From: Mikhail Gavrilov <mikhail.v.gavrilov@gmail.com>
To: Tom St Denis <tom.stdenis@amd.com>
Cc: "Wentland, Harry" <Harry.Wentland@amd.com>,
"Deucher, Alexander" <Alexander.Deucher@amd.com>,
"Koenig, Christian" <Christian.Koenig@amd.com>,
"linux-mm@kvack.org" <linux-mm@kvack.org>,
"amd-gfx@lists.freedesktop.org" <amd-gfx@lists.freedesktop.org>,
"Grodzovsky, Andrey" <Andrey.Grodzovsky@amd.com>
Subject: Re: After Vega 56/64 GPU hang I unable reboot system
Date: Thu, 10 Jan 2019 20:22:56 +0500 [thread overview]
Message-ID: <CABXGCsN4NwYG4UeJ9a-P_8cEd1qg_sC24z3L3Ek9Wn0JttkZ=g@mail.gmail.com> (raw)
In-Reply-To: <CABXGCsPPjz57=Et-V-_iGyY0GrEwfcK2QcRJcqiujUp90zaz-g@mail.gmail.com>
[-- Attachment #1: Type: text/plain, Size: 611 bytes --]
On Thu, 10 Jan 2019 at 00:36, Mikhail Gavrilov
<mikhail.v.gavrilov@gmail.com> wrote:
>
> All new one logs attached here.
>
> Thanks.
>
> P.S. This time I had to terminate command `./umr -O verbose,follow -R
> gfx[.] > gfx.log 2>&1` cause it tried to write log infinitely.
> I also had to terminate command `./umr -O verbose,follow -R gfx[.] >
> gfx.log 2>&1` cause it stuck for a long time.
>
>
It became clear why umr stuck at the gfx dump. I ran umr under gdb and
I got a segfault at a moment when umr was stuck earlier.
Tom, are you here? Can you look attached backtrace?
--
Best Regards,
Mike Gavrilov.
[-- Attachment #2: umr-backtrace.txt --]
[-- Type: text/plain, Size: 11754 bytes --]
pgm[1@0x80010e520500 + 0x12c4] = 0x043a0f18 v_sub_f32_e32 v29, v24, v7
pgm[1@0x80010e520500 + 0x12c8] = 0x0a3c3d0d v_mul_f32_e32 v30, v13, v30
pgm[1@0x80010e520500 + 0x12cc] = 0x2c3c3b11 v_mac_f32_e32 v30, v17, v29
pgm[1@0x80010e520500 + 0x12d0] = 0x0238391e v_add_f32_e32 v28, v30, v28
pgm[1@0x80010e520500 + 0x12d4] = 0xd105801c v_mul_f32_e64 v28, v21, v28 clamp
pgm[1@0x80010e520500 + 0x12d8] = 0x00023915 ;;
pgm[1@0x80010e520500 + 0x12dc] = 0xd1c1001e v_mad_f32 v30, v28, v13, v6
pgm[1@0x80010e520500 + 0x12e0] = 0x041a1b1c ;;
pgm[1@0x80010e520500 + 0x12e4] = 0xd1c1001d v_mad_f32 v29, v28, v17, v7
pgm[1@0x80010e520500 + 0x12e8] = 0x041e231c ;;
pgm[1@0x80010e520500 + 0x12ec] = 0x0432331e v_sub_f32_e32 v25, v30, v25
pgm[1@0x80010e520500 + 0x12f0] = 0xd1c1001c v_mad_f32 v28, v28, v10, v3
pgm[1@0x80010e520500 + 0x12f4] = 0x040e151c ;;
pgm[1@0x80010e520500 + 0x12f8] = 0x0430311d v_sub_f32_e32 v24, v29, v24
pgm[1@0x80010e520500 + 0x12fc] = 0x0a323319 v_mul_f32_e32 v25, v25, v25
pgm[1@0x80010e520500 + 0x1300] = 0x0434351c v_sub_f32_e32 v26, v28, v26
pgm[1@0x80010e520500 + 0x1304] = 0x2c323118 v_mac_f32_e32 v25, v24, v24
pgm[1@0x80010e520500 + 0x1308] = 0x2c32351a v_mac_f32_e32 v25, v26, v26
pgm[1@0x80010e520500 + 0x130c] = 0x7c8c331b v_cmp_ge_f32_e32 vcc, v27, v25
pgm[1@0x80010e520500 + 0x1310] = 0xd100001e v_cndmask_b32_e64 v30, 0, -1, vcc
pgm[1@0x80010e520500 + 0x1314] = 0x01a98280 ;;
pgm[1@0x80010e520500 + 0x1318] = 0x87fe007e s_or_b64 exec, exec, s[0:1]
pgm[1@0x80010e520500 + 0x131c] = 0x26302e9f v_and_b32_e32 v24, 31, v23
