linux/drivers/net/wireless/ath/ath6kl/htc.h

616 lines
15 KiB
C
Raw Normal View History

Add ath6kl cleaned up driver Last May we started working on cleaning up ath6kl driver which is currently in staging. The work has happened in a separate ath6kl-cleanup tree: http://git.kernel.org/?p=linux/kernel/git/kvalo/ath6kl-cleanup.git;a=summary After over 1100 (!) patches we have now reached a state where I would like to start discussing about pushing the driver to the wireless trees and replacing the staging driver. The driver is now a lot smaller and looks like a proper Linux driver. The size of the driver (measured with simple wc -l) dropped from 49 kLOC to 18 kLOC and the number of the .c and .h files dropped from 107 to 22. Most importantly the number of subdirectories reduced from 26 to zero :) There are two remaining checkpatch warnings in the driver which we decided to omit for now: drivers/net/wireless/ath/ath6kl/debug.c:31: WARNING: printk() should include KERN_ facility level drivers/net/wireless/ath/ath6kl/sdio.c:527: WARNING: msleep < 20ms can sleep for up to 20ms; see Documentation/timers/timers-howto.txt The driver has endian annotations for all the hardware specific structures and there are no sparse errors. Unfortunately I don't have any big endian hardware to test that right now. We have been testing the driver both on x86 and arm platforms. The code is also compiled with sparc and parisc cross compilers. Notable missing features compared to the current staging driver are: o HCI over SDIO support o nl80211 testmode o firmware logging o suspend support Testmode, firmware logging and suspend support will be added soon. HCI over SDIO support will be more difficult as the HCI driver needs to share code with the wifi driver. This is something we need to research more. Also I want to point out the changes I did for signed endian support. As I wasn't able to find any support for signed endian annotations I decided to follow what NTFS has done and added my own. Grep for sle16 and sle32, especially from wmi.h. Various people have been working on the cleanup, the hall of fame based on number of patches is: 543 Vasanthakumar Thiagarajan 403 Raja Mani 252 Kalle Valo 16 Vivek Natarajan 12 Suraj Sumangala 3 Joe Perches 2 Jouni Malinen Signed-off-by: Vasanthakumar Thiagarajan <vthiagar@qca.qualcomm.com> Signed-off-by: Raja Mani <rmani@qca.qualcomm.com> Signed-off-by: Vivek Natarajan <nataraja@qca.qualcomm.com> Signed-off-by: Suraj Sumangala <surajs@qca.qualcomm.com> Signed-off-by: Joe Perches <joe@perches.com> Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
2011-07-18 05:22:30 +08:00
/*
* Copyright (c) 2004-2011 Atheros Communications Inc.
*
* Permission to use, copy, modify, and/or distribute this software for any
* purpose with or without fee is hereby granted, provided that the above
* copyright notice and this permission notice appear in all copies.
*
* THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL WARRANTIES
* WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF
* MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR
* ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES
* WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN
* ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF
* OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
*/
#ifndef HTC_H
#define HTC_H
#include "common.h"
/* frame header flags */
/* send direction */
#define HTC_FLAGS_NEED_CREDIT_UPDATE (1 << 0)
#define HTC_FLAGS_SEND_BUNDLE (1 << 1)
/* receive direction */
#define HTC_FLG_RX_UNUSED (1 << 0)
#define HTC_FLG_RX_TRAILER (1 << 1)
/* Bundle count maske and shift */
#define HTC_FLG_RX_BNDL_CNT (0xF0)
#define HTC_FLG_RX_BNDL_CNT_S 4
#define HTC_HDR_LENGTH (sizeof(struct htc_frame_hdr))
#define HTC_MAX_PAYLOAD_LENGTH (4096 - sizeof(struct htc_frame_hdr))
/* HTC control message IDs */
#define HTC_MSG_READY_ID 1
#define HTC_MSG_CONN_SVC_ID 2
#define HTC_MSG_CONN_SVC_RESP_ID 3
#define HTC_MSG_SETUP_COMPLETE_ID 4
#define HTC_MSG_SETUP_COMPLETE_EX_ID 5
#define HTC_MAX_CTRL_MSG_LEN 256
#define HTC_VERSION_2P0 0x00
#define HTC_VERSION_2P1 0x01
#define HTC_SERVICE_META_DATA_MAX_LENGTH 128
