[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [PULL 11/24] tcg/optimize: Use tcg_constant_internal with constant f
|
From: |
Alistair Francis |
|
Subject: |
Re: [PULL 11/24] tcg/optimize: Use tcg_constant_internal with constant folding |
|
Date: |
Fri, 15 Jan 2021 15:03:12 -0800 |
On Wed, Jan 13, 2021 at 6:32 PM Richard Henderson
<richard.henderson@linaro.org> wrote:
>
> Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
This patch results in a QEMU seg fault when starting userspace on RISC-V 32-bit.
This is the full backtrace:
```
#0 0x0000555555a67c4d in ts_are_copies (ts2=0x7fffa8008008,
ts1=0x7fffa8001e40) at ../tcg/optimize.c:163
#1 tcg_opt_gen_mov (s=s@entry=0x7fffa8000b60,
op=op@entry=0x7fffa81ac778, dst=140736011968064, src=140736011993096)
at ../tcg/optimize.c:191
#2 0x0000555555a67dcb in tcg_opt_gen_movi (s=s@entry=0x7fffa8000b60,
temps_used=temps_used@entry=0x7ffff1cb92c0,
op=op@entry=0x7fffa81ac778, dst=<optimized out>, val=<optimized out>)
at ../tcg/optimize.c:249
#3 0x0000555555a6914d in tcg_optimize (s=s@entry=0x7fffa8000b60) at
../tcg/optimize.c:1242
#4 0x0000555555abb248 in tcg_gen_code (s=0x7fffa8000b60,
tb=tb@entry=0x7fffae84edc0 <code_gen_buffer+42266003>) at
../tcg/tcg.c:4406
#5 0x0000555555a7f4d5 in tb_gen_code (cpu=cpu@entry=0x7ffff7fac930,
pc=160234, cs_base=0, flags=16640, cflags=-16252928) at
../accel/tcg/translate-all.c:1952
#6 0x0000555555ae4fe4 in tb_find (cf_mask=<optimized out>, tb_exit=0,
last_tb=0x0, cpu=0x7ffff7fac930) at ../accel/tcg/cpu-exec.c:454
#7 cpu_exec (cpu=cpu@entry=0x7ffff7fac930) at ../accel/tcg/cpu-exec.c:810
#8 0x0000555555aa6513 in tcg_cpus_exec (cpu=cpu@entry=0x7ffff7fac930)
at ../accel/tcg/tcg-cpus.c:57
#9 0x0000555555a8c7a3 in mttcg_cpu_thread_fn
(arg=arg@entry=0x7ffff7fac930) at ../accel/tcg/tcg-cpus-mttcg.c:69
#10 0x0000555555c94209 in qemu_thread_start (args=0x7ffff1cb96d0) at
../util/qemu-thread-posix.c:521
#11 0x00007ffff673a3e9 in start_thread () at /usr/lib/libpthread.so.0
#12 0x00007ffff6395293 in clone () at /usr/lib/libc.so.6
```
I run QEMU with these arguments:
./build/riscv32-softmmu/qemu-system-riscv32 \
-machine virt -serial mon:stdio -serial null -nographic \
-append "root=/dev/vda rw highres=off console=ttyS0 ip=dhcp earlycon=sbi" \
-device virtio-net-device,netdev=net0,mac=52:54:00:12:34:02
-netdev user,id=net0 \
-object rng-random,filename=/dev/urandom,id=rng0 -device
virtio-rng-device,rng=rng0 \
-smp 4 -d guest_errors -m 256M \
-kernel ./Image \
-drive
id=disk0,file=./core-image-minimal-qemuriscv32.ext4,if=none,format=raw
\
-device virtio-blk-device,drive=disk0 \
-bios default
I am uploading the images to:
https://nextcloud.alistair23.me/index.php/s/MQFyGGNLPZjLZPH
Although apparently it will take a few hours to upload the 2GB rootFS.
