linux-kernel - Re: [PATCH v7 12/19] dmaengine: tegra-apb: Remove handling of unrealistic error condition

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

Message-ID: <e89b7319-74c6-4a88-4634-f127aca86851@gmail.com>
Date:   Wed, 5 Feb 2020 01:17:29 +0300
From:   Dmitry Osipenko <digetx@...il.com>
To:     Jon Hunter <jonathanh@...dia.com>,
        Laxman Dewangan <ldewangan@...dia.com>,
        Vinod Koul <vkoul@...nel.org>,
        Dan Williams <dan.j.williams@...el.com>,
        Thierry Reding <thierry.reding@...il.com>,
        Michał Mirosław <mirq-linux@...e.qmqm.pl>
Cc:     dmaengine@...r.kernel.org, linux-tegra@...r.kernel.org,
        linux-kernel@...r.kernel.org
Subject: Re: [PATCH v7 12/19] dmaengine: tegra-apb: Remove handling of
 unrealistic error condition

04.02.2020 14:52, Jon Hunter пишет:
> 
> On 02/02/2020 22:28, Dmitry Osipenko wrote:
>> The pending_sg_req list can't ever be empty because:
>>
>> 1. If it was empty, then handle_cont_sngl_cycle_dma_done() shall crash
>>    before of handle_continuous_head_request() invocation.
>>
>> 2. The handle_cont_sngl_cycle_dma_done() can't happen after stopping DMA.
> 
> By this you mean calling terminate_all?

Yes, and also the handle_continuous_head_request() itself because it
stops DMA on error, which clears interrupt status, and thus, ISR handle
returns IRQ_NONE without handling next interrupt.

>> Signed-off-by: Dmitry Osipenko <digetx@...il.com>
>> ---
>>  drivers/dma/tegra20-apb-dma.c | 6 ------
>>  1 file changed, 6 deletions(-)
>>
>> diff --git a/drivers/dma/tegra20-apb-dma.c b/drivers/dma/tegra20-apb-dma.c
>> index 62d181bd5e62..c7dc27ef1856 100644
>> --- a/drivers/dma/tegra20-apb-dma.c
>> +++ b/drivers/dma/tegra20-apb-dma.c
>> @@ -564,12 +564,6 @@ static bool handle_continuous_head_request(struct tegra_dma_channel *tdc,
>>  {
>>  	struct tegra_dma_sg_req *hsgreq;
>>  
>> -	if (list_empty(&tdc->pending_sg_req)) {
>> -		dev_err(tdc2dev(tdc), "DMA is running without req\n");
>> -		tegra_dma_stop(tdc);
>> -		return false;
>> -	}
>> -
>>  	/*
>>  	 * Check that head req on list should be in flight.
>>  	 * If it is not in flight then abort transfer as
>>
> 
> There is also a list_empty() check in tdc_configure_next_head_desc()
> which is also redundant and could be removed here as well.

Good catch :) I'll squash all these list_empty() removals into a single
patch.