intr -> ath5k_hw_set_imr (As in get_isr we set imr here, not "intr") * Move ath5k_hw_setup_rx_desc on ah->ah_setup_rx_desc so we can include support for different rx descriptors in the future * Further cleanups so that checkpatch doesn't complain (only some > 80 col warnings for eeprom.h and reg.h as usual due to comments) Tested on 5211 and 5213 cards and works ok. Changes-licensed-under: ISC Signed-off-by: Nick Kossifidis <mickflemm@gmail.com> Acked-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-08-30 03:45:39 +08:00
/* Clear descriptor */
memset(&desc->ud.ds_tx5212, 0, sizeof(struct ath5k_hw_5212_tx_desc));
/* Setup control descriptor */
/* Verify and set frame length */
/* remove padding we might have added before */
ath5k: Fix TX/RX padding for all frames Currently, the padding position is based on ieee80211_get_hdrlen_from_skb(). This is not correct since the HW does padding on RX (and expect the same padding to be present on TX) at the following position : - management : 24 + 6 if 4-addr format - control : 24 + 6 if 4-addr format - data : 24 + 6 if 4-addr format + 2 if QoS - invalid : 24 + 6 if 4-addr format whereas ieee80211_get_hdrlen_from_skb() is : - management : 24 - control : 16 except for ACK/CTS where it is 10 - data : 24 + 6 if 4-addr format + 2 if QoS + 2 if QoS & order - invalid : 24 So, correct frames are not affected : management frames do not use 4-addr format, control frames have no body and invalid frames are ... not valid by definition. However, in order to use monitor interface for debugging purpose, one must be able to send/receive any frames, be it correct or not. Such frames are affected by incorrect padding. Moreover, since padding is added on TX, we need to remove it before calling ieee80211_tx_status. This affect TX packets received by monitor interfaces. It has been tested between an ath5k based card (AR5212) and an ar9170usb based card (netgear WNDA3100) using a frame generator and a monitor interface for each card. v2: Added ath5k_add_padding / ath5k_remove_padding Signed-off-by: Benoit Papillault <benoit.papillault@free.fr> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-02-28 06:05:26 +08:00
frame_len = pkt_len - padsize + FCS_LEN;
ath5k: HW code cleanup * No code changes... * Split hw.c to multiple files for better maintenance and add some documentation on each file code is going to grow soon (eeprom.c for example is going to get much stuff currently developed on ath_info) so it's better this way. * Rename following functions to maintain naming scheme: ah_setup_xtx_desc -> ah_setup_mrr_tx_desc (Because xtx doesn't say much, it's actually a multi-rate-retry tx descriptor) ath5k_hw_put_tx/rx_buf - > ath5k_hw_set_tx/rxdp ath5k_hw_get_tx/rx_buf -> ath5k_hw_get_tx/rxdp (We don't put any "buf" we set descriptor pointers on hw) ath5k_hw_tx_start -> ath5k_hw_start_tx_dma ath5k_hw_start_rx -> ath5k_hw_start_rx_dma ath5k_hw_stop_pcu_recv -> ath5k_hw_stop_rx_pcu (It's easier this way to identify them, we also have ath5k_hw_start_rx_pcu which completes the set) ath5k_hw_set_intr -> ath5k_hw_set_imr (As in get_isr we set imr here, not "intr") * Move ath5k_hw_setup_rx_desc on ah->ah_setup_rx_desc so we can include support for different rx descriptors in the future * Further cleanups so that checkpatch doesn't complain (only some > 80 col warnings for eeprom.h and reg.h as usual due to comments) Tested on 5211 and 5213 cards and works ok. Changes-licensed-under: ISC Signed-off-by: Nick Kossifidis <mickflemm@gmail.com> Acked-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-08-30 03:45:39 +08:00
if (frame_len & ~AR5K_4W_TX_DESC_CTL0_FRAME_LEN)
return -EINVAL;
tx_ctl->tx_control_0 = frame_len & AR5K_4W_TX_DESC_CTL0_FRAME_LEN;
/* Verify and set buffer length */
/* NB: beacon's BufLen must be a multiple of 4 bytes */
if (type == AR5K_PKT_TYPE_BEACON)
pkt_len = roundup(pkt_len, 4);
if (pkt_len & ~AR5K_4W_TX_DESC_CTL1_BUF_LEN)
return -EINVAL;
tx_ctl->tx_control_1 = pkt_len & AR5K_4W_TX_DESC_CTL1_BUF_LEN;
tx_ctl->tx_control_0 |=
AR5K_REG_SM(tx_power, AR5K_4W_TX_DESC_CTL0_XMIT_POWER) |
AR5K_REG_SM(antenna_mode, AR5K_4W_TX_DESC_CTL0_ANT_MODE_XMIT);
tx_ctl->tx_control_1 |= AR5K_REG_SM(type,
AR5K_4W_TX_DESC_CTL1_FRAME_TYPE);
tx_ctl->tx_control_2 = AR5K_REG_SM(tx_tries0 + AR5K_TUNE_HWTXTRIES,
AR5K_4W_TX_DESC_CTL2_XMIT_TRIES0);
tx_ctl->tx_control_3 = tx_rate0 & AR5K_4W_TX_DESC_CTL3_XMIT_RATE0;
#define _TX_FLAGS(_c, _flag) \
if (flags & AR5K_TXDESC_##_flag) { \
tx_ctl->tx_control_##_c |= \
AR5K_4W_TX_DESC_CTL##_c##_##_flag; \
}
_TX_FLAGS(0, CLRDMASK);
_TX_FLAGS(0, VEOL);
_TX_FLAGS(0, INTREQ);
_TX_FLAGS(0, RTSENA);
_TX_FLAGS(0, CTSENA);
_TX_FLAGS(1, NOACK);
#undef _TX_FLAGS
/*
* WEP crap
*/
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);
}
/*
* RTS/CTS
*/
if (flags & (AR5K_TXDESC_RTSENA | AR5K_TXDESC_CTSENA)) {
if ((flags & AR5K_TXDESC_RTSENA) &&
(flags & AR5K_TXDESC_CTSENA))
return -EINVAL;
tx_ctl->tx_control_2 |= rtscts_duration &
AR5K_4W_TX_DESC_CTL2_RTS_DURATION;
tx_ctl->tx_control_3 |= AR5K_REG_SM(rtscts_rate,
AR5K_4W_TX_DESC_CTL3_RTS_CTS_RATE);
}
return 0;
}
/*
* Initialize a 4-word multi rate retry tx control descriptor on 5212
*/
static 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;
/*
* 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
* it continously sends noise on the channel, so it is important to
* avoid this.
