cregit-Linux how code gets into the kernel

Release 4.11 drivers/net/fddi/skfp/hwmtm.c

/******************************************************************************
 *
 *      (C)Copyright 1998,1999 SysKonnect,
 *      a business unit of Schneider & Koch & Co. Datensysteme GmbH.
 *
 *      See the file "skfddi.c" for further information.
 *
 *      This program is free software; you can redistribute it and/or modify
 *      it under the terms of the GNU General Public License as published by
 *      the Free Software Foundation; either version 2 of the License, or
 *      (at your option) any later version.
 *
 *      The information in this file is provided "AS IS" without warranty.
 *
 ******************************************************************************/

#ifndef	lint

static char const ID_sccs[] = "@(#)hwmtm.c    1.40 99/05/31 (C) SK" ;
#endif


#define	HWMTM

#ifndef FDDI

#define	FDDI
#endif

#include "h/types.h"
#include "h/fddi.h"
#include "h/smc.h"
#include "h/supern_2.h"
#include "h/skfbiinc.h"

/*
        -------------------------------------------------------------
        DOCUMENTATION
        -------------------------------------------------------------
        BEGIN_MANUAL_ENTRY(DOCUMENTATION)

                        T B D

        END_MANUAL_ENTRY
*/
/*
        -------------------------------------------------------------
        LOCAL VARIABLES:
        -------------------------------------------------------------
*/
#ifdef COMMON_MB_POOL

static	SMbuf *mb_start = 0 ;

static	SMbuf *mb_free = 0 ;

static	int mb_init = FALSE ;

static	int call_count = 0 ;
#endif

/*
        -------------------------------------------------------------
        EXTERNE VARIABLES:
        -------------------------------------------------------------
*/

#ifdef	DEBUG
#ifndef	DEBUG_BRD
extern	struct smt_debug	debug ;
#endif
#endif

#ifdef	NDIS_OS2
extern	u_char	offDepth ;
extern	u_char	force_irq_pending ;
#endif

/*
        -------------------------------------------------------------
        LOCAL FUNCTIONS:
        -------------------------------------------------------------
*/

static void queue_llc_rx(struct s_smc *smc, SMbuf *mb);
static void smt_to_llc(struct s_smc *smc, SMbuf *mb);
static void init_txd_ring(struct s_smc *smc);
static void init_rxd_ring(struct s_smc *smc);
static void queue_txd_mb(struct s_smc *smc, SMbuf *mb);
static u_long init_descr_ring(struct s_smc *smc, union s_fp_descr volatile *start,
			      int count);
static u_long repair_txd_ring(struct s_smc *smc, struct s_smt_tx_queue *queue);
static u_long repair_rxd_ring(struct s_smc *smc, struct s_smt_rx_queue *queue);
static SMbuf* get_llc_rx(struct s_smc *smc);
static SMbuf* get_txd_mb(struct s_smc *smc);
static void mac_drv_clear_txd(struct s_smc *smc);

/*
        -------------------------------------------------------------
        EXTERNAL FUNCTIONS:
        -------------------------------------------------------------
*/
/*      The external SMT functions are listed in cmtdef.h */

extern void* mac_drv_get_space(struct s_smc *smc, unsigned int size);
extern void* mac_drv_get_desc_mem(struct s_smc *smc, unsigned int size);
extern void mac_drv_fill_rxd(struct s_smc *smc);
extern void mac_drv_tx_complete(struct s_smc *smc,
				volatile struct s_smt_fp_txd *txd);
extern void mac_drv_rx_complete(struct s_smc *smc,
				volatile struct s_smt_fp_rxd *rxd,
				int frag_count, int len);
extern void mac_drv_requeue_rxd(struct s_smc *smc, 
				volatile struct s_smt_fp_rxd *rxd,
				int frag_count);
extern void mac_drv_clear_rxd(struct s_smc *smc,
			      volatile struct s_smt_fp_rxd *rxd, int frag_count);

#ifdef	USE_OS_CPY
extern void hwm_cpy_rxd2mb(void);
extern void hwm_cpy_txd2mb(void);
#endif

#ifdef	ALL_RX_COMPLETE
extern void mac_drv_all_receives_complete(void);
#endif

extern u_long mac_drv_virt2phys(struct s_smc *smc, void *virt);
extern u_long dma_master(struct s_smc *smc, void *virt, int len, int flag);

#ifdef	NDIS_OS2
extern void post_proc(void);
#else
extern void dma_complete(struct s_smc *smc, volatile union s_fp_descr *descr,
			 int flag);
#endif

extern int mac_drv_rx_init(struct s_smc *smc, int len, int fc, char *look_ahead,
			   int la_len);

/*
        -------------------------------------------------------------
        PUBLIC FUNCTIONS:
        -------------------------------------------------------------
*/
void process_receive(struct s_smc *smc);
void fddi_isr(struct s_smc *smc);
void smt_free_mbuf(struct s_smc *smc, SMbuf *mb);
void init_driver_fplus(struct s_smc *smc);
void mac_drv_rx_mode(struct s_smc *smc, int mode);
void init_fddi_driver(struct s_smc *smc, u_char *mac_addr);
void mac_drv_clear_tx_queue(struct s_smc *smc);
void mac_drv_clear_rx_queue(struct s_smc *smc);
void hwm_tx_frag(struct s_smc *smc, char far *virt, u_long phys, int len,
		 int frame_status);
void hwm_rx_frag(struct s_smc *smc, char far *virt, u_long phys, int len,
		 int frame_status);

int mac_drv_init(struct s_smc *smc);
int hwm_tx_init(struct s_smc *smc, u_char fc, int frag_count, int frame_len,
		int frame_status);

u_int mac_drv_check_space(void);

SMbuf* smt_get_mbuf(struct s_smc *smc);

#ifdef DEBUG
	void mac_drv_debug_lev(struct s_smc *smc, int flag, int lev);
#endif

/*
        -------------------------------------------------------------
        MACROS:
        -------------------------------------------------------------
*/
#ifndef	UNUSED
#ifdef	lint

#define UNUSED(x)	(x) = (x)
#else

#define UNUSED(x)
#endif
#endif

#ifdef	USE_CAN_ADDR

#define MA		smc->hw.fddi_canon_addr.a

#define	GROUP_ADDR_BIT	0x01
#else

#define	MA		smc->hw.fddi_home_addr.a

#define	GROUP_ADDR_BIT	0x80
#endif


#define RXD_TXD_COUNT	(HWM_ASYNC_TXD_COUNT+HWM_SYNC_TXD_COUNT+\
                        SMT_R1_RXD_COUNT+SMT_R2_RXD_COUNT)

#ifdef	MB_OUTSIDE_SMC

#define	EXT_VIRT_MEM	((RXD_TXD_COUNT+1)*sizeof(struct s_smt_fp_txd) +\
                        MAX_MBUF*sizeof(SMbuf))

#define	EXT_VIRT_MEM_2	((RXD_TXD_COUNT+1)*sizeof(struct s_smt_fp_txd))
#else

#define	EXT_VIRT_MEM	((RXD_TXD_COUNT+1)*sizeof(struct s_smt_fp_txd))
#endif

	/*
         * define critical read for 16 Bit drivers
         */
#if	defined(NDIS_OS2) || defined(ODI2)

#define CR_READ(var)	((var) & 0xffff0000 | ((var) & 0xffff))
#else

#define CR_READ(var)	(__le32)(var)
#endif


#define IMASK_SLOW	(IS_PLINT1 | IS_PLINT2 | IS_TIMINT | IS_TOKEN | \
                         IS_MINTR1 | IS_MINTR2 | IS_MINTR3 | IS_R1_P | \
                         IS_R1_C | IS_XA_C | IS_XS_C)

/*
        -------------------------------------------------------------
        INIT- AND SMT FUNCTIONS:
        -------------------------------------------------------------
*/


/*
 *      BEGIN_MANUAL_ENTRY(mac_drv_check_space)
 *      u_int mac_drv_check_space()
 *
 *      function        DOWNCALL        (drvsr.c)
 *                      This function calculates the needed non virtual
 *                      memory for MBufs, RxD and TxD descriptors etc.
 *                      needed by the driver.
 *
 *      return          u_int   memory in bytes
 *
 *      END_MANUAL_ENTRY
 */

u_int mac_drv_check_space(void) { #ifdef MB_OUTSIDE_SMC #ifdef COMMON_MB_POOL call_count++ ; if (call_count == 1) { return EXT_VIRT_MEM; } else { return EXT_VIRT_MEM_2; } #else return EXT_VIRT_MEM; #endif #else return 0; #endif }

Contributors

PersonTokensPropCommitsCommitProp
Linus Torvalds (pre-git)4493.62%150.00%
Stephen Hemminger36.38%150.00%
Total47100.00%2100.00%

/* * BEGIN_MANUAL_ENTRY(mac_drv_init) * void mac_drv_init(smc) * * function DOWNCALL (drvsr.c) * In this function the hardware module allocates it's * memory. * The operating system dependent module should call * mac_drv_init once, after the adatper is detected. * END_MANUAL_ENTRY */
int mac_drv_init(struct s_smc *smc) { if (sizeof(struct s_smt_fp_rxd) % 16) { SMT_PANIC(smc,HWM_E0001,HWM_E0001_MSG) ; } if (sizeof(struct s_smt_fp_txd) % 16) { SMT_PANIC(smc,HWM_E0002,HWM_E0002_MSG) ; } /* * get the required memory for the RxDs and TxDs */ if (!(smc->os.hwm.descr_p = (union s_fp_descr volatile *) mac_drv_get_desc_mem(smc,(u_int) (RXD_TXD_COUNT+1)*sizeof(struct s_smt_fp_txd)))) { return 1; /* no space the hwm modul can't work */ } /* * get the memory for the SMT MBufs */ #ifndef MB_OUTSIDE_SMC smc->os.hwm.mbuf_pool.mb_start=(SMbuf *)(&smc->os.hwm.mbuf_pool.mb[0]) ; #else #ifndef COMMON_MB_POOL if (!(smc->os.hwm.mbuf_pool.mb_start = (SMbuf *) mac_drv_get_space(smc, MAX_MBUF*sizeof(SMbuf)))) { return 1; /* no space the hwm modul can't work */ } #else if (!mb_start) { if (!(mb_start = (SMbuf *) mac_drv_get_space(smc, MAX_MBUF*sizeof(SMbuf)))) { return 1; /* no space the hwm modul can't work */ } } #endif #endif return 0; }

