From mboxrd@z Thu Jan 1 00:00:00 1970 From: Xiaotian Feng Subject: [PATCH] mmc: s3cmci: use tasklet_kill in device remove/release process Date: Wed, 31 Oct 2012 18:48:45 +0800 Message-ID: <1351680525-27253-1-git-send-email-xtfeng@gmail.com> Return-path: Received: from mail-da0-f46.google.com ([209.85.210.46]:37677 "EHLO mail-da0-f46.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752283Ab2JaKsI (ORCPT ); Wed, 31 Oct 2012 06:48:08 -0400 Sender: linux-mmc-owner@vger.kernel.org List-Id: linux-mmc@vger.kernel.org To: linux-kernel@vger.kernel.org Cc: Xiaotian Feng , Xiaotian Feng , Ben Dooks , Chris Ball , linux-arm-kernel@lists.infradead.org, linux-mmc@vger.kernel.org Some driver uses tasklet_disable in device remove/release process, tasklet_disable will inc tasklet->count and return. If the tasklet is not handled yet under some softirq pressure, the tasklet will be placed on the tasklet_vec, never have a chance to be excuted. This might lead to a heavy loaded ksoftirqd, wakeup with pending_softirq, but tasklet is disabled. tasklet_kill should be used in this case. Signed-off-by: Xiaotian Feng Cc: Ben Dooks Cc: Chris Ball Cc: linux-arm-kernel@lists.infradead.org Cc: linux-mmc@vger.kernel.org --- drivers/mmc/host/s3cmci.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/mmc/host/s3cmci.c b/drivers/mmc/host/s3cmci.c index 4638dda..d70d0cc 100644 --- a/drivers/mmc/host/s3cmci.c +++ b/drivers/mmc/host/s3cmci.c @@ -1830,7 +1830,7 @@ static int __devexit s3cmci_remove(struct platform_device *pdev) clk_put(host->clk); - tasklet_disable(&host->pio_tasklet); + tasklet_kill(&host->pio_tasklet); if (s3cmci_host_usedma(host)) s3c2410_dma_free(host->dma, &s3cmci_dma_client); -- 1.7.9.5 From mboxrd@z Thu Jan 1 00:00:00 1970 From: xtfeng@gmail.com (Xiaotian Feng) Date: Wed, 31 Oct 2012 18:48:45 +0800 Subject: [PATCH] mmc: s3cmci: use tasklet_kill in device remove/release process Message-ID: <1351680525-27253-1-git-send-email-xtfeng@gmail.com> To: linux-arm-kernel@lists.infradead.org List-Id: linux-arm-kernel.lists.infradead.org Some driver uses tasklet_disable in device remove/release process, tasklet_disable will inc tasklet->count and return. If the tasklet is not handled yet under some softirq pressure, the tasklet will be placed on the tasklet_vec, never have a chance to be excuted. This might lead to a heavy loaded ksoftirqd, wakeup with pending_softirq, but tasklet is disabled. tasklet_kill should be used in this case. Signed-off-by: Xiaotian Feng Cc: Ben Dooks Cc: Chris Ball Cc: linux-arm-kernel at lists.infradead.org Cc: linux-mmc at vger.kernel.org --- drivers/mmc/host/s3cmci.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/mmc/host/s3cmci.c b/drivers/mmc/host/s3cmci.c index 4638dda..d70d0cc 100644 --- a/drivers/mmc/host/s3cmci.c +++ b/drivers/mmc/host/s3cmci.c @@ -1830,7 +1830,7 @@ static int __devexit s3cmci_remove(struct platform_device *pdev) clk_put(host->clk); - tasklet_disable(&host->pio_tasklet); + tasklet_kill(&host->pio_tasklet); if (s3cmci_host_usedma(host)) s3c2410_dma_free(host->dma, &s3cmci_dma_client); -- 1.7.9.5