*/
if (unlikely((tx_rate1 == 0 && tx_tries1 != 0) ||
(tx_rate2 == 0 && tx_tries2 != 0) ||
(tx_rate3 == 0 && tx_tries3 != 0))) {
ATH5K_ERR(ah->ah_sc, "zero rate\n");
WARN_ON(1);
return -EINVAL;
}
if (ah->ah_version == AR5K_AR5212) {
tx_ctl = &desc->ud.ds_tx5212.tx_ctl;
#define _XTX_TRIES(_n) \
if (tx_tries##_n) { \
tx_ctl->tx_control_2 |= \
AR5K_REG_SM(tx_tries##_n, \
AR5K_4W_TX_DESC_CTL2_XMIT_TRIES##_n); \
tx_ctl->tx_control_3 |= \
AR5K_REG_SM(tx_rate##_n, \
AR5K_4W_TX_DESC_CTL3_XMIT_RATE##_n); \
}
_XTX_TRIES(1);
_XTX_TRIES(2);
_XTX_TRIES(3);
#undef _XTX_TRIES
return 1;
}
return 0;
}
ath5k: implement multi-rate retry support, fix tx status reporting Clean up the tx status reporting, fix retry counters (short retries are virtual collisions, not actual retries). Implement multi-rate retry support. This also fixes strong throughput fluctuations with rc80211_pid Signed-off-by: Felix Fietkau <nbd@openwrt.org> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-10-06 00:05:48 +08:00
/* 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;
}
ath5k: HW code cleanup * No code changes... * Split hw.c to multiple files for better maintenance and add some documentation on each file code is going to grow soon (eeprom.c for example is going to get much stuff currently developed on ath_info) so it's better this way. * Rename following functions to maintain naming scheme: ah_setup_xtx_desc -> ah_setup_mrr_tx_desc (Because xtx doesn't say much, it's actually a multi-rate-retry tx descriptor) ath5k_hw_put_tx/rx_buf - > ath5k_hw_set_tx/rxdp ath5k_hw_get_tx/rx_buf -> ath5k_hw_get_tx/rxdp (We don't put any "buf" we set descriptor pointers on hw) ath5k_hw_tx_start -> ath5k_hw_start_tx_dma ath5k_hw_start_rx -> ath5k_hw_start_rx_dma ath5k_hw_stop_pcu_recv -> ath5k_hw_stop_rx_pcu (It's easier this way to identify them, we also have ath5k_hw_start_rx_pcu which completes the set) ath5k_hw_set_intr -> ath5k_hw_set_imr (As in get_isr we set imr here, not "intr") * Move ath5k_hw_setup_rx_desc on ah->ah_setup_rx_desc so we can include support for different rx descriptors in the future * Further cleanups so that checkpatch doesn't complain (only some > 80 col warnings for eeprom.h and reg.h as usual due to comments) Tested on 5211 and 5213 cards and works ok. Changes-licensed-under: ISC Signed-off-by: Nick Kossifidis <mickflemm@gmail.com> Acked-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-08-30 03:45:39 +08:00
/*
* Proccess the tx status descriptor on 5210/5211
*/
static int ath5k_hw_proc_2word_tx_status(struct ath5k_hw *ah,
struct ath5k_desc *desc, struct ath5k_tx_status *ts)
{
struct ath5k_hw_2w_tx_ctl *tx_ctl;
struct ath5k_hw_tx_status *tx_status;
ATH5K_TRACE(ah->ah_sc);
tx_ctl = &desc->ud.ds_tx5210.tx_ctl;
tx_status = &desc->ud.ds_tx5210.tx_stat;
/* No frame has been send or error */
if (unlikely((tx_status->tx_status_1 & AR5K_DESC_TX_STATUS1_DONE) == 0))
return -EINPROGRESS;
/*
* Get descriptor status
*/
ts->ts_tstamp = AR5K_REG_MS(tx_status->tx_status_0,
AR5K_DESC_TX_STATUS0_SEND_TIMESTAMP);
ts->ts_shortretry = AR5K_REG_MS(tx_status->tx_status_0,
AR5K_DESC_TX_STATUS0_SHORT_RETRY_COUNT);
ts->ts_longretry = AR5K_REG_MS(tx_status->tx_status_0,
AR5K_DESC_TX_STATUS0_LONG_RETRY_COUNT);
/*TODO: ts->ts_virtcol + test*/
ts->ts_seqnum = AR5K_REG_MS(tx_status->tx_status_1,
AR5K_DESC_TX_STATUS1_SEQ_NUM);
ts->ts_rssi = AR5K_REG_MS(tx_status->tx_status_1,
AR5K_DESC_TX_STATUS1_ACK_SIG_STRENGTH);
ts->ts_antenna = 1;
ts->ts_status = 0;
ath5k: implement multi-rate retry support, fix tx status reporting Clean up the tx status reporting, fix retry counters (short retries are virtual collisions, not actual retries). Implement multi-rate retry support. This also fixes strong throughput fluctuations with rc80211_pid Signed-off-by: Felix Fietkau <nbd@openwrt.org> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-10-06 00:05:48 +08:00
ts->ts_rate[0] = AR5K_REG_MS(tx_ctl->tx_control_0,