Contributors

PersonTokensPropCommitsCommitProp
Linus Torvalds (pre-git)21899.54%150.00%
Stephen Hemminger10.46%150.00%
Total219100.00%2100.00%

/* * BEGIN_MANUAL_ENTRY(init_driver_fplus) * init_driver_fplus(smc) * * Sets hardware modul specific values for the mode register 2 * (e.g. the byte alignment for the received frames, the position of the * least significant byte etc.) * END_MANUAL_ENTRY */
void init_driver_fplus(struct s_smc *smc) { smc->hw.fp.mdr2init = FM_LSB | FM_BMMODE | FM_ENNPRQ | FM_ENHSRQ | 3 ; #ifdef PCI smc->hw.fp.mdr2init |= FM_CHKPAR | FM_PARITY ; #endif smc->hw.fp.mdr3init = FM_MENRQAUNLCK | FM_MENRS ; #ifdef USE_CAN_ADDR /* enable address bit swapping */ smc->hw.fp.frselreg_init = FM_ENXMTADSWAP | FM_ENRCVADSWAP ; #endif }

Contributors

PersonTokensPropCommitsCommitProp
Linus Torvalds (pre-git)7498.67%150.00%
Stephen Hemminger11.33%150.00%
Total75100.00%2100.00%


static u_long init_descr_ring(struct s_smc *smc, union s_fp_descr volatile *start, int count) { int i ; union s_fp_descr volatile *d1 ; union s_fp_descr volatile *d2 ; u_long phys ; DB_GEN(3, "descr ring starts at = %p", start); for (i=count-1, d1=start; i ; i--) { d2 = d1 ; d1++ ; /* descr is owned by the host */ d2->r.rxd_rbctrl = cpu_to_le32(BMU_CHECK) ; d2->r.rxd_next = &d1->r ; phys = mac_drv_virt2phys(smc,(void *)d1) ; d2->r.rxd_nrdadr = cpu_to_le32(phys) ; } DB_GEN(3, "descr ring ends at = %p", d1); d1->r.rxd_rbctrl = cpu_to_le32(BMU_CHECK) ; d1->r.rxd_next = &start->r ; phys = mac_drv_virt2phys(smc,(void *)start) ; d1->r.rxd_nrdadr = cpu_to_le32(phys) ; for (i=count, d1=start; i ; i--) { DRV_BUF_FLUSH(&d1->r,DDI_DMA_SYNC_FORDEV) ; d1++; } return phys; }

Contributors

PersonTokensPropCommitsCommitProp
Linus Torvalds (pre-git)19593.75%125.00%
Joe Perches62.88%125.00%
Al Viro41.92%125.00%
Stephen Hemminger31.44%125.00%
Total208100.00%4100.00%


static void init_txd_ring(struct s_smc *smc) { struct s_smt_fp_txd volatile *ds ; struct s_smt_tx_queue *queue ; u_long phys ; /* * initialize the transmit descriptors */ ds = (struct s_smt_fp_txd volatile *) ((char *)smc->os.hwm.descr_p + SMT_R1_RXD_COUNT*sizeof(struct s_smt_fp_rxd)) ; queue = smc->hw.fp.tx[QUEUE_A0] ; DB_GEN(3, "Init async TxD ring, %d TxDs", HWM_ASYNC_TXD_COUNT); (void)init_descr_ring(smc,(union s_fp_descr volatile *)ds, HWM_ASYNC_TXD_COUNT) ; phys = le32_to_cpu(ds->txd_ntdadr) ; ds++ ; queue->tx_curr_put = queue->tx_curr_get = ds ; ds-- ; queue->tx_free = HWM_ASYNC_TXD_COUNT ; queue->tx_used = 0 ; outpd(ADDR(B5_XA_DA),phys) ; ds = (struct s_smt_fp_txd volatile *) ((char *)ds + HWM_ASYNC_TXD_COUNT*sizeof(struct s_smt_fp_txd)) ; queue = smc->hw.fp.tx[QUEUE_S] ; DB_GEN(3, "Init sync TxD ring, %d TxDs", HWM_SYNC_TXD_COUNT); (void)init_descr_ring(smc,(union s_fp_descr volatile *)ds, HWM_SYNC_TXD_COUNT) ; phys = le32_to_cpu(ds->txd_ntdadr) ; ds++ ; queue->tx_curr_put = queue->tx_curr_get = ds ; queue->tx_free = HWM_SYNC_TXD_COUNT ; queue->tx_used = 0 ; outpd(ADDR(B5_XS_DA),phys) ; }

Contributors

PersonTokensPropCommitsCommitProp
Linus Torvalds (pre-git)24296.41%125.00%
Joe Perches62.39%125.00%
Al Viro20.80%125.00%
Stephen Hemminger10.40%125.00%
Total251100.00%4100.00%


static void init_rxd_ring(struct s_smc *smc) { struct s_smt_fp_rxd volatile *ds ; struct s_smt_rx_queue *queue ; u_long phys ; /* * initialize the receive descriptors */ ds = (struct s_smt_fp_rxd volatile *) smc->os.hwm.descr_p ; queue = smc->hw.fp.rx[QUEUE_R1] ; DB_GEN(3, "Init RxD ring, %d RxDs", SMT_R1_RXD_COUNT); (void)init_descr_ring(smc,(union s_fp_descr volatile *)ds, SMT_R1_RXD_COUNT) ; phys = le32_to_cpu(ds->rxd_nrdadr) ; ds++ ; queue->rx_curr_put = queue->rx_curr_get = ds ; queue->rx_free = SMT_R1_RXD_COUNT ; queue->rx_used = 0 ; outpd(ADDR(B4_R1_DA),phys) ; }

Contributors

PersonTokensPropCommitsCommitProp
Linus Torvalds (pre-git)12196.03%125.00%
Joe Perches32.38%125.00%
Stephen Hemminger10.79%125.00%
Al Viro10.79%125.00%
Total126100.00%4100.00%

/* * BEGIN_MANUAL_ENTRY(init_fddi_driver) * void init_fddi_driver(smc,mac_addr) * * initializes the driver and it's variables * * END_MANUAL_ENTRY */
void init_fddi_driver(struct s_smc *smc, u_char *mac_addr) { SMbuf *mb ; int i ; init_board(smc,mac_addr) ; (void)init_fplus(smc) ; /* * initialize the SMbufs for the SMT */ #ifndef COMMON_MB_POOL mb = smc->os.hwm.mbuf_pool.mb_start ; smc->os.hwm.mbuf_pool.mb_free = (SMbuf *)NULL ; for (i = 0; i < MAX_MBUF; i++) { mb->sm_use_count = 1 ; smt_free_mbuf(smc,mb) ; mb++ ; } #else mb = mb_start ; if (!mb_init) { mb_free = 0 ; for (i = 0; i < MAX_MBUF; i++) { mb->sm_use_count = 1 ; smt_free_mbuf(smc,mb) ; mb++ ; } mb_init = TRUE ; } #endif /* * initialize the other variables */ smc->os.hwm.llc_rx_pipe = smc->os.hwm.llc_rx_tail = (SMbuf *)NULL ; smc->os.hwm.txd_tx_pipe = smc->os.hwm.txd_tx_tail = NULL ; smc->os.hwm.pass_SMT = smc->os.hwm.pass_NSA = smc->os.hwm.pass_DB = 0 ; smc->os.hwm.pass_llc_promisc = TRUE ; smc->os.hwm.queued_rx_frames = smc->os.hwm.queued_txd_mb = 0 ; smc->os.hwm.detec_count = 0 ; smc->os.hwm.rx_break = 0 ; smc->os.hwm.rx_len_error = 0 ; smc->os.hwm.isr_flag = FALSE ; /* * make sure that the start pointer is 16 byte aligned */ i = 16 - ((long)smc->os.hwm.descr_p & 0xf) ; if (i != 16) { DB_GEN(3, "i = %d", i); smc->os.hwm.descr_p = (union s_fp_descr volatile *) ((char *)smc->os.hwm.descr_p+i) ; } DB_GEN(3, "pt to descr area = %p", smc->os.hwm.descr_p); init_txd_ring(smc) ; init_rxd_ring(smc) ; mac_drv_fill_rxd(smc) ; init_plc(smc) ; }

Contributors

PersonTokensPropCommitsCommitProp
Linus Torvalds (pre-git)38398.21%250.00%
Joe Perches51.28%125.00%
Stephen Hemminger20.51%125.00%
Total390100.00%4100.00%


SMbuf *smt_get_mbuf(struct s_smc *smc) { register SMbuf *mb ; #ifndef COMMON_MB_POOL mb = smc->os.hwm.mbuf_pool.mb_free ; #else mb = mb_free ; #endif if (mb) { #ifndef COMMON_MB_POOL smc->os.hwm.mbuf_pool.mb_free = mb->sm_next ; #else mb_free = mb->sm_next ; #endif mb->sm_off = 8 ; mb->sm_use_count = 1 ; } DB_GEN(3, "get SMbuf: mb = %p", mb); return mb; /* May be NULL */ }

Contributors

PersonTokensPropCommitsCommitProp
Linus Torvalds (pre-git)9395.88%133.33%
Joe Perches33.09%133.33%
Stephen Hemminger11.03%133.33%
Total97100.00%3100.00%


void smt_free_mbuf(struct s_smc *smc, SMbuf *mb) { if (mb) { mb->sm_use_count-- ; DB_GEN(3, "free_mbuf: sm_use_count = %d", mb->sm_use_count); /* * If the use_count is != zero the MBuf is queued * more than once and must not queued into the * free MBuf queue */ if (!mb->sm_use_count) { DB_GEN(3, "free SMbuf: mb = %p", mb); #ifndef COMMON_MB_POOL mb->sm_next = smc->os.hwm.mbuf_pool.mb_free ; smc->os.hwm.mbuf_pool.mb_free = mb ; #else mb->sm_next = mb_free ; mb_free = mb ; #endif } } else SMT_PANIC(smc,HWM_E0003,HWM_E0003_MSG) ; }

