From: Richard Henderson <richard.henderson@linaro.org>
To: qemu-devel@nongnu.org
Cc: peter.maydell@linux.org
Subject: [Qemu-devel] [PULL 13/31] tcg/aarch64: Implement tcg_out_dupm_vec
Date: Mon, 13 May 2019 17:05:22 -0700 [thread overview]
Message-ID: <20190514000540.4313-14-richard.henderson@linaro.org> (raw)
In-Reply-To: <20190514000540.4313-1-richard.henderson@linaro.org>
The LD1R instruction does all the work. Note that the only
useful addressing mode is a base register with no offset.
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
---
tcg/aarch64/tcg-target.inc.c | 37 ++++++++++++++++++++++++++++++++++--
1 file changed, 35 insertions(+), 2 deletions(-)
diff --git a/tcg/aarch64/tcg-target.inc.c b/tcg/aarch64/tcg-target.inc.c
index 4a3cfa778a..b4585724d3 100644
--- a/tcg/aarch64/tcg-target.inc.c
+++ b/tcg/aarch64/tcg-target.inc.c
@@ -381,6 +381,9 @@ typedef enum {
I3207_BLR = 0xd63f0000,
I3207_RET = 0xd65f0000,
+ /* AdvSIMD load/store single structure. */
+ I3303_LD1R = 0x0d40c000,
+
/* Load literal for loading the address at pc-relative offset */
I3305_LDR = 0x58000000,
I3305_LDR_v64 = 0x5c000000,
@@ -566,7 +569,14 @@ static inline uint32_t tcg_in32(TCGContext *s)
#define tcg_out_insn(S, FMT, OP, ...) \
glue(tcg_out_insn_,FMT)(S, glue(glue(glue(I,FMT),_),OP), ## __VA_ARGS__)
-static void tcg_out_insn_3305(TCGContext *s, AArch64Insn insn, int imm19, TCGReg rt)
+static void tcg_out_insn_3303(TCGContext *s, AArch64Insn insn, bool q,
+ TCGReg rt, TCGReg rn, unsigned size)
+{
+ tcg_out32(s, insn | (rt & 0x1f) | (rn << 5) | (size << 10) | (q << 30));
+}
+
+static void tcg_out_insn_3305(TCGContext *s, AArch64Insn insn,
+ int imm19, TCGReg rt)
{
tcg_out32(s, insn | (imm19 & 0x7ffff) << 5 | rt);
}
@@ -825,7 +835,30 @@ static bool tcg_out_dup_vec(TCGContext *s, TCGType type, unsigned vece,
static bool tcg_out_dupm_vec(TCGContext *s, TCGType type, unsigned vece,
TCGReg r, TCGReg base, intptr_t offset)
{
- return false;
+ TCGReg temp = TCG_REG_TMP;
+
+ if (offset < -0xffffff || offset > 0xffffff) {
+ tcg_out_movi(s, TCG_TYPE_PTR, temp, offset);
+ tcg_out_insn(s, 3502, ADD, 1, temp, temp, base);
+ base = temp;
+ } else {
+ AArch64Insn add_insn = I3401_ADDI;
+
+ if (offset < 0) {
+ add_insn = I3401_SUBI;
+ offset = -offset;
+ }
+ if (offset & 0xfff000) {
+ tcg_out_insn_3401(s, add_insn, 1, temp, base, offset & 0xfff000);
+ base = temp;
+ }
+ if (offset & 0xfff) {
+ tcg_out_insn_3401(s, add_insn, 1, temp, base, offset & 0xfff);
+ base = temp;
+ }
+ }
+ tcg_out_insn(s, 3303, LD1R, type == TCG_TYPE_V128, r, base, vece);
+ return true;
}
static void tcg_out_movi(TCGContext *s, TCGType type, TCGReg rd,
--
2.17.1
next prev parent reply other threads:[~2019-05-14 0:09 UTC|newest]
Thread overview: 33+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-05-14 0:05 [Qemu-devel] [PULL 00/31] tcg: gvec improvments Richard Henderson
2019-05-14 0:05 ` [Qemu-devel] [PULL 01/31] tcg: Implement tcg_gen_gvec_3i() Richard Henderson
2019-05-14 0:05 ` [Qemu-devel] [PULL 02/31] tcg: Do not recreate INDEX_op_neg_vec unless supported Richard Henderson
2019-05-14 0:05 ` [Qemu-devel] [PULL 03/31] tcg: Allow add_vec, sub_vec, neg_vec, not_vec to be expanded Richard Henderson
