[v11,13/77] target/riscv: rvv-1.0: add fractional LMUL

Message ID	20211210075704.23951-14-frank.chang@sifive.com (mailing list archive)
State	New, archived
Headers	show Return-Path: <qemu-devel-bounces+qemu-devel=archiver.kernel.org@nongnu.org> From: frank.chang@sifive.com To: qemu-devel@nongnu.org Subject: [PATCH v11 13/77] target/riscv: rvv-1.0: add fractional LMUL Date: Fri, 10 Dec 2021 15:55:59 +0800 Message-Id: <20211210075704.23951-14-frank.chang@sifive.com> In-Reply-To: <20211210075704.23951-1-frank.chang@sifive.com> References: <20211210075704.23951-1-frank.chang@sifive.com> MIME-Version: 1.0 Content-Transfer-Encoding: 8bit Received-SPF: pass client-ip=2607:f8b0:4864:20::102d; envelope-from=frank.chang@sifive.com; helo=mail-pj1-x102d.google.com X-Spam_score_int: -12 X-Spam_score: -1.3 X-Spam_bar: - X-Spam_report: (-1.3 / 5.0 requ) BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, RCVD_IN_DNSWL_NONE=-0.0001, RDNS_NONE=0.793, SPF_HELO_NONE=0.001, SPF_PASS=-0.001 autolearn=no autolearn_force=no X-Spam_action: no action Precedence: list Cc: qemu-riscv@nongnu.org, Frank Chang <frank.chang@sifive.com>, Bin Meng <bin.meng@windriver.com>, Richard Henderson <richard.henderson@linaro.org>, Alistair Francis <alistair.francis@wdc.com>, Palmer Dabbelt <palmer@dabbelt.com> Errors-To: qemu-devel-bounces+qemu-devel=archiver.kernel.org@nongnu.org Sender: "Qemu-devel" <qemu-devel-bounces+qemu-devel=archiver.kernel.org@nongnu.org>
Series	support vector extension v1.0 \| expand [v11,00/77] support vector extension v1.0 [v11,01/77] target/riscv: drop vector 0.7.1 and add 1.0 support [v11,02/77] target/riscv: Use FIELD_EX32() to extract wd field [v11,03/77] target/riscv: rvv-1.0: add mstatus VS field [v11,04/77] target/riscv: rvv-1.0: set mstatus.SD bit if mstatus.VS is dirty [v11,05/77] target/riscv: rvv-1.0: add sstatus VS field [v11,06/77] target/riscv: rvv-1.0: introduce writable misa.v field [v11,07/77] target/riscv: rvv-1.0: add translation-time vector context status [v11,08/77] target/riscv: rvv-1.0: remove rvv related codes from fcsr registers [v11,09/77] target/riscv: rvv-1.0: add vcsr register [v11,10/77] target/riscv: rvv-1.0: add vlenb register [v11,11/77] target/riscv: rvv-1.0: check MSTATUS_VS when accessing vector csr registers [v11,12/77] target/riscv: rvv-1.0: remove MLEN calculations [v11,13/77] target/riscv: rvv-1.0: add fractional LMUL [v11,14/77] target/riscv: rvv-1.0: add VMA and VTA [v11,15/77] target/riscv: rvv-1.0: update check functions [v11,16/77] target/riscv: introduce more imm value modes in translator functions [v11,17/77] target/riscv: rvv:1.0: add translation-time nan-box helper function [v11,18/77] target/riscv: rvv-1.0: remove amo operations instructions [v11,19/77] target/riscv: rvv-1.0: configure instructions [v11,20/77] target/riscv: rvv-1.0: stride load and store instructions [v11,21/77] target/riscv: rvv-1.0: index load and store instructions [v11,22/77] target/riscv: rvv-1.0: fix address index overflow bug of indexed load/store insns [v11,23/77] target/riscv: rvv-1.0: fault-only-first unit stride load [v11,24/77] target/riscv: rvv-1.0: load/store whole register instructions [v11,25/77] target/riscv: rvv-1.0: update vext_max_elems() for load/store insns [v11,26/77] target/riscv: rvv-1.0: take fractional LMUL into vector max elements calculation [v11,27/77] target/riscv: rvv-1.0: floating-point square-root instruction [v11,28/77] target/riscv: rvv-1.0: floating-point classify instructions [v11,29/77] target/riscv: rvv-1.0: count population in mask