pgm[1@0x80010e520500 + 0x1320] = 0xd1120018 v_lshlrev_b32_e64 v24, v24, 1
pgm[1@0x80010e520500 + 0x1324] = 0x00010318 ;;
pgm[1@0x80010e520500 + 0x1328] = 0x7e32570e v_not_b32_e32 v25, v14
pgm[1@0x80010e520500 + 0x132c] = 0x26303318 v_and_b32_e32 v24, v24, v25
pgm[1@0x80010e520500 + 0x1330] = 0x7d9a3c80 v_cmp_ne_u32_e32 vcc, 0, v30
pgm[1@0x80010e520500 + 0x1334] = 0x00363080 v_cndmask_b32_e32 v27, 0, v24, vcc
pgm[1@0x80010e520500 + 0x1338] = 0x2a30370e v_xor_b32_e32 v24, v14, v27
pgm[1@0x80010e520500 + 0x133c] = 0x682e2e81 v_add_u32_e32 v23, 1, v23
pgm[1@0x80010e520500 + 0x1340] = 0x682c2c90 v_add_u32_e32 v22, 16, v22
pgm[1@0x80010e520500 + 0x1344] = 0x7e320280 v_mov_b32_e32 v25, 0
pgm[1@0x80010e520500 + 0x1348] = 0xbf820000 s_branch 0
pgm[1@0x80010e520500 + 0x134c] = 0x7d9a3280 v_cmp_ne_u32_e32 vcc, 0, v25
pgm[1@0x80010e520500 + 0x1350] = 0x86ea6a7e s_and_b64 vcc, exec, vcc
pgm[1@0x80010e520500 + 0x1354] = 0xbf86fd70 s_cbranch_vccz 64880
pgm[1@0x80010e520500 + 0x1358] = 0xb0038000 s_movk_i32 s3, 0x8000
pgm[1@0x80010e520500 + 0x135c] = 0xc00e0001 s_load_dwordx8 s[0:7], s[2:3], 0x0
pgm[1@0x80010e520500 + 0x1360] = 0x00000000 ;;
pgm[1@0x80010e520500 + 0x1364] = 0x7e1e030e v_mov_b32_e32 v15, v14
pgm[1@0x80010e520500 + 0x1368] = 0x7e20030e v_mov_b32_e32 v16, v14
pgm[1@0x80010e520500 + 0x136c] = 0x7e22030e v_mov_b32_e32 v17, v14
pgm[1@0x80010e520500 + 0x1370] = 0xbf8cc07f s_waitcnt lgkmcnt(0)
pgm[1@0x80010e520500 + 0x1374] = 0xf0201f00 image_store v[14:17], v0, s[0:7] dmask:0xf unorm
pgm[1@0x80010e520500 + 0x1378] = 0x00000e00 ;;
pgm[1@0x80010e520500 + 0x137c] = 0xbf810000 s_endpgm
End of disassembly.
Disassembly of shader 1@0x80010010b400 of length 161292 bytes from IB[1@0x80010e555000 + 0x1078]
Program received signal SIGSEGV, Segmentation fault.
0x0000000000494220 in llvm::MCAssembler::getAtom(llvm::MCSymbol const&) const [clone .cold.240] ()
(gdb) thread apply all bt full
Thread 1 (Thread 0x7ffff7a20740 (LWP 5975)):
#0 0x0000000000494220 in llvm::MCAssembler::getAtom(llvm::MCSymbol const&) const [clone .cold.240] ()
No symbol table info available.
#1 0x00007fff3fffffff in ?? ()
No symbol table info available.
#2 0x0000000000000902 in ?? ()
No symbol table info available.
#3 0x0000000000533a41 in llvm::AMDGPUDisassembler::decodeOperand_VReg_96(unsigned int) const ()
No symbol table info available.
#4 0x0000000000537243 in DecodeSReg_128RegisterClass(llvm::MCInst&, unsigned int, unsigned long, void const*) [clone .isra.60] ()
No symbol table info available.
#5 0x00000000005464b2 in llvm::MCDisassembler::DecodeStatus llvm::decodeToMCInst<unsigned long>(llvm::MCDisassembler::DecodeStatus, unsigned int, unsigned long, llvm::MCInst&, unsigned long, void const*, bool&) ()
No symbol table info available.
#6 0x000000000054e1ad in llvm::AMDGPUDisassembler::tryDecodeInst(unsigned char const*, llvm::MCInst&, unsigned long, unsigned long) const ()
No symbol table info available.
