when trying to install amdgpu-pro_16.30.3-315407.tar.xz on ubuntu 16.10 (Yakkety Yak) /var/lib/dkms/amdgpu-pro/16.30.3-315407/build/make.log CC [M] /var/lib/dkms/amdgpu-pro/16.30.3-315407/build/amd/amdgpu/../dal/dc/calcs/bandwidth_calcs.o /var/lib/dkms/amdgpu-pro/16.30.3-315407/build/amd/amdgpu/../dal/dc/calcs/bandwidth_calcs.c: In function ‘calculate_bandwidth’: /var/lib/dkms/amdgpu-pro/16.30.3-315407/build/amd/amdgpu/../dal/dc/calcs/bandwidth_calcs.c:3162:7: error: array subscript is above array bounds [-Werror=array-bounds] bw_max2( ^~~~~~~~ results->blackout_recovery_time, I looked at the code and gave up before even trying to understand it. the function call or macro was covering 43 lines. That is it's not a function that is implemented in 43 lines it's the actual call that is 43 lines of code.
results->blackout_recovery_time = bw_max2( results->blackout_recovery_time, bw_div( (bw_add( bw_mul( bw_div( bw_mul( results->display_bandwidth[k], results->useful_bytes_per_request[k]), results->bytes_per_request[k]), vbios->blackout_duration), bw_sub( bw_div( bw_mul( bw_mul( bw_mul( (bw_add( bw_add( bw_mul( bw_int_to_fixed( 2), vbios->mcifwrmc_urgent_latency), results->dmif_burst_time[i][j]), results->mcifwr_burst_time[results->y_clk_level][results->sclk_level])), results->dispclk), bw_int_to_fixed( results->bytes_per_pixel[k])), results->lines_interleaved_in_mem_access[k]), results->latency_hiding_lines[k]), results->adjusted_data_buffer_size[k]))), (bw_sub( bw_div( bw_mul( bw_mul( results->dispclk, bw_int_to_fixed( results->bytes_per_pixel[k])), results->lines_interleaved_in_mem_access[k]), results->latency_hiding_lines[k]), bw_div( bw_mul( results->display_bandwidth[k], results->useful_bytes_per_request[k]), results->bytes_per_request[k])))));
Created attachment 126449 [details] dkms build log
Closing, as this is pretty old, and only Ubuntu LTS is supported (e.g. 16.04, 18.04).
Use of freedesktop.org services, including Bugzilla, is subject to our Code of Conduct. How we collect and use information is described in our Privacy Policy.