2019-05-14 0:05 ` [Qemu-devel] [PULL 04/31] tcg: Specify optional vector requirements with a list Richard Henderson
2019-05-14 0:05 ` [Qemu-devel] [PULL 05/31] tcg: Assert fixed_reg is read-only Richard Henderson
2019-05-14 0:05 ` [Qemu-devel] [PULL 06/31] tcg/arm: Use tcg_out_mov_reg in tcg_out_mov Richard Henderson
2019-05-14 0:05 ` [Qemu-devel] [PULL 07/31] tcg: Return bool success from tcg_out_mov Richard Henderson
2019-05-14 0:05 ` [Qemu-devel] [PULL 08/31] tcg: Support cross-class moves without instruction support Richard Henderson
2019-05-14 0:05 ` [Qemu-devel] [PULL 09/31] tcg: Promote tcg_out_{dup, dupi}_vec to backend interface Richard Henderson
2019-05-14 0:05 ` [Qemu-devel] [PULL 10/31] tcg: Manually expand INDEX_op_dup_vec Richard Henderson
2019-05-14 0:05 ` [Qemu-devel] [PULL 11/31] tcg: Add tcg_out_dupm_vec to the backend interface Richard Henderson
2019-05-14 0:05 ` [Qemu-devel] [PULL 12/31] tcg/i386: Implement tcg_out_dupm_vec Richard Henderson
2019-05-14 0:05 ` Richard Henderson [this message]
2019-05-14 0:05 ` [Qemu-devel] [PULL 14/31] tcg: Add INDEX_op_dupm_vec Richard Henderson
2019-05-14 0:05 ` [Qemu-devel] [PULL 15/31] tcg: Add gvec expanders for variable shift Richard Henderson
2019-05-14 0:05 ` [Qemu-devel] [PULL 16/31] tcg/i386: Support vector variable shift opcodes Richard Henderson
2019-05-14 0:05 ` [Qemu-devel] [PULL 17/31] tcg/aarch64: " Richard Henderson
2019-05-14 0:05 ` [Qemu-devel] [PULL 18/31] tcg: Add gvec expanders for vector shift by scalar Richard Henderson
2019-05-14 0:05 ` [Qemu-devel] [PULL 19/31] tcg/i386: Support vector scalar shift opcodes Richard Henderson
2019-05-14 0:05 ` [Qemu-devel] [PULL 20/31] tcg: Add support for integer absolute value Richard Henderson
2019-05-14 0:05 ` [Qemu-devel] [PULL 21/31] tcg: Add support for vector " Richard Henderson
2019-05-14 0:05 ` [Qemu-devel] [PULL 22/31] tcg/i386: Support " Richard Henderson
2019-05-14 0:05 ` [Qemu-devel] [PULL 23/31] tcg/aarch64: " Richard Henderson
2019-05-14 0:05 ` [Qemu-devel] [PULL 24/31] target/arm: Use tcg_gen_abs_i64 and tcg_gen_gvec_abs Richard Henderson
2019-05-14 0:05 ` [Qemu-devel] [PULL 25/31] target/cris: Use tcg_gen_abs_tl Richard Henderson
2019-05-14 0:05 ` [Qemu-devel] [PULL 26/31] target/ppc: Use tcg_gen_abs_i32 Richard Henderson
2019-05-14 0:05 ` [Qemu-devel] [PULL 27/31] target/ppc: Use tcg_gen_abs_tl Richard Henderson
2019-05-14 0:05 ` [Qemu-devel] [PULL 28/31] target/s390x: Use tcg_gen_abs_i64 Richard Henderson
2019-05-14 0:05 ` [Qemu-devel] [PULL 29/31] target/tricore: Use tcg_gen_abs_tl Richard Henderson
2019-05-14 0:05 ` [Qemu-devel] [PULL 30/31] target/xtensa: Use tcg_gen_abs_i32 Richard Henderson
2019-05-14 0:05 ` [Qemu-devel] [PULL 31/31] tcg/aarch64: Do not advertise minmax for MO_64 Richard Henderson
2019-05-14 12:35 ` [Qemu-devel] [PULL 00/31] tcg: gvec improvments Peter Maydell
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=20190514000540.4313-14-richard.henderson@linaro.org \
--to=richard.henderson@linaro.org \
--cc=peter.maydell@linux.org \
--cc=qemu-devel@nongnu.org \
/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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).