From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail.linuxfoundation.org ([140.211.169.12]:56160 "EHLO mail.linuxfoundation.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S936363AbeFRIZD (ORCPT ); Mon, 18 Jun 2018 04:25:03 -0400 From: Greg Kroah-Hartman To: linux-kernel@vger.kernel.org Cc: Greg Kroah-Hartman , stable@vger.kernel.org, Mike Marciniszyn , "Michael J. Ruhl" , Sebastian Sanchez , Dennis Dalessandro , Doug Ledford , Sasha Levin Subject: [PATCH 4.16 175/279] IB/hfi1: Fix memory leak in exception path in get_irq_affinity() Date: Mon, 18 Jun 2018 10:12:40 +0200 Message-Id: <20180618080616.155373602@linuxfoundation.org> In-Reply-To: <20180618080608.851973560@linuxfoundation.org> References: <20180618080608.851973560@linuxfoundation.org> MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Sender: stable-owner@vger.kernel.org List-ID: 4.16-stable review patch. If anyone has any objections, please let me know. ------------------ From: Sebastian Sanchez [ Upstream commit 59482a14918b282ca2a98f38c69da5ebeb1107d2 ] When IRQ affinity is set and the interrupt type is unknown, a cpu mask allocated within the function is never freed. Fix this memory leak by allocating memory within the scope where it is used. Reviewed-by: Mike Marciniszyn Reviewed-by: Michael J. Ruhl Signed-off-by: Sebastian Sanchez Signed-off-by: Dennis Dalessandro Signed-off-by: Doug Ledford Signed-off-by: Sasha Levin Signed-off-by: Greg Kroah-Hartman --- drivers/infiniband/hw/hfi1/affinity.c | 11 +++++------ 1 file changed, 5 insertions(+), 6 deletions(-) --- a/drivers/infiniband/hw/hfi1/affinity.c +++ b/drivers/infiniband/hw/hfi1/affinity.c @@ -412,7 +412,6 @@ static void hfi1_cleanup_sdma_notifier(s static int get_irq_affinity(struct hfi1_devdata *dd, struct hfi1_msix_entry *msix) { - int ret; cpumask_var_t diff; struct hfi1_affinity_node *entry; struct cpu_mask_set *set = NULL; @@ -424,10 +423,6 @@ static int get_irq_affinity(struct hfi1_ extra[0] = '\0'; cpumask_clear(&msix->mask); - ret = zalloc_cpumask_var(&diff, GFP_KERNEL); - if (!ret) - return -ENOMEM; - entry = node_affinity_lookup(dd->node); switch (msix->type) { @@ -458,6 +453,9 @@ static int get_irq_affinity(struct hfi1_ * finds its CPU here. */ if (cpu == -1 && set) { + if (!zalloc_cpumask_var(&diff, GFP_KERNEL)) + return -ENOMEM; + if (cpumask_equal(&set->mask, &set->used)) { /* * We've used up all the CPUs, bump up the generation @@ -469,6 +467,8 @@ static int get_irq_affinity(struct hfi1_ cpumask_andnot(diff, &set->mask, &set->used); cpu = cpumask_first(diff); cpumask_set_cpu(cpu, &set->used); + + free_cpumask_var(diff); } cpumask_set_cpu(cpu, &msix->mask); @@ -482,7 +482,6 @@ static int get_irq_affinity(struct hfi1_ hfi1_setup_sdma_notifier(msix); } - free_cpumask_var(diff); return 0; }