instruction [v11,30/77] target/riscv: rvv-1.0: find-first-set mask bit instruction [v11,31/77] target/riscv: rvv-1.0: set-X-first mask bit instructions [v11,32/77] target/riscv: rvv-1.0: iota instruction [v11,33/77] target/riscv: rvv-1.0: element index instruction [v11,34/77] target/riscv: rvv-1.0: allow load element with sign-extended [v11,35/77] target/riscv: rvv-1.0: register gather instructions [v11,36/77] target/riscv: rvv-1.0: integer scalar move instructions [v11,37/77] target/riscv: rvv-1.0: floating-point move instruction [v11,38/77] target/riscv: rvv-1.0: floating-point scalar move instructions [v11,39/77] target/riscv: rvv-1.0: whole register move instructions [v11,40/77] target/riscv: rvv-1.0: integer extension instructions [v11,41/77] target/riscv: rvv-1.0: single-width averaging add and subtract instructions [v11,42/77] target/riscv: rvv-1.0: single-width bit shift instructions [v11,43/77] target/riscv: rvv-1.0: integer add-with-carry/subtract-with-borrow [v11,44/77] target/riscv: rvv-1.0: narrowing integer right shift instructions [v11,45/77] target/riscv: rvv-1.0: widening integer multiply-add instructions [v11,46/77] target/riscv: rvv-1.0: single-width saturating add and subtract instructions [v11,47/77] target/riscv: rvv-1.0: integer comparison instructions [v11,48/77] target/riscv: rvv-1.0: floating-point compare instructions [v11,49/77] target/riscv: rvv-1.0: mask-register logical instructions [v11,50/77] target/riscv: rvv-1.0: slide instructions [v11,51/77] target/riscv: rvv-1.0: floating-point slide instructions [v11,52/77] target/riscv: rvv-1.0: narrowing fixed-point clip instructions [v11,53/77] target/riscv: rvv-1.0: single-width floating-point reduction [v11,54/77] target/riscv: rvv-1.0: widening floating-point reduction instructions [v11,55/77] target/riscv: rvv-1.0: single-width scaling shift instructions [v11,56/77] target/riscv: rvv-1.0: remove widening saturating scaled multiply-add [v11,57/77] target/riscv: rvv-1.0: remove vmford.vv and vmford.vf [v11,58/77] target/riscv: rvv-1.0: remove integer extract instruction [v11,59/77] target/riscv: rvv-1.0: floating-point min/max instructions [v11,60/77] target/riscv: introduce floating-point rounding mode enum [v11,61/77] target/riscv: rvv-1.0: floating-point/integer type-convert instructions [v11,62/77] target/riscv: rvv-1.0: widening floating-point/integer type-convert [v11,63/77] target/riscv: add "set round to odd" rounding mode helper function [v11,64/77] target/riscv: rvv-1.0: narrowing floating-point/integer type-convert [v11,65/77] target/riscv: rvv-1.0: relax RV_VLEN_MAX to 1024-bits [v11,66/77] target/riscv: rvv-1.0: implement vstart CSR [v11,67/77] target/riscv: rvv-1.0: trigger illegal instruction exception if frm is not valid [v11,68/77] target/riscv: gdb: support vector registers for rv64 & rv32 [v11,69/77] target/riscv: rvv-1.0: floating-point reciprocal square-root estimate instruction [v11,70/77] target/riscv: rvv-1.0: floating-point reciprocal estimate instruction [v11,71/77] target/riscv: rvv-1.0: rename r2_zimm to r2_zimm11 [v11,72/77] target/riscv: rvv-1.0: add vsetivli instruction [v11,73/77] target/riscv: rvv-1.0: add evl parameter to vext_ldst_us() [v11,74/77] target/riscv: rvv-1.0: add vector unit-stride mask load/store insns [v11,75/77] target/riscv: rvv-1.0: rename vmandnot.mm and vmornot.mm to vmandn.mm and vmorn.mm [v11,76/77] target/riscv: rvv-1.0: update opivv_vadc_check() comment [v11,77/77] target/riscv: rvv-1.0: Add ELEN checks for widening and narrowing instructions

