linux-kernel - Re: [PATCH] dmaengine: ti: Fix refcount leak in ti_dra7_xbar_route

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-ID: <b659bba8-0de9-4159-4311-bb13036bef05@gmail.com>
Date:   Mon, 16 May 2022 13:56:39 +0800
From:   Miaoqian Lin <linmq006@...il.com>
To:     Dave Jiang <dave.jiang@...el.com>,
        Peter Ujfalusi <peter.ujfalusi@...il.com>,
        Vinod Koul <vkoul@...nel.org>, dmaengine@...r.kernel.org,
        linux-kernel@...r.kernel.org
Subject: Re: [PATCH] dmaengine: ti: Fix refcount leak in
 ti_dra7_xbar_route_allocate

Hi, Dave

On 2022/5/13 1:03, Dave Jiang wrote:
>
> On 5/11/2022 10:18 PM, Miaoqian Lin wrote:
>> 1. of_find_device_by_node() takes reference, we should use put_device()
>> to release it when not need anymore.
>> 2. of_parse_phandle() returns a node pointer with refcount
>> incremented, we should use of_node_put() on it when not needed anymore.
>>
>> Add put_device() and of_node_put() in some error paths to fix.
> Sounds like you need 2 patches for this? One just for the put_device() and the other for the of_node_put()?

Thanks for your response, I will split it into 2 patches.


>>
>> Fixes: ec9bfa1e1a79 ("dmaengine: ti-dma-crossbar: dra7: Use bitops instead of idr")
>> Fixes: a074ae38f859 ("dmaengine: Add driver for TI DMA crossbar on DRA7x")
>> Signed-off-by: Miaoqian Lin <linmq006@...il.com>
>> ---
>>   drivers/dma/ti/dma-crossbar.c | 5 +++++
>>   1 file changed, 5 insertions(+)
>>
>> diff --git a/drivers/dma/ti/dma-crossbar.c b/drivers/dma/ti/dma-crossbar.c
>> index 71d24fc07c00..f744ddbbbad7 100644
>> --- a/drivers/dma/ti/dma-crossbar.c
>> +++ b/drivers/dma/ti/dma-crossbar.c
>> @@ -245,6 +245,7 @@ static void *ti_dra7_xbar_route_allocate(struct of_phandle_args *dma_spec,
>>       if (dma_spec->args[0] >= xbar->xbar_requests) {
>>           dev_err(&pdev->dev, "Invalid XBAR request number: %d\n",
>>               dma_spec->args[0]);
>> +        put_device(&pdev->dev);
>>           return ERR_PTR(-EINVAL);
>>       }
>>   @@ -252,12 +253,14 @@ static void *ti_dra7_xbar_route_allocate(struct of_phandle_args *dma_spec,
>>       dma_spec->np = of_parse_phandle(ofdma->of_node, "dma-masters", 0);
>>       if (!dma_spec->np) {
>>           dev_err(&pdev->dev, "Can't get DMA master\n");
>> +        put_device(&pdev->dev);
>>           return ERR_PTR(-EINVAL);
>>       }
>>         map = kzalloc(sizeof(*map), GFP_KERNEL);
>>       if (!map) {
>>           of_node_put(dma_spec->np);
>> +        put_device(&pdev->dev);
>>           return ERR_PTR(-ENOMEM);
>>       }
>>   @@ -268,6 +271,8 @@ static void *ti_dra7_xbar_route_allocate(struct of_phandle_args *dma_spec,
>>           mutex_unlock(&xbar->mutex);
>>           dev_err(&pdev->dev, "Run out of free DMA requests\n");
>>           kfree(map);
>> +        of_node_put(dma_spec->np);
>> +        put_device(&pdev->dev);
>>           return ERR_PTR(-ENOMEM);
>>       }
>>       set_bit(map->xbar_out, xbar->dma_inuse);