Re: [PATCH] net/mlx4_en: ensure rx_desc updating reaches HW before prod db updating

[Tariq Toukan]

On 21/01/2018 11:31 AM, Tariq Toukan wrote:
> On 19/01/2018 5:49 PM, Eric Dumazet wrote:
> > On Fri, 2018-01-19 at 23:16 +0800, jianchao.wang wrote:
> > > Hi Tariq
> > >
> > > Very sad that the crash was reproduced again after applied the patch.

Memory barriers vary for different Archs, can you please share more 
details regarding arch and repro steps?

> > > --- a/drivers/net/ethernet/mellanox/mlx4/en_rx.c
> > > +++ b/drivers/net/ethernet/mellanox/mlx4/en_rx.c
> > > @@ -252,6 +252,7 @@ static inline bool mlx4_en_is_ring_empty(struct 
> > > mlx4_en_rx_ring *ring)
> > > Â static inline void mlx4_en_update_rx_prod_db(struct mlx4_en_rx_ring 
> > > *ring)
> > > Â {
> > > +ÂÂÂ dma_wmb();
> >
> > So... is wmb() here fixing the issue ?
> >
> > > ÂÂÂÂÂ *ring->wqres.db.db = cpu_to_be32(ring->prod & 0xffff);
> > > Â }
> > >
> > > I analyzed the kdump, it should be a memory corruption.
> > >
> > > Thanks
> > > Jianchao
>
> Hmm, this is actually consistent with the example below [1].
>
> AIU from the example, it seems that the dma_wmb/dma_rmb barriers are 
> good for synchronizing cpu/device accesses to the "Streaming DMA mapped" 
> buffers (the descriptors, went through the dma_map_page() API), but not 
> for the doorbell (a coherent memory, typically allocated via 
> dma_alloc_coherent) that requires using the stronger wmb() barrier.
>
>
> [1] Documentation/memory-barriers.txt
>
>  Â(*) dma_wmb();
>  Â(*) dma_rmb();
>
>  ÂÂÂÂ These are for use with consistent memory to guarantee the ordering
>  ÂÂÂÂ of writes or reads of shared memory accessible to both the CPU and a
>  ÂÂÂÂ DMA capable device.
>
>  ÂÂÂÂ For example, consider a device driver that shares memory with a 
> device
>  ÂÂÂÂ and uses a descriptor status value to indicate if the descriptor 
> belongs
>  ÂÂÂÂ to the device or the CPU, and a doorbell to notify it when new
>  ÂÂÂÂ descriptors are available:
>
>  ÂÂÂÂif (desc->status != DEVICE_OWN) {
>  ÂÂÂÂÂÂÂ /* do not read data until we own descriptor */
>  ÂÂÂÂÂÂÂ dma_rmb();
>
>  ÂÂÂÂÂÂÂ /* read/modify data */
>  ÂÂÂÂÂÂÂ read_data = desc->data;
>  ÂÂÂÂÂÂÂ desc->data = write_data;
>
>  ÂÂÂÂÂÂÂ /* flush modifications before status update */
>  ÂÂÂÂÂÂÂ dma_wmb();
>
>  ÂÂÂÂÂÂÂ /* assign ownership */
>  ÂÂÂÂÂÂÂ desc->status = DEVICE_OWN;
>
>  ÂÂÂÂÂÂÂ /* force memory to sync before notifying device via MMIO */
>  ÂÂÂÂÂÂÂ wmb();
>
>  ÂÂÂÂÂÂÂ /* notify device of new descriptors */
>  ÂÂÂÂÂÂÂ writel(DESC_NOTIFY, doorbell);
>  ÂÂÂÂ}
>
>  ÂÂÂÂ The dma_rmb() allows us guarantee the device has released ownership
>  ÂÂÂÂ before we read the data from the descriptor, and the dma_wmb() allows
>  ÂÂÂÂ us to guarantee the data is written to the descriptor before the 
> device
>   can see it now has ownership. The wmb() is needed to guarantee 
> that the
>  ÂÂÂÂ cache coherent memory writes have completed before attempting a 
> write to
>  ÂÂÂÂ the cache incoherent MMIO region.
>
>  ÂÂÂÂ See Documentation/DMA-API.txt for more information on consistent 
> memory.
>
>
> > > On 01/15/2018 01:50 PM, jianchao.wang wrote:
> > > > Hi Tariq
> > > >
> > > > Thanks for your kindly response.
> > > >
> > > > On 01/14/2018 05:47 PM, Tariq Toukan wrote:
> > > > > Thanks Jianchao for your patch.
> > > > >
> > > > > And Thank you guys for your reviews, much appreciated.
> > > > > I was off-work on Friday and Saturday.
> > > > >
> > > > > On 14/01/2018 4:40 AM, jianchao.wang wrote:
> > > > > > Dear all
> > > > > >
> > > > > > Thanks for the kindly response and reviewing. That's really 
> > > > > > appreciated.
> > > > > >
> > > > > > On 01/13/2018 12:46 AM, Eric Dumazet wrote:
> > > > > > > > Does this need to be dma_wmb(), and should it be in
> > > > > > > > mlx4_en_update_rx_prod_db ?
> > > > > > >
> > > > > > > +1 on dma_wmb()
> > > > > > >
> > > > > > > On what architecture bug was observed ?
> > > > > >
> > > > > > This issue was observed on x86-64.
> > > > > > And I will send a new patch, in which replace wmb() with 
> > > > > > dma_wmb(), to customer
> > > > > > to confirm.
> > > > >
> > > > > +1 on dma_wmb, let us know once customer confirms.
> > > > > Please place it within mlx4_en_update_rx_prod_db as suggested.
> > > >
> > > > Yes, I have recommended it to customer.
> > > > Once I get the result, I will share it here.
> > > > > All other calls to mlx4_en_update_rx_prod_db are in control/slow 
> > > > > path so I prefer being on the safe side, and care less about 
> > > > > bulking the barrier.
> > > > >
> > > > > Thanks,
> > > > > Tariq
> > --
> > To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
> > the body of a message to majordomo@xxxxxxxxxxxxxxx
> > More majordomo info at http://vger.kernel.org/majordomo-info.html