Re: powerpc: add kernel parameter iommu_alloc_quiet

2016-10-03 Thread Mauricio Faria de Oliveira
Hi Michael, On 09/21/2016 11:34 AM, Mauricio Faria de Oliveira wrote: So, the intent is to have a single/common hack that upstream is OK w/, then apply that downstream in the multiple distros. Of course, if you are not OK w/ this patch (which is obviously fair) we'll have to try it downstream o

Re: powerpc: add kernel parameter iommu_alloc_quiet

2016-09-21 Thread Mauricio Faria de Oliveira
Hi Michael, Thanks for the review. On 09/21/2016 10:51 AM, Michael Ellerman wrote: That is important/requirement for the distribution kernels - where the DMA_ATTR_NO_WARN changes to 'enum dma_attr' are not acceptable because it breaks the kernel ABI. [snip] > Removing an entry from an enum wou

Re: powerpc: add kernel parameter iommu_alloc_quiet

2016-09-21 Thread Michael Ellerman
On Thu, 2016-01-09 at 12:56:42 UTC, Mauricio Faria de Oliveira wrote: > This patch introduces the 'iommu_alloc_quiet=driver_name' parameter > to suppress the 'iommu_alloc failures' messages for that one driver. > > This is an additional approach for this 'problem' of flooding logs, > not as fine-g

Re: [PATCH] powerpc: add kernel parameter iommu_alloc_quiet

2016-09-01 Thread Mauricio Faria de Oliveira
On 09/01/2016 10:39 AM, Torsten Duwe wrote: JFYI, my strongly preferred solution would still be to just dev_dbg() the whole thing. Which group of people would be interested in these messages, after all? Certainly understandable. Ben didn't like the idea to convert the messages to dynamic debu

Re: [PATCH] powerpc: add kernel parameter iommu_alloc_quiet

2016-09-01 Thread Torsten Duwe
On Thu, Sep 01, 2016 at 09:56:42AM -0300, Mauricio Faria de Oliveira wrote: > This patch introduces the 'iommu_alloc_quiet=driver_name' parameter > to suppress the 'iommu_alloc failures' messages for that one driver. > > This is an additional approach for this 'problem' of flooding logs, > not as

Re: [PATCH] powerpc: add kernel parameter iommu_alloc_quiet

2016-09-01 Thread Mauricio Faria de Oliveira
Michael / Ben, On 09/01/2016 09:56 AM, Mauricio Faria de Oliveira wrote: +#define IOMMU_ALLOC_QUIET_LEN 16 /* includes '\0' */ Guilherme suggested MAX_PARAM_PREFIX_LEN for this, which looks better (a few extra bytes). Would you mind to s/16/MAX_PARAM_PREFIX_LEN/ if you like that? I can send

[PATCH] powerpc: add kernel parameter iommu_alloc_quiet

2016-09-01 Thread Mauricio Faria de Oliveira
This patch introduces the 'iommu_alloc_quiet=driver_name' parameter to suppress the 'iommu_alloc failures' messages for that one driver. This is an additional approach for this 'problem' of flooding logs, not as fine-grained and not enabled by default as DMA_ATTR_NO_WARN, but it has the advantage