#define HTC_CONN_FLGS_THRESH_LVL_QUAT 0x0
#define HTC_CONN_FLGS_THRESH_LVL_HALF 0x1
#define HTC_CONN_FLGS_THRESH_LVL_THREE_QUAT 0x2
#define HTC_CONN_FLGS_REDUCE_CRED_DRIB 0x4
#define HTC_CONN_FLGS_THRESH_MASK 0x3
/* connect response status codes */
#define HTC_SERVICE_SUCCESS 0
#define HTC_SERVICE_NOT_FOUND 1
#define HTC_SERVICE_FAILED 2
/* no resources (i.e. no more endpoints) */
#define HTC_SERVICE_NO_RESOURCES 3
/* specific service is not allowing any more endpoints */
#define HTC_SERVICE_NO_MORE_EP 4
/* report record IDs */
#define HTC_RECORD_NULL 0
#define HTC_RECORD_CREDITS 1
#define HTC_RECORD_LOOKAHEAD 2
#define HTC_RECORD_LOOKAHEAD_BUNDLE 3
#define HTC_SETUP_COMP_FLG_RX_BNDL_EN (1 << 0)
#define MAKE_SERVICE_ID(group, index) \
(int)(((int)group << 8) | (int)(index))
/* NOTE: service ID of 0x0000 is reserved and should never be used */
#define HTC_CTRL_RSVD_SVC MAKE_SERVICE_ID(RSVD_SERVICE_GROUP, 1)
#define WMI_CONTROL_SVC MAKE_SERVICE_ID(WMI_SERVICE_GROUP, 0)
#define WMI_DATA_BE_SVC MAKE_SERVICE_ID(WMI_SERVICE_GROUP, 1)
#define WMI_DATA_BK_SVC MAKE_SERVICE_ID(WMI_SERVICE_GROUP, 2)
#define WMI_DATA_VI_SVC MAKE_SERVICE_ID(WMI_SERVICE_GROUP, 3)
#define WMI_DATA_VO_SVC MAKE_SERVICE_ID(WMI_SERVICE_GROUP, 4)
#define WMI_MAX_SERVICES 5
/* reserved and used to flush ALL packets */
#define HTC_TX_PACKET_TAG_ALL 0
#define HTC_SERVICE_TX_PACKET_TAG 1
#define HTC_TX_PACKET_TAG_USER_DEFINED (HTC_SERVICE_TX_PACKET_TAG + 9)
/* more packets on this endpoint are being fetched */
#define HTC_RX_FLAGS_INDICATE_MORE_PKTS (1 << 0)
/* TODO.. for BMI */
#define ENDPOINT1 0
/* TODO -remove me, but we have to fix BMI first */
#define HTC_MAILBOX_NUM_MAX 4
/* enable send bundle padding for this endpoint */
#define HTC_FLGS_TX_BNDL_PAD_EN (1 << 0)
#define HTC_EP_ACTIVE ((u32) (1u << 31))
/* HTC operational parameters */
#define HTC_TARGET_RESPONSE_TIMEOUT 2000 /* in ms */
#define HTC_TARGET_DEBUG_INTR_MASK 0x01
#define HTC_TARGET_CREDIT_INTR_MASK 0xF0
#define HTC_HOST_MAX_MSG_PER_BUNDLE 8
#define HTC_MIN_HTC_MSGS_TO_BUNDLE 2
/* packet flags */
#define HTC_RX_PKT_IGNORE_LOOKAHEAD (1 << 0)
#define HTC_RX_PKT_REFRESH_HDR (1 << 1)
#define HTC_RX_PKT_PART_OF_BUNDLE (1 << 2)
#define HTC_RX_PKT_NO_RECYCLE (1 << 3)
#define NUM_CONTROL_BUFFERS 8
#define NUM_CONTROL_TX_BUFFERS 2
#define NUM_CONTROL_RX_BUFFERS (NUM_CONTROL_BUFFERS - NUM_CONTROL_TX_BUFFERS)
#define HTC_RECV_WAIT_BUFFERS (1 << 0)
#define HTC_OP_STATE_STOPPING (1 << 0)
/*
* The frame header length and message formats defined herein were selected
* to accommodate optimal alignment for target processing. This reduces
* code size and improves performance. Any changes to the header length may
* alter the alignment and cause exceptions on the target. When adding to
* the messagestructures insure that fields are properly aligned.
*/
/* HTC frame header
*
* NOTE: do not remove or re-arrange the fields, these are minimally
* required to take advantage of 4-byte lookaheads in some hardware
* implementations.
*/
struct htc_frame_hdr {
u8 eid;
u8 flags;
/* length of data (including trailer) that follows the header */
__le16 payld_len;
/* end of 4-byte lookahead */
u8 ctrl[2];
} __packed;
/* HTC ready message */
struct htc_ready_msg {
__le16 msg_id;
__le16 cred_cnt;
__le16 cred_sz;
u8 max_ep;
u8 pad;
} __packed;
/* extended HTC ready message */
struct htc_ready_ext_msg {
struct htc_ready_msg ver2_0_info;
u8 htc_ver;
u8 msg_per_htc_bndl;
} __packed;
/* connect service */
struct htc_conn_service_msg {
__le16 msg_id;
__le16 svc_id;
__le16 conn_flags;
u8 svc_meta_len;
u8 pad;
} __packed;
/* connect response */
struct htc_conn_service_resp {
__le16 msg_id;
__le16 svc_id;
u8 status;
u8 eid;
__le16 max_msg_sz;
u8 svc_meta_len;
u8 pad;
} __packed;
struct htc_setup_comp_msg {
__le16 msg_id;
} __packed;
/* extended setup completion message */
struct htc_setup_comp_ext_msg {
__le16 msg_id;
__le32 flags;
u8 msg_per_rxbndl;
u8 Rsvd[3];
} __packed;
struct htc_record_hdr {
u8 rec_id;
u8 len;
} __packed;
struct htc_credit_report {
u8 eid;
u8 credits;
} __packed;
/*
* NOTE: The lk_ahd array is guarded by a pre_valid
* and Post Valid guard bytes. The pre_valid bytes must
* equal the inverse of the post_valid byte.