Contributors

PersonTokensPropCommitsCommitProp
Linus Torvalds (pre-git)10193.52%133.33%
Joe Perches54.63%133.33%
Stephen Hemminger21.85%133.33%
Total108100.00%3100.00%

/* * BEGIN_MANUAL_ENTRY(mac_drv_repair_descr) * void mac_drv_repair_descr(smc) * * function called from SMT (HWM / hwmtm.c) * The BMU is idle when this function is called. * Mac_drv_repair_descr sets up the physical address * for all receive and transmit queues where the BMU * should continue. * It may be that the BMU was reseted during a fragmented * transfer. In this case there are some fragments which will * never completed by the BMU. The OWN bit of this fragments * must be switched to be owned by the host. * * Give a start command to the receive BMU. * Start the transmit BMUs if transmit frames pending. * * END_MANUAL_ENTRY */
void mac_drv_repair_descr(struct s_smc *smc) { u_long phys ; if (smc->hw.hw_state != STOPPED) { SK_BREAK() ; SMT_PANIC(smc,HWM_E0013,HWM_E0013_MSG) ; return ; } /* * repair tx queues: don't start */ phys = repair_txd_ring(smc,smc->hw.fp.tx[QUEUE_A0]) ; outpd(ADDR(B5_XA_DA),phys) ; if (smc->hw.fp.tx_q[QUEUE_A0].tx_used) { outpd(ADDR(B0_XA_CSR),CSR_START) ; } phys = repair_txd_ring(smc,smc->hw.fp.tx[QUEUE_S]) ; outpd(ADDR(B5_XS_DA),phys) ; if (smc->hw.fp.tx_q[QUEUE_S].tx_used) { outpd(ADDR(B0_XS_CSR),CSR_START) ; } /* * repair rx queues */ phys = repair_rxd_ring(smc,smc->hw.fp.rx[QUEUE_R1]) ; outpd(ADDR(B4_R1_DA),phys) ; outpd(ADDR(B0_R1_CSR),CSR_START) ; }

Contributors

PersonTokensPropCommitsCommitProp
Linus Torvalds (pre-git)18799.47%150.00%
Stephen Hemminger10.53%150.00%
Total188100.00%2100.00%


static u_long repair_txd_ring(struct s_smc *smc, struct s_smt_tx_queue *queue) { int i ; int tx_used ; u_long phys ; u_long tbctrl ; struct s_smt_fp_txd volatile *t ; SK_UNUSED(smc) ; t = queue->tx_curr_get ; tx_used = queue->tx_used ; for (i = tx_used+queue->tx_free-1 ; i ; i-- ) { t = t->txd_next ; } phys = le32_to_cpu(t->txd_ntdadr) ; t = queue->tx_curr_get ; while (tx_used) { DRV_BUF_FLUSH(t,DDI_DMA_SYNC_FORCPU) ; tbctrl = le32_to_cpu(t->txd_tbctrl) ; if (tbctrl & BMU_OWN) { if (tbctrl & BMU_STF) { break ; /* exit the loop */ } else { /* * repair the descriptor */ t->txd_tbctrl &= ~cpu_to_le32(BMU_OWN) ; } } phys = le32_to_cpu(t->txd_ntdadr) ; DRV_BUF_FLUSH(t,DDI_DMA_SYNC_FORDEV) ; t = t->txd_next ; tx_used-- ; } return phys; }

Contributors

PersonTokensPropCommitsCommitProp
Linus Torvalds (pre-git)16695.95%133.33%
Al Viro52.89%133.33%
Stephen Hemminger21.16%133.33%
Total173100.00%3100.00%

/* * Repairs the receive descriptor ring and returns the physical address * where the BMU should continue working. * * o The physical address where the BMU was stopped has to be * determined. This is the next RxD after rx_curr_get with an OWN * bit set. * o The BMU should start working at beginning of the next frame. * RxDs with an OWN bit set but with a reset STF bit should be * skipped and owned by the driver (OWN = 0). */
static u_long repair_rxd_ring(struct s_smc *smc, struct s_smt_rx_queue *queue) { int i ; int rx_used ; u_long phys ; u_long rbctrl ; struct s_smt_fp_rxd volatile *r ; SK_UNUSED(smc) ; r = queue->rx_curr_get ; rx_used = queue->rx_used ; for (i = SMT_R1_RXD_COUNT-1 ; i ; i-- ) { r = r->rxd_next ; } phys = le32_to_cpu(r->rxd_nrdadr) ; r = queue->rx_curr_get ; while (rx_used) { DRV_BUF_FLUSH(r,DDI_DMA_SYNC_FORCPU) ; rbctrl = le32_to_cpu(r->rxd_rbctrl) ; if (rbctrl & BMU_OWN) { if (rbctrl & BMU_STF) { break ; /* exit the loop */ } else { /* * repair the descriptor */ r->rxd_rbctrl &= ~cpu_to_le32(BMU_OWN) ; } } phys = le32_to_cpu(r->rxd_nrdadr) ; DRV_BUF_FLUSH(r,DDI_DMA_SYNC_FORDEV) ; r = r->rxd_next ; rx_used-- ; } return phys; }

Contributors

PersonTokensPropCommitsCommitProp
Linus Torvalds (pre-git)16295.86%133.33%
Al Viro52.96%133.33%
Stephen Hemminger21.18%133.33%
Total169100.00%3100.00%

/* ------------------------------------------------------------- INTERRUPT SERVICE ROUTINE: ------------------------------------------------------------- */ /* * BEGIN_MANUAL_ENTRY(fddi_isr) * void fddi_isr(smc) * * function DOWNCALL (drvsr.c) * interrupt service routine, handles the interrupt requests * generated by the FDDI adapter. * * NOTE: The operating system dependent module must guarantee that the * interrupts of the adapter are disabled when it calls fddi_isr. * * About the USE_BREAK_ISR mechanismn: * * The main requirement of this mechanismn is to force an timer IRQ when * leaving process_receive() with leave_isr set. process_receive() may * be called at any time from anywhere! * To be sure we don't miss such event we set 'force_irq' per default. * We have to force and Timer IRQ if 'smc->os.hwm.leave_isr' AND * 'force_irq' are set. 'force_irq' may be reset if a receive complete * IRQ is pending. * * END_MANUAL_ENTRY */
void fddi_isr(struct s_smc *smc) { u_long is ; /* ISR source */ u_short stu, stl ; SMbuf *mb ; #ifdef USE_BREAK_ISR int force_irq ; #endif #ifdef ODI2 if (smc->os.hwm.rx_break) { mac_drv_fill_rxd(smc) ; if (smc->hw.fp.rx_q[QUEUE_R1].rx_used > 0) { smc->os.hwm.rx_break = 0 ; process_receive(smc) ; } else { smc->os.hwm.detec_count = 0 ; smt_force_irq(smc) ; } } #endif smc->os.hwm.isr_flag = TRUE ; #ifdef USE_BREAK_ISR force_irq = TRUE ; if (smc->os.hwm.leave_isr) { smc->os.hwm.leave_isr = FALSE ; process_receive(smc) ; } #endif while ((is = GET_ISR() & ISR_MASK)) { NDD_TRACE("CH0B",is,0,0) ; DB_GEN(7, "ISA = 0x%lx", is); if (is & IMASK_SLOW) { NDD_TRACE("CH1b",is,0,0) ; if (is & IS_PLINT1) { /* PLC1 */ plc1_irq(smc) ; } if (is & IS_PLINT2) { /* PLC2 */ plc2_irq(smc) ; } if (is & IS_MINTR1) { /* FORMAC+ STU1(U/L) */ stu = inpw(FM_A(FM_ST1U)) ; stl = inpw(FM_A(FM_ST1L)) ; DB_GEN(6, "Slow transmit complete"); mac1_irq(smc,stu,stl) ; } if (is & IS_MINTR2) { /* FORMAC+ STU2(U/L) */ stu= inpw(FM_A(FM_ST2U)) ; stl= inpw(FM_A(FM_ST2L)) ; DB_GEN(6, "Slow receive complete"); DB_GEN(7, "stl = %x : stu = %x", stl, stu); mac2_irq(smc,stu,stl) ; } if (is & IS_MINTR3) { /* FORMAC+ STU3(U/L) */ stu= inpw(FM_A(FM_ST3U)) ; stl= inpw(FM_A(FM_ST3L)) ; DB_GEN(6, "FORMAC Mode Register 3"); mac3_irq(smc,stu,stl) ; } if (is & IS_TIMINT) { /* Timer 82C54-2 */ timer_irq(smc) ; #ifdef NDIS_OS2 force_irq_pending = 0 ; #endif /* * out of RxD detection */ if (++smc->os.hwm.detec_count > 4) { /* * check out of RxD condition */ process_receive(smc) ; } } if (is & IS_TOKEN) { /* Restricted Token Monitor */ rtm_irq(smc) ; } if (is & IS_R1_P) { /* Parity error rx queue 1 */ /* clear IRQ */ outpd(ADDR(B4_R1_CSR),CSR_IRQ_CL_P) ; SMT_PANIC(smc,HWM_E0004,HWM_E0004_MSG) ; } if (is & IS_R1_C) { /* Encoding error rx queue 1 */ /* clear IRQ */ outpd(ADDR(B4_R1_CSR),CSR_IRQ_CL_C) ; SMT_PANIC(smc,HWM_E0005,HWM_E0005_MSG) ; } if (is & IS_XA_C) { /* Encoding error async tx q */ /* clear IRQ */ outpd(ADDR(B5_XA_CSR),CSR_IRQ_CL_C) ; SMT_PANIC(smc,HWM_E0006,HWM_E0006_MSG) ; } if (is & IS_XS_C) { /* Encoding error sync tx q */ /* clear IRQ */ outpd(ADDR(B5_XS_CSR),CSR_IRQ_CL_C) ; SMT_PANIC(smc,HWM_E0007,HWM_E0007_MSG) ; } } /* * Fast Tx complete Async/Sync Queue (BMU service) */ if (is & (IS_XS_F|IS_XA_F)) { DB_GEN(6, "Fast tx complete queue"); /* * clear IRQ, Note: no IRQ is lost, because * we always service both queues */ outpd(ADDR(B5_XS_CSR),CSR_IRQ_CL_F) ; outpd(ADDR(B5_XA_CSR),CSR_IRQ_CL_F) ; mac_drv_clear_txd(smc) ; llc_restart_tx(smc) ; } /* * Fast Rx Complete (BMU service) */ if (is & IS_R1_F) { DB_GEN(6, "Fast receive complete"); /* clear IRQ */ #ifndef USE_BREAK_ISR outpd(ADDR(B4_R1_CSR),CSR_IRQ_CL_F) ; process_receive(smc) ; #else process_receive(smc) ; if (smc->os.hwm.leave_isr) { force_irq = FALSE ; } else { outpd(ADDR(B4_R1_CSR),CSR_IRQ_CL_F) ; process_receive(smc) ; } #endif } #ifndef NDIS_OS2 while ((mb = get_llc_rx(smc))) { smt_to_llc(smc,mb) ; } #else if (offDepth) post_proc() ; while (!offDepth && (mb = get_llc_rx(smc))) { smt_to_llc(smc,mb) ; } if (!offDepth && smc->os.hwm.rx_break) { process_receive(smc) ; } #endif if (smc->q.ev_get != smc->q.ev_put) { NDD_TRACE("CH2a",0,0,0) ; ev_dispatcher(smc) ; } #ifdef NDIS_OS2 post_proc() ; if (offDepth) { /* leave fddi_isr because */ break ; /* indications not allowed */ } #endif #ifdef USE_BREAK_ISR if (smc->os.hwm.leave_isr) { break ; /* leave fddi_isr */ } #endif /* NOTE: when the isr is left, no rx is pending */ } /* end of interrupt source polling loop */ #ifdef USE_BREAK_ISR if (smc->os.hwm.leave_isr && force_irq) { smt_force_irq(smc) ; } #endif smc->os.hwm.isr_flag = FALSE ; NDD_TRACE("CH0E",0,0,0) ; }