Alistair
> ---
> tcg/optimize.c | 108 ++++++++++++++++++++++---------------------------
> 1 file changed, 49 insertions(+), 59 deletions(-)
>
> diff --git a/tcg/optimize.c b/tcg/optimize.c
> index 49bf1386c7..bda727d5ed 100644
> --- a/tcg/optimize.c
> +++ b/tcg/optimize.c
> @@ -178,37 +178,6 @@ static bool args_are_copies(TCGArg arg1, TCGArg arg2)
> return ts_are_copies(arg_temp(arg1), arg_temp(arg2));
> }
>
> -static void tcg_opt_gen_movi(TCGContext *s, TCGOp *op, TCGArg dst, uint64_t
> val)
> -{
> - const TCGOpDef *def;
> - TCGOpcode new_op;
> - uint64_t mask;
> - TempOptInfo *di = arg_info(dst);
> -
> - def = &tcg_op_defs[op->opc];
> - if (def->flags & TCG_OPF_VECTOR) {
> - new_op = INDEX_op_dupi_vec;
> - } else if (def->flags & TCG_OPF_64BIT) {
> - new_op = INDEX_op_movi_i64;
> - } else {
> - new_op = INDEX_op_movi_i32;
> - }
> - op->opc = new_op;
> - /* TCGOP_VECL and TCGOP_VECE remain unchanged. */
> - op->args[0] = dst;
> - op->args[1] = val;
> -
> - reset_temp(dst);
> - di->is_const = true;
> - di->val = val;
> - mask = val;
> - if (TCG_TARGET_REG_BITS > 32 && new_op == INDEX_op_movi_i32) {
> - /* High bits of the destination are now garbage. */
> - mask |= ~0xffffffffull;
> - }
> - di->mask = mask;
> -}
> -
> static void tcg_opt_gen_mov(TCGContext *s, TCGOp *op, TCGArg dst, TCGArg src)
> {
> TCGTemp *dst_ts = arg_temp(dst);
> @@ -259,6 +228,27 @@ static void tcg_opt_gen_mov(TCGContext *s, TCGOp *op,
> TCGArg dst, TCGArg src)
> }
> }
>
> +static void tcg_opt_gen_movi(TCGContext *s, TCGTempSet *temps_used,
> + TCGOp *op, TCGArg dst, uint64_t val)
> +{
> + const TCGOpDef *def = &tcg_op_defs[op->opc];
> + TCGType type;
> + TCGTemp *tv;
> +
> + if (def->flags & TCG_OPF_VECTOR) {
> + type = TCGOP_VECL(op) + TCG_TYPE_V64;
> + } else if (def->flags & TCG_OPF_64BIT) {
> + type = TCG_TYPE_I64;
> + } else {
> + type = TCG_TYPE_I32;
> + }
> +
> + /* Convert movi to mov with constant temp. */
> + tv = tcg_constant_internal(type, val);
> + init_ts_info(temps_used, tv);
> + tcg_opt_gen_mov(s, op, dst, temp_arg(tv));
> +}
> +
> static uint64_t do_constant_folding_2(TCGOpcode op, uint64_t x, uint64_t y)
> {
> uint64_t l64, h64;
> @@ -622,7 +612,7 @@ void tcg_optimize(TCGContext *s)
> nb_temps = s->nb_temps;
> nb_globals = s->nb_globals;
>
> - bitmap_zero(temps_used.l, nb_temps);
> + memset(&temps_used, 0, sizeof(temps_used));
> for (i = 0; i < nb_temps; ++i) {
> s->temps[i].state_ptr = NULL;
> }
> @@ -727,7 +717,7 @@ void tcg_optimize(TCGContext *s)
> CASE_OP_32_64(rotr):
> if (arg_is_const(op->args[1])
> && arg_info(op->args[1])->val == 0) {
> - tcg_opt_gen_movi(s, op, op->args[0], 0);
> + tcg_opt_gen_movi(s, &temps_used, op, op->args[0], 0);
> continue;
> }
> break;
> @@ -1054,7 +1044,7 @@ void tcg_optimize(TCGContext *s)
>
> if (partmask == 0) {
> tcg_debug_assert(nb_oargs == 1);
> - tcg_opt_gen_movi(s, op, op->args[0], 0);
> + tcg_opt_gen_movi(s, &temps_used, op, op->args[0], 0);
> continue;