Message ID

20211210075704.23951-14-frank.chang@sifive.com (mailing list archive)

State

New, archived

Headers

From: frank.chang@sifive.com
To: qemu-devel@nongnu.org
Subject: [PATCH v11 13/77] target/riscv: rvv-1.0: add fractional LMUL
Date: Fri, 10 Dec 2021 15:55:59 +0800
Message-Id: <20211210075704.23951-14-frank.chang@sifive.com>
In-Reply-To: <20211210075704.23951-1-frank.chang@sifive.com>
References: <20211210075704.23951-1-frank.chang@sifive.com>
MIME-Version: 1.0
Content-Transfer-Encoding: 8bit
Received-SPF: pass client-ip=2607:f8b0:4864:20::102d;
 envelope-from=frank.chang@sifive.com; helo=mail-pj1-x102d.google.com
X-Spam_score_int: -12
X-Spam_score: -1.3
X-Spam_bar: -
X-Spam_report: (-1.3 / 5.0 requ) BAYES_00=-1.9, DKIM_SIGNED=0.1,
 DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1,
 RCVD_IN_DNSWL_NONE=-0.0001, RDNS_NONE=0.793, SPF_HELO_NONE=0.001,
 SPF_PASS=-0.001 autolearn=no autolearn_force=no
X-Spam_action: no action
X-BeenThere: qemu-devel@nongnu.org
X-Mailman-Version: 2.1.29
Precedence: list
List-Id: <qemu-devel.nongnu.org>
List-Unsubscribe: <https://lists.nongnu.org/mailman/options/qemu-devel>,
 <mailto:qemu-devel-request@nongnu.org?subject=unsubscribe>
List-Archive: <https://lists.nongnu.org/archive/html/qemu-devel>
List-Post: <mailto:qemu-devel@nongnu.org>
List-Help: <mailto:qemu-devel-request@nongnu.org?subject=help>
List-Subscribe: <https://lists.nongnu.org/mailman/listinfo/qemu-devel>,
 <mailto:qemu-devel-request@nongnu.org?subject=subscribe>
Cc: qemu-riscv@nongnu.org, Frank Chang <frank.chang@sifive.com>,
 Bin Meng <bin.meng@windriver.com>,
 Richard Henderson <richard.henderson@linaro.org>,
 Alistair Francis <alistair.francis@wdc.com>,
 Palmer Dabbelt <palmer@dabbelt.com>
Errors-To: qemu-devel-bounces+qemu-devel=archiver.kernel.org@nongnu.org
Sender: "Qemu-devel"
 <qemu-devel-bounces+qemu-devel=archiver.kernel.org@nongnu.org>

Series

support vector extension v1.0 | expand

Commit Message

Frank Chang Dec. 10, 2021, 7:55 a.m. UTC

From: Frank Chang <frank.chang@sifive.com>

Introduce the concepts of fractional LMUL for RVV 1.0.
In RVV 1.0, LMUL bits are contiguous in vtype register.

Also rearrange rvv bits in TB_FLAGS to skip MSTATUS_VS (0x600)
and MSTATUS_FS (0x6000) bits.

Signed-off-by: Frank Chang <frank.chang@sifive.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
---
 target/riscv/cpu.h           | 26 ++++++++++++++------------
 target/riscv/translate.c     | 16 ++++++++++++++--
 target/riscv/vector_helper.c | 16 ++++++++++++++--
 3 files changed, 42 insertions(+), 16 deletions(-)

diff --git a/target/riscv/cpu.h b/target/riscv/cpu.h
index eee2a2b19e..17414bfff7 100644
--- a/target/riscv/cpu.h
+++ b/target/riscv/cpu.h
@@ -102,10 +102,10 @@  typedef struct CPURISCVState CPURISCVState;
 
 #define RV_VLEN_MAX 256
 
-FIELD(VTYPE, VLMUL, 0, 2)
-FIELD(VTYPE, VSEW, 2, 3)
-FIELD(VTYPE, VEDIV, 5, 2)
-FIELD(VTYPE, RESERVED, 7, sizeof(target_ulong) * 8 - 9)
+FIELD(VTYPE, VLMUL, 0, 3)
+FIELD(VTYPE, VSEW, 3, 3)
+FIELD(VTYPE, VEDIV, 8, 2)
+FIELD(VTYPE, RESERVED, 10, sizeof(target_ulong) * 8 - 11)
 FIELD(VTYPE, VILL, sizeof(target_ulong) * 8 - 1, 1)
 
 struct CPURISCVState {
@@ -403,18 +403,20 @@  typedef RISCVCPU ArchCPU;
 #include "exec/cpu-all.h"
 
