/kernel/linux/linux-5.10/drivers/gpu/drm/amd/amdkfd/ |
H A D | cwsr_trap_handler_gfx9.asm | 716 buffer_load_dword v0, v0, s_restore_buf_rsrc0, s_restore_mem_offset lds:1 // first 64DW 717 buffer_load_dword v0, v0, s_restore_buf_rsrc0, s_restore_mem_offset lds:1 offset:256 // second 64DW 718 s_add_u32 m0, m0, 256*2 // 128 DW 719 s_add_u32 s_restore_mem_offset, s_restore_mem_offset, 256*2 //mem offset increased by 128DW 1001 // SQ LDS granularity is 64DW, while PGM_RSRC2.lds_size is in granularity 128DW 1003 s_lshl_b32 s_lds_size_byte, s_lds_size_byte, 8 //LDS size in dwords = lds_size * 64 *4Bytes // granularity 64DW
|
H A D | cwsr_trap_handler_gfx10.asm | 725 buffer_load_dword v0, v0, s_restore_buf_rsrc0, s_restore_mem_offset lds:1 // first 64DW 726 s_add_u32 m0, m0, 128 // 128 DW 727 s_add_u32 s_restore_mem_offset, s_restore_mem_offset, 128 //mem offset increased by 128DW 733 buffer_load_dword v0, v0, s_restore_buf_rsrc0, s_restore_mem_offset lds:1 // first 64DW 734 s_add_u32 m0, m0, 256 // 256 DW 735 s_add_u32 s_restore_mem_offset, s_restore_mem_offset, 256 //mem offset increased by 256DW 1061 s_lshl_b32 s_lds_size_byte, s_lds_size_byte, 8 //LDS size in dwords = lds_size * 64 *4Bytes // granularity 64DW
|
H A D | cwsr_trap_handler_gfx8.asm | 517 buffer_load_dword v0, v0, s_restore_buf_rsrc0, s_restore_mem_offset lds:1 // first 64DW 518 buffer_load_dword v0, v0, s_restore_buf_rsrc0, s_restore_mem_offset lds:1 offset:256 // second 64DW 519 s_add_u32 m0, m0, 256*2 // 128 DW 520 s_add_u32 s_restore_mem_offset, s_restore_mem_offset, 256*2 //mem offset increased by 128DW 719 // SQ LDS granularity is 64DW, while PGM_RSRC2.lds_size is in granularity 128DW 721 s_lshl_b32 s_lds_size_byte, s_lds_size_byte, 8 //LDS size in dwords = lds_size * 64 *4Bytes // granularity 64DW
|
/kernel/linux/linux-6.6/drivers/gpu/drm/amd/amdkfd/ |
H A D | cwsr_trap_handler_gfx9.asm | 749 buffer_load_dword v0, v0, s_restore_buf_rsrc0, s_restore_mem_offset lds:1 // first 64DW 750 buffer_load_dword v0, v0, s_restore_buf_rsrc0, s_restore_mem_offset lds:1 offset:256 // second 64DW 751 s_add_u32 m0, m0, 256*2 // 128 DW 752 s_add_u32 s_restore_mem_offset, s_restore_mem_offset, 256*2 //mem offset increased by 128DW 1020 // SQ LDS granularity is 64DW, while PGM_RSRC2.lds_size is in granularity 128DW 1022 s_lshl_b32 s_lds_size_byte, s_lds_size_byte, 8 //LDS size in dwords = lds_size * 64 *4Bytes // granularity 64DW
|
H A D | cwsr_trap_handler_gfx8.asm | 517 buffer_load_dword v0, v0, s_restore_buf_rsrc0, s_restore_mem_offset lds:1 // first 64DW 518 buffer_load_dword v0, v0, s_restore_buf_rsrc0, s_restore_mem_offset lds:1 offset:256 // second 64DW 519 s_add_u32 m0, m0, 256*2 // 128 DW 520 s_add_u32 s_restore_mem_offset, s_restore_mem_offset, 256*2 //mem offset increased by 128DW 720 // SQ LDS granularity is 64DW, while PGM_RSRC2.lds_size is in granularity 128DW 722 s_lshl_b32 s_lds_size_byte, s_lds_size_byte, 8 //LDS size in dwords = lds_size * 64 *4Bytes // granularity 64DW