#7 0x000000000054ec0f in llvm::AMDGPUDisassembler::getInstruction(llvm::MCInst&, unsigned long&, llvm::ArrayRef<unsigned char>, unsigned long, llvm::raw_ostream&, llvm::raw_ostream&) const ()
No symbol table info available.
#8 0x00000000007d0a5e in LLVMDisasmInstruction ()
No symbol table info available.
#9 0x00000000004c24b1 in umr_llvm_disasm (disasm_text=0x7ffff7952010, PC=140741784417280, inst_bytes=161292, inst=0x4599e30 "\202", asic=<optimized out>) at /home/mikhail/packaging-work/umr/src/lib/umr_llvm_disasm.c:81
i = 4532
tmp = "\tv_madak_f32 v255, v255, v255, 0xfffffff0\000\071\070, attr49.z\000isn't aligned 3\000\061\065 offen offset:2521 glc\000\061 glc tfe ; Error: unknown operand encoding 125\000oding 125\000\000\000\200\000\000\000\321\002\000\000\340\023\000\000v\002\000\000\000\000\000\000N\000\000\000\000\000\000\000\t\000\000\000\000\000\000\000\274\000\000\000}\000\000\000"...
cpuname = <optimized out>
disasm_ref = 0x454a080
x = 18128
z = <optimized out>
n = <optimized out>
disasm_ref = <optimized out>
x = <optimized out>
z = <optimized out>
i = <optimized out>
n = <optimized out>
tmp = <optimized out>
cpuname = <optimized out>
#10 umr_llvm_disasm (asic=<optimized out>, inst=0x4599e30 "\202", inst_bytes=161292, PC=140741784417280, disasm_text=0x7ffff7952010) at /home/mikhail/packaging-work/umr/src/lib/umr_llvm_disasm.c:38
disasm_ref = <optimized out>
x = <optimized out>
z = <optimized out>
i = <optimized out>
n = <optimized out>
tmp = <optimized out>
cpuname = <optimized out>
#11 0x00000000004c26d1 in umr_vm_disasm (asic=asic@entry=0x1c08a50, vmid=1, addr=140741784417280, PC=PC@entry=0, size=<optimized out>, start_offset=start_offset@entry=0, wd=0x1c0b6b0) at /home/mikhail/packaging-work/umr/src/lib/umr_llvm_disasm.c:170
opcodes = 0x4599e30
x = <optimized out>
y = <optimized out>
--Type <RET> for more, q to quit, c to continue without paging--c
nwave = <optimized out>
wavehits = 0
opcode_strs = 0x7ffff7952010
pwd = <optimized out>
r = 0
#12 0x00000000004b998f in umr_dump_shaders (asic=asic@entry=0x1c08a50, decoder=decoder@entry=0x1c0b5d0, wd=wd@entry=0x1c0b6b0) at /home/mikhail/packaging-work/umr/src/lib/dump_ib.c:88
pshader = <optimized out>
shader = 0x1c094c0
#13 0x00000000004b056e in umr_read_ring (asic=asic@entry=0x1c08a50, ringpath=<optimized out>) at /home/mikhail/packaging-work/umr/src/app/ring_read.c:141
ringname = "gfx", '\000' <repeats 28 times>
from = ".", '\000' <repeats 30 times>
to = '\000' <repeats 31 times>
use_decoder = <optimized out>
enable_decoder = 1
wptr = 3072
rptr = 2044
drv_wptr = 3072
ringsize = 8192
start = 3076
end = 3072
value = <optimized out>
ring_data = <optimized out>
decoder = {pm = 4, src = {addr = 0, vmid = 0, ib_addr = 0}, pm4 = {cur_opcode = 34, pkt_type = 3, n_words = 4, cur_word = 0, control = 1074790400, next_ib_state = {ib_addr_lo = 0, ib_addr_hi = 0, ib_size = 0, ib_vmid = 0, tally = 0}, next_write_mem = {type = 570425344, addr_lo = 4082, addr_hi = 0, value = 0}, nop = {pktlen = 0, pkttype = 0, magic = 0, str = 0x0}}, sdma = {cur_opcode = 4294967295, cur_sub_opcode = 0, n_words = 0, cur_word = 0, header_dw = 0, next_write_mem = 0, next_ib_state = {ib_addr_lo = 0, ib_addr_hi = 0, csa_addr_lo = 0, csa_addr_hi = 0, ib_size = 0, ib_vmid = 0}}, next_ib = 0x1c0b410, next_ib_info = {ib_addr = 0, vm_base_addr = 0, vmid = 0, size = 0, addr = 768}, shader = 0x0}
pdecoder = 0x1c0b5d0
ppdecoder = <optimized out>
wd = 0x1c0b6b0
#14 0x0000000000496c68 in main (argc=<optimized out>, argv=<optimized out>) at /home/mikhail/packaging-work/umr/src/app/main.c:339
i = 3
j = <optimized out>
k = <optimized out>
l = <optimized out>
asic = 0x1c08a50
blockname = <optimized out>
str = <optimized out>
str2 = <optimized out>
asicname = "\000\000\000\000\004", '\000' <repeats 19 times>, "\001", '\000' <repeats 11 times>, "\004", '\000' <repeats 19 times>, "\a", '\000' <repeats 11 times>, "\004", '\000' <repeats 31 times>, "\004", '\000' <repeats 31 times>, "\004", '\000' <repeats 31 times>, "\004", '\000' <repeats 31 times>, "\004", '\000' <repeats 31 times>...