Contributors

PersonTokensPropCommitsCommitProp
Linus Torvalds (pre-git)85798.17%133.33%
Joe Perches151.72%133.33%
Stephen Hemminger10.11%133.33%
Total873100.00%3100.00%

/* ------------------------------------------------------------- RECEIVE FUNCTIONS: ------------------------------------------------------------- */ #ifndef NDIS_OS2 /* * BEGIN_MANUAL_ENTRY(mac_drv_rx_mode) * void mac_drv_rx_mode(smc,mode) * * function DOWNCALL (fplus.c) * Corresponding to the parameter mode, the operating system * dependent module can activate several receive modes. * * para mode = 1: RX_ENABLE_ALLMULTI enable all multicasts * = 2: RX_DISABLE_ALLMULTI disable "enable all multicasts" * = 3: RX_ENABLE_PROMISC enable promiscuous * = 4: RX_DISABLE_PROMISC disable promiscuous * = 5: RX_ENABLE_NSA enable rec. of all NSA frames * (disabled after 'driver reset' & 'set station address') * = 6: RX_DISABLE_NSA disable rec. of all NSA frames * * = 21: RX_ENABLE_PASS_SMT ( see description ) * = 22: RX_DISABLE_PASS_SMT ( " " ) * = 23: RX_ENABLE_PASS_NSA ( " " ) * = 24: RX_DISABLE_PASS_NSA ( " " ) * = 25: RX_ENABLE_PASS_DB ( " " ) * = 26: RX_DISABLE_PASS_DB ( " " ) * = 27: RX_DISABLE_PASS_ALL ( " " ) * = 28: RX_DISABLE_LLC_PROMISC ( " " ) * = 29: RX_ENABLE_LLC_PROMISC ( " " ) * * * RX_ENABLE_PASS_SMT / RX_DISABLE_PASS_SMT * * If the operating system dependent module activates the * mode RX_ENABLE_PASS_SMT, the hardware module * duplicates all SMT frames with the frame control * FC_SMT_INFO and passes them to the LLC receive channel * by calling mac_drv_rx_init. * The SMT Frames which are sent by the local SMT and the NSA * frames whose A- and C-Indicator is not set are also duplicated * and passed. * The receive mode RX_DISABLE_PASS_SMT disables the passing * of SMT frames. * * RX_ENABLE_PASS_NSA / RX_DISABLE_PASS_NSA * * If the operating system dependent module activates the * mode RX_ENABLE_PASS_NSA, the hardware module * duplicates all NSA frames with frame control FC_SMT_NSA * and a set A-Indicator and passed them to the LLC * receive channel by calling mac_drv_rx_init. * All NSA Frames which are sent by the local SMT * are also duplicated and passed. * The receive mode RX_DISABLE_PASS_NSA disables the passing * of NSA frames with the A- or C-Indicator set. * * NOTE: For fear that the hardware module receives NSA frames with * a reset A-Indicator, the operating system dependent module * has to call mac_drv_rx_mode with the mode RX_ENABLE_NSA * before activate the RX_ENABLE_PASS_NSA mode and after every * 'driver reset' and 'set station address'. * * RX_ENABLE_PASS_DB / RX_DISABLE_PASS_DB * * If the operating system dependent module activates the * mode RX_ENABLE_PASS_DB, direct BEACON frames * (FC_BEACON frame control) are passed to the LLC receive * channel by mac_drv_rx_init. * The receive mode RX_DISABLE_PASS_DB disables the passing * of direct BEACON frames. * * RX_DISABLE_PASS_ALL * * Disables all special receives modes. It is equal to * call mac_drv_set_rx_mode successively with the * parameters RX_DISABLE_NSA, RX_DISABLE_PASS_SMT, * RX_DISABLE_PASS_NSA and RX_DISABLE_PASS_DB. * * RX_ENABLE_LLC_PROMISC * * (default) all received LLC frames and all SMT/NSA/DBEACON * frames depending on the attitude of the flags * PASS_SMT/PASS_NSA/PASS_DBEACON will be delivered to the * LLC layer * * RX_DISABLE_LLC_PROMISC * * all received SMT/NSA/DBEACON frames depending on the * attitude of the flags PASS_SMT/PASS_NSA/PASS_DBEACON * will be delivered to the LLC layer. * all received LLC frames with a directed address, Multicast * or Broadcast address will be delivered to the LLC * layer too. * * END_MANUAL_ENTRY */
void mac_drv_rx_mode(struct s_smc *smc, int mode) { switch(mode) { case RX_ENABLE_PASS_SMT: smc->os.hwm.pass_SMT = TRUE ; break ; case RX_DISABLE_PASS_SMT: smc->os.hwm.pass_SMT = FALSE ; break ; case RX_ENABLE_PASS_NSA: smc->os.hwm.pass_NSA = TRUE ; break ; case RX_DISABLE_PASS_NSA: smc->os.hwm.pass_NSA = FALSE ; break ; case RX_ENABLE_PASS_DB: smc->os.hwm.pass_DB = TRUE ; break ; case RX_DISABLE_PASS_DB: smc->os.hwm.pass_DB = FALSE ; break ; case RX_DISABLE_PASS_ALL: smc->os.hwm.pass_SMT = smc->os.hwm.pass_NSA = FALSE ; smc->os.hwm.pass_DB = FALSE ; smc->os.hwm.pass_llc_promisc = TRUE ; mac_set_rx_mode(smc,RX_DISABLE_NSA) ; break ; case RX_DISABLE_LLC_PROMISC: smc->os.hwm.pass_llc_promisc = FALSE ; break ; case RX_ENABLE_LLC_PROMISC: smc->os.hwm.pass_llc_promisc = TRUE ; break ; case RX_ENABLE_ALLMULTI: case RX_DISABLE_ALLMULTI: case RX_ENABLE_PROMISC: case RX_DISABLE_PROMISC: case RX_ENABLE_NSA: case RX_DISABLE_NSA: default: mac_set_rx_mode(smc,mode) ; break ; } }

Contributors

PersonTokensPropCommitsCommitProp
Linus Torvalds (pre-git)20599.03%150.00%
Stephen Hemminger20.97%150.00%
Total207100.00%2100.00%

