[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <a448fecd-96b1-614c-1bbf-1421d1b36e6f@huawei.com>
Date: Tue, 30 Mar 2021 17:24:24 +0800
From: "yekai(A)" <yekai13@...wei.com>
To: Joe Perches <joe@...ches.com>, <herbert@...dor.apana.org.au>
CC: <linux-crypto@...r.kernel.org>, <linux-kernel@...r.kernel.org>,
<wangzhou1@...ilicon.com>
Subject: Re: [PATCH v2 3/5] crypto: hisilicon/sgl - add some dfx logs
However, I think this log can be used to quickly locate the function or
module if dma alloc failed.
On 2021/3/30 15:56, Joe Perches wrote:
> On Tue, 2021-03-30 at 15:39 +0800, Kai Ye wrote:
>> Add some dfx logs in some abnormal exit situations.
> []
>> diff --git a/drivers/crypto/hisilicon/sgl.c b/drivers/crypto/hisilicon/sgl.c
> []
>> @@ -87,8 +87,10 @@ struct hisi_acc_sgl_pool *hisi_acc_create_sgl_pool(struct device *dev,
>> block[i].sgl = dma_alloc_coherent(dev, block_size,
>> &block[i].sgl_dma,
>> GFP_KERNEL);
>> - if (!block[i].sgl)
>> + if (!block[i].sgl) {
>> + dev_err(dev, "Fail to allocate hw SG buffer!\n");
> This doesn't seem useful as dma_alloc_coherent does a dump_stack
> by default on OOM.
>
>
> .
>
Powered by blists - more mailing lists