ipname = '\000' <repeats 56 times>, "-options", '\000' <repeats 120 times>, "\037", '\000' <repeats 31 times>...
regname = "\000\000\000\000\000 ", '\000' <repeats 18 times>, "\017\004", '\000' <repeats 11 times>, " ", '\000' <repeats 18 times>, "\220\377\377\377\377\377\377\377", '\000' <repeats 16 times>, "\031", '\000' <repeats 15 times>, "\a\000\000\000\000\000\000\000\037\000\000\000\000\000\000\000\003\000\000\000\000\000\000\000\030\220\275\001\000\000\000\000P\000\000\000\000\000\000\000\220\377\377\377\377\377\377\377\000\000\000\000\000\000\000\000\003\000\000\000w\000\000\000[\000\000\000\060", '\000' <repeats 27 times>, "n\000\000\000|", '\000' <repeats 19 times>...
req = {tv_sec = 0, tv_nsec = 7312272888393198945}
(gdb)
(gdb)
(gdb)
WARNING: multiple messages have this Message-ID (diff)
From: Mikhail Gavrilov <mikhail.v.gavrilov-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
To: Tom St Denis <tom.stdenis-5C7GfCeVMHo@public.gmane.org>
Cc: "Grodzovsky,
Andrey" <Andrey.Grodzovsky-5C7GfCeVMHo@public.gmane.org>,
"amd-gfx-PD4FTy7X32lNgt0PjOBp9y5qC8QIuHrW@public.gmane.org"
<amd-gfx-PD4FTy7X32lNgt0PjOBp9y5qC8QIuHrW@public.gmane.org>,
"linux-mm-Bw31MaZKKs3YtjvyW6yDsg@public.gmane.org"
<linux-mm-Bw31MaZKKs3YtjvyW6yDsg@public.gmane.org>,
"Deucher,
Alexander" <Alexander.Deucher-5C7GfCeVMHo@public.gmane.org>,
"Wentland, Harry" <Harry.Wentland-5C7GfCeVMHo@public.gmane.org>,
"Koenig,
Christian" <Christian.Koenig-5C7GfCeVMHo@public.gmane.org>
Subject: Re: After Vega 56/64 GPU hang I unable reboot system
Date: Thu, 10 Jan 2019 20:22:56 +0500 [thread overview]
Message-ID: <CABXGCsN4NwYG4UeJ9a-P_8cEd1qg_sC24z3L3Ek9Wn0JttkZ=g@mail.gmail.com> (raw)
In-Reply-To: <CABXGCsPPjz57=Et-V-_iGyY0GrEwfcK2QcRJcqiujUp90zaz-g-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
[-- Attachment #1: Type: text/plain, Size: 641 bytes --]
On Thu, 10 Jan 2019 at 00:36, Mikhail Gavrilov
<mikhail.v.gavrilov-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> wrote:
>
> All new one logs attached here.
>
> Thanks.
>
> P.S. This time I had to terminate command `./umr -O verbose,follow -R
> gfx[.] > gfx.log 2>&1` cause it tried to write log infinitely.
> I also had to terminate command `./umr -O verbose,follow -R gfx[.] >
> gfx.log 2>&1` cause it stuck for a long time.
>
>
It became clear why umr stuck at the gfx dump. I ran umr under gdb and
I got a segfault at a moment when umr was stuck earlier.
Tom, are you here? Can you look attached backtrace?
--
Best Regards,
Mike Gavrilov.
