linux-kernel - Re: [PATCH v4 09/15] dmaengine: ti: New driver for K3 UDMA

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]

Date:   Mon, 11 Nov 2019 10:33:12 +0200
From:   Peter Ujfalusi <peter.ujfalusi@...com>
To:     Vinod Koul <vkoul@...nel.org>
CC:     <robh+dt@...nel.org>, <nm@...com>, <ssantosh@...nel.org>,
        <dan.j.williams@...el.com>, <dmaengine@...r.kernel.org>,
        <linux-arm-kernel@...ts.infradead.org>,
        <devicetree@...r.kernel.org>, <linux-kernel@...r.kernel.org>,
        <grygorii.strashko@...com>, <lokeshvutla@...com>,
        <t-kristo@...com>, <tony@...mide.com>, <j-keerthy@...com>
Subject: Re: [PATCH v4 09/15] dmaengine: ti: New driver for K3 UDMA - split#1:
 defines, structs, io func



On 11/11/2019 7.28, Vinod Koul wrote:
> On 01-11-19, 10:41, Peter Ujfalusi wrote:
> 
>> +struct udma_chan {
>> +	struct virt_dma_chan vc;
>> +	struct dma_slave_config	cfg;
>> +	struct udma_dev *ud;
>> +	struct udma_desc *desc;
>> +	struct udma_desc *terminated_desc;
> 
> descriptor and not a list?

Yes, not a list. I have only one transfer (if any) submitted to
hardware. This is mostly due to the packet mode RX operation: no
prelinked support in UDMAP so I need to have as many descriptors queued
up as the number of sg elements.

I need to keep the terminated descriptor around to be able to free it up
_after_ UDMAP returned it to avoid it modifying released memory.

>> +	struct udma_static_tr static_tr;
>> +	char *name;
>> +
>> +	struct udma_tchan *tchan;
>> +	struct udma_rchan *rchan;
>> +	struct udma_rflow *rflow;
>> +
>> +	bool psil_paired;
>> +
>> +	int irq_num_ring;
>> +	int irq_num_udma;
>> +
>> +	bool cyclic;
>> +	bool paused;
>> +
>> +	enum udma_chan_state state;
>> +	struct completion teardown_completed;
>> +
>> +	u32 bcnt; /* number of bytes completed since the start of the channel */
>> +	u32 in_ring_cnt; /* number of descriptors in flight */
>> +
>> +	bool pkt_mode; /* TR or packet */
>> +	bool needs_epib; /* EPIB is needed for the communication or not */
>> +	u32 psd_size; /* size of Protocol Specific Data */
>> +	u32 metadata_size; /* (needs_epib ? 16:0) + psd_size */
>> +	u32 hdesc_size; /* Size of a packet descriptor in packet mode */
>> +	bool notdpkt; /* Suppress sending TDC packet */
>> +	int remote_thread_id;
>> +	u32 src_thread;
>> +	u32 dst_thread;
>> +	enum psil_endpoint_type ep_type;
>> +	bool enable_acc32;
>> +	bool enable_burst;
>> +	enum udma_tp_level channel_tpl; /* Channel Throughput Level */
>> +
>> +	/* dmapool for packet mode descriptors */
>> +	bool use_dma_pool;
>> +	struct dma_pool *hdesc_pool;
>> +
>> +	u32 id;
>> +	enum dma_transfer_direction dir;
> 
> why does channel have this, it already exists in descriptor

The channel can not change role, it is set when it was requested. In the
prep callbacks I do check if the direction matches with the channel's
direction.

>> +static irqreturn_t udma_udma_irq_handler(int irq, void *data)
>> +{
>> +	struct udma_chan *uc = data;
>> +
>> +	udma_tr_event_callback(uc);
> 
> any reason why we want to call a fn and not code here..?

No particular reason, I'll move them.

- Péter

Texas Instruments Finland Oy, Porkkalankatu 22, 00180 Helsinki.
Y-tunnus/Business ID: 0615521-4. Kotipaikka/Domicile: Helsinki