can bcm: fix incomplete tx_setup fix
Subsystems (3)
Oliver Hartkopp
maintainerPATCH 1/1 can bcm: fix incomplete tx_setup fix
The commit aabdcb0b553b9c9547b1a506b34d55a764745870 ("can bcm: fix tx_setup off-by-one errors") fixed only a part of the original problem reported by Andre Naujoks. It turned out that the original code needed to be re-ordered to reduce complexity and to finally fix the reported frame counting issues.
---
@Andre: I checked the changes (which make the code much clearer now) against your test program and some of mine i modified to test all flags like STARTTIMER, SETTIMER, TX_ANNOUNCE in different configurations. Please double check with your setup and send a Tested-by: if it is ok for you.
@Dave: Again sorry for this second patch. The first patch looked sane and fixed the reported issue - but obviously not the whole problem :-(
bcm.c | 48 +++++++++++++++++++++--------------------------- 1 file changed, 21 insertions(+), 27 deletions(-)
net/can/bcm.c +21 -27
@@ -344,6 +344,18 @@ static void bcm_send_to_user(struct bcm_op *op, struct bcm_msg_head *head, } } +static void bcm_tx_start_timer (struct bcm_op *op) +{ +if (op->kt_ival1.tv64 && op->count) +hrtimer_start (&op->timer, +ktime_add (ktime_get (), op->kt_ival1), +HRTIMER_MODE_ABS ); +else if (op->kt_ival2.tv64) +hrtimer_start (&op->timer, +ktime_add (ktime_get (), op->kt_ival2), +HRTIMER_MODE_ABS ); +} +static void bcm_tx_timeout_tsklet (unsigned long data) {struct bcm_op *op = (struct bcm_op *)data;
@@ -365,23 +377,12 @@ static void bcm_tx_timeout_tsklet(unsigned long data)bcm_send_to_user (op, &msg_head,NULL ,0 ); } - -/* send (next) frame */ bcm_can_tx (op); -hrtimer_start (&op->timer, -ktime_add (ktime_get (), op->kt_ival1), -HRTIMER_MODE_ABS ); - }else { -if (op->kt_ival2.tv64) { + }else if (op->kt_ival2.tv64) +bcm_can_tx (op); -/* send (next) frame */ -bcm_can_tx (op); -hrtimer_start (&op->timer, -ktime_add (ktime_get (), op->kt_ival2), -HRTIMER_MODE_ABS ); - } - } +bcm_tx_start_timer (op); }/*
@@ -961,28 +962,21 @@ static int bcm_tx_setup(struct bcm_msg_head *msg_head, struct msghdr *msg,hrtimer_cancel (&op->timer); } -if ((op->flags &STARTTIMER ) && - ((op->kt_ival1.tv64 && op->count) || op->kt_ival2.tv64)) { - +if (op->flags &STARTTIMER ) { +hrtimer_cancel (&op->timer);/* spec: send can_frame when starting timer */ op->flags |=TX_ANNOUNCE ; - -/* only start timer when having more frames than sent below */ -if (op->kt_ival1.tv64 && (op->count >1 )) { -/* op->count-- is done in bcm_tx_timeout_tsklet */ -hrtimer_start (&op->timer, op->kt_ival1, -HRTIMER_MODE_REL ); - }else -hrtimer_start (&op->timer, op->kt_ival2, -HRTIMER_MODE_REL ); }if (op->flags &TX_ANNOUNCE ) {bcm_can_tx (op); -if (op->kt_ival1.tv64 && (op->count >0 )) +if (op->count) op->count--; } +if (op->flags &STARTTIMER ) +bcm_tx_start_timer (op); +return msg_head->nframes *CFSIZ +MHSIZ ; }
From: Oliver Hartkopp <[email protected]>
Date: Thu, 29 Sep 2011 21:19:21 +0200
Quoted 6 lines
The commit
aabdcb0b553b9c9547b1a506b34d55a764745870("can bcm: fixtx_setupoff-by-one errors") fixed only a part of the original problem reported by Andre Naujoks. It turned out that the original code needed to be re-ordered to reduce complexity and to finally fix the reported frame counting issues.Signed-off-by Oliver Hartkopp <[email protected]>
Applied, thanks.