ath5k: HW code cleanup * No code changes... * Split hw.c to multiple files for better maintenance and add some documentation on each file code is going to grow soon (eeprom.c for example is going to get much stuff currently developed on ath_info) so it's better this way. * Rename following functions to maintain naming scheme: ah_setup_xtx_desc -> ah_setup_mrr_tx_desc (Because xtx doesn't say much, it's actually a multi-rate-retry tx descriptor) ath5k_hw_put_tx/rx_buf - > ath5k_hw_set_tx/rxdp ath5k_hw_get_tx/rx_buf -> ath5k_hw_get_tx/rxdp (We don't put any "buf" we set descriptor pointers on hw) ath5k_hw_tx_start -> ath5k_hw_start_tx_dma ath5k_hw_start_rx -> ath5k_hw_start_rx_dma ath5k_hw_stop_pcu_recv -> ath5k_hw_stop_rx_pcu (It's easier this way to identify them, we also have ath5k_hw_start_rx_pcu which completes the set) ath5k_hw_set_intr -> ath5k_hw_set_imr (As in get_isr we set imr here, not "intr") * Move ath5k_hw_setup_rx_desc on ah->ah_setup_rx_desc so we can include support for different rx descriptors in the future * Further cleanups so that checkpatch doesn't complain (only some > 80 col warnings for eeprom.h and reg.h as usual due to comments) Tested on 5211 and 5213 cards and works ok. Changes-licensed-under: ISC Signed-off-by: Nick Kossifidis <mickflemm@gmail.com> Acked-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-08-30 03:45:39 +08:00
AR5K_2W_TX_DESC_CTL0_XMIT_RATE);
ath5k: implement multi-rate retry support, fix tx status reporting Clean up the tx status reporting, fix retry counters (short retries are virtual collisions, not actual retries). Implement multi-rate retry support. This also fixes strong throughput fluctuations with rc80211_pid Signed-off-by: Felix Fietkau <nbd@openwrt.org> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-10-06 00:05:48 +08:00
ts->ts_retry[0] = ts->ts_longretry;
ts->ts_final_idx = 0;
ath5k: HW code cleanup * No code changes... * Split hw.c to multiple files for better maintenance and add some documentation on each file code is going to grow soon (eeprom.c for example is going to get much stuff currently developed on ath_info) so it's better this way. * Rename following functions to maintain naming scheme: ah_setup_xtx_desc -> ah_setup_mrr_tx_desc (Because xtx doesn't say much, it's actually a multi-rate-retry tx descriptor) ath5k_hw_put_tx/rx_buf - > ath5k_hw_set_tx/rxdp ath5k_hw_get_tx/rx_buf -> ath5k_hw_get_tx/rxdp (We don't put any "buf" we set descriptor pointers on hw) ath5k_hw_tx_start -> ath5k_hw_start_tx_dma ath5k_hw_start_rx -> ath5k_hw_start_rx_dma ath5k_hw_stop_pcu_recv -> ath5k_hw_stop_rx_pcu (It's easier this way to identify them, we also have ath5k_hw_start_rx_pcu which completes the set) ath5k_hw_set_intr -> ath5k_hw_set_imr (As in get_isr we set imr here, not "intr") * Move ath5k_hw_setup_rx_desc on ah->ah_setup_rx_desc so we can include support for different rx descriptors in the future * Further cleanups so that checkpatch doesn't complain (only some > 80 col warnings for eeprom.h and reg.h as usual due to comments) Tested on 5211 and 5213 cards and works ok. Changes-licensed-under: ISC Signed-off-by: Nick Kossifidis <mickflemm@gmail.com> Acked-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-08-30 03:45:39 +08:00
if (!(tx_status->tx_status_0 & AR5K_DESC_TX_STATUS0_FRAME_XMIT_OK)) {
if (tx_status->tx_status_0 &
AR5K_DESC_TX_STATUS0_EXCESSIVE_RETRIES)
ts->ts_status |= AR5K_TXERR_XRETRY;
if (tx_status->tx_status_0 & AR5K_DESC_TX_STATUS0_FIFO_UNDERRUN)
ts->ts_status |= AR5K_TXERR_FIFO;
if (tx_status->tx_status_0 & AR5K_DESC_TX_STATUS0_FILTERED)
ts->ts_status |= AR5K_TXERR_FILT;
}
return 0;
}
/*
* Proccess a tx status descriptor on 5212
*/
static int ath5k_hw_proc_4word_tx_status(struct ath5k_hw *ah,
struct ath5k_desc *desc, struct ath5k_tx_status *ts)
{
struct ath5k_hw_4w_tx_ctl *tx_ctl;
struct ath5k_hw_tx_status *tx_status;
ATH5K_TRACE(ah->ah_sc);
tx_ctl = &desc->ud.ds_tx5212.tx_ctl;
tx_status = &desc->ud.ds_tx5212.tx_stat;
/* No frame has been send or error */
if (unlikely(!(tx_status->tx_status_1 & AR5K_DESC_TX_STATUS1_DONE)))
return -EINPROGRESS;
/*
* Get descriptor status
*/
ts->ts_tstamp = AR5K_REG_MS(tx_status->tx_status_0,
AR5K_DESC_TX_STATUS0_SEND_TIMESTAMP);
ts->ts_shortretry = AR5K_REG_MS(tx_status->tx_status_0,
AR5K_DESC_TX_STATUS0_SHORT_RETRY_COUNT);
ts->ts_longretry = AR5K_REG_MS(tx_status->tx_status_0,
AR5K_DESC_TX_STATUS0_LONG_RETRY_COUNT);
ts->ts_seqnum = AR5K_REG_MS(tx_status->tx_status_1,
AR5K_DESC_TX_STATUS1_SEQ_NUM);
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;
ts->ts_status = 0;
ath5k: implement multi-rate retry support, fix tx status reporting Clean up the tx status reporting, fix retry counters (short retries are virtual collisions, not actual retries). Implement multi-rate retry support. This also fixes strong throughput fluctuations with rc80211_pid Signed-off-by: Felix Fietkau <nbd@openwrt.org> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-10-06 00:05:48 +08:00