*/
struct htc_lookahead_report {
u8 pre_valid;
u8 lk_ahd[4];
u8 post_valid;
} __packed;
struct htc_bundle_lkahd_rpt {
u8 lk_ahd[4];
} __packed;
/* Current service IDs */
enum htc_service_grp_ids {
RSVD_SERVICE_GROUP = 0,
WMI_SERVICE_GROUP = 1,
HTC_TEST_GROUP = 254,
HTC_SERVICE_GROUP_LAST = 255
};
/* ------ endpoint IDS ------ */
enum htc_endpoint_id {
ENDPOINT_UNUSED = -1,
ENDPOINT_0 = 0,
ENDPOINT_1 = 1,
ENDPOINT_2 = 2,
ENDPOINT_3,
ENDPOINT_4,
ENDPOINT_5,
ENDPOINT_6,
ENDPOINT_7,
ENDPOINT_8,
ENDPOINT_MAX,
};
struct htc_tx_packet_info {
u16 tag;
int cred_used;
u8 flags;
int seqno;
};
struct htc_rx_packet_info {
u32 exp_hdr;
u32 rx_flags;
u32 indicat_flags;
};
struct htc_target;
/* wrapper around endpoint-specific packets */
struct htc_packet {
struct list_head list;
/* caller's per packet specific context */
void *pkt_cntxt;
/*
* the true buffer start , the caller can store the real
* buffer start here. In receive callbacks, the HTC layer
* sets buf to the start of the payload past the header.
* This field allows the caller to reset buf when it recycles
* receive packets back to HTC.
*/
u8 *buf_start;
/*
* Pointer to the start of the buffer. In the transmit
* direction this points to the start of the payload. In the
* receive direction, however, the buffer when queued up
* points to the start of the HTC header but when returned
* to the caller points to the start of the payload
*/
u8 *buf;
u32 buf_len;
/* actual length of payload */
u32 act_len;
/* endpoint that this packet was sent/recv'd from */
enum htc_endpoint_id endpoint;
/* completion status */
int status;
union {
struct htc_tx_packet_info tx;
struct htc_rx_packet_info rx;
} info;
void (*completion) (struct htc_target *, struct htc_packet *);
struct htc_target *context;
};
enum htc_send_full_action {
HTC_SEND_FULL_KEEP = 0,
HTC_SEND_FULL_DROP = 1,
};
struct htc_ep_callbacks {
void (*rx) (struct htc_target *, struct htc_packet *);
void (*rx_refill) (struct htc_target *, enum htc_endpoint_id endpoint);
enum htc_send_full_action (*tx_full) (struct htc_target *,
struct htc_packet *);
struct htc_packet *(*rx_allocthresh) (struct htc_target *,
enum htc_endpoint_id, int);
int rx_alloc_thresh;
int rx_refill_thresh;
};
/* service connection information */
struct htc_service_connect_req {
u16 svc_id;
u16 conn_flags;
struct htc_ep_callbacks ep_cb;
int max_txq_depth;
u32 flags;
unsigned int max_rxmsg_sz;
};
/* service connection response information */
struct htc_service_connect_resp {
u8 buf_len;
u8 act_len;
enum htc_endpoint_id endpoint;
unsigned int len_max;
u8 resp_code;
};
/* endpoint distributionstructure */
struct htc_endpoint_credit_dist {
struct list_head list;
/* Service ID (set by HTC) */
u16 svc_id;
/* endpoint for this distributionstruct (set by HTC) */
enum htc_endpoint_id endpoint;
u32 dist_flags;
/*
* credits for normal operation, anything above this
* indicates the endpoint is over-subscribed.
*/
int cred_norm;
/* floor for credit distribution */
int cred_min;
int cred_assngd;
/* current credits available */
int credits;
/*
* pending credits to distribute on this endpoint, this
* is set by HTC when credit reports arrive. The credit
* distribution functions sets this to zero when it distributes
* the credits.
*/
int cred_to_dist;
/*
* the number of credits that the current pending TX packet needs
* to transmit. This is set by HTC when endpoint needs credits in
* order to transmit.
*/
int seek_cred;
/* size in bytes of each credit */
int cred_sz;
/* credits required for a maximum sized messages */
int cred_per_msg;
/* reserved for HTC use */
struct htc_endpoint *htc_ep;
Add ath6kl cleaned up driver Last May we started working on cleaning up ath6kl driver which is currently in staging. The work has happened in a separate ath6kl-cleanup tree: http://git.kernel.org/?p=linux/kernel/git/kvalo/ath6kl-cleanup.git;a=summary After over 1100 (!) patches we have now reached a state where I would like to start discussing about pushing the driver to the wireless trees and replacing the staging driver. The driver is now a lot smaller and looks like a proper Linux driver. The size of the driver (measured with simple wc -l) dropped from 49 kLOC to 18 kLOC and the number of the .c and .h files dropped from 107 to 22. Most importantly the number of subdirectories reduced from 26 to zero :) There are two remaining checkpatch warnings in the driver which we decided to omit for now: drivers/net/wireless/ath/ath6kl/debug.c:31: WARNING: printk() should include KERN_ facility level drivers/net/wireless/ath/ath6kl/sdio.c:527: WARNING: msleep < 20ms can sleep for up to 20ms; see Documentation/timers/timers-howto.txt The driver has endian annotations for all the hardware specific structures and there are no sparse errors. Unfortunately I don't have any big endian hardware to test that right now. We have been testing the driver both on x86 and arm platforms. The code is also compiled with sparc and parisc cross compilers. Notable missing features compared to the current staging driver are: o HCI over SDIO support o nl80211 testmode o firmware logging o suspend support Testmode, firmware logging and suspend support will be added soon. HCI over SDIO support will be more difficult as the HCI driver needs to share code with the wifi driver. This is something we need to research more. Also I want to point out the changes I did for signed endian support. As I wasn't able to find any support for signed endian annotations I decided to follow what NTFS has done and added my own. Grep for sle16 and sle32, especially from wmi.h. Various people have been working on the cleanup, the hall of fame based on number of patches is: 543 Vasanthakumar Thiagarajan 403 Raja Mani 252 Kalle Valo 16 Vivek Natarajan 12 Suraj Sumangala 3 Joe Perches 2 Jouni Malinen Signed-off-by: Vasanthakumar Thiagarajan <vthiagar@qca.qualcomm.com> Signed-off-by: Raja Mani <rmani@qca.qualcomm.com> Signed-off-by: Vivek Natarajan <nataraja@qca.qualcomm.com> Signed-off-by: Suraj Sumangala <surajs@qca.qualcomm.com> Signed-off-by: Joe Perches <joe@perches.com> Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
2011-07-18 05:22:30 +08:00
/*
* current depth of TX queue , i.e. messages waiting for credits
* This field is valid only when HTC_CREDIT_DIST_ACTIVITY_CHANGE
* or HTC_CREDIT_DIST_SEND_COMPLETE is indicated on an endpoint
* that has non-zero credits to recover.