[-- Attachment #2: umr-backtrace.txt --]
[-- Type: text/plain, Size: 11754 bytes --]
pgm[1@0x80010e520500 + 0x12c4] = 0x043a0f18 v_sub_f32_e32 v29, v24, v7
pgm[1@0x80010e520500 + 0x12c8] = 0x0a3c3d0d v_mul_f32_e32 v30, v13, v30
pgm[1@0x80010e520500 + 0x12cc] = 0x2c3c3b11 v_mac_f32_e32 v30, v17, v29
pgm[1@0x80010e520500 + 0x12d0] = 0x0238391e v_add_f32_e32 v28, v30, v28
pgm[1@0x80010e520500 + 0x12d4] = 0xd105801c v_mul_f32_e64 v28, v21, v28 clamp
pgm[1@0x80010e520500 + 0x12d8] = 0x00023915 ;;
pgm[1@0x80010e520500 + 0x12dc] = 0xd1c1001e v_mad_f32 v30, v28, v13, v6
pgm[1@0x80010e520500 + 0x12e0] = 0x041a1b1c ;;
pgm[1@0x80010e520500 + 0x12e4] = 0xd1c1001d v_mad_f32 v29, v28, v17, v7
pgm[1@0x80010e520500 + 0x12e8] = 0x041e231c ;;
pgm[1@0x80010e520500 + 0x12ec] = 0x0432331e v_sub_f32_e32 v25, v30, v25
pgm[1@0x80010e520500 + 0x12f0] = 0xd1c1001c v_mad_f32 v28, v28, v10, v3
pgm[1@0x80010e520500 + 0x12f4] = 0x040e151c ;;
pgm[1@0x80010e520500 + 0x12f8] = 0x0430311d v_sub_f32_e32 v24, v29, v24
pgm[1@0x80010e520500 + 0x12fc] = 0x0a323319 v_mul_f32_e32 v25, v25, v25
pgm[1@0x80010e520500 + 0x1300] = 0x0434351c v_sub_f32_e32 v26, v28, v26
pgm[1@0x80010e520500 + 0x1304] = 0x2c323118 v_mac_f32_e32 v25, v24, v24
pgm[1@0x80010e520500 + 0x1308] = 0x2c32351a v_mac_f32_e32 v25, v26, v26
pgm[1@0x80010e520500 + 0x130c] = 0x7c8c331b v_cmp_ge_f32_e32 vcc, v27, v25
pgm[1@0x80010e520500 + 0x1310] = 0xd100001e v_cndmask_b32_e64 v30, 0, -1, vcc
pgm[1@0x80010e520500 + 0x1314] = 0x01a98280 ;;
pgm[1@0x80010e520500 + 0x1318] = 0x87fe007e s_or_b64 exec, exec, s[0:1]
pgm[1@0x80010e520500 + 0x131c] = 0x26302e9f v_and_b32_e32 v24, 31, v23
pgm[1@0x80010e520500 + 0x1320] = 0xd1120018 v_lshlrev_b32_e64 v24, v24, 1
pgm[1@0x80010e520500 + 0x1324] = 0x00010318 ;;
pgm[1@0x80010e520500 + 0x1328] = 0x7e32570e v_not_b32_e32 v25, v14
pgm[1@0x80010e520500 + 0x132c] = 0x26303318 v_and_b32_e32 v24, v24, v25
pgm[1@0x80010e520500 + 0x1330] = 0x7d9a3c80 v_cmp_ne_u32_e32 vcc, 0, v30
pgm[1@0x80010e520500 + 0x1334] = 0x00363080 v_cndmask_b32_e32 v27, 0, v24, vcc
pgm[1@0x80010e520500 + 0x1338] = 0x2a30370e v_xor_b32_e32 v24, v14, v27
pgm[1@0x80010e520500 + 0x133c] = 0x682e2e81 v_add_u32_e32 v23, 1, v23
pgm[1@0x80010e520500 + 0x1340] = 0x682c2c90 v_add_u32_e32 v22, 16, v22
pgm[1@0x80010e520500 + 0x1344] = 0x7e320280 v_mov_b32_e32 v25, 0
pgm[1@0x80010e520500 + 0x1348] = 0xbf820000 s_branch 0
pgm[1@0x80010e520500 + 0x134c] = 0x7d9a3280 v_cmp_ne_u32_e32 vcc, 0, v25
pgm[1@0x80010e520500 + 0x1350] = 0x86ea6a7e s_and_b64 vcc, exec, vcc
pgm[1@0x80010e520500 + 0x1354] = 0xbf86fd70 s_cbranch_vccz 64880
pgm[1@0x80010e520500 + 0x1358] = 0xb0038000 s_movk_i32 s3, 0x8000
pgm[1@0x80010e520500 + 0x135c] = 0xc00e0001 s_load_dwordx8 s[0:7], s[2:3], 0x0
pgm[1@0x80010e520500 + 0x1360] = 0x00000000 ;;
pgm[1@0x80010e520500 + 0x1364] = 0x7e1e030e v_mov_b32_e32 v15, v14
pgm[1@0x80010e520500 + 0x1368] = 0x7e20030e v_mov_b32_e32 v16, v14
pgm[1@0x80010e520500 + 0x136c] = 0x7e22030e v_mov_b32_e32 v17, v14
pgm[1@0x80010e520500 + 0x1370] = 0xbf8cc07f s_waitcnt lgkmcnt(0)
pgm[1@0x80010e520500 + 0x1374] = 0xf0201f00 image_store v[14:17], v0, s[0:7] dmask:0xf unorm
pgm[1@0x80010e520500 + 0x1378] = 0x00000e00 ;;
pgm[1@0x80010e520500 + 0x137c] = 0xbf810000 s_endpgm
End of disassembly.