ts->ts_final_idx = AR5K_REG_MS(tx_status->tx_status_1,
AR5K_DESC_TX_STATUS1_FINAL_TS_INDEX);
/* The longretry counter has the number of un-acked retries
* for the final rate. To get the total number of retries
* we have to add the retry counters for the other rates
* as well
*/
ts->ts_retry[ts->ts_final_idx] = ts->ts_longretry;
switch (ts->ts_final_idx) {
case 3:
ts->ts_rate[3] = AR5K_REG_MS(tx_ctl->tx_control_3,
AR5K_4W_TX_DESC_CTL3_XMIT_RATE3);
ts->ts_retry[2] = AR5K_REG_MS(tx_ctl->tx_control_2,
AR5K_4W_TX_DESC_CTL2_XMIT_TRIES2);
ts->ts_longretry += ts->ts_retry[2];
/* fall through */
case 2:
ts->ts_rate[2] = AR5K_REG_MS(tx_ctl->tx_control_3,
AR5K_4W_TX_DESC_CTL3_XMIT_RATE2);
ts->ts_retry[1] = AR5K_REG_MS(tx_ctl->tx_control_2,
AR5K_4W_TX_DESC_CTL2_XMIT_TRIES1);
ts->ts_longretry += ts->ts_retry[1];
/* fall through */
ath5k: HW code cleanup * No code changes... * Split hw.c to multiple files for better maintenance and add some documentation on each file code is going to grow soon (eeprom.c for example is going to get much stuff currently developed on ath_info) so it's better this way. * Rename following functions to maintain naming scheme: ah_setup_xtx_desc -> ah_setup_mrr_tx_desc (Because xtx doesn't say much, it's actually a multi-rate-retry tx descriptor) ath5k_hw_put_tx/rx_buf - > ath5k_hw_set_tx/rxdp ath5k_hw_get_tx/rx_buf -> ath5k_hw_get_tx/rxdp (We don't put any "buf" we set descriptor pointers on hw) ath5k_hw_tx_start -> ath5k_hw_start_tx_dma ath5k_hw_start_rx -> ath5k_hw_start_rx_dma ath5k_hw_stop_pcu_recv -> ath5k_hw_stop_rx_pcu (It's easier this way to identify them, we also have ath5k_hw_start_rx_pcu which completes the set) ath5k_hw_set_intr -> ath5k_hw_set_imr (As in get_isr we set imr here, not "intr") * Move ath5k_hw_setup_rx_desc on ah->ah_setup_rx_desc so we can include support for different rx descriptors in the future * Further cleanups so that checkpatch doesn't complain (only some > 80 col warnings for eeprom.h and reg.h as usual due to comments) Tested on 5211 and 5213 cards and works ok. Changes-licensed-under: ISC Signed-off-by: Nick Kossifidis <mickflemm@gmail.com> Acked-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-08-30 03:45:39 +08:00
case 1:
ath5k: implement multi-rate retry support, fix tx status reporting Clean up the tx status reporting, fix retry counters (short retries are virtual collisions, not actual retries). Implement multi-rate retry support. This also fixes strong throughput fluctuations with rc80211_pid Signed-off-by: Felix Fietkau <nbd@openwrt.org> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-10-06 00:05:48 +08:00
ts->ts_rate[1] = AR5K_REG_MS(tx_ctl->tx_control_3,
ath5k: HW code cleanup * No code changes... * Split hw.c to multiple files for better maintenance and add some documentation on each file code is going to grow soon (eeprom.c for example is going to get much stuff currently developed on ath_info) so it's better this way. * Rename following functions to maintain naming scheme: ah_setup_xtx_desc -> ah_setup_mrr_tx_desc (Because xtx doesn't say much, it's actually a multi-rate-retry tx descriptor) ath5k_hw_put_tx/rx_buf - > ath5k_hw_set_tx/rxdp ath5k_hw_get_tx/rx_buf -> ath5k_hw_get_tx/rxdp (We don't put any "buf" we set descriptor pointers on hw) ath5k_hw_tx_start -> ath5k_hw_start_tx_dma ath5k_hw_start_rx -> ath5k_hw_start_rx_dma ath5k_hw_stop_pcu_recv -> ath5k_hw_stop_rx_pcu (It's easier this way to identify them, we also have ath5k_hw_start_rx_pcu which completes the set) ath5k_hw_set_intr -> ath5k_hw_set_imr (As in get_isr we set imr here, not "intr") * Move ath5k_hw_setup_rx_desc on ah->ah_setup_rx_desc so we can include support for different rx descriptors in the future * Further cleanups so that checkpatch doesn't complain (only some > 80 col warnings for eeprom.h and reg.h as usual due to comments) Tested on 5211 and 5213 cards and works ok. Changes-licensed-under: ISC Signed-off-by: Nick Kossifidis <mickflemm@gmail.com> Acked-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-08-30 03:45:39 +08:00
AR5K_4W_TX_DESC_CTL3_XMIT_RATE1);
