qemu-devel
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

RE: simple example of pci driver with dma


From: Shaked Matzner
Subject: RE: simple example of pci driver with dma
Date: Thu, 29 Oct 2020 20:32:39 +0000

Hey Peter,

Currently I have this test in the driver which allocates data, assign it to default value config source as the BASE address of dma, dest as the physical address which I got from dma_alloc_coherent, set the count and assign the command to raise interrupt and read,the piece of code used is something like this(this a test performed from the probe function of the driver)
                                                vaddr_to = dma_alloc_coherent(&(dev->dev), 4, &dma_handle_to, GFP_ATOMIC |GFP_KERNEL);

            *((volatile int*)vaddr_to) = 0xff;

                                                test->vaddr_to = vaddr_to;

                                                dev_info(&(dev->dev), "vaddr_to = %px\n", vaddr_to);

                                                dev_info(&(dev->dev), "dma_handle_to = %llx\n", (unsigned long long)dma_handle_to);

                                                iowrite32(DMA_BASE, mmio + IO_DMA_SRC);

                                                iowrite32((u32)dma_handle_to, mmio + IO_DMA_DST);

                                                iowrite32(SIZE, mmio + IO_DMA_CNT);

                                               iowrite32(DMA_CMD | DMA_FROM_DEV | DMA_IRQ, mmio + IO_DMA_CMD);

Where and when should the pci_set_master hould be called?
Thanks,

             Shaked Matzner

 

From: Peter Maydell <peter.maydell@linaro.org>
Sent: Thursday, October 29, 2020 5:46 PM
To: Shaked Matzner <shaked.m@neuroblade.ai>
Cc: qemu-devel@nongnu.org
Subject: Re: simple example of pci driver with dma

 

IRONSCALES couldn't recognize this email as this is the first time you received an email from this sender peter.maydell@linaro.org

 

[EXTERNAL]

On Thu, 29 Oct 2020 at 14:59, Shaked Matzner wrote:
> however the value I get is still 255(0xff) and not 18(0x12) probably I've missed something but when the interrupt is called the transfer to the RAM address should be completed, however it seems like the dma_write_buffer function from the device does not perform any transfer. What Am I missing?

The usual mistake is forgetting in the guest code to program the
PCI device to enable bus mastering by setting the Bus Master bit
in the Command register in the PCI config space registers for
the device. Unless you do that then all DMA attempts will fail
(same as on real h/w). In the Linux kernel the function for this
is pci_set_master(), I think.

thanks
-- PMM


The contents of this email message and any attachments are intended solely for the addressee(s) and may contain confidential and/or privileged information and may be legally protected from disclosure. If you are not the intended recipient of this message or their agent, or if this message has been addressed to you in error, please immediately alert the sender by reply email and then delete this message and any attachments. If you are not the intended recipient, you are hereby notified that any use, dissemination, copying, or storage of this message or its attachments is strictly prohibited.


reply via email to

[Prev in Thread] Current Thread [Next in Thread]