Re: [PATCH 1/3] dmaengine: virt-dma: don't always free descriptor upon completion

2015-09-24 Thread Robert Jarzmik
Vinod Koul writes: > On Sun, Sep 06, 2015 at 01:40:52PM +0200, Robert Jarzmik wrote: >> @@ -29,7 +29,7 @@ dma_cookie_t vchan_tx_submit(struct >> dma_async_tx_descriptor *tx) >> spin_lock_irqsave(&vc->lock, flags); >> cookie = dma_cookie_assign(tx); >> >> -list_add_tail(&vd->node,

Re: [PATCH 1/3] dmaengine: virt-dma: don't always free descriptor upon completion

2015-09-24 Thread Vinod Koul
On Sun, Sep 06, 2015 at 01:40:52PM +0200, Robert Jarzmik wrote: > @@ -29,7 +29,7 @@ dma_cookie_t vchan_tx_submit(struct dma_async_tx_descriptor > *tx) > spin_lock_irqsave(&vc->lock, flags); > cookie = dma_cookie_assign(tx); > > - list_add_tail(&vd->node, &vc->desc_submitted); > +

Re: [PATCH 1/3] dmaengine: virt-dma: don't always free descriptor upon completion

2015-09-16 Thread Robert Jarzmik
Robert Jarzmik writes: > Robert Jarzmik writes: > Hi Jun, Lars-Peter and Vinod, > > The revert of the former patch of this type, 8c8fe97b2b8a ("Revert "dmaengine: > virt-dma: don't always free descriptor upon completion"") broke pxa_dma > driver. So ... what's the plan, Vinod ? I need to fix px

Re: [PATCH 1/3] dmaengine: virt-dma: don't always free descriptor upon completion

2015-09-12 Thread Robert Jarzmik
Robert Jarzmik writes: > This patch attempts to enhance the case of a transfer submitted multiple > times, and where the cost of creating the descriptors chain is not > negligible. > > This happens with big video buffers (several megabytes, ie. several > thousands of linked descriptors in one sca