*/
int txq_depth;
};
/*
* credit distibution code that is passed into the distrbution function,
* there are mandatory and optional codes that must be handled
*/
enum htc_credit_dist_reason {
HTC_CREDIT_DIST_SEND_COMPLETE = 0,
HTC_CREDIT_DIST_ACTIVITY_CHANGE = 1,
HTC_CREDIT_DIST_SEEK_CREDITS,
};
struct ath6kl_htc_credit_info {
Add ath6kl cleaned up driver Last May we started working on cleaning up ath6kl driver which is currently in staging. The work has happened in a separate ath6kl-cleanup tree: http://git.kernel.org/?p=linux/kernel/git/kvalo/ath6kl-cleanup.git;a=summary After over 1100 (!) patches we have now reached a state where I would like to start discussing about pushing the driver to the wireless trees and replacing the staging driver. The driver is now a lot smaller and looks like a proper Linux driver. The size of the driver (measured with simple wc -l) dropped from 49 kLOC to 18 kLOC and the number of the .c and .h files dropped from 107 to 22. Most importantly the number of subdirectories reduced from 26 to zero :) There are two remaining checkpatch warnings in the driver which we decided to omit for now: drivers/net/wireless/ath/ath6kl/debug.c:31: WARNING: printk() should include KERN_ facility level drivers/net/wireless/ath/ath6kl/sdio.c:527: WARNING: msleep < 20ms can sleep for up to 20ms; see Documentation/timers/timers-howto.txt The driver has endian annotations for all the hardware specific structures and there are no sparse errors. Unfortunately I don't have any big endian hardware to test that right now. We have been testing the driver both on x86 and arm platforms. The code is also compiled with sparc and parisc cross compilers. Notable missing features compared to the current staging driver are: o HCI over SDIO support o nl80211 testmode o firmware logging o suspend support Testmode, firmware logging and suspend support will be added soon. HCI over SDIO support will be more difficult as the HCI driver needs to share code with the wifi driver. This is something we need to research more. Also I want to point out the changes I did for signed endian support. As I wasn't able to find any support for signed endian annotations I decided to follow what NTFS has done and added my own. Grep for sle16 and sle32, especially from wmi.h. Various people have been working on the cleanup, the hall of fame based on number of patches is: 543 Vasanthakumar Thiagarajan 403 Raja Mani 252 Kalle Valo 16 Vivek Natarajan 12 Suraj Sumangala 3 Joe Perches 2 Jouni Malinen Signed-off-by: Vasanthakumar Thiagarajan <vthiagar@qca.qualcomm.com> Signed-off-by: Raja Mani <rmani@qca.qualcomm.com> Signed-off-by: Vivek Natarajan <nataraja@qca.qualcomm.com> Signed-off-by: Suraj Sumangala <surajs@qca.qualcomm.com> Signed-off-by: Joe Perches <joe@perches.com> Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
2011-07-18 05:22:30 +08:00
int total_avail_credits;
int cur_free_credits;
/* list of lowest priority endpoints */
Add ath6kl cleaned up driver Last May we started working on cleaning up ath6kl driver which is currently in staging. The work has happened in a separate ath6kl-cleanup tree: http://git.kernel.org/?p=linux/kernel/git/kvalo/ath6kl-cleanup.git;a=summary After over 1100 (!) patches we have now reached a state where I would like to start discussing about pushing the driver to the wireless trees and replacing the staging driver. The driver is now a lot smaller and looks like a proper Linux driver. The size of the driver (measured with simple wc -l) dropped from 49 kLOC to 18 kLOC and the number of the .c and .h files dropped from 107 to 22. Most importantly the number of subdirectories reduced from 26 to zero :) There are two remaining checkpatch warnings in the driver which we decided to omit for now: drivers/net/wireless/ath/ath6kl/debug.c:31: WARNING: printk() should include KERN_ facility level drivers/net/wireless/ath/ath6kl/sdio.c:527: WARNING: msleep < 20ms can sleep for up to 20ms; see Documentation/timers/timers-howto.txt The driver has endian annotations for all the hardware specific structures and there are no sparse errors. Unfortunately I don't have any big endian hardware to test that right now. We have been testing the driver both on x86 and arm platforms. The code is also compiled with sparc and parisc cross compilers. Notable missing features compared to the current staging driver are: o HCI over SDIO support o nl80211 testmode o firmware logging o suspend support Testmode, firmware logging and suspend support will be added soon. HCI over SDIO support will be more difficult as the HCI driver needs to share code with the wifi driver. This is something we need to research more. Also I want to point out the changes I did for signed endian support. As I wasn't able to find any support for signed endian annotations I decided to follow what NTFS has done and added my own. Grep for sle16 and sle32, especially from wmi.h. Various people have been working on the cleanup, the hall of fame based on number of patches is: 543 Vasanthakumar Thiagarajan 403 Raja Mani 252 Kalle Valo 16 Vivek Natarajan 12 Suraj Sumangala 3 Joe Perches 2 Jouni Malinen Signed-off-by: Vasanthakumar Thiagarajan <vthiagar@qca.qualcomm.com> Signed-off-by: Raja Mani <rmani@qca.qualcomm.com> Signed-off-by: Vivek Natarajan <nataraja@qca.qualcomm.com> Signed-off-by: Suraj Sumangala <surajs@qca.qualcomm.com> Signed-off-by: Joe Perches <joe@perches.com> Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