> }
> if (affected == 0) {
> @@ -1071,7 +1061,7 @@ void tcg_optimize(TCGContext *s)
> CASE_OP_32_64(mulsh):
> if (arg_is_const(op->args[2])
> && arg_info(op->args[2])->val == 0) {
> - tcg_opt_gen_movi(s, op, op->args[0], 0);
> + tcg_opt_gen_movi(s, &temps_used, op, op->args[0], 0);
> continue;
> }
> break;
> @@ -1098,7 +1088,7 @@ void tcg_optimize(TCGContext *s)
> CASE_OP_32_64_VEC(sub):
> CASE_OP_32_64_VEC(xor):
> if (args_are_copies(op->args[1], op->args[2])) {
> - tcg_opt_gen_movi(s, op, op->args[0], 0);
> + tcg_opt_gen_movi(s, &temps_used, op, op->args[0], 0);
> continue;
> }
> break;
> @@ -1115,14 +1105,14 @@ void tcg_optimize(TCGContext *s)
> break;
> CASE_OP_32_64(movi):
> case INDEX_op_dupi_vec:
> - tcg_opt_gen_movi(s, op, op->args[0], op->args[1]);
> + tcg_opt_gen_movi(s, &temps_used, op, op->args[0], op->args[1]);
> break;
>
> case INDEX_op_dup_vec:
> if (arg_is_const(op->args[1])) {
> tmp = arg_info(op->args[1])->val;
> tmp = dup_const(TCGOP_VECE(op), tmp);
> - tcg_opt_gen_movi(s, op, op->args[0], tmp);
> + tcg_opt_gen_movi(s, &temps_used, op, op->args[0], tmp);
> break;
> }
> goto do_default;
> @@ -1132,7 +1122,7 @@ void tcg_optimize(TCGContext *s)
> if (arg_is_const(op->args[1]) && arg_is_const(op->args[2])) {
> tmp = arg_info(op->args[1])->val;
> if (tmp == arg_info(op->args[2])->val) {
> - tcg_opt_gen_movi(s, op, op->args[0], tmp);
> + tcg_opt_gen_movi(s, &temps_used, op, op->args[0], tmp);
> break;
> }
> } else if (args_are_copies(op->args[1], op->args[2])) {
> @@ -1160,7 +1150,7 @@ void tcg_optimize(TCGContext *s)
> case INDEX_op_extrh_i64_i32:
> if (arg_is_const(op->args[1])) {
> tmp = do_constant_folding(opc, arg_info(op->args[1])->val,
> 0);
> - tcg_opt_gen_movi(s, op, op->args[0], tmp);
> + tcg_opt_gen_movi(s, &temps_used, op, op->args[0], tmp);
> break;
> }
> goto do_default;
> @@ -1190,7 +1180,7 @@ void tcg_optimize(TCGContext *s)
> if (arg_is_const(op->args[1]) && arg_is_const(op->args[2])) {
> tmp = do_constant_folding(opc, arg_info(op->args[1])->val,
> arg_info(op->args[2])->val);
> - tcg_opt_gen_movi(s, op, op->args[0], tmp);
> + tcg_opt_gen_movi(s, &temps_used, op, op->args[0], tmp);
> break;
> }
> goto do_default;
> @@ -1201,7 +1191,7 @@ void tcg_optimize(TCGContext *s)
> TCGArg v = arg_info(op->args[1])->val;
> if (v != 0) {
> tmp = do_constant_folding(opc, v, 0);
> - tcg_opt_gen_movi(s, op, op->args[0], tmp);
> + tcg_opt_gen_movi(s, &temps_used, op, op->args[0], tmp);
> } else {
> tcg_opt_gen_mov(s, op, op->args[0], op->args[2]);
> }
> @@ -1214,7 +1204,7 @@ void tcg_optimize(TCGContext *s)
> tmp = deposit64(arg_info(op->args[1])->val,
> op->args[3], op->args[4],
> arg_info(op->args[2])->val);
> - tcg_opt_gen_movi(s, op, op->args[0], tmp);
> + tcg_opt_gen_movi(s, &temps_used, op, op->args[0], tmp);
> break;
> }
> goto do_default;
> @@ -1223,7 +1213,7 @@ void tcg_optimize(TCGContext *s)
> if (arg_is_const(op->args[1])) {
> tmp = extract64(arg_info(op->args[1])->val,