Disassembly of shader 1@0x80010010b400 of length 161292 bytes from IB[1@0x80010e555000 + 0x1078]
Program received signal SIGSEGV, Segmentation fault.
0x0000000000494220 in llvm::MCAssembler::getAtom(llvm::MCSymbol const&) const [clone .cold.240] ()
(gdb) thread apply all bt full
Thread 1 (Thread 0x7ffff7a20740 (LWP 5975)):
#0 0x0000000000494220 in llvm::MCAssembler::getAtom(llvm::MCSymbol const&) const [clone .cold.240] ()
No symbol table info available.
#1 0x00007fff3fffffff in ?? ()
No symbol table info available.
#2 0x0000000000000902 in ?? ()
No symbol table info available.
#3 0x0000000000533a41 in llvm::AMDGPUDisassembler::decodeOperand_VReg_96(unsigned int) const ()
No symbol table info available.
#4 0x0000000000537243 in DecodeSReg_128RegisterClass(llvm::MCInst&, unsigned int, unsigned long, void const*) [clone .isra.60] ()
No symbol table info available.
#5 0x00000000005464b2 in llvm::MCDisassembler::DecodeStatus llvm::decodeToMCInst<unsigned long>(llvm::MCDisassembler::DecodeStatus, unsigned int, unsigned long, llvm::MCInst&, unsigned long, void const*, bool&) ()
No symbol table info available.
#6 0x000000000054e1ad in llvm::AMDGPUDisassembler::tryDecodeInst(unsigned char const*, llvm::MCInst&, unsigned long, unsigned long) const ()
No symbol table info available.
#7 0x000000000054ec0f in llvm::AMDGPUDisassembler::getInstruction(llvm::MCInst&, unsigned long&, llvm::ArrayRef<unsigned char>, unsigned long, llvm::raw_ostream&, llvm::raw_ostream&) const ()
No symbol table info available.
#8 0x00000000007d0a5e in LLVMDisasmInstruction ()
No symbol table info available.
#9 0x00000000004c24b1 in umr_llvm_disasm (disasm_text=0x7ffff7952010, PC=140741784417280, inst_bytes=161292, inst=0x4599e30 "\202", asic=<optimized out>) at /home/mikhail/packaging-work/umr/src/lib/umr_llvm_disasm.c:81
i = 4532
tmp = "\tv_madak_f32 v255, v255, v255, 0xfffffff0\000\071\070, attr49.z\000isn't aligned 3\000\061\065 offen offset:2521 glc\000\061 glc tfe ; Error: unknown operand encoding 125\000oding 125\000\000\000\200\000\000\000\321\002\000\000\340\023\000\000v\002\000\000\000\000\000\000N\000\000\000\000\000\000\000\t\000\000\000\000\000\000\000\274\000\000\000}\000\000\000"...
