From mboxrd@z Thu Jan 1 00:00:00 1970 From: Nishanth Aravamudan Date: Wed, 15 Sep 2004 18:13:30 +0000 Subject: [Kernel-janitors] [PATCH 3/5] atm/he: replace schedule_timeout() Message-Id: <20040915181330.GD4363@us.ibm.com> MIME-Version: 1 Content-Type: multipart/mixed; boundary="===============50988199177227034==" List-Id: To: kernel-janitors@vger.kernel.org --===============50988199177227034== Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Any comments would be appreciated. This is a re-push of a patch I sent 19 July but the changes haven't appared in either 2.6.9-rc1-mm5 or 2.6.9-rc2. Description: msleep() is used instead of schedule_timeout() to guarantee the task delays as expected. Also, an extraneous set_current_state() is removed, as the task is guaranteed to be in TASK_RUNNING upon returning from schedule_timeout(). Signed-off-by: Nishanth Aravamudan --- 2.6.9-rc2-vanilla/drivers/atm/he.c 2004-09-13 17:16:06.000000000 -0700 +++ 2.6.9-rc2/drivers/atm/he.c 2004-09-14 11:25:55.000000000 -0700 @@ -2569,7 +2569,6 @@ he_close(struct atm_vcc *vcc) timeout = schedule_timeout(30*HZ); remove_wait_queue(&he_vcc->rx_waitq, &wait); - set_current_state(TASK_RUNNING); if (timeout == 0) hprintk("close rx timeout cid 0x%x\n", cid); @@ -2595,8 +2594,7 @@ he_close(struct atm_vcc *vcc) while (((tx_inuse = atomic_read(&vcc->sk->sk_wmem_alloc)) > 0) && (retry < MAX_RETRY)) { - set_current_state(TASK_UNINTERRUPTIBLE); - (void) schedule_timeout(sleep); + msleep(jiffies_to_msecs(sleep)); if (sleep < HZ) sleep = sleep * 2; --===============50988199177227034== 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 --===============50988199177227034==--