> op->args[2], op->args[3]);
> - tcg_opt_gen_movi(s, op, op->args[0], tmp);
> + tcg_opt_gen_movi(s, &temps_used, op, op->args[0], tmp);
> break;
> }
> goto do_default;
> @@ -1232,7 +1222,7 @@ void tcg_optimize(TCGContext *s)
> if (arg_is_const(op->args[1])) {
> tmp = sextract64(arg_info(op->args[1])->val,
> op->args[2], op->args[3]);
> - tcg_opt_gen_movi(s, op, op->args[0], tmp);
> + tcg_opt_gen_movi(s, &temps_used, op, op->args[0], tmp);
> break;
> }
> goto do_default;
> @@ -1249,7 +1239,7 @@ void tcg_optimize(TCGContext *s)
> tmp = (int32_t)(((uint32_t)v1 >> shr) |
> ((uint32_t)v2 << (32 - shr)));
> }
> - tcg_opt_gen_movi(s, op, op->args[0], tmp);
> + tcg_opt_gen_movi(s, &temps_used, op, op->args[0], tmp);
> break;
> }
> goto do_default;
> @@ -1258,7 +1248,7 @@ void tcg_optimize(TCGContext *s)
> tmp = do_constant_folding_cond(opc, op->args[1],
> op->args[2], op->args[3]);
> if (tmp != 2) {
> - tcg_opt_gen_movi(s, op, op->args[0], tmp);
> + tcg_opt_gen_movi(s, &temps_used, op, op->args[0], tmp);
> break;
> }
> goto do_default;
> @@ -1268,7 +1258,7 @@ void tcg_optimize(TCGContext *s)
> op->args[1], op->args[2]);
> if (tmp != 2) {
> if (tmp) {
> - bitmap_zero(temps_used.l, nb_temps);
> + memset(&temps_used, 0, sizeof(temps_used));
> op->opc = INDEX_op_br;
> op->args[0] = op->args[3];
> } else {
> @@ -1314,7 +1304,7 @@ void tcg_optimize(TCGContext *s)
> uint64_t a = ((uint64_t)ah << 32) | al;
> uint64_t b = ((uint64_t)bh << 32) | bl;
> TCGArg rl, rh;
> - TCGOp *op2 = tcg_op_insert_before(s, op, INDEX_op_movi_i32);
> + TCGOp *op2 = tcg_op_insert_before(s, op, INDEX_op_mov_i32);
>
> if (opc == INDEX_op_add2_i32) {
> a += b;
> @@ -1324,8 +1314,8 @@ void tcg_optimize(TCGContext *s)
>
> rl = op->args[0];
> rh = op->args[1];
> - tcg_opt_gen_movi(s, op, rl, (int32_t)a);
> - tcg_opt_gen_movi(s, op2, rh, (int32_t)(a >> 32));
> + tcg_opt_gen_movi(s, &temps_used, op, rl, (int32_t)a);
> + tcg_opt_gen_movi(s, &temps_used, op2, rh, (int32_t)(a >>
> 32));
> break;
> }
> goto do_default;
> @@ -1336,12 +1326,12 @@ void tcg_optimize(TCGContext *s)
> uint32_t b = arg_info(op->args[3])->val;
> uint64_t r = (uint64_t)a * b;
> TCGArg rl, rh;
> - TCGOp *op2 = tcg_op_insert_before(s, op, INDEX_op_movi_i32);
> + TCGOp *op2 = tcg_op_insert_before(s, op, INDEX_op_mov_i32);
>
> rl = op->args[0];
> rh = op->args[1];
> - tcg_opt_gen_movi(s, op, rl, (int32_t)r);
> - tcg_opt_gen_movi(s, op2, rh, (int32_t)(r >> 32));
> + tcg_opt_gen_movi(s, &temps_used, op, rl, (int32_t)r);
> + tcg_opt_gen_movi(s, &temps_used, op2, rh, (int32_t)(r >>
> 32));
> break;
> }
> goto do_default;
> @@ -1352,7 +1342,7 @@ void tcg_optimize(TCGContext *s)
> if (tmp != 2) {
> if (tmp) {
> do_brcond_true:
> - bitmap_zero(temps_used.l, nb_temps);
> + memset(&temps_used, 0, sizeof(temps_used));
> op->opc = INDEX_op_br;
> op->args[0] = op->args[5];
> } else {
> @@ -1368,7 +1358,7 @@ void tcg_optimize(TCGContext *s)