2011-07-18 05:22:30 +08:00
struct list_head lowestpri_ep_dist;
};
/* endpoint statistics */
struct htc_endpoint_stats {
/*
* number of times the host set the credit-low flag in a send
* message on this endpoint
*/
u32 cred_low_indicate;
u32 tx_issued;
u32 tx_pkt_bundled;
u32 tx_bundles;
u32 tx_dropped;
/* running count of total credit reports received for this endpoint */
u32 tx_cred_rpt;
/* credit reports received from this endpoint's RX packets */
u32 cred_rpt_from_rx;
/* credit reports received from RX packets of other endpoints */
u32 cred_rpt_from_other;
/* credit reports received from endpoint 0 RX packets */
u32 cred_rpt_ep0;
/* count of credits received via Rx packets on this endpoint */
u32 cred_from_rx;
/* count of credits received via another endpoint */
u32 cred_from_other;
/* count of credits received via another endpoint */
u32 cred_from_ep0;
/* count of consummed credits */
u32 cred_cosumd;
/* count of credits returned */
u32 cred_retnd;
u32 rx_pkts;
/* count of lookahead records found in Rx msg */
u32 rx_lkahds;
/* count of recv packets received in a bundle */
u32 rx_bundl;
/* count of number of bundled lookaheads */
u32 rx_bundle_lkahd;
/* count of the number of bundle indications from the HTC header */
u32 rx_bundle_from_hdr;
/* the number of times the recv allocation threshold was hit */
u32 rx_alloc_thresh_hit;
/* total number of bytes */
u32 rxalloc_thresh_byte;
};
struct htc_endpoint {
enum htc_endpoint_id eid;
u16 svc_id;
struct list_head txq;
struct list_head rx_bufq;
struct htc_endpoint_credit_dist cred_dist;
struct htc_ep_callbacks ep_cb;
int max_txq_depth;
int len_max;
int tx_proc_cnt;
int rx_proc_cnt;
struct htc_target *target;
u8 seqno;
u32 conn_flags;
struct htc_endpoint_stats ep_st;
};
struct htc_control_buffer {
struct htc_packet packet;
u8 *buf;
};
struct ath6kl_device;
/* our HTC target state */
struct htc_target {
struct htc_endpoint endpoint[ENDPOINT_MAX];
/* contains struct htc_endpoint_credit_dist */
Add ath6kl cleaned up driver Last May we started working on cleaning up ath6kl driver which is currently in staging. The work has happened in a separate ath6kl-cleanup tree: http://git.kernel.org/?p=linux/kernel/git/kvalo/ath6kl-cleanup.git;a=summary After over 1100 (!) patches we have now reached a state where I would like to start discussing about pushing the driver to the wireless trees and replacing the staging driver. The driver is now a lot smaller and looks like a proper Linux driver. The size of the driver (measured with simple wc -l) dropped from 49 kLOC to 18 kLOC and the number of the .c and .h files dropped from 107 to 22. Most importantly the number of subdirectories reduced from 26 to zero :) There are two remaining checkpatch warnings in the driver which we decided to omit for now: drivers/net/wireless/ath/ath6kl/debug.c:31: WARNING: printk() should include KERN_ facility level drivers/net/wireless/ath/ath6kl/sdio.c:527: WARNING: msleep < 20ms can sleep for up to 20ms; see Documentation/timers/timers-howto.txt The driver has endian annotations for all the hardware specific structures and there are no sparse errors. Unfortunately I don't have any big endian hardware to test that right now. We have been testing the driver both on x86 and arm platforms. The code is also compiled with sparc and parisc cross compilers. Notable missing features compared to the current staging driver are: o HCI over SDIO support o nl80211 testmode o firmware logging o suspend support Testmode, firmware logging and suspend support will be added soon. HCI over SDIO support will be more difficult as the HCI driver needs to share code with the wifi driver. This is something we need to research more. Also I want to point out the changes I did for signed endian support. As I wasn't able to find any support for signed endian annotations I decided to follow what NTFS has done and added my own. Grep for sle16 and sle32, especially from wmi.h. Various people have been working on the cleanup, the hall of fame based on number of patches is: 543 Vasanthakumar Thiagarajan 403 Raja Mani 252 Kalle Valo 16 Vivek Natarajan 12 Suraj Sumangala 3 Joe Perches 2 Jouni Malinen Signed-off-by: Vasanthakumar Thiagarajan <vthiagar@qca.qualcomm.com> Signed-off-by: Raja Mani <rmani@qca.qualcomm.com> Signed-off-by: Vivek Natarajan <nataraja@qca.qualcomm.com> Signed-off-by: Suraj Sumangala <surajs@qca.qualcomm.com> Signed-off-by: Joe Perches <joe@perches.com> Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
2011-07-18 05:22:30 +08:00
struct list_head cred_dist_list;
