Re: [Mesa-dev] [PATCH v2 1/2] intel: gen-decoder: rework how we handle groups

2017-06-06 Thread Lionel Landwerlin
On 05/06/17 23:54, Rafael Antognolli wrote: realloc() does not initialize the newly added memory, unlike calloc(). So you can't depend on those new pointers to be zero. It doesn't look like you depend on that because you have the nfields variable anyways. But if you really don't care about initi

Re: [Mesa-dev] [PATCH v2 1/2] intel: gen-decoder: rework how we handle groups

2017-06-05 Thread Rafael Antognolli
Hi Lionel, I went through this patch more than once, but I think I understood what you did, and I think it highly simplifies the code, not necessarily in lines of code but at least the logic is way easier to follow. Just a minor nitpick below. On Wed, May 31, 2017 at 10:32:52AM +0100, Lionel Lan

[Mesa-dev] [PATCH v2 1/2] intel: gen-decoder: rework how we handle groups

2017-05-31 Thread Lionel Landwerlin
The current way of handling groups doesn't seem to be able to handle MI_LOAD_REGISTER_* with more than one register. This change reworks the way we handle groups by building a traversal list on loading the GENXML files. Let's say you have Instruction { Field0 Field1 Field2 Group0 (count=2