Re: [PATCH] dma-coherent: fix rmem_dma_device_init regression

[Christoph Hellwig]

Thanks a lot Arnd,

applied to the dma-mapping tree for 4.14.

Re: [PATCH] dma-coherent: fix rmem_dma_device_init regression

[Roy Pledge]

On 9/15/2017 11:08 AM, Arnd Bergmann wrote:
> My recent bug fix introduced another bug, which caused rmem_dma_device_init
> to always fail, as rmem->priv is never set to anything.
> 
> This restores the previous behavior, calling dma_init_coherent_memory()
> whenever ->priv is NULL.
> 
> Fixes: d35b0996fef3 ("dma-coherent: fix dma_declare_coherent_memory() logic 
> error")
> Reported-by: Roy Pledge 
> Signed-off-by: Arnd Bergmann 
> ---
> Roy, can you test this new fix?
> ---
>   drivers/base/dma-coherent.c | 19 +--
>   1 file changed, 9 insertions(+), 10 deletions(-)
> 
> diff --git a/drivers/base/dma-coherent.c b/drivers/base/dma-coherent.c
> index a39b2166b145..744f64f43454 100644
> --- a/drivers/base/dma-coherent.c
> +++ b/drivers/base/dma-coherent.c
> @@ -348,16 +348,15 @@ static int rmem_dma_device_init(struct reserved_mem 
> *rmem, struct device *dev)
>   struct dma_coherent_mem *mem = rmem->priv;
>   int ret;
>   
> - if (!mem)
> - return -ENODEV;
> -
> - ret = dma_init_coherent_memory(rmem->base, rmem->base, rmem->size,
> -DMA_MEMORY_EXCLUSIVE, &mem);
> -
> - if (ret) {
> - pr_err("Reserved memory: failed to init DMA memory pool at %pa, 
> size %ld MiB\n",
> - &rmem->base, (unsigned long)rmem->size / SZ_1M);
> - return ret;
> + if (!mem) {
> + ret = dma_init_coherent_memory(rmem->base, rmem->base,
> +rmem->size,
> +DMA_MEMORY_EXCLUSIVE, &mem);
> + if (ret) {
> + pr_err("Reserved memory: failed to init DMA memory pool 
> at %pa, size %ld MiB\n",
> + &rmem->base, (unsigned long)rmem->size / SZ_1M);
> + return ret;
> + }
>   }
>   mem->use_dev_dma_pfn_offset = true;
>   rmem->priv = mem;
> 
I tried this and it solves the issue I was seeing. Thanks for the quick fix.

Tested-by:  Roy Pledge 

[PATCH] dma-coherent: fix rmem_dma_device_init regression

[Arnd Bergmann]

My recent bug fix introduced another bug, which caused rmem_dma_device_init
to always fail, as rmem->priv is never set to anything.

This restores the previous behavior, calling dma_init_coherent_memory()
whenever ->priv is NULL.

Fixes: d35b0996fef3 ("dma-coherent: fix dma_declare_coherent_memory() logic 
error")
Reported-by: Roy Pledge 
Signed-off-by: Arnd Bergmann 
---
Roy, can you test this new fix?
---
 drivers/base/dma-coherent.c | 19 +--
 1 file changed, 9 insertions(+), 10 deletions(-)

diff --git a/drivers/base/dma-coherent.c b/drivers/base/dma-coherent.c
index a39b2166b145..744f64f43454 100644
--- a/drivers/base/dma-coherent.c
+++ b/drivers/base/dma-coherent.c
@@ -348,16 +348,15 @@ static int rmem_dma_device_init(struct reserved_mem 
*rmem, struct device *dev)
struct dma_coherent_mem *mem = rmem->priv;
int ret;
 
-   if (!mem)
-   return -ENODEV;
-
-   ret = dma_init_coherent_memory(rmem->base, rmem->base, rmem->size,
-  DMA_MEMORY_EXCLUSIVE, &mem);
-
-   if (ret) {
-   pr_err("Reserved memory: failed to init DMA memory pool at %pa, 
size %ld MiB\n",
-   &rmem->base, (unsigned long)rmem->size / SZ_1M);
-   return ret;
+   if (!mem) {
+   ret = dma_init_coherent_memory(rmem->base, rmem->base,
+  rmem->size,
+  DMA_MEMORY_EXCLUSIVE, &mem);
+   if (ret) {
+   pr_err("Reserved memory: failed to init DMA memory pool 
at %pa, size %ld MiB\n",
+   &rmem->base, (unsigned long)rmem->size / SZ_1M);
+   return ret;
+   }
}
mem->use_dev_dma_pfn_offset = true;
rmem->priv = mem;
-- 
2.9.0