Add ath6kl cleaned up driver Last May we started working on cleaning up ath6kl driver which is currently in staging. The work has happened in a separate ath6kl-cleanup tree: http://git.kernel.org/?p=linux/kernel/git/kvalo/ath6kl-cleanup.git;a=summary After over 1100 (!) patches we have now reached a state where I would like to start discussing about pushing the driver to the wireless trees and replacing the staging driver. The driver is now a lot smaller and looks like a proper Linux driver. The size of the driver (measured with simple wc -l) dropped from 49 kLOC to 18 kLOC and the number of the .c and .h files dropped from 107 to 22. Most importantly the number of subdirectories reduced from 26 to zero :) There are two remaining checkpatch warnings in the driver which we decided to omit for now: drivers/net/wireless/ath/ath6kl/debug.c:31: WARNING: printk() should include KERN_ facility level drivers/net/wireless/ath/ath6kl/sdio.c:527: WARNING: msleep < 20ms can sleep for up to 20ms; see Documentation/timers/timers-howto.txt The driver has endian annotations for all the hardware specific structures and there are no sparse errors. Unfortunately I don't have any big endian hardware to test that right now. We have been testing the driver both on x86 and arm platforms. The code is also compiled with sparc and parisc cross compilers. Notable missing features compared to the current staging driver are: o HCI over SDIO support o nl80211 testmode o firmware logging o suspend support Testmode, firmware logging and suspend support will be added soon. HCI over SDIO support will be more difficult as the HCI driver needs to share code with the wifi driver. This is something we need to research more. Also I want to point out the changes I did for signed endian support. As I wasn't able to find any support for signed endian annotations I decided to follow what NTFS has done and added my own. Grep for sle16 and sle32, especially from wmi.h. Various people have been working on the cleanup, the hall of fame based on number of patches is: 543 Vasanthakumar Thiagarajan 403 Raja Mani 252 Kalle Valo 16 Vivek Natarajan 12 Suraj Sumangala 3 Joe Perches 2 Jouni Malinen Signed-off-by: Vasanthakumar Thiagarajan <vthiagar@qca.qualcomm.com> Signed-off-by: Raja Mani <rmani@qca.qualcomm.com> Signed-off-by: Vivek Natarajan <nataraja@qca.qualcomm.com> Signed-off-by: Suraj Sumangala <surajs@qca.qualcomm.com> Signed-off-by: Joe Perches <joe@perches.com> Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
2011-07-18 05:22:30 +08:00
struct list_head free_ctrl_txbuf;
struct list_head free_ctrl_rxbuf;
struct ath6kl_htc_credit_info *credit_info;
Add ath6kl cleaned up driver Last May we started working on cleaning up ath6kl driver which is currently in staging. The work has happened in a separate ath6kl-cleanup tree: http://git.kernel.org/?p=linux/kernel/git/kvalo/ath6kl-cleanup.git;a=summary After over 1100 (!) patches we have now reached a state where I would like to start discussing about pushing the driver to the wireless trees and replacing the staging driver. The driver is now a lot smaller and looks like a proper Linux driver. The size of the driver (measured with simple wc -l) dropped from 49 kLOC to 18 kLOC and the number of the .c and .h files dropped from 107 to 22. Most importantly the number of subdirectories reduced from 26 to zero :) There are two remaining checkpatch warnings in the driver which we decided to omit for now: drivers/net/wireless/ath/ath6kl/debug.c:31: WARNING: printk() should include KERN_ facility level drivers/net/wireless/ath/ath6kl/sdio.c:527: WARNING: msleep < 20ms can sleep for up to 20ms; see Documentation/timers/timers-howto.txt The driver has endian annotations for all the hardware specific structures and there are no sparse errors. Unfortunately I don't have any big endian hardware to test that right now. We have been testing the driver both on x86 and arm platforms. The code is also compiled with sparc and parisc cross compilers. Notable missing features compared to the current staging driver are: o HCI over SDIO support o nl80211 testmode o firmware logging o suspend support Testmode, firmware logging and suspend support will be added soon. HCI over SDIO support will be more difficult as the HCI driver needs to share code with the wifi driver. This is something we need to research more. Also I want to point out the changes I did for signed endian support. As I wasn't able to find any support for signed endian annotations I decided to follow what NTFS has done and added my own. Grep for sle16 and sle32, especially from wmi.h. Various people have been working on the cleanup, the hall of fame based on number of patches is: 543 Vasanthakumar Thiagarajan 403 Raja Mani 252 Kalle Valo 16 Vivek Natarajan 12 Suraj Sumangala 3 Joe Perches 2 Jouni Malinen Signed-off-by: Vasanthakumar Thiagarajan <vthiagar@qca.qualcomm.com> Signed-off-by: Raja Mani <rmani@qca.qualcomm.com> Signed-off-by: Vivek Natarajan <nataraja@qca.qualcomm.com> Signed-off-by: Suraj Sumangala <surajs@qca.qualcomm.com> Signed-off-by: Joe Perches <joe@perches.com> Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