cpuname = <optimized out>
disasm_ref = 0x454a080
x = 18128
z = <optimized out>
n = <optimized out>
disasm_ref = <optimized out>
x = <optimized out>
z = <optimized out>
i = <optimized out>
n = <optimized out>
tmp = <optimized out>
cpuname = <optimized out>
#10 umr_llvm_disasm (asic=<optimized out>, inst=0x4599e30 "\202", inst_bytes=161292, PC=140741784417280, disasm_text=0x7ffff7952010) at /home/mikhail/packaging-work/umr/src/lib/umr_llvm_disasm.c:38
disasm_ref = <optimized out>
x = <optimized out>
z = <optimized out>
i = <optimized out>
n = <optimized out>
tmp = <optimized out>
cpuname = <optimized out>
#11 0x00000000004c26d1 in umr_vm_disasm (asic=asic@entry=0x1c08a50, vmid=1, addr=140741784417280, PC=PC@entry=0, size=<optimized out>, start_offset=start_offset@entry=0, wd=0x1c0b6b0) at /home/mikhail/packaging-work/umr/src/lib/umr_llvm_disasm.c:170
opcodes = 0x4599e30
x = <optimized out>
y = <optimized out>
--Type <RET> for more, q to quit, c to continue without paging--c
nwave = <optimized out>
wavehits = 0
opcode_strs = 0x7ffff7952010
pwd = <optimized out>
r = 0
#12 0x00000000004b998f in umr_dump_shaders (asic=asic@entry=0x1c08a50, decoder=decoder@entry=0x1c0b5d0, wd=wd@entry=0x1c0b6b0) at /home/mikhail/packaging-work/umr/src/lib/dump_ib.c:88
pshader = <optimized out>
shader = 0x1c094c0
#13 0x00000000004b056e in umr_read_ring (asic=asic@entry=0x1c08a50, ringpath=<optimized out>) at /home/mikhail/packaging-work/umr/src/app/ring_read.c:141
ringname = "gfx", '\000' <repeats 28 times>
from = ".", '\000' <repeats 30 times>
to = '\000' <repeats 31 times>
use_decoder = <optimized out>
enable_decoder = 1
wptr = 3072
rptr = 2044
drv_wptr = 3072
ringsize = 8192
start = 3076
end = 3072
value = <optimized out>
ring_data = <optimized out>
decoder = {pm = 4, src = {addr = 0, vmid = 0, ib_addr = 0}, pm4 = {cur_opcode = 34, pkt_type = 3, n_words = 4, cur_word = 0, control = 1074790400, next_ib_state = {ib_addr_lo = 0, ib_addr_hi = 0, ib_size = 0, ib_vmid = 0, tally = 0}, next_write_mem = {type = 570425344, addr_lo = 4082, addr_hi = 0, value = 0}, nop = {pktlen = 0, pkttype = 0, magic = 0, str = 0x0}}, sdma = {cur_opcode = 4294967295, cur_sub_opcode = 0, n_words = 0, cur_word = 0, header_dw = 0, next_write_mem = 0, next_ib_state = {ib_addr_lo = 0, ib_addr_hi = 0, csa_addr_lo = 0, csa_addr_hi = 0, ib_size = 0, ib_vmid = 0}}, next_ib = 0x1c0b410, next_ib_info = {ib_addr = 0, vm_base_addr = 0, vmid = 0, size = 0, addr = 768}, shader = 0x0}
pdecoder = 0x1c0b5d0
ppdecoder = <optimized out>
wd = 0x1c0b6b0
#14 0x0000000000496c68 in main (argc=<optimized out>, argv=<optimized out>) at /home/mikhail/packaging-work/umr/src/app/main.c:339
i = 3
j = <optimized out>
k = <optimized out>
l = <optimized out>
asic = 0x1c08a50
blockname = <optimized out>
str = <optimized out>
str2 = <optimized out>
asicname = "\000\000\000\000\004", '\000' <repeats 19 times>, "\001", '\000' <repeats 11 times>, "\004", '\000' <repeats 19 times>, "\a", '\000' <repeats 11 times>, "\004", '\000' <repeats 31 times>, "\004", '\000' <repeats 31 times>, "\004", '\000' <repeats 31 times>, "\004", '\000' <repeats 31 times>, "\004", '\000' <repeats 31 times>...
ipname = '\000' <repeats 56 times>, "-options", '\000' <repeats 120 times>, "\037", '\000' <repeats 31 times>...
regname = "\000\000\000\000\000 ", '\000' <repeats 18 times>, "\017\004", '\000' <repeats 11 times>, " ", '\000' <repeats 18 times>, "\220\377\377\377\377\377\377\377", '\000' <repeats 16 times>, "\031", '\000' <repeats 15 times>, "\a\000\000\000\000\000\000\000\037\000\000\000\000\000\000\000\003\000\000\000\000\000\000\000\030\220\275\001\000\000\000\000P\000\000\000\000\000\000\000\220\377\377\377\377\377\377\377\000\000\000\000\000\000\000\000\003\000\000\000w\000\000\000[\000\000\000\060", '\000' <repeats 27 times>, "n\000\000\000|", '\000' <repeats 19 times>...
