From mboxrd@z Thu Jan 1 00:00:00 1970 From: Nishanth Aravamudan Date: Mon, 24 Jan 2005 22:50:12 +0000 Subject: [KJ] [UPDATE PATCH 2/40] atm/zatm: replace sleep_on() with Message-Id: <20050124225012.GE2685@us.ibm.com> MIME-Version: 1 Content-Type: multipart/mixed; boundary="===============45453578141360951==" List-Id: To: kernel-janitors@vger.kernel.org --===============45453578141360951== Content-Type: text/plain; charset=us-ascii Content-Disposition: inline On Thu, Jan 20, 2005 at 10:17:42AM -0800, Nishanth Aravamudan wrote: > Hi, > > Please consider applying. > > Description: Use wait_event() instead of the deprecated sleep_on(). The first > replacement is relatively non-controversial, as the message is only printed > once, and still is. In the second, though, the message is printed every > iteration of the loop. This functionality has been lost with wait_event(). > Removed the unused variable once (which made sure that certain things were done > on the first iteration of the loop, but not after; now there is only one > "iteration"). Also fixed some strange tabbing. > > Signed-off-by: Nishanth Aravamudan > > --- 2.6.11-rc1-kj-v/drivers/atm/zatm.c 2005-01-15 16:55:43.000000000 -0800 > +++ 2.6.11-rc1-kj/drivers/atm/zatm.c 2005-01-19 17:41:21.000000000 -0800 > + wait_event(zatm_vcc->tx_wait, !skb_peek(&zatm_vcc->backlog)); > + printk("waiting for TX queue to drain ... %p\n",skb); This line results in a compile failure. Fixed (sort of) below... Description: Use wait_event() instead of the deprecated sleep_on(). The first replacement is relatively non-controversial, as the message is only printed once, and still is. In the second, though, the message is printed every iteration of the loop. This functionality has been lost with wait_event(). Removed the unused variable once (which made sure that certain things were done on the first iteration of the loop, but not after; now there is only one "iteration"). Also fixed some strange tabbing. The final printk had to be modified as well, since it references a pointer which has not yet been assigned. Since the printk only happens once now (instead of repeatedly in a loop as before), I think this should be ok. Signed-off-by: Nishanth Aravamudan --- 2.6.11-rc1-kj-v/drivers/atm/zatm.c 2005-01-15 16:55:43.000000000 -0800 +++ 2.6.11-rc1-kj/drivers/atm/zatm.c 2005-01-19 17:41:21.000000000 -0800 @@ -22,6 +22,7 @@ #include #include #include +#include #include #include #include @@ -867,32 +868,19 @@ static void close_tx(struct atm_vcc *vcc struct zatm_vcc *zatm_vcc; unsigned long flags; int chan; -struct sk_buff *skb; -int once = 1; + struct sk_buff *skb; zatm_vcc = ZATM_VCC(vcc); zatm_dev = ZATM_DEV(vcc->dev); chan = zatm_vcc->tx_chan; if (!chan) return; DPRINTK("close_tx\n"); - while (skb_peek(&zatm_vcc->backlog)) { -if (once) { -printk("waiting for backlog to drain ...\n"); -event_dump(); -once = 0; -} - sleep_on(&zatm_vcc->tx_wait); - } -once = 1; - while ((skb = skb_peek(&zatm_vcc->tx_queue))) { -if (once) { -printk("waiting for TX queue to drain ... %p\n",skb); -event_dump(); -once = 0; -} - DPRINTK("waiting for TX queue to drain ... %p\n",skb); - sleep_on(&zatm_vcc->tx_wait); - } + printk("waiting for backlog to drain ...\n"); + event_dump(); + wait_event(zatm_vcc->tx_wait, !skb_peek(&zatm_vcc->backlog)); + printk("waiting for TX queue to drain ...\n"); + event_dump(); + wait_event(zatm_vcc->tx_wait, !(skb = skb_peek(&zatm_vcc->tx_queue))); spin_lock_irqsave(&zatm_dev->lock, flags); #if 0 zwait; --===============45453578141360951== Content-Type: text/plain; charset="iso-8859-1" MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Content-Disposition: inline _______________________________________________ Kernel-janitors mailing list Kernel-janitors@lists.osdl.org http://lists.osdl.org/mailman/listinfo/kernel-janitors --===============45453578141360951==--