 FIELD(TB_FLAGS, MEM_IDX, 0, 3)
-FIELD(TB_FLAGS, VL_EQ_VLMAX, 3, 1)
-FIELD(TB_FLAGS, LMUL, 4, 2)
+FIELD(TB_FLAGS, LMUL, 3, 3)
 FIELD(TB_FLAGS, SEW, 6, 3)
-FIELD(TB_FLAGS, VILL, 9, 1)
+/* Skip MSTATUS_VS (0x600) bits */
+FIELD(TB_FLAGS, VL_EQ_VLMAX, 11, 1)
+FIELD(TB_FLAGS, VILL, 12, 1)
+/* Skip MSTATUS_FS (0x6000) bits */
 /* Is a Hypervisor instruction load/store allowed? */
-FIELD(TB_FLAGS, HLSX, 10, 1)
-FIELD(TB_FLAGS, MSTATUS_HS_FS, 11, 2)
-FIELD(TB_FLAGS, MSTATUS_HS_VS, 13, 2)
+FIELD(TB_FLAGS, HLSX, 15, 1)
+FIELD(TB_FLAGS, MSTATUS_HS_FS, 16, 2)
+FIELD(TB_FLAGS, MSTATUS_HS_VS, 18, 2)
 /* The combination of MXL/SXL/UXL that applies to the current cpu mode. */
-FIELD(TB_FLAGS, XL, 15, 2)
+FIELD(TB_FLAGS, XL, 20, 2)
 /* If PointerMasking should be applied */
-FIELD(TB_FLAGS, PM_ENABLED, 17, 1)
+FIELD(TB_FLAGS, PM_ENABLED, 22, 1)
 
 #ifdef TARGET_RISCV32
 #define riscv_cpu_mxl(env)  ((void)(env), MXL_RV32)
diff --git a/target/riscv/translate.c b/target/riscv/translate.c
index 17101d854b..3688e80d03 100644
--- a/target/riscv/translate.c
+++ b/target/riscv/translate.c
@@ -80,7 +80,19 @@  typedef struct DisasContext {
     bool hlsx;
     /* vector extension */
     bool vill;
-    uint8_t lmul;
+    /*
+     * Encode LMUL to lmul as follows:
+     *     LMUL    vlmul    lmul
+     *      1       000       0
+     *      2       001       1
+     *      4       010       2
+     *      8       011       3
+     *      -       100       -
+     *     1/8      101      -3
+     *     1/4      110      -2
+     *     1/2      111      -1
+     */
+    int8_t lmul;
     uint8_t sew;
     uint16_t vlen;
     bool vl_eq_vlmax;
@@ -690,7 +702,7 @@  static void riscv_tr_init_disas_context(DisasContextBase *dcbase, CPUState *cs)
     ctx->hlsx = FIELD_EX32(tb_flags, TB_FLAGS, HLSX);
     ctx->vill = FIELD_EX32(tb_flags, TB_FLAGS, VILL);
     ctx->sew = FIELD_EX32(tb_flags, TB_FLAGS, SEW);
-    ctx->lmul = FIELD_EX32(tb_flags, TB_FLAGS, LMUL);
+    ctx->lmul = sextract32(FIELD_EX32(tb_flags, TB_FLAGS, LMUL), 0, 3);
     ctx->vl_eq_vlmax = FIELD_EX32(tb_flags, TB_FLAGS, VL_EQ_VLMAX);
     ctx->xl = FIELD_EX32(tb_flags, TB_FLAGS, XL);
     ctx->cs = cs;
diff --git a/target/riscv/vector_helper.c b/target/riscv/vector_helper.c
index ac6c7298bc..866bfb3571 100644
--- a/target/riscv/vector_helper.c
+++ b/target/riscv/vector_helper.c
@@ -86,9 +86,21 @@  static inline uint32_t vext_vm(uint32_t desc)
     return FIELD_EX32(simd_data(desc), VDATA, VM);
 }
 
-static inline uint32_t vext_lmul(uint32_t desc)
+/*
+ * Encode LMUL to lmul as following:
+ *     LMUL    vlmul    lmul
+ *      1       000       0
+ *      2       001       1
+ *      4       010       2
+ *      8       011       3
+ *      -       100       -
+ *     1/8      101      -3
+ *     1/4      110      -2
+ *     1/2      111      -1
+ */
+static inline int32_t vext_lmul(uint32_t desc)
 {
-    return FIELD_EX32(simd_data(desc), VDATA, LMUL);
+    return sextract32(FIELD_EX32(simd_data(desc), VDATA, LMUL), 0, 3);
 }
 
 static uint32_t vext_wd(uint32_t desc)

[v11,13/77] target/riscv: rvv-1.0: add fractional LMUL

Commit Message

Patch