2011-07-18 05:22:30 +08:00
int tgt_creds;
unsigned int tgt_cred_sz;
spinlock_t htc_lock;
spinlock_t rx_lock;
spinlock_t tx_lock;
struct ath6kl_device *dev;
u32 htc_flags;
u32 rx_st_flags;
enum htc_endpoint_id ep_waiting;
u8 htc_tgt_ver;
/* max messages per bundle for HTC */
int msg_per_bndl_max;
bool tx_bndl_enable;
int rx_bndl_enable;
int max_rx_bndl_sz;
int max_tx_bndl_sz;
u32 block_sz;
u32 block_mask;
int max_scat_entries;
int max_xfer_szper_scatreq;
int chk_irq_status_cnt;
Add ath6kl cleaned up driver Last May we started working on cleaning up ath6kl driver which is currently in staging. The work has happened in a separate ath6kl-cleanup tree: http://git.kernel.org/?p=linux/kernel/git/kvalo/ath6kl-cleanup.git;a=summary After over 1100 (!) patches we have now reached a state where I would like to start discussing about pushing the driver to the wireless trees and replacing the staging driver. The driver is now a lot smaller and looks like a proper Linux driver. The size of the driver (measured with simple wc -l) dropped from 49 kLOC to 18 kLOC and the number of the .c and .h files dropped from 107 to 22. Most importantly the number of subdirectories reduced from 26 to zero :) There are two remaining checkpatch warnings in the driver which we decided to omit for now: drivers/net/wireless/ath/ath6kl/debug.c:31: WARNING: printk() should include KERN_ facility level drivers/net/wireless/ath/ath6kl/sdio.c:527: WARNING: msleep < 20ms can sleep for up to 20ms; see Documentation/timers/timers-howto.txt The driver has endian annotations for all the hardware specific structures and there are no sparse errors. Unfortunately I don't have any big endian hardware to test that right now. We have been testing the driver both on x86 and arm platforms. The code is also compiled with sparc and parisc cross compilers. Notable missing features compared to the current staging driver are: o HCI over SDIO support o nl80211 testmode o firmware logging o suspend support Testmode, firmware logging and suspend support will be added soon. HCI over SDIO support will be more difficult as the HCI driver needs to share code with the wifi driver. This is something we need to research more. Also I want to point out the changes I did for signed endian support. As I wasn't able to find any support for signed endian annotations I decided to follow what NTFS has done and added my own. Grep for sle16 and sle32, especially from wmi.h. Various people have been working on the cleanup, the hall of fame based on number of patches is: 543 Vasanthakumar Thiagarajan 403 Raja Mani 252 Kalle Valo 16 Vivek Natarajan 12 Suraj Sumangala 3 Joe Perches 2 Jouni Malinen Signed-off-by: Vasanthakumar Thiagarajan <vthiagar@qca.qualcomm.com> Signed-off-by: Raja Mani <rmani@qca.qualcomm.com> Signed-off-by: Vivek Natarajan <nataraja@qca.qualcomm.com> Signed-off-by: Suraj Sumangala <surajs@qca.qualcomm.com> Signed-off-by: Joe Perches <joe@perches.com> Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
2011-07-18 05:22:30 +08:00
};
void *ath6kl_htc_create(struct ath6kl *ar);
void ath6kl_htc_set_credit_dist(struct htc_target *target,
struct ath6kl_htc_credit_info *cred_info,
u16 svc_pri_order[], int len);
int ath6kl_htc_wait_target(struct htc_target *target);
int ath6kl_htc_start(struct htc_target *target);
int ath6kl_htc_conn_service(struct htc_target *target,
struct htc_service_connect_req *req,
struct htc_service_connect_resp *resp);
int ath6kl_htc_tx(struct htc_target *target, struct htc_packet *packet);
void ath6kl_htc_stop(struct htc_target *target);
void ath6kl_htc_cleanup(struct htc_target *target);
void ath6kl_htc_flush_txep(struct htc_target *target,
enum htc_endpoint_id endpoint, u16 tag);
void ath6kl_htc_flush_rx_buf(struct htc_target *target);
void ath6kl_htc_indicate_activity_change(struct htc_target *target,
enum htc_endpoint_id endpoint,
bool active);
int ath6kl_htc_get_rxbuf_num(struct htc_target *target,
enum htc_endpoint_id endpoint);
int ath6kl_htc_add_rxbuf_multiple(struct htc_target *target,
struct list_head *pktq);
int ath6kl_htc_rxmsg_pending_handler(struct htc_target *target,
u32 msg_look_ahead, int *n_pkts);
Add ath6kl cleaned up driver Last May we started working on cleaning up ath6kl driver which is currently in staging. The work has happened in a separate ath6kl-cleanup tree: http://git.kernel.org/?p=linux/kernel/git/kvalo/ath6kl-cleanup.git;a=summary After over 1100 (!) patches we have now reached a state where I would like to start discussing about pushing the driver to the wireless trees and replacing the staging driver. The driver is now a lot smaller and looks like a proper Linux driver. The size of the driver (measured with simple wc -l) dropped from 49 kLOC to 18 kLOC and the number of the .c and .h files dropped from 107 to 22. Most importantly the number of subdirectories reduced from 26 to zero :) There are two remaining checkpatch warnings in the driver which we decided to omit for now: drivers/net/wireless/ath/ath6kl/debug.c:31: WARNING: printk() should include KERN_ facility level