On 3/21/25 23:26, Bert Karwatzki wrote: >>> >> >> I am not an expert in amdgpu or gtt_mgr, but I wonder if some of the deletes >> are coming >> from forceful eviction of memory during allocation? >> >> Have you filed a bug report for the nokaslr case? >> >> Balbir Singh > > I did some more monitoring and in the BAD case ttm_global_swapout() is called > 585 times leading to 54495 calls to ttm_bo_swapout_cb() while in the GOOD both > function are called 0 times. >
Could you please check if under /sys/kernel/debug You can find a ttm directory under that hierarchy and a ttm_shrink file? Is anything writing to the file (like the game?) With nokaslr, do you see a change in the game characteristics, like more simulated objects (an increase in fleet size, simulated objects)? It might be interesting to debug the ttm_pages_limit to see if that changes when nokaslr is enabled as well Thanks! Balbir