ath5k: implement multi-rate retry support, fix tx status reporting Clean up the tx status reporting, fix retry counters (short retries are virtual collisions, not actual retries). Implement multi-rate retry support. This also fixes strong throughput fluctuations with rc80211_pid Signed-off-by: Felix Fietkau <nbd@openwrt.org> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-10-06 00:05:48 +08:00
ts->ts_retry[0] = AR5K_REG_MS(tx_ctl->tx_control_2,
ath5k: HW code cleanup * No code changes... * Split hw.c to multiple files for better maintenance and add some documentation on each file code is going to grow soon (eeprom.c for example is going to get much stuff currently developed on ath_info) so it's better this way. * Rename following functions to maintain naming scheme: ah_setup_xtx_desc -> ah_setup_mrr_tx_desc (Because xtx doesn't say much, it's actually a multi-rate-retry tx descriptor) ath5k_hw_put_tx/rx_buf - > ath5k_hw_set_tx/rxdp ath5k_hw_get_tx/rx_buf -> ath5k_hw_get_tx/rxdp (We don't put any "buf" we set descriptor pointers on hw) ath5k_hw_tx_start -> ath5k_hw_start_tx_dma ath5k_hw_start_rx -> ath5k_hw_start_rx_dma ath5k_hw_stop_pcu_recv -> ath5k_hw_stop_rx_pcu (It's easier this way to identify them, we also have ath5k_hw_start_rx_pcu which completes the set) ath5k_hw_set_intr -> ath5k_hw_set_imr (As in get_isr we set imr here, not "intr") * Move ath5k_hw_setup_rx_desc on ah->ah_setup_rx_desc so we can include support for different rx descriptors in the future * Further cleanups so that checkpatch doesn't complain (only some > 80 col warnings for eeprom.h and reg.h as usual due to comments) Tested on 5211 and 5213 cards and works ok. Changes-licensed-under: ISC Signed-off-by: Nick Kossifidis <mickflemm@gmail.com> Acked-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-08-30 03:45:39 +08:00
AR5K_4W_TX_DESC_CTL2_XMIT_TRIES1);
ath5k: implement multi-rate retry support, fix tx status reporting Clean up the tx status reporting, fix retry counters (short retries are virtual collisions, not actual retries). Implement multi-rate retry support. This also fixes strong throughput fluctuations with rc80211_pid Signed-off-by: Felix Fietkau <nbd@openwrt.org> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-10-06 00:05:48 +08:00
ts->ts_longretry += ts->ts_retry[0];
/* fall through */
case 0:
ts->ts_rate[0] = tx_ctl->tx_control_3 &
AR5K_4W_TX_DESC_CTL3_XMIT_RATE0;
ath5k: HW code cleanup * No code changes... * Split hw.c to multiple files for better maintenance and add some documentation on each file code is going to grow soon (eeprom.c for example is going to get much stuff currently developed on ath_info) so it's better this way. * Rename following functions to maintain naming scheme: ah_setup_xtx_desc -> ah_setup_mrr_tx_desc (Because xtx doesn't say much, it's actually a multi-rate-retry tx descriptor) ath5k_hw_put_tx/rx_buf - > ath5k_hw_set_tx/rxdp ath5k_hw_get_tx/rx_buf -> ath5k_hw_get_tx/rxdp (We don't put any "buf" we set descriptor pointers on hw) ath5k_hw_tx_start -> ath5k_hw_start_tx_dma ath5k_hw_start_rx -> ath5k_hw_start_rx_dma ath5k_hw_stop_pcu_recv -> ath5k_hw_stop_rx_pcu (It's easier this way to identify them, we also have ath5k_hw_start_rx_pcu which completes the set) ath5k_hw_set_intr -> ath5k_hw_set_imr (As in get_isr we set imr here, not "intr") * Move ath5k_hw_setup_rx_desc on ah->ah_setup_rx_desc so we can include support for different rx descriptors in the future * Further cleanups so that checkpatch doesn't complain (only some > 80 col warnings for eeprom.h and reg.h as usual due to comments) Tested on 5211 and 5213 cards and works ok. Changes-licensed-under: ISC Signed-off-by: Nick Kossifidis <mickflemm@gmail.com> Acked-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-08-30 03:45:39 +08:00
break;
}
/* TX error */
if (!(tx_status->tx_status_0 & AR5K_DESC_TX_STATUS0_FRAME_XMIT_OK)) {
if (tx_status->tx_status_0 &
AR5K_DESC_TX_STATUS0_EXCESSIVE_RETRIES)
ts->ts_status |= AR5K_TXERR_XRETRY;
if (tx_status->tx_status_0 & AR5K_DESC_TX_STATUS0_FIFO_UNDERRUN)
ts->ts_status |= AR5K_TXERR_FIFO;
if (tx_status->tx_status_0 & AR5K_DESC_TX_STATUS0_FILTERED)
ts->ts_status |= AR5K_TXERR_FILT;
}
return 0;
}
/*
* RX Descriptors
*/
/*
* 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)
{
struct ath5k_hw_rx_ctl *rx_ctl;
ATH5K_TRACE(ah->ah_sc);
rx_ctl = &desc->ud.ds_rx.rx_ctl;
/*
* Clear the descriptor
* If we don't clean the status descriptor,
* while scanning we get too many results,
* most of them virtual, after some secs
* of scanning system hangs. M.F.