drivers/net/wireless/ath/ath6kl/sdio.c:527: WARNING: msleep < 20ms can sleep for up to 20ms; see Documentation/timers/timers-howto.txt The driver has endian annotations for all the hardware specific structures and there are no sparse errors. Unfortunately I don't have any big endian hardware to test that right now. We have been testing the driver both on x86 and arm platforms. The code is also compiled with sparc and parisc cross compilers. Notable missing features compared to the current staging driver are: o HCI over SDIO support o nl80211 testmode o firmware logging o suspend support Testmode, firmware logging and suspend support will be added soon. HCI over SDIO support will be more difficult as the HCI driver needs to share code with the wifi driver. This is something we need to research more. Also I want to point out the changes I did for signed endian support. As I wasn't able to find any support for signed endian annotations I decided to follow what NTFS has done and added my own. Grep for sle16 and sle32, especially from wmi.h. Various people have been working on the cleanup, the hall of fame based on number of patches is: 543 Vasanthakumar Thiagarajan 403 Raja Mani 252 Kalle Valo 16 Vivek Natarajan 12 Suraj Sumangala 3 Joe Perches 2 Jouni Malinen Signed-off-by: Vasanthakumar Thiagarajan <vthiagar@qca.qualcomm.com> Signed-off-by: Raja Mani <rmani@qca.qualcomm.com> Signed-off-by: Vivek Natarajan <nataraja@qca.qualcomm.com> Signed-off-by: Suraj Sumangala <surajs@qca.qualcomm.com> Signed-off-by: Joe Perches <joe@perches.com> Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
2011-07-18 05:22:30 +08:00
int ath6kl_credit_setup(void *htc_handle,
struct ath6kl_htc_credit_info *cred_info);
Add ath6kl cleaned up driver Last May we started working on cleaning up ath6kl driver which is currently in staging. The work has happened in a separate ath6kl-cleanup tree: http://git.kernel.org/?p=linux/kernel/git/kvalo/ath6kl-cleanup.git;a=summary After over 1100 (!) patches we have now reached a state where I would like to start discussing about pushing the driver to the wireless trees and replacing the staging driver. The driver is now a lot smaller and looks like a proper Linux driver. The size of the driver (measured with simple wc -l) dropped from 49 kLOC to 18 kLOC and the number of the .c and .h files dropped from 107 to 22. Most importantly the number of subdirectories reduced from 26 to zero :) There are two remaining checkpatch warnings in the driver which we decided to omit for now: drivers/net/wireless/ath/ath6kl/debug.c:31: WARNING: printk() should include KERN_ facility level drivers/net/wireless/ath/ath6kl/sdio.c:527: WARNING: msleep < 20ms can sleep for up to 20ms; see Documentation/timers/timers-howto.txt The driver has endian annotations for all the hardware specific structures and there are no sparse errors. Unfortunately I don't have any big endian hardware to test that right now. We have been testing the driver both on x86 and arm platforms. The code is also compiled with sparc and parisc cross compilers. Notable missing features compared to the current staging driver are: o HCI over SDIO support o nl80211 testmode o firmware logging o suspend support Testmode, firmware logging and suspend support will be added soon. HCI over SDIO support will be more difficult as the HCI driver needs to share code with the wifi driver. This is something we need to research more. Also I want to point out the changes I did for signed endian support. As I wasn't able to find any support for signed endian annotations I decided to follow what NTFS has done and added my own. Grep for sle16 and sle32, especially from wmi.h. Various people have been working on the cleanup, the hall of fame based on number of patches is: 543 Vasanthakumar Thiagarajan 403 Raja Mani 252 Kalle Valo 16 Vivek Natarajan 12 Suraj Sumangala 3 Joe Perches 2 Jouni Malinen Signed-off-by: Vasanthakumar Thiagarajan <vthiagar@qca.qualcomm.com> Signed-off-by: Raja Mani <rmani@qca.qualcomm.com> Signed-off-by: Vivek Natarajan <nataraja@qca.qualcomm.com> Signed-off-by: Suraj Sumangala <surajs@qca.qualcomm.com> Signed-off-by: Joe Perches <joe@perches.com> Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
2011-07-18 05:22:30 +08:00
static inline void set_htc_pkt_info(struct htc_packet *packet, void *context,
u8 *buf, unsigned int len,
enum htc_endpoint_id eid, u16 tag)
{
packet->pkt_cntxt = context;
packet->buf = buf;
packet->act_len = len;
packet->endpoint = eid;
packet->info.tx.tag = tag;
}
static inline void htc_rxpkt_reset(struct htc_packet *packet)
{
packet->buf = packet->buf_start;
packet->act_len = 0;
}
static inline void set_htc_rxpkt_info(struct htc_packet *packet, void *context,
u8 *buf, unsigned long len,
enum htc_endpoint_id eid)
{
packet->pkt_cntxt = context;
packet->buf = buf;
packet->buf_start = buf;
packet->buf_len = len;
packet->endpoint = eid;
}
static inline int get_queue_depth(struct list_head *queue)
{
struct list_head *tmp_list;
int depth = 0;
list_for_each(tmp_list, queue)
depth++;
return depth;
}
#endif