1;; Entirely zeroed kernel descriptor (for GFX10). 2 3; RUN: llvm-mc %s --triple=amdgcn-amd-amdhsa -mcpu=gfx1010 -mattr=-xnack -filetype=obj -o %t 4; RUN: llvm-objdump -s -d -j .text %t | FileCheck --check-prefix=OBJDUMP %s 5 6;; TODO: 7;; This file and kd-zeroed-raw.s should produce the same output for the kernel 8;; descriptor - a block of 64 zeroed bytes. But looks like the assembler sets 9;; the FWD_PROGRESS bit in COMPUTE_PGM_RSRC1 to 1 even when the directive 10;; mentions 0 (see line 36). 11 12;; Check the raw bytes right now. 13 14; OBJDUMP-LABEL: Contents of section .text: 15; OBJDUMP-NEXT: 0000 00000000 00000000 00000000 00000000 16; OBJDUMP-NEXT: 0010 00000000 00000000 00000000 00000000 17; OBJDUMP-NEXT: 0020 00000000 00000000 00000000 00000000 18; OBJDUMP-NEXT: 0030 01000000 00000000 00000000 00000000 19; OBJDUMP-EMPTY: 20 21; OBJDUMP-LABEL: Disassembly of section .text: 22; OBJDUMP-EMPTY: 23; OBJDUMP-NEXT: 0000000000000000 <my_kernel.kd>: 24; OBJDUMP-NEXT: .amdhsa_kernel my_kernel 25; OBJDUMP-NEXT: .amdhsa_group_segment_fixed_size 0 26; OBJDUMP-NEXT: .amdhsa_private_segment_fixed_size 0 27; OBJDUMP-NEXT: .amdhsa_kernarg_size 0 28; OBJDUMP-NEXT: .amdhsa_shared_vgpr_count 0 29; OBJDUMP-NEXT: .amdhsa_next_free_vgpr 8 30; OBJDUMP-NEXT: .amdhsa_reserve_vcc 0 31; OBJDUMP-NEXT: .amdhsa_reserve_flat_scratch 0 32; OBJDUMP-NEXT: .amdhsa_reserve_xnack_mask 0 33; OBJDUMP-NEXT: .amdhsa_next_free_sgpr 8 34; OBJDUMP-NEXT: .amdhsa_float_round_mode_32 0 35; OBJDUMP-NEXT: .amdhsa_float_round_mode_16_64 0 36; OBJDUMP-NEXT: .amdhsa_float_denorm_mode_32 0 37; OBJDUMP-NEXT: .amdhsa_float_denorm_mode_16_64 0 38; OBJDUMP-NEXT: .amdhsa_dx10_clamp 0 39; OBJDUMP-NEXT: .amdhsa_ieee_mode 0 40; OBJDUMP-NEXT: .amdhsa_fp16_overflow 0 41; OBJDUMP-NEXT: .amdhsa_workgroup_processor_mode 0 42; OBJDUMP-NEXT: .amdhsa_memory_ordered 0 43; OBJDUMP-NEXT: .amdhsa_forward_progress 0 44; OBJDUMP-NEXT: .amdhsa_system_sgpr_private_segment_wavefront_offset 0 45; OBJDUMP-NEXT: .amdhsa_system_sgpr_workgroup_id_x 0 46; OBJDUMP-NEXT: .amdhsa_system_sgpr_workgroup_id_y 0 47; OBJDUMP-NEXT: .amdhsa_system_sgpr_workgroup_id_z 0 48; OBJDUMP-NEXT: .amdhsa_system_sgpr_workgroup_info 0 49; OBJDUMP-NEXT: .amdhsa_system_vgpr_workitem_id 0 50; OBJDUMP-NEXT: .amdhsa_exception_fp_ieee_invalid_op 0 51; OBJDUMP-NEXT: .amdhsa_exception_fp_denorm_src 0 52; OBJDUMP-NEXT: .amdhsa_exception_fp_ieee_div_zero 0 53; OBJDUMP-NEXT: .amdhsa_exception_fp_ieee_overflow 0 54; OBJDUMP-NEXT: .amdhsa_exception_fp_ieee_underflow 0 55; OBJDUMP-NEXT: .amdhsa_exception_fp_ieee_inexact 0 56; OBJDUMP-NEXT: .amdhsa_exception_int_div_zero 0 57; OBJDUMP-NEXT: .amdhsa_user_sgpr_private_segment_buffer 0 58; OBJDUMP-NEXT: .amdhsa_user_sgpr_dispatch_ptr 0 59; OBJDUMP-NEXT: .amdhsa_user_sgpr_queue_ptr 0 60; OBJDUMP-NEXT: .amdhsa_user_sgpr_kernarg_segment_ptr 0 61; OBJDUMP-NEXT: .amdhsa_user_sgpr_dispatch_id 0 62; OBJDUMP-NEXT: .amdhsa_user_sgpr_flat_scratch_init 0 63; OBJDUMP-NEXT: .amdhsa_user_sgpr_private_segment_size 0 64; OBJDUMP-NEXT: .amdhsa_wavefront_size32 0 65; OBJDUMP-NEXT: .amdhsa_uses_dynamic_stack 0 66; OBJDUMP-NEXT: .end_amdhsa_kernel 67 68.amdhsa_code_object_version 5 69.amdhsa_kernel my_kernel 70 .amdhsa_group_segment_fixed_size 0 71 .amdhsa_private_segment_fixed_size 0 72 .amdhsa_next_free_vgpr 8 73 .amdhsa_reserve_vcc 0 74 .amdhsa_reserve_flat_scratch 0 75 .amdhsa_reserve_xnack_mask 0 76 .amdhsa_next_free_sgpr 8 77 .amdhsa_float_round_mode_32 0 78 .amdhsa_float_round_mode_16_64 0 79 .amdhsa_float_denorm_mode_32 0 80 .amdhsa_float_denorm_mode_16_64 0 81 .amdhsa_dx10_clamp 0 82 .amdhsa_ieee_mode 0 83 .amdhsa_fp16_overflow 0 84 .amdhsa_workgroup_processor_mode 0 85 .amdhsa_memory_ordered 0 86 .amdhsa_forward_progress 0 87 .amdhsa_system_sgpr_private_segment_wavefront_offset 0 88 .amdhsa_system_sgpr_workgroup_id_x 0 89 .amdhsa_system_sgpr_workgroup_id_y 0 90 .amdhsa_system_sgpr_workgroup_id_z 0 91 .amdhsa_system_sgpr_workgroup_info 0 92 .amdhsa_system_vgpr_workitem_id 0 93 .amdhsa_exception_fp_ieee_invalid_op 0 94 .amdhsa_exception_fp_denorm_src 0 95 .amdhsa_exception_fp_ieee_div_zero 0 96 .amdhsa_exception_fp_ieee_overflow 0 97 .amdhsa_exception_fp_ieee_underflow 0 98 .amdhsa_exception_fp_ieee_inexact 0 99 .amdhsa_exception_int_div_zero 0 100 .amdhsa_user_sgpr_private_segment_buffer 0 101 .amdhsa_user_sgpr_dispatch_ptr 0 102 .amdhsa_user_sgpr_queue_ptr 0 103 .amdhsa_user_sgpr_kernarg_segment_ptr 0 104 .amdhsa_user_sgpr_dispatch_id 0 105 .amdhsa_user_sgpr_flat_scratch_init 0 106 .amdhsa_user_sgpr_private_segment_size 0 107 .amdhsa_wavefront_size32 0 108.end_amdhsa_kernel 109