*/
memset(&desc->ud.ds_rx, 0, sizeof(struct ath5k_hw_all_rx_desc));
/* 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;
return 0;
}
/*
* Proccess the rx status descriptor on 5210/5211
*/
static int ath5k_hw_proc_5210_rx_status(struct ath5k_hw *ah,
struct ath5k_desc *desc, struct ath5k_rx_status *rs)
{
struct ath5k_hw_rx_status *rx_status;
rx_status = &desc->ud.ds_rx.u.rx_stat;
/* No frame received / not ready */
if (unlikely(!(rx_status->rx_status_1 &
AR5K_5210_RX_DESC_STATUS1_DONE)))
return -EINPROGRESS;
/*
* Frame receive status
*/
rs->rs_datalen = rx_status->rx_status_0 &
AR5K_5210_RX_DESC_STATUS0_DATA_LEN;
rs->rs_rssi = AR5K_REG_MS(rx_status->rx_status_0,
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);
ath5k: correct handling of rx status fields ath5k_rx_status fields rs_antenna and rs_more are u8s, but we were setting them with bitwise ANDs of 32-bit values. As a consequence, jumbo frames would not be discarded as intended. Then, because the hw rate value of such frames is zero, and, since "ath5k: rates cleanup", we do not fall back to the basic rate, such packets would trigger the following WARN_ON: ------------[ cut here ]------------ WARNING: at net/mac80211/rx.c:2192 __ieee80211_rx+0x4d/0x57e [mac80211]() Modules linked in: ath5k af_packet sha256_generic aes_i586 aes_generic cbc loop i915 drm binfmt_misc acpi_cpufreq fan container nls_utf8 hfsplus dm_crypt dm_mod kvm_intel kvm fuse sbp2 snd_hda_intel snd_pcm_oss snd_pcm snd_mixer_oss snd_seq_dummy snd_seq_oss arc4 joydev hid_apple ecb snd_seq_midi snd_rawmidi snd_seq_midi_event snd_seq snd_timer snd_seq_device usbhid appletouch mac80211 sky2 snd ehci_hcd ohci1394 bitrev crc32 sr_mod cdrom rtc sg uhci_hcd snd_page_alloc cfg80211 ieee1394 thermal ac battery processor button evdev unix [last unloaded: ath5k] Pid: 0, comm: swapper Tainted: G W 2.6.28-rc2-wl #14 Call Trace: [<c0123d1e>] warn_on_slowpath+0x41/0x5b [<c012005d>] ? sched_debug_show+0x31e/0x9c6 [<c012489f>] ? vprintk+0x369/0x389 [<c0309539>] ? _spin_unlock_irqrestore+0x54/0x58 [<c011cd8f>] ? try_to_wake_up+0x14f/0x15a [<f81918cb>] __ieee80211_rx+0x4d/0x57e [mac80211] [<f828872a>] ath5k_tasklet_rx+0x5a1/0x5e4 [ath5k] [<c013b9cd>] ? clockevents_program_event+0xd4/0xe3 [<c01283a9>] tasklet_action+0x94/0xfd [<c0127d19>] __do_softirq+0x8c/0x13e [<c0127e04>] do_softirq+0x39/0x55 [<c0128082>] irq_exit+0x46/0x85 [<c010576c>] do_IRQ+0x9a/0xb2 [<c010461c>] common_interrupt+0x28/0x30 [<f80e934a>] ? acpi_idle_enter_bm+0x2ad/0x31b [processor] [<c02976bf>] cpuidle_idle_call+0x65/0x9a [<c010262c>] cpu_idle+0x76/0xa6 [<c02fb402>] rest_init+0x62/0x64 Signed-off-by: Bob Copeland <me@bobcopeland.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-11-04 11:14:00 +08:00
rs->rs_antenna = AR5K_REG_MS(rx_status->rx_status_0,
AR5K_5210_RX_DESC_STATUS0_RECEIVE_ANTENNA);
rs->rs_more = !!(rx_status->rx_status_0 &
AR5K_5210_RX_DESC_STATUS0_MORE);
ath5k: HW code cleanup * No code changes... * Split hw.c to multiple files for better maintenance and add some documentation on each file code is going to grow soon (eeprom.c for example is going to get much stuff currently developed on ath_info) so it's better this way. * Rename following functions to maintain naming scheme: ah_setup_xtx_desc -> ah_setup_mrr_tx_desc (Because xtx doesn't say much, it's actually a multi-rate-retry tx descriptor) ath5k_hw_put_tx/rx_buf - > ath5k_hw_set_tx/rxdp ath5k_hw_get_tx/rx_buf -> ath5k_hw_get_tx/rxdp (We don't put any "buf" we set descriptor pointers on hw) ath5k_hw_tx_start -> ath5k_hw_start_tx_dma ath5k_hw_start_rx -> ath5k_hw_start_rx_dma ath5k_hw_stop_pcu_recv -> ath5k_hw_stop_rx_pcu (It's easier this way to identify them, we also have ath5k_hw_start_rx_pcu which completes the set) ath5k_hw_set_intr -> ath5k_hw_set_imr (As in get_isr we set imr here, not "intr") * Move ath5k_hw_setup_rx_desc on ah->ah_setup_rx_desc so we can include support for different rx descriptors in the future * Further cleanups so that checkpatch doesn't complain (only some > 80 col warnings for eeprom.h and reg.h as usual due to comments) Tested on 5211 and 5213 cards and works ok. Changes-licensed-under: ISC Signed-off-by: Nick Kossifidis <mickflemm@gmail.com> Acked-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-08-30 03:45:39 +08:00