|
H A D | cwsr_trap_handler_gfx10.asm | 824 buffer_load_dword v0, v0, s_restore_buf_rsrc0, s_restore_mem_offset lds:1 // first 64DW 830 s_add_u32 m0, m0, 128 // 128 DW 831 s_add_u32 s_restore_mem_offset, s_restore_mem_offset, 128 //mem offset increased by 128DW 838 buffer_load_dword v0, v0, s_restore_buf_rsrc0, s_restore_mem_offset lds:1 // first 64DW 844 s_add_u32 m0, m0, 256 // 256 DW 845 s_add_u32 s_restore_mem_offset, s_restore_mem_offset, 256 //mem offset increased by 256DW 1175 s_lshl_b32 s_lds_size_byte, s_lds_size_byte, 8 //LDS size in dwords = lds_size * 64 *4Bytes // granularity 64DW
|
/kernel/linux/linux-5.10/arch/x86/crypto/ |
H A D | aesni-intel_avx-x86_64.S | 921 vpsrldq $4, \T2, \T5 # shift-R T5 1 DW 1556 vpsrldq $4, \T2, \T1 # shift-R T1 1 DW 1744 vpsrldq $4, \T2, \T1 # shift-R T1 1 DW 1903 vpsrldq $4, \T2, \T2 # shift-R T2 1 DW (Shift-R only 1-DW to obtain 2-DWs shift-R) 1906 vpslldq $4, \GH, \GH # shift-L GH 1 DW (Shift-L 1-DW to obtain result with no shifts) 2526 vpsrldq $4, \T2, \T2 # shift-R xmm2 1 DW (Shift-R only 1-DW to obtain 2-DWs shift-R) 2529 vpslldq $4, \T4, \T4 # shift-L xmm0 1 DW (Shif [all...] |
H A D | aesni-intel_asm.S | 544 psrldq $4, \TMP5 # right shift TMP5 1 DW 1161 psrldq $4, \TMP5 # right shift T5 1 DW 1373 psrldq $4, \TMP5 # right shift T5 1 DW 1472 psrldq $4, \TMP7 # right shift TMP7 1 DW
|
/kernel/linux/linux-6.6/arch/x86/crypto/ |
H A D | aesni-intel_avx-x86_64.S | 889 vpsrldq $4, \T2, \T5 # shift-R T5 1 DW 1524 vpsrldq $4, \T2, \T1 # shift-R T1 1 DW 1712 vpsrldq $4, \T2, \T1 # shift-R T1 1 DW 1871 vpsrldq $4, \T2, \T2 # shift-R T2 1 DW (Shift-R only 1-DW to obtain 2-DWs shift-R) 1874 vpslldq $4, \GH, \GH # shift-L GH 1 DW (Shift-L 1-DW to obtain result with no shifts) 2494 vpsrldq $4, \T2, \T2 # shift-R xmm2 1 DW (Shift-R only 1-DW to obtain 2-DWs shift-R) 2497 vpslldq $4, \T4, \T4 # shift-L xmm0 1 DW (Shif [all...] |
H A D | aesni-intel_asm.S | 540 psrldq $4, \TMP5 # right shift TMP5 1 DW 1157 psrldq $4, \TMP5 # right shift T5 1 DW 1369 psrldq $4, \TMP5 # right shift T5 1 DW 1468 psrldq $4, \TMP7 # right shift TMP7 1 DW
|
/kernel/linux/linux-5.10/drivers/atm/ |
H A D | horizon.h | 463 #define DW(x) do{ x } while(0) macro
|
/kernel/linux/linux-5.10/kernel/bpf/ |
H A D | core.c | 1330 INSN_3(STX, MEM, DW), \ 1332 INSN_3(STX, XADD, DW), \ 1337 INSN_3(ST, MEM, DW), \ 1343 INSN_3(LDX, MEM, DW), \ 1345 INSN_3(LD, IMM, DW) 1671 LDST(DW, u64) in ___bpf_prog_run()
|
/kernel/linux/linux-6.6/kernel/bpf/ |
H A D | core.c | 1624 INSN_3(STX, MEM, DW), \ 1626 INSN_3(STX, ATOMIC, DW), \ 1631 INSN_3(ST, MEM, DW), \ 1637 INSN_3(LDX, MEM, DW), \ 1642 INSN_3(LD, IMM, DW) 2084 LDST(DW, u64) in ___bpf_prog_run()
|