#endif /* ifndef NDIS_OS2 */ /* * process receive queue */
void process_receive(struct s_smc *smc) { int i ; int n ; int frag_count ; /* number of RxDs of the curr rx buf */ int used_frags ; /* number of RxDs of the curr frame */ struct s_smt_rx_queue *queue ; /* points to the queue ctl struct */ struct s_smt_fp_rxd volatile *r ; /* rxd pointer */ struct s_smt_fp_rxd volatile *rxd ; /* first rxd of rx frame */ u_long rbctrl ; /* receive buffer control word */ u_long rfsw ; /* receive frame status word */ u_short rx_used ; u_char far *virt ; char far *data ; SMbuf *mb ; u_char fc ; /* Frame control */ int len ; /* Frame length */ smc->os.hwm.detec_count = 0 ; queue = smc->hw.fp.rx[QUEUE_R1] ; NDD_TRACE("RHxB",0,0,0) ; for ( ; ; ) { r = queue->rx_curr_get ; rx_used = queue->rx_used ; frag_count = 0 ; #ifdef USE_BREAK_ISR if (smc->os.hwm.leave_isr) { goto rx_end ; } #endif #ifdef NDIS_OS2 if (offDepth) { smc->os.hwm.rx_break = 1 ; goto rx_end ; } smc->os.hwm.rx_break = 0 ; #endif #ifdef ODI2 if (smc->os.hwm.rx_break) { goto rx_end ; } #endif n = 0 ; do { DB_RX(5, "Check RxD %p for OWN and EOF", r); DRV_BUF_FLUSH(r,DDI_DMA_SYNC_FORCPU) ; rbctrl = le32_to_cpu(CR_READ(r->rxd_rbctrl)); if (rbctrl & BMU_OWN) { NDD_TRACE("RHxE",r,rfsw,rbctrl) ; DB_RX(4, "End of RxDs"); goto rx_end ; } /* * out of RxD detection */ if (!rx_used) { SK_BREAK() ; SMT_PANIC(smc,HWM_E0009,HWM_E0009_MSG) ; /* Either we don't have an RxD or all * RxDs are filled. Therefore it's allowed * for to set the STOPPED flag */ smc->hw.hw_state = STOPPED ; mac_drv_clear_rx_queue(smc) ; smc->hw.hw_state = STARTED ; mac_drv_fill_rxd(smc) ; smc->os.hwm.detec_count = 0 ; goto rx_end ; } rfsw = le32_to_cpu(r->rxd_rfsw) ; if ((rbctrl & BMU_STF) != ((rbctrl & BMU_ST_BUF) <<5)) { /* * The BMU_STF bit is deleted, 1 frame is * placed into more than 1 rx buffer * * skip frame by setting the rx len to 0 * * if fragment count == 0 * The missing STF bit belongs to the * current frame, search for the * EOF bit to complete the frame * else * the fragment belongs to the next frame, * exit the loop and process the frame */ SK_BREAK() ; rfsw = 0 ; if (frag_count) { break ; } } n += rbctrl & 0xffff ; r = r->rxd_next ; frag_count++ ; rx_used-- ; } while (!(rbctrl & BMU_EOF)) ; used_frags = frag_count ; DB_RX(5, "EOF set in RxD, used_frags = %d", used_frags); /* may be next 2 DRV_BUF_FLUSH() can be skipped, because */ /* BMU_ST_BUF will not be changed by the ASIC */ DRV_BUF_FLUSH(r,DDI_DMA_SYNC_FORCPU) ; while (rx_used && !(r->rxd_rbctrl & cpu_to_le32(BMU_ST_BUF))) { DB_RX(5, "Check STF bit in %p", r); r = r->rxd_next ; DRV_BUF_FLUSH(r,DDI_DMA_SYNC_FORCPU) ; frag_count++ ; rx_used-- ; } DB_RX(5, "STF bit found"); /* * The received frame is finished for the process receive */ rxd = queue->rx_curr_get ; queue->rx_curr_get = r ; queue->rx_free += frag_count ; queue->rx_used = rx_used ; /* * ASIC Errata no. 7 (STF - Bit Bug) */ rxd->rxd_rbctrl &= cpu_to_le32(~BMU_STF) ; for (r=rxd, i=frag_count ; i ; r=r->rxd_next, i--){ DB_RX(5, "dma_complete for RxD %p", r); dma_complete(smc,(union s_fp_descr volatile *)r,DMA_WR); } smc->hw.fp.err_stats.err_valid++ ; smc->mib.m[MAC0].fddiMACCopied_Ct++ ; /* the length of the data including the FC */ len = (rfsw & RD_LENGTH) - 4 ; DB_RX(4, "frame length = %d", len); /* * check the frame_length and all error flags */ if (rfsw & (RX_MSRABT|RX_FS_E|RX_FS_CRC|RX_FS_IMPL)){ if (rfsw & RD_S_MSRABT) { DB_RX(2, "Frame aborted by the FORMAC"); smc->hw.fp.err_stats.err_abort++ ; } /* * check frame status */ if (rfsw & RD_S_SEAC2) { DB_RX(2, "E-Indicator set"); smc->hw.fp.err_stats.err_e_indicator++ ; } if (rfsw & RD_S_SFRMERR) { DB_RX(2, "CRC error"); smc->hw.fp.err_stats.err_crc++ ; } if (rfsw & RX_FS_IMPL) { DB_RX(2, "Implementer frame"); smc->hw.fp.err_stats.err_imp_frame++ ; } goto abort_frame ; } if (len > FDDI_RAW_MTU-4) { DB_RX(2, "Frame too long error"); smc->hw.fp.err_stats.err_too_long++ ; goto abort_frame ; } /* * SUPERNET 3 Bug: FORMAC delivers status words * of aborded frames to the BMU */ if (len <= 4) { DB_RX(2, "Frame length = 0"); goto abort_frame ; } if (len != (n-4)) { DB_RX(4, "BMU: rx len differs: [%d:%d]", len, n); smc->os.hwm.rx_len_error++ ; goto abort_frame ; } /* * Check SA == MA */ virt = (u_char far *) rxd->rxd_virt ; DB_RX(2, "FC = %x", *virt); if (virt[12] == MA[5] && virt[11] == MA[4] && virt[10] == MA[3] && virt[9] == MA[2] && virt[8] == MA[1] && (virt[7] & ~GROUP_ADDR_BIT) == MA[0]) { goto abort_frame ; } /* * test if LLC frame */ if (rfsw & RX_FS_LLC) { /* * if pass_llc_promisc is disable * if DA != Multicast or Broadcast or DA!=MA * abort the frame */ if (!smc->os.hwm.pass_llc_promisc) { if(!(virt[1] & GROUP_ADDR_BIT)) { if (virt[6] != MA[5] || virt[5] != MA[4] || virt[4] != MA[3] || virt[3] != MA[2] || virt[2] != MA[1] || virt[1] != MA[0]) { DB_RX(2, "DA != MA and not multi- or broadcast"); goto abort_frame ; } } } /* * LLC frame received */ DB_RX(4, "LLC - receive"); mac_drv_rx_complete(smc,rxd,frag_count,len) ; } else { if (!(mb = smt_get_mbuf(smc))) { smc->hw.fp.err_stats.err_no_buf++ ; DB_RX(4, "No SMbuf; receive terminated"); goto abort_frame ; } data = smtod(mb,char *) - 1 ; /* * copy the frame into a SMT_MBuf */ #ifdef USE_OS_CPY hwm_cpy_rxd2mb(rxd,data,len) ; #else for (r=rxd, i=used_frags ; i ; r=r->rxd_next, i--){ n = le32_to_cpu(r->rxd_rbctrl) & RD_LENGTH ; DB_RX(6, "cp SMT frame to mb: len = %d", n); memcpy(data,r->rxd_virt,n) ; data += n ; } data = smtod(mb,char *) - 1 ; #endif fc = *(char *)mb->sm_data = *data ; mb->sm_len = len - 1 ; /* len - fc */ data++ ; /* * SMT frame received */ switch(fc) { case FC_SMT_INFO : smc->hw.fp.err_stats.err_smt_frame++ ; DB_RX(5, "SMT frame received"); if (smc->os.hwm.pass_SMT) { DB_RX(5, "pass SMT frame"); mac_drv_rx_complete(smc, rxd, frag_count,len) ; } else { DB_RX(5, "requeue RxD"); mac_drv_requeue_rxd(smc,rxd,frag_count); } smt_received_pack(smc,mb,(int)(rfsw>>25)) ; break ; case FC_SMT_NSA : smc->hw.fp.err_stats.err_smt_frame++ ; DB_RX(5, "SMT frame received"); /* if pass_NSA set pass the NSA frame or */ /* pass_SMT set and the A-Indicator */ /* is not set, pass the NSA frame */ if (smc->os.hwm.pass_NSA || (smc->os.hwm.pass_SMT && !(rfsw & A_INDIC))) { DB_RX(5, "pass SMT frame"); mac_drv_rx_complete(smc, rxd, frag_count,len) ; } else { DB_RX(5, "requeue RxD"); mac_drv_requeue_rxd(smc,rxd,frag_count); } smt_received_pack(smc,mb,(int)(rfsw>>25)) ; break ; case FC_BEACON : if (smc->os.hwm.pass_DB) { DB_RX(5, "pass DB frame"); mac_drv_rx_complete(smc, rxd, frag_count,len) ; } else { DB_RX(5, "requeue RxD"); mac_drv_requeue_rxd(smc,rxd,frag_count); } smt_free_mbuf(smc,mb) ; break ; default : /* * unknown FC abord the frame */ DB_RX(2, "unknown FC error"); smt_free_mbuf(smc,mb) ; DB_RX(5, "requeue RxD"); mac_drv_requeue_rxd(smc,rxd,frag_count) ; if ((fc & 0xf0) == FC_MAC) smc->hw.fp.err_stats.err_mac_frame++ ; else smc->hw.fp.err_stats.err_imp_frame++ ; break ; } } DB_RX(3, "next RxD is %p", queue->rx_curr_get); NDD_TRACE("RHx1",queue->rx_curr_get,0,0) ; continue ; /*--------------------------------------------------------------------*/ abort_frame: DB_RX(5, "requeue RxD"); mac_drv_requeue_rxd(smc,rxd,frag_count) ; DB_RX(3, "next RxD is %p", queue->rx_curr_get); NDD_TRACE("RHx2",queue->rx_curr_get,0,0) ; } rx_end: #ifdef ALL_RX_COMPLETE mac_drv_all_receives_complete(smc) ; #endif return ; /* lint bug: needs return detect end of function */ }

Contributors

PersonTokensPropCommitsCommitProp
Linus Torvalds (pre-git)153395.16%120.00%
Joe Perches704.35%120.00%
Al Viro60.37%120.00%
Stephen Hemminger10.06%120.00%
Paulius Zaleckas10.06%120.00%
Total1611100.00%5100.00%


static void smt_to_llc(struct s_smc *smc, SMbuf *mb) { u_char fc ; DB_RX(4, "send a queued frame to the llc layer"); smc->os.hwm.r.len = mb->sm_len ; smc->os.hwm.r.mb_pos = smtod(mb,char *) ; fc = *smc->os.hwm.r.mb_pos ; (void)mac_drv_rx_init(smc,(int)mb->sm_len,(int)fc, smc->os.hwm.r.mb_pos,(int)mb->sm_len) ; smt_free_mbuf(smc,mb) ; }

Contributors

PersonTokensPropCommitsCommitProp
Linus Torvalds (pre-git)11096.49%133.33%
Joe Perches21.75%133.33%
Stephen Hemminger21.75%133.33%
Total114100.00%3100.00%