> /* Simplify LT/GE comparisons vs zero to a single compare
> vs the high word of the input. */
> do_brcond_high:
> - bitmap_zero(temps_used.l, nb_temps);
> + memset(&temps_used, 0, sizeof(temps_used));
> op->opc = INDEX_op_brcond_i32;
> op->args[0] = op->args[1];
> op->args[1] = op->args[3];
> @@ -1394,7 +1384,7 @@ void tcg_optimize(TCGContext *s)
> goto do_default;
> }
> do_brcond_low:
> - bitmap_zero(temps_used.l, nb_temps);
> + memset(&temps_used, 0, sizeof(temps_used));
> op->opc = INDEX_op_brcond_i32;
> op->args[1] = op->args[2];
> op->args[2] = op->args[4];
> @@ -1429,7 +1419,7 @@ void tcg_optimize(TCGContext *s)
> op->args[5]);
> if (tmp != 2) {
> do_setcond_const:
> - tcg_opt_gen_movi(s, op, op->args[0], tmp);
> + tcg_opt_gen_movi(s, &temps_used, op, op->args[0], tmp);
> } else if ((op->args[5] == TCG_COND_LT
> || op->args[5] == TCG_COND_GE)
> && arg_is_const(op->args[3])
> @@ -1514,7 +1504,7 @@ void tcg_optimize(TCGContext *s)
> block, otherwise we only trash the output args. "mask" is
> the non-zero bits mask for the first output arg. */
> if (def->flags & TCG_OPF_BB_END) {
> - bitmap_zero(temps_used.l, nb_temps);
> + memset(&temps_used, 0, sizeof(temps_used));
> } else {
> do_reset_output:
> for (i = 0; i < nb_oargs; i++) {
> --
> 2.25.1
>
>
- [PULL 05/24] tcg: Expand TCGTemp.val to 64-bits, (continued)
- [PULL 05/24] tcg: Expand TCGTemp.val to 64-bits, Richard Henderson, 2021/01/13
- [PULL 06/24] tcg: Rename struct tcg_temp_info to TempOptInfo, Richard Henderson, 2021/01/13
- [PULL 07/24] tcg: Expand TempOptInfo to 64-bits, Richard Henderson, 2021/01/13
- [PULL 08/24] tcg: Introduce TYPE_CONST temporaries, Richard Henderson, 2021/01/13
- [PULL 09/24] tcg/optimize: Improve find_better_copy, Richard Henderson, 2021/01/13
- [PULL 10/24] tcg/optimize: Adjust TempOptInfo allocation, Richard Henderson, 2021/01/13
- [PULL 13/24] tcg: Use tcg_constant_i32 with icount expander, Richard Henderson, 2021/01/13
- [PULL 12/24] tcg: Convert tcg_gen_dupi_vec to TCG_CONST, Richard Henderson, 2021/01/13
- [PULL 19/24] tcg: Add tcg_reg_alloc_dup2, Richard Henderson, 2021/01/13
- [PULL 11/24] tcg/optimize: Use tcg_constant_internal with constant folding, Richard Henderson, 2021/01/13
- Re: [PULL 11/24] tcg/optimize: Use tcg_constant_internal with constant folding,
Alistair Francis <=
[PULL 14/24] tcg: Use tcg_constant_{i32,i64} with tcg int expanders, Richard Henderson, 2021/01/13
[PULL 15/24] tcg: Use tcg_constant_{i32,i64} with tcg plugins, Richard Henderson, 2021/01/13
[PULL 18/24] tcg: Remove movi and dupi opcodes, Richard Henderson, 2021/01/13
[PULL 16/24] tcg: Use tcg_constant_{i32,i64,vec} with gvec expanders, Richard Henderson, 2021/01/13
[PULL 22/24] tcg/ppc: Use tcg_constant_vec with tcg vec expanders, Richard Henderson, 2021/01/13
[PULL 20/24] tcg/i386: Use tcg_constant_vec with tcg vec expanders, Richard Henderson, 2021/01/13
[PULL 21/24] tcg: Remove tcg_gen_dup{8,16,32,64}i_vec, Richard Henderson, 2021/01/13