/* 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);
rs->rs_status = 0;
rs->rs_phyerr = 0;
/*
* Key table status
*/
if (rx_status->rx_status_1 & AR5K_5210_RX_DESC_STATUS1_KEY_INDEX_VALID)
rs->rs_keyix = AR5K_REG_MS(rx_status->rx_status_1,
AR5K_5210_RX_DESC_STATUS1_KEY_INDEX);
else
rs->rs_keyix = AR5K_RXKEYIX_INVALID;
/*
* Receive/descriptor errors
*/
if (!(rx_status->rx_status_1 &
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)
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,
AR5K_5210_RX_DESC_STATUS1_PHY_ERROR);
}
if (rx_status->rx_status_1 &
AR5K_5210_RX_DESC_STATUS1_DECRYPT_CRC_ERROR)
rs->rs_status |= AR5K_RXERR_DECRYPT;
}
return 0;
}
/*
* 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_hw_rx_status *rx_status;
struct ath5k_hw_rx_error *rx_err;
ATH5K_TRACE(ah->ah_sc);
rx_status = &desc->ud.ds_rx.u.rx_stat;
/* Overlay on error */
rx_err = &desc->ud.ds_rx.u.rx_err;
/* No frame received / not ready */
if (unlikely(!(rx_status->rx_status_1 &
AR5K_5212_RX_DESC_STATUS1_DONE)))
return -EINPROGRESS;
/*
* Frame receive status
*/
rs->rs_datalen = rx_status->rx_status_0 &
AR5K_5212_RX_DESC_STATUS0_DATA_LEN;
rs->rs_rssi = AR5K_REG_MS(rx_status->rx_status_0,
AR5K_5212_RX_DESC_STATUS0_RECEIVE_SIGNAL);
rs->rs_rate = AR5K_REG_MS(rx_status->rx_status_0,
AR5K_5212_RX_DESC_STATUS0_RECEIVE_RATE);
ath5k: correct handling of rx status fields ath5k_rx_status fields rs_antenna and rs_more are u8s, but we were setting them with bitwise ANDs of 32-bit values. As a consequence, jumbo frames would not be discarded as intended. Then, because the hw rate value of such frames is zero, and, since "ath5k: rates cleanup", we do not fall back to the basic rate, such packets would trigger the following WARN_ON: ------------[ cut here ]------------ WARNING: at net/mac80211/rx.c:2192 __ieee80211_rx+0x4d/0x57e [mac80211]() Modules linked in: ath5k af_packet sha256_generic aes_i586 aes_generic cbc loop i915 drm binfmt_misc acpi_cpufreq fan container nls_utf8 hfsplus dm_crypt dm_mod kvm_intel kvm fuse sbp2 snd_hda_intel snd_pcm_oss snd_pcm snd_mixer_oss snd_seq_dummy snd_seq_oss arc4 joydev hid_apple ecb snd_seq_midi snd_rawmidi snd_seq_midi_event snd_seq snd_timer snd_seq_device usbhid appletouch mac80211 sky2 snd ehci_hcd ohci1394 bitrev crc32 sr_mod cdrom rtc sg uhci_hcd snd_page_alloc cfg80211 ieee1394 thermal ac battery processor button evdev unix [last unloaded: ath5k] Pid: 0, comm: swapper Tainted: G W 2.6.28-rc2-wl #14 Call Trace: [<c0123d1e>] warn_on_slowpath+0x41/0x5b [<c012005d>] ? sched_debug_show+0x31e/0x9c6 [<c012489f>] ? vprintk+0x369/0x389 [<c0309539>] ? _spin_unlock_irqrestore+0x54/0x58 [<c011cd8f>] ? try_to_wake_up+0x14f/0x15a [<f81918cb>] __ieee80211_rx+0x4d/0x57e [mac80211] [<f828872a>] ath5k_tasklet_rx+0x5a1/0x5e4 [ath5k] [<c013b9cd>] ? clockevents_program_event+0xd4/0xe3 [<c01283a9>] tasklet_action+0x94/0xfd [<c0127d19>] __do_softirq+0x8c/0x13e [<c0127e04>] do_softirq+0x39/0x55 [<c0128082>] irq_exit+0x46/0x85 [<c010576c>] do_IRQ+0x9a/0xb2 [<c010461c>] common_interrupt+0x28/0x30 [<f80e934a>] ? acpi_idle_enter_bm+0x2ad/0x31b [processor] [<c02976bf>] cpuidle_idle_call+0x65/0x9a [<c010262c>] cpu_idle+0x76/0xa6 [<c02fb402>] rest_init+0x62/0x64 Signed-off-by: Bob Copeland <me@bobcopeland.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-11-04 11:14:00 +08:00
rs->rs_antenna = AR5K_REG_MS(rx_status->rx_status_0,
AR5K_5212_RX_DESC_STATUS0_RECEIVE_ANTENNA);
rs->rs_more = !!(rx_status->rx_status_0 &
AR5K_5212_RX_DESC_STATUS0_MORE);