/* * BEGIN_MANUAL_ENTRY(hwm_rx_frag) * void hwm_rx_frag(smc,virt,phys,len,frame_status) * * function MACRO (hardware module, hwmtm.h) * This function calls dma_master for preparing the * system hardware for the DMA transfer and initializes * the current RxD with the length and the physical and * virtual address of the fragment. Furthermore, it sets the * STF and EOF bits depending on the frame status byte, * switches the OWN flag of the RxD, so that it is owned by the * adapter and issues an rx_start. * * para virt virtual pointer to the fragment * len the length of the fragment * frame_status status of the frame, see design description * * NOTE: It is possible to call this function with a fragment length * of zero. * * END_MANUAL_ENTRY */
void hwm_rx_frag(struct s_smc *smc, char far *virt, u_long phys, int len, int frame_status) { struct s_smt_fp_rxd volatile *r ; __le32 rbctrl; NDD_TRACE("RHfB",virt,len,frame_status) ; DB_RX(2, "hwm_rx_frag: len = %d, frame_status = %x", len, frame_status); r = smc->hw.fp.rx_q[QUEUE_R1].rx_curr_put ; r->rxd_virt = virt ; r->rxd_rbadr = cpu_to_le32(phys) ; rbctrl = cpu_to_le32( (((__u32)frame_status & (FIRST_FRAG|LAST_FRAG))<<26) | (((u_long) frame_status & FIRST_FRAG) << 21) | BMU_OWN | BMU_CHECK | BMU_EN_IRQ_EOF | len) ; r->rxd_rbctrl = rbctrl ; DRV_BUF_FLUSH(r,DDI_DMA_SYNC_FORDEV) ; outpd(ADDR(B0_R1_CSR),CSR_START) ; smc->hw.fp.rx_q[QUEUE_R1].rx_free-- ; smc->hw.fp.rx_q[QUEUE_R1].rx_used++ ; smc->hw.fp.rx_q[QUEUE_R1].rx_curr_put = r->rxd_next ; NDD_TRACE("RHfE",r,le32_to_cpu(r->rxd_rbadr),0) ; }

Contributors

PersonTokensPropCommitsCommitProp
Linus Torvalds (pre-git)19993.87%125.00%
Al Viro52.36%125.00%
Stephen Hemminger52.36%125.00%
Joe Perches31.42%125.00%
Total212100.00%4100.00%

/* * BEGINN_MANUAL_ENTRY(mac_drv_clear_rx_queue) * * void mac_drv_clear_rx_queue(smc) * struct s_smc *smc ; * * function DOWNCALL (hardware module, hwmtm.c) * mac_drv_clear_rx_queue is called by the OS-specific module * after it has issued a card_stop. * In this case, the frames in the receive queue are obsolete and * should be removed. For removing mac_drv_clear_rx_queue * calls dma_master for each RxD and mac_drv_clear_rxd for each * receive buffer. * * NOTE: calling sequence card_stop: * CLI_FBI(), card_stop(), * mac_drv_clear_tx_queue(), mac_drv_clear_rx_queue(), * * NOTE: The caller is responsible that the BMUs are idle * when this function is called. * * END_MANUAL_ENTRY */
void mac_drv_clear_rx_queue(struct s_smc *smc) { struct s_smt_fp_rxd volatile *r ; struct s_smt_fp_rxd volatile *next_rxd ; struct s_smt_rx_queue *queue ; int frag_count ; int i ; if (smc->hw.hw_state != STOPPED) { SK_BREAK() ; SMT_PANIC(smc,HWM_E0012,HWM_E0012_MSG) ; return ; } queue = smc->hw.fp.rx[QUEUE_R1] ; DB_RX(5, "clear_rx_queue"); /* * dma_complete and mac_drv_clear_rxd for all RxDs / receive buffers */ r = queue->rx_curr_get ; while (queue->rx_used) { DRV_BUF_FLUSH(r,DDI_DMA_SYNC_FORCPU) ; DB_RX(5, "switch OWN bit of RxD 0x%p", r); r->rxd_rbctrl &= ~cpu_to_le32(BMU_OWN) ; frag_count = 1 ; DRV_BUF_FLUSH(r,DDI_DMA_SYNC_FORDEV) ; r = r->rxd_next ; DRV_BUF_FLUSH(r,DDI_DMA_SYNC_FORCPU) ; while (r != queue->rx_curr_put && !(r->rxd_rbctrl & cpu_to_le32(BMU_ST_BUF))) { DB_RX(5, "Check STF bit in %p", r); r->rxd_rbctrl &= ~cpu_to_le32(BMU_OWN) ; DRV_BUF_FLUSH(r,DDI_DMA_SYNC_FORDEV) ; r = r->rxd_next ; DRV_BUF_FLUSH(r,DDI_DMA_SYNC_FORCPU) ; frag_count++ ; } DB_RX(5, "STF bit found"); next_rxd = r ; for (r=queue->rx_curr_get,i=frag_count; i ; r=r->rxd_next,i--){ DB_RX(5, "dma_complete for RxD %p", r); dma_complete(smc,(union s_fp_descr volatile *)r,DMA_WR); } DB_RX(5, "mac_drv_clear_rxd: RxD %p frag_count %d", queue->rx_curr_get, frag_count); mac_drv_clear_rxd(smc,queue->rx_curr_get,frag_count) ; queue->rx_curr_get = next_rxd ; queue->rx_used -= frag_count ; queue->rx_free += frag_count ; } }

Contributors

PersonTokensPropCommitsCommitProp
Linus Torvalds (pre-git)28792.88%125.00%
Joe Perches165.18%125.00%
Al Viro51.62%125.00%
Stephen Hemminger10.32%125.00%
Total309100.00%4100.00%

/* ------------------------------------------------------------- SEND FUNCTIONS: ------------------------------------------------------------- */ /* * BEGIN_MANUAL_ENTRY(hwm_tx_init) * int hwm_tx_init(smc,fc,frag_count,frame_len,frame_status) * * function DOWN_CALL (hardware module, hwmtm.c) * hwm_tx_init checks if the frame can be sent through the * corresponding send queue. * * para fc the frame control. To determine through which * send queue the frame should be transmitted. * 0x50 - 0x57: asynchronous LLC frame * 0xD0 - 0xD7: synchronous LLC frame * 0x41, 0x4F: SMT frame to the network * 0x42: SMT frame to the network and to the local SMT * 0x43: SMT frame to the local SMT * frag_count count of the fragments for this frame * frame_len length of the frame * frame_status status of the frame, the send queue bit is already * specified * * return frame_status * * END_MANUAL_ENTRY */
int hwm_tx_init(struct s_smc *smc, u_char fc, int frag_count, int frame_len, int frame_status) { NDD_TRACE("THiB",fc,frag_count,frame_len) ; smc->os.hwm.tx_p = smc->hw.fp.tx[frame_status & QUEUE_A0] ; smc->os.hwm.tx_descr = TX_DESCRIPTOR | (((u_long)(frame_len-1)&3)<<27) ; smc->os.hwm.tx_len = frame_len ; DB_TX(3, "hwm_tx_init: fc = %x, len = %d", fc, frame_len); if ((fc & ~(FC_SYNC_BIT|FC_LLC_PRIOR)) == FC_ASYNC_LLC) { frame_status |= LAN_TX ; } else { switch (fc) { case FC_SMT_INFO : case FC_SMT_NSA : frame_status |= LAN_TX ; break ; case FC_SMT_LOC : frame_status |= LOC_TX ; break ; case FC_SMT_LAN_LOC : frame_status |= LAN_TX | LOC_TX ; break ; default : SMT_PANIC(smc,HWM_E0010,HWM_E0010_MSG) ; } } if (!smc->hw.mac_ring_is_up) { frame_status &= ~LAN_TX ; frame_status |= RING_DOWN ; DB_TX(2, "Ring is down: terminate LAN_TX"); } if (frag_count > smc->os.hwm.tx_p->tx_free) { #ifndef NDIS_OS2 mac_drv_clear_txd(smc) ; if (frag_count > smc->os.hwm.tx_p->tx_free) { DB_TX(2, "Out of TxDs, terminate LAN_TX"); frame_status &= ~LAN_TX ; frame_status |= OUT_OF_TXD ; } #else DB_TX(2, "Out of TxDs, terminate LAN_TX"); frame_status &= ~LAN_TX ; frame_status |= OUT_OF_TXD ; #endif } DB_TX(3, "frame_status = %x", frame_status); NDD_TRACE("THiE",frame_status,smc->os.hwm.tx_p->tx_free,0) ; return frame_status; }

Contributors

PersonTokensPropCommitsCommitProp
Linus Torvalds (pre-git)29095.08%133.33%
Joe Perches103.28%133.33%
Stephen Hemminger51.64%133.33%
Total305100.00%3100.00%