req = {tv_sec = 0, tv_nsec = 7312272888393198945}
(gdb)
(gdb)
(gdb)
[-- Attachment #3: Type: text/plain, Size: 154 bytes --]
_______________________________________________
amd-gfx mailing list
amd-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/amd-gfx
next prev parent reply other threads:[~2019-01-10 15:23 UTC|newest]
Thread overview: 70+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-12-04 22:18 After Vega 56/64 GPU hang I unable reboot system Mikhail Gavrilov
2018-12-04 22:18 ` Mikhail Gavrilov
2018-12-14 19:36 ` Wentland, Harry
2018-12-14 19:36 ` Wentland, Harry
2018-12-15 9:42 ` Mikhail Gavrilov
2018-12-15 9:42 ` Mikhail Gavrilov
2018-12-17 18:51 ` Wentland, Harry
2018-12-17 18:51 ` Wentland, Harry
2018-12-17 19:08 ` Grodzovsky, Andrey
2018-12-17 19:08 ` Grodzovsky, Andrey
2018-12-19 18:35 ` Mikhail Gavrilov
2018-12-19 18:35 ` Mikhail Gavrilov
2018-12-19 20:21 ` Grodzovsky, Andrey
2018-12-19 20:21 ` Grodzovsky, Andrey
2018-12-19 20:51 ` StDenis, Tom
2018-12-19 20:51 ` StDenis, Tom
2018-12-19 20:56 ` StDenis, Tom
2018-12-19 20:56 ` StDenis, Tom
2018-12-19 21:14 ` Mikhail Gavrilov
2018-12-19 21:14 ` Mikhail Gavrilov
2018-12-19 21:21 ` Mikhail Gavrilov
2018-12-19 21:21 ` Mikhail Gavrilov
2018-12-19 22:41 ` StDenis, Tom
2018-12-19 22:41 ` StDenis, Tom
2018-12-20 3:29 ` Mikhail Gavrilov
2018-12-20 3:29 ` Mikhail Gavrilov
2018-12-20 11:17 ` StDenis, Tom
2018-12-20 11:17 ` StDenis, Tom
2018-12-20 11:45 ` Mikhail Gavrilov
2018-12-20 11:45 ` Mikhail Gavrilov
2018-12-20 14:06 ` StDenis, Tom
2018-12-20 14:06 ` StDenis, Tom
2018-12-20 14:08 ` StDenis, Tom
2018-12-20 14:08 ` StDenis, Tom
2018-12-20 14:19 ` StDenis, Tom
2018-12-20 14:19 ` StDenis, Tom
[not found] ` <96c70496-ce62-b162-187c-ff34ebb84ec2-5C7GfCeVMHo@public.gmane.org>
2018-12-20 16:07 ` Mikhail Gavrilov
2018-12-20 16:20 ` StDenis, Tom
2018-12-20 16:20 ` StDenis, Tom
2018-12-22 10:28 ` Mikhail Gavrilov
2018-12-22 10:28 ` Mikhail Gavrilov
2018-12-22 10:28 ` Mikhail Gavrilov
2019-01-03 20:23 ` Mikhail Gavrilov
2019-01-03 20:23 ` Mikhail Gavrilov
2019-01-04 17:50 ` Mikhail Gavrilov
2019-01-07 18:46 ` Grodzovsky, Andrey
2019-01-07 18:46 ` Grodzovsky, Andrey
2019-01-09 19:36 ` Mikhail Gavrilov
2019-01-09 19:36 ` Mikhail Gavrilov
2019-01-09 20:35 ` Grodzovsky, Andrey
2019-01-09 20:35 ` Grodzovsky, Andrey
2019-01-09 21:12 ` Mikhail Gavrilov
2019-01-09 21:12 ` Mikhail Gavrilov
2019-01-09 21:12 ` Mikhail Gavrilov
2019-01-09 21:48 ` Grodzovsky, Andrey
2019-01-09 21:48 ` Grodzovsky, Andrey
2019-01-10 4:04 ` Mikhail Gavrilov
2019-01-10 4:04 ` Mikhail Gavrilov
2019-01-10 8:54 ` Michel Dänzer
2019-01-10 8:54 ` Michel Dänzer
2019-01-10 9:42 ` Mikhail Gavrilov
2019-01-10 9:42 ` Mikhail Gavrilov
2019-01-10 9:42 ` Mikhail Gavrilov
2019-01-10 10:38 ` Michel Dänzer
2019-01-10 10:38 ` Michel Dänzer
2019-01-10 15:22 ` Mikhail Gavrilov [this message]
2019-01-10 15:22 ` Mikhail Gavrilov
2019-01-10 15:22 ` Mikhail Gavrilov
2019-01-10 15:31 ` StDenis, Tom
2019-01-10 15:31 ` StDenis, Tom
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to='CABXGCsN4NwYG4UeJ9a-P_8cEd1qg_sC24z3L3Ek9Wn0JttkZ=g@mail.gmail.com' \
--to=mikhail.v.gavrilov@gmail.com \
--cc=Alexander.Deucher@amd.com \
--cc=Andrey.Grodzovsky@amd.com \
--cc=Christian.Koenig@amd.com \
--cc=Harry.Wentland@amd.com \
--cc=amd-gfx@lists.freedesktop.org \
--cc=linux-mm@kvack.org \
--cc=tom.stdenis@amd.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.