ath5k: HW code cleanup * No code changes... * Split hw.c to multiple files for better maintenance and add some documentation on each file code is going to grow soon (eeprom.c for example is going to get much stuff currently developed on ath_info) so it's better this way. * Rename following functions to maintain naming scheme: ah_setup_xtx_desc -> ah_setup_mrr_tx_desc (Because xtx doesn't say much, it's actually a multi-rate-retry tx descriptor) ath5k_hw_put_tx/rx_buf - > ath5k_hw_set_tx/rxdp ath5k_hw_get_tx/rx_buf -> ath5k_hw_get_tx/rxdp (We don't put any "buf" we set descriptor pointers on hw) ath5k_hw_tx_start -> ath5k_hw_start_tx_dma ath5k_hw_start_rx -> ath5k_hw_start_rx_dma ath5k_hw_stop_pcu_recv -> ath5k_hw_stop_rx_pcu (It's easier this way to identify them, we also have ath5k_hw_start_rx_pcu which completes the set) ath5k_hw_set_intr -> ath5k_hw_set_imr (As in get_isr we set imr here, not "intr") * Move ath5k_hw_setup_rx_desc on ah->ah_setup_rx_desc so we can include support for different rx descriptors in the future * Further cleanups so that checkpatch doesn't complain (only some > 80 col warnings for eeprom.h and reg.h as usual due to comments) Tested on 5211 and 5213 cards and works ok. Changes-licensed-under: ISC Signed-off-by: Nick Kossifidis <mickflemm@gmail.com> Acked-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-08-30 03:45:39 +08:00
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
*/
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);
else
rs->rs_keyix = AR5K_RXKEYIX_INVALID;
/*
* Receive/descriptor errors
*/
if (!(rx_status->rx_status_1 &
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;
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);
}
if (rx_status->rx_status_1 &
AR5K_5212_RX_DESC_STATUS1_DECRYPT_CRC_ERROR)
rs->rs_status |= AR5K_RXERR_DECRYPT;
if (rx_status->rx_status_1 &
AR5K_5212_RX_DESC_STATUS1_MIC_ERROR)
rs->rs_status |= AR5K_RXERR_MIC;
}
return 0;
}
/*
* Init function pointers inside ath5k_hw struct
*/
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_setup_tx_desc = ath5k_hw_setup_2word_tx_desc;
ath5k: implement multi-rate retry support, fix tx status reporting Clean up the tx status reporting, fix retry counters (short retries are virtual collisions, not actual retries). Implement multi-rate retry support. This also fixes strong throughput fluctuations with rc80211_pid Signed-off-by: Felix Fietkau <nbd@openwrt.org> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-10-06 00:05:48 +08:00
ah->ah_setup_mrr_tx_desc = ath5k_hw_setup_no_mrr;
ath5k: HW code cleanup * No code changes... * Split hw.c to multiple files for better maintenance and add some documentation on each file code is going to grow soon (eeprom.c for example is going to get much stuff currently developed on ath_info) so it's better this way. * Rename following functions to maintain naming scheme: ah_setup_xtx_desc -> ah_setup_mrr_tx_desc (Because xtx doesn't say much, it's actually a multi-rate-retry tx descriptor) ath5k_hw_put_tx/rx_buf - > ath5k_hw_set_tx/rxdp ath5k_hw_get_tx/rx_buf -> ath5k_hw_get_tx/rxdp (We don't put any "buf" we set descriptor pointers on hw) ath5k_hw_tx_start -> ath5k_hw_start_tx_dma ath5k_hw_start_rx -> ath5k_hw_start_rx_dma ath5k_hw_stop_pcu_recv -> ath5k_hw_stop_rx_pcu (It's easier this way to identify them, we also have ath5k_hw_start_rx_pcu which completes the set) ath5k_hw_set_intr -> ath5k_hw_set_imr (As in get_isr we set imr here, not "intr") * Move ath5k_hw_setup_rx_desc on ah->ah_setup_rx_desc so we can include support for different rx descriptors in the future * Further cleanups so that checkpatch doesn't complain (only some > 80 col warnings for eeprom.h and reg.h as usual due to comments) Tested on 5211 and 5213 cards and works ok. Changes-licensed-under: ISC Signed-off-by: Nick Kossifidis <mickflemm@gmail.com> Acked-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-08-30 03:45:39 +08:00
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;
return 0;
}