/* * BEGIN_MANUAL_ENTRY(hwm_tx_frag) * void hwm_tx_frag(smc,virt,phys,len,frame_status) * * function DOWNCALL (hardware module, hwmtm.c) * If the frame should be sent to the LAN, this function calls * dma_master, fills the current TxD with the virtual and the * physical address, sets the STF and EOF bits dependent on * the frame status, and requests the BMU to start the * transmit. * If the frame should be sent to the local SMT, an SMT_MBuf * is allocated if the FIRST_FRAG bit is set in the frame_status. * The fragment of the frame is copied into the SMT MBuf. * The function smt_received_pack is called if the LAST_FRAG * bit is set in the frame_status word. * * para virt virtual pointer to the fragment * len the length of the fragment * frame_status status of the frame, see design description * * return nothing returned, no parameter is modified * * NOTE: It is possible to invoke this macro with a fragment length * of zero. * * END_MANUAL_ENTRY */
void hwm_tx_frag(struct s_smc *smc, char far *virt, u_long phys, int len, int frame_status) { struct s_smt_fp_txd volatile *t ; struct s_smt_tx_queue *queue ; __le32 tbctrl ; queue = smc->os.hwm.tx_p ; NDD_TRACE("THfB",virt,len,frame_status) ; /* Bug fix: AF / May 31 1999 (#missing) * snmpinfo problem reported by IBM is caused by invalid * t-pointer (txd) if LAN_TX is not set but LOC_TX only. * Set: t = queue->tx_curr_put here ! */ t = queue->tx_curr_put ; DB_TX(2, "hwm_tx_frag: len = %d, frame_status = %x", len, frame_status); if (frame_status & LAN_TX) { /* '*t' is already defined */ DB_TX(3, "LAN_TX: TxD = %p, virt = %p", t, virt); t->txd_virt = virt ; t->txd_txdscr = cpu_to_le32(smc->os.hwm.tx_descr) ; t->txd_tbadr = cpu_to_le32(phys) ; tbctrl = cpu_to_le32((((__u32)frame_status & (FIRST_FRAG|LAST_FRAG|EN_IRQ_EOF))<< 26) | BMU_OWN|BMU_CHECK |len) ; t->txd_tbctrl = tbctrl ; #ifndef AIX DRV_BUF_FLUSH(t,DDI_DMA_SYNC_FORDEV) ; outpd(queue->tx_bmu_ctl,CSR_START) ; #else /* ifndef AIX */ DRV_BUF_FLUSH(t,DDI_DMA_SYNC_FORDEV) ; if (frame_status & QUEUE_A0) { outpd(ADDR(B0_XA_CSR),CSR_START) ; } else { outpd(ADDR(B0_XS_CSR),CSR_START) ; } #endif queue->tx_free-- ; queue->tx_used++ ; queue->tx_curr_put = t->txd_next ; if (frame_status & LAST_FRAG) { smc->mib.m[MAC0].fddiMACTransmit_Ct++ ; } } if (frame_status & LOC_TX) { DB_TX(3, "LOC_TX:"); if (frame_status & FIRST_FRAG) { if(!(smc->os.hwm.tx_mb = smt_get_mbuf(smc))) { smc->hw.fp.err_stats.err_no_buf++ ; DB_TX(4, "No SMbuf; transmit terminated"); } else { smc->os.hwm.tx_data = smtod(smc->os.hwm.tx_mb,char *) - 1 ; #ifdef USE_OS_CPY #ifdef PASS_1ST_TXD_2_TX_COMP hwm_cpy_txd2mb(t,smc->os.hwm.tx_data, smc->os.hwm.tx_len) ; #endif #endif } } if (smc->os.hwm.tx_mb) { #ifndef USE_OS_CPY DB_TX(3, "copy fragment into MBuf"); memcpy(smc->os.hwm.tx_data,virt,len) ; smc->os.hwm.tx_data += len ; #endif if (frame_status & LAST_FRAG) { #ifdef USE_OS_CPY #ifndef PASS_1ST_TXD_2_TX_COMP /* * hwm_cpy_txd2mb(txd,data,len) copies 'len' * bytes from the virtual pointer in 'rxd' * to 'data'. The virtual pointer of the * os-specific tx-buffer should be written * in the LAST txd. */ hwm_cpy_txd2mb(t,smc->os.hwm.tx_data, smc->os.hwm.tx_len) ; #endif /* nPASS_1ST_TXD_2_TX_COMP */ #endif /* USE_OS_CPY */ smc->os.hwm.tx_data = smtod(smc->os.hwm.tx_mb,char *) - 1 ; *(char *)smc->os.hwm.tx_mb->sm_data = *smc->os.hwm.tx_data ; smc->os.hwm.tx_data++ ; smc->os.hwm.tx_mb->sm_len = smc->os.hwm.tx_len - 1 ; DB_TX(3, "pass LLC frame to SMT"); smt_received_pack(smc,smc->os.hwm.tx_mb, RD_FS_LOCAL) ; } } } NDD_TRACE("THfE",t,queue->tx_free,0) ; }

Contributors

PersonTokensPropCommitsCommitProp
Linus Torvalds (pre-git)56195.90%240.00%
Joe Perches142.39%120.00%
Al Viro50.85%120.00%
Stephen Hemminger50.85%120.00%
Total585100.00%5100.00%

/* * queues a receive for later send */
static void queue_llc_rx(struct s_smc *smc, SMbuf *mb) { DB_GEN(4, "queue_llc_rx: mb = %p", mb); smc->os.hwm.queued_rx_frames++ ; mb->sm_next = (SMbuf *)NULL ; if (smc->os.hwm.llc_rx_pipe == NULL) { smc->os.hwm.llc_rx_pipe = mb ; } else { smc->os.hwm.llc_rx_tail->sm_next = mb ; } smc->os.hwm.llc_rx_tail = mb ; /* * force an timer IRQ to receive the data */ if (!smc->os.hwm.isr_flag) { smt_force_irq(smc) ; } }

Contributors

PersonTokensPropCommitsCommitProp
Linus Torvalds (pre-git)10594.59%125.00%
Joe Perches32.70%125.00%
Stephen Hemminger21.80%125.00%
Al Viro10.90%125.00%
Total111100.00%4100.00%

/* * get a SMbuf from the llc_rx_queue */
static SMbuf *get_llc_rx(struct s_smc *smc) { SMbuf *mb ; if ((mb = smc->os.hwm.llc_rx_pipe)) { smc->os.hwm.queued_rx_frames-- ; smc->os.hwm.llc_rx_pipe = mb->sm_next ; } DB_GEN(4, "get_llc_rx: mb = 0x%p", mb); return mb; }

Contributors

PersonTokensPropCommitsCommitProp
Linus Torvalds (pre-git)6193.85%133.33%
Joe Perches34.62%133.33%
Stephen Hemminger11.54%133.33%
Total65100.00%3100.00%

/* * queues a transmit SMT MBuf during the time were the MBuf is * queued the TxD ring */
static void queue_txd_mb(struct s_smc *smc, SMbuf *mb) { DB_GEN(4, "_rx: queue_txd_mb = %p", mb); smc->os.hwm.queued_txd_mb++ ; mb->sm_next = (SMbuf *)NULL ; if (smc->os.hwm.txd_tx_pipe == NULL) { smc->os.hwm.txd_tx_pipe = mb ; } else { smc->os.hwm.txd_tx_tail->sm_next = mb ; } smc->os.hwm.txd_tx_tail = mb ; }

Contributors

PersonTokensPropCommitsCommitProp
Linus Torvalds (pre-git)8693.48%125.00%
Joe Perches33.26%125.00%
Stephen Hemminger22.17%125.00%
Al Viro11.09%125.00%
Total92100.00%4100.00%

/* * get a SMbuf from the txd_tx_queue */
static SMbuf *get_txd_mb(struct s_smc *smc) { SMbuf *mb ; if ((mb = smc->os.hwm.txd_tx_pipe)) { smc->os.hwm.queued_txd_mb-- ; smc->os.hwm.txd_tx_pipe = mb->sm_next ; } DB_GEN(4, "get_txd_mb: mb = 0x%p", mb); return mb; }

Contributors

PersonTokensPropCommitsCommitProp
Linus Torvalds (pre-git)6193.85%133.33%
Joe Perches34.62%133.33%
Stephen Hemminger11.54%133.33%
Total65100.00%3100.00%

/* * SMT Send function */
void smt_send_mbuf(struct s_smc *smc, SMbuf *mb, int fc) { char far *data ; int len ; int n ; int i ; int frag_count ; int frame_status ; SK_LOC_DECL(char far,*virt[3]) ; int frag_len[3] ; struct s_smt_tx_queue *queue ; struct s_smt_fp_txd volatile *t ; u_long phys ; __le32 tbctrl; NDD_TRACE("THSB",mb,fc,0) ; DB_TX(4, "smt_send_mbuf: mb = 0x%p, fc = 0x%x", mb, fc); mb->sm_off-- ; /* set to fc */ mb->sm_len++ ; /* + fc */ data = smtod(mb,char *) ; *data = fc ; if (fc == FC_SMT_LOC) *data = FC_SMT_INFO ; /* * determine the frag count and the virt addresses of the frags */ frag_count = 0 ; len = mb->sm_len ; while (len) { n = SMT_PAGESIZE - ((long)data & (SMT_PAGESIZE-1)) ; if (n >= len) { n = len ; } DB_TX(5, "frag: virt/len = 0x%p/%d", data, n); virt[frag_count] = data ; frag_len[frag_count] = n ; frag_count++ ; len -= n ; data += n ; } /* * determine the frame status */ queue = smc->hw.fp.tx[QUEUE_A0] ; if (fc == FC_BEACON || fc == FC_SMT_LOC) { frame_status = LOC_TX ; } else { frame_status = LAN_TX ; if ((smc->os.hwm.pass_NSA &&(fc == FC_SMT_NSA)) || (smc->os.hwm.pass_SMT &&(fc == FC_SMT_INFO))) frame_status |= LOC_TX ; } if (!smc->hw.mac_ring_is_up || frag_count > queue->tx_free) { frame_status &= ~LAN_TX; if (frame_status) { DB_TX(2, "Ring is down: terminate LAN_TX"); } else { DB_TX(2, "Ring is down: terminate transmission"); smt_free_mbuf(smc,mb) ; return ; } } DB_TX(5, "frame_status = 0x%x", frame_status); if ((frame_status & LAN_TX) && (frame_status & LOC_TX)) { mb->sm_use_count = 2 ; } if (frame_status & LAN_TX) { t = queue->tx_curr_put ; frame_status |= FIRST_FRAG ; for (i = 0; i < frag_count; i++) { DB_TX(5, "init TxD = 0x%p", t); if (i == frag_count-1) { frame_status |= LAST_FRAG ; t->txd_txdscr = cpu_to_le32(TX_DESCRIPTOR | (((__u32)(mb->sm_len-1)&3) << 27)) ; } t->txd_virt = virt[i] ; phys = dma_master(smc, (void far *)virt[i], frag_len[i], DMA_RD|SMT_BUF) ; t->txd_tbadr = cpu_to_le32(phys) ; tbctrl = cpu_to_le32((((__u32)frame_status & (FIRST_FRAG|LAST_FRAG)) << 26) | BMU_OWN | BMU_CHECK | BMU_SMT_TX |frag_len[i]) ; t->txd_tbctrl = tbctrl ; #ifndef AIX DRV_BUF_FLUSH(t,DDI_DMA_SYNC_FORDEV) ; outpd(queue->tx_bmu_ctl,CSR_START) ; #else DRV_BUF_FLUSH(t,DDI_DMA_SYNC_FORDEV) ; outpd(ADDR(B0_XA_CSR),CSR_START) ; #endif frame_status &= ~FIRST_FRAG ; queue->tx_curr_put = t = t->txd_next ; queue->tx_free-- ; queue->tx_used++ ; } smc->mib.m[MAC0].fddiMACTransmit_Ct++ ; queue_txd_mb(smc,mb) ; } if (frame_status & LOC_TX) { DB_TX(5, "pass Mbuf to LLC queue"); queue_llc_rx(smc,mb) ; } /* * We need to unqueue the free SMT_MBUFs here, because it may * be that the SMT want's to send more than 1 frame for one down call */ mac_drv_clear_txd(smc) ; NDD_TRACE("THSE",t,queue->tx_free,frag_count) ; }

Contributors

PersonTokensPropCommitsCommitProp
Linus Torvalds (pre-git)63195.32%228.57%
Joe Perches172.57%114.29%
Al Viro60.91%114.29%
Randy Dunlap40.60%114.29%
Stephen Hemminger30.45%114.29%
Colin Ian King10.15%114.29%
Total662100.00%7100.00%

/* BEGIN_MANUAL_ENTRY(mac_drv_clear_txd) * void mac_drv_clear_txd(smc) * * function DOWNCALL (hardware module, hwmtm.c) * mac_drv_clear_txd searches in both send queues for TxD's * which were finished by the adapter. It calls dma_complete * for each TxD. If the last fragment of an LLC frame is * reached, it calls mac_drv_tx_complete to release the * send buffer. * * return nothing * * END_MANUAL_ENTRY */
static void mac_drv_clear_txd(struct s_smc *smc) { struct s_smt_tx_queue *queue ; struct s_smt_fp_txd volatile *t1 ; struct s_smt_fp_txd volatile *t2 = NULL ; SMbuf *mb ; u_long tbctrl ; int i ; int frag_count ; int n ; NDD_TRACE("THcB",0,0,0) ; for (i = QUEUE_S; i <= QUEUE_A0; i++) { queue = smc->hw.fp.tx[i] ; t1 = queue->tx_curr_get ; DB_TX(5, "clear_txd: QUEUE = %d (0=sync/1=async)", i); for ( ; ; ) { frag_count = 0 ; do { DRV_BUF_FLUSH(t1,DDI_DMA_SYNC_FORCPU) ; DB_TX(5, "check OWN/EOF bit of TxD 0x%p", t1); tbctrl = le32_to_cpu(CR_READ(t1->txd_tbctrl)); if (tbctrl & BMU_OWN || !queue->tx_used){ DB_TX(4, "End of TxDs queue %d", i); goto free_next_queue ; /* next queue */ } t1 = t1->txd_next ; frag_count++ ; } while (!(tbctrl & BMU_EOF)) ; t1 = queue->tx_curr_get ; for (n = frag_count; n; n--) { tbctrl = le32_to_cpu(t1->txd_tbctrl) ; dma_complete(smc, (union s_fp_descr volatile *) t1, (int) (DMA_RD | ((tbctrl & BMU_SMT_TX) >> 18))) ; t2 = t1 ; t1 = t1->txd_next ; } if (tbctrl & BMU_SMT_TX) { mb = get_txd_mb(smc) ; smt_free_mbuf(smc,mb) ; } else { #ifndef PASS_1ST_TXD_2_TX_COMP DB_TX(4, "mac_drv_tx_comp for TxD 0x%p", t2); mac_drv_tx_complete(smc,t2) ; #else DB_TX(4, "mac_drv_tx_comp for TxD 0x%x", queue->tx_curr_get); mac_drv_tx_complete(smc,queue->tx_curr_get) ; #endif } queue->tx_curr_get = t1 ; queue->tx_free += frag_count ; queue->tx_used -= frag_count ; } free_next_queue: ; } NDD_TRACE("THcE",0,0,0) ; }

Contributors

PersonTokensPropCommitsCommitProp
Linus Torvalds (pre-git)33794.93%114.29%
Joe Perches102.82%114.29%
Al Viro41.13%228.57%
Colin Ian King20.56%114.29%
Stephen Hemminger10.28%114.29%
Adrian Bunk10.28%114.29%
Total355100.00%7100.00%

/* * BEGINN_MANUAL_ENTRY(mac_drv_clear_tx_queue) * * void mac_drv_clear_tx_queue(smc) * struct s_smc *smc ; * * function DOWNCALL (hardware module, hwmtm.c) * mac_drv_clear_tx_queue is called from the SMT when * the RMT state machine has entered the ISOLATE state. * This function is also called by the os-specific module * after it has called the function card_stop(). * In this case, the frames in the send queues are obsolete and * should be removed. * * note calling sequence: * CLI_FBI(), card_stop(), * mac_drv_clear_tx_queue(), mac_drv_clear_rx_queue(), * * NOTE: The caller is responsible that the BMUs are idle * when this function is called. * * END_MANUAL_ENTRY */
void mac_drv_clear_tx_queue(struct s_smc *smc) { struct s_smt_fp_txd volatile *t ; struct s_smt_tx_queue *queue ; int tx_used ; int i ; if (smc->hw.hw_state != STOPPED) { SK_BREAK() ; SMT_PANIC(smc,HWM_E0011,HWM_E0011_MSG) ; return ; } for (i = QUEUE_S; i <= QUEUE_A0; i++) { queue = smc->hw.fp.tx[i] ; DB_TX(5, "clear_tx_queue: QUEUE = %d (0=sync/1=async)", i); /* * switch the OWN bit of all pending frames to the host */ t = queue->tx_curr_get ; tx_used = queue->tx_used ; while (tx_used) { DRV_BUF_FLUSH(t,DDI_DMA_SYNC_FORCPU) ; DB_TX(5, "switch OWN bit of TxD 0x%p", t); t->txd_tbctrl &= ~cpu_to_le32(BMU_OWN) ; DRV_BUF_FLUSH(t,DDI_DMA_SYNC_FORDEV) ; t = t->txd_next ; tx_used-- ; } } /* * release all TxD's for both send queues */ mac_drv_clear_txd(smc) ; for (i = QUEUE_S; i <= QUEUE_A0; i++) { queue = smc->hw.fp.tx[i] ; t = queue->tx_curr_get ; /* * write the phys pointer of the NEXT descriptor into the * BMU's current address descriptor pointer and set * tx_curr_get and tx_curr_put to this position */ if (i == QUEUE_S) { outpd(ADDR(B5_XS_DA),le32_to_cpu(t->txd_ntdadr)) ; } else { outpd(ADDR(B5_XA_DA),le32_to_cpu(t->txd_ntdadr)) ; } queue->tx_curr_put = queue->tx_curr_get->txd_next ; queue->tx_curr_get = queue->tx_curr_put ; } }

Contributors

PersonTokensPropCommitsCommitProp
Linus Torvalds (pre-git)24096.00%125.00%
Joe Perches52.00%125.00%
Al Viro41.60%125.00%
Stephen Hemminger10.40%125.00%
Total250100.00%4100.00%

/* ------------------------------------------------------------- TEST FUNCTIONS: ------------------------------------------------------------- */ #ifdef DEBUG /* * BEGIN_MANUAL_ENTRY(mac_drv_debug_lev) * void mac_drv_debug_lev(smc,flag,lev) * * function DOWNCALL (drvsr.c) * To get a special debug info the user can assign a debug level * to any debug flag. * * para flag debug flag, possible values are: * = 0: reset all debug flags (the defined level is * ignored) * = 1: debug.d_smtf * = 2: debug.d_smt * = 3: debug.d_ecm * = 4: debug.d_rmt * = 5: debug.d_cfm * = 6: debug.d_pcm * * = 10: debug.d_os.hwm_rx (hardware module receive path) * = 11: debug.d_os.hwm_tx(hardware module transmit path) * = 12: debug.d_os.hwm_gen(hardware module general flag) * * lev debug level * * END_MANUAL_ENTRY */
void mac_drv_debug_lev(struct s_smc *smc, int flag, int lev) { switch(flag) { case (int)NULL: DB_P.d_smtf = DB_P.d_smt = DB_P.d_ecm = DB_P.d_rmt = 0 ; DB_P.d_cfm = 0 ; DB_P.d_os.hwm_rx = DB_P.d_os.hwm_tx = DB_P.d_os.hwm_gen = 0 ; #ifdef SBA DB_P.d_sba = 0 ; #endif #ifdef ESS DB_P.d_ess = 0 ; #endif break ; case DEBUG_SMTF: DB_P.d_smtf = lev ; break ; case DEBUG_SMT: DB_P.d_smt = lev ; break ; case DEBUG_ECM: DB_P.d_ecm = lev ; break ; case DEBUG_RMT: DB_P.d_rmt = lev ; break ; case DEBUG_CFM: DB_P.d_cfm = lev ; break ; case DEBUG_PCM: DB_P.d_pcm = lev ; break ; case DEBUG_SBA: #ifdef SBA DB_P.d_sba = lev ; #endif break ; case DEBUG_ESS: #ifdef ESS DB_P.d_ess = lev ; #endif break ; case DB_HWM_RX: DB_P.d_os.hwm_rx = lev ; break ; case DB_HWM_TX: DB_P.d_os.hwm_tx = lev ; break ; case DB_HWM_GEN: DB_P.d_os.hwm_gen = lev ; break ; default: break ; } }

Contributors

PersonTokensPropCommitsCommitProp
Linus Torvalds (pre-git)22098.65%150.00%
Stephen Hemminger31.35%150.00%
Total223100.00%2100.00%

#endif

Overall Contributors

PersonTokensPropCommitsCommitProp
Linus Torvalds (pre-git)811291.00%215.38%
Stephen Hemminger5155.78%17.69%
Joe Perches2122.38%215.38%
Al Viro550.62%323.08%
Adrian Bunk110.12%17.69%
Randy Dunlap40.04%17.69%
Colin Ian King30.03%17.69%
Lucas De Marchi10.01%17.69%
Paulius Zaleckas10.01%17.69%
Total8914100.00%13100.00%
Information contained on this website is for historical information purposes only and does not indicate or represent copyright ownership.
Created with cregit.