You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Copy file name to clipboardExpand all lines: backends/instructions_appendix/all_instructions.golden.adoc
+6-9Lines changed: 6 additions & 9 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -569,8 +569,6 @@ Encoding::
569
569
Description::
570
570
This instruction implements the rotation, SubBytes and Round Constant addition steps of the AES
571
571
block cipher Key Schedule.
572
-
573
-
574
572
`rnum` must be in the range `0x0..0xA`. The values `0xB..0xF` are reserved.
575
573
576
574
@@ -7268,7 +7266,7 @@ Encoding::
7268
7266
....
7269
7267
7270
7268
Description::
7271
-
The xref:insts:fadd_d.adoc#udb:doc:inst:fadd_d[fadd.d] instruction is analogous to xref:insts:fadd_s.adoc#udb:doc:inst:fadd_s[fadd.s] and performs double-precision floating-point addition between
7269
+
The xref:insts:fadd_d.adoc#udb:doc:inst:fadd_d[fadd.d] instruction is analogous to xref:insts:fadd_s.adoc#udb:doc:inst:fadd_s[fadd.s] and performs double-precision floating-point addition of
7272
7270
`fs1` and `fs2` and writes the final result to `fd`.
7273
7271
7274
7272
@@ -7387,7 +7385,7 @@ Encoding::
7387
7385
....
7388
7386
7389
7387
Description::
7390
-
The xref:insts:fadd_s.adoc#udb:doc:inst:fadd_s[fadd.s] instruction performs single-precision floating-point addition of `xs1` and `xs2`
7388
+
The xref:insts:fadd_s.adoc#udb:doc:inst:fadd_s[fadd.s] instruction performs single-precision floating-point addition of `fs1` and `fs2`
7391
7389
and writes the final result to `fd`.
7392
7390
7393
7391
@@ -10622,8 +10620,7 @@ Encoding::
10622
10620
....
10623
10621
10624
10622
Description::
10625
-
The
10626
-
The xref:insts:fli_q.adoc#udb:doc:inst:fli_q[fli.q] instruction loads one of 32 quad-precision floating-point constants, encoded in the `rs1`
10623
+
The xref:insts:fli_q.adoc#udb:doc:inst:fli_q[fli.q] instruction loads one of 32 quad-precision floating-point constants, encoded in the `xs1`
10627
10624
field, into floating-point register `rd`.
10628
10625
xref:insts:fli_q.adoc#udb:doc:inst:fli_q[fli.q] is encoded like xref:insts:fmv_w_x.adoc#udb:doc:inst:fmv_w_x[fmv.w.x], but with _fmt_ = Q.
10629
10626
@@ -12246,7 +12243,7 @@ Included in::
12246
12243
== fmv.d.x
12247
12244
12248
12245
Synopsis::
12249
-
Floating-Point Move from Integer Register to Double-Precision Register
12246
+
Floating-Point Move Double-Precision from Integer Register
12250
12247
12251
12248
Assembly::
12252
12249
fmv.d.x fd, xs1
@@ -12365,7 +12362,7 @@ Included in::
12365
12362
== fmv.x.d
12366
12363
12367
12364
Synopsis::
12368
-
Floating-Point Move from Double-Precision Register to Integer Register
12365
+
Floating-Point Move Double-Precision to Integer Register
12369
12366
12370
12367
Assembly::
12371
12368
fmv.x.d xd, fs1
@@ -12631,7 +12628,7 @@ Encoding::
12631
12628
Description::
12632
12629
The xref:insts:fmvp_q_x.adoc#udb:doc:inst:fmvp_q_x[fmvp.q.x] instruction moves a double-precision number from a pair of integer registers into
12633
12630
a floating-point register.
12634
-
Integer registers xs1 and xs2 supply bits 63:0 and 127:64, respectively; the result is written to
12631
+
Integer registers `xs1` and `xs2` supply bits 63:0 and 127:64, respectively; the result is written to
12635
12632
floating-point register `fd`.
12636
12633
xref:insts:fmvp_q_x.adoc#udb:doc:inst:fmvp_q_x[fmvp.q.x] is encoded in the OP-FP major opcode with _funct3_=0 and _funct7_=1011011.
* On a trap into HS-mode, hardware writes 1 when the prior mode was VS-mode or VU-mode, and 0 otherwise.
186
+
* On a trap into HS-mode, hardware writes 1 when the prior mode was VS-mode or VU-mode, and 0 otherwise.
212
187
213
-
Can also be written by software without immediate side-effect.
188
+
Can also be written by software without immediate side-effect.
214
189
215
-
Affects execution by:
190
+
Affects execution by:
216
191
217
-
* When an `sret` instruction in executed in HS-mode or M-mode,
218
-
control returns to VS-mode or VU-mode (as selected by `mstatus.SPP`) when
219
-
`hstatus.SPV` is 1 and to HS-mode or U-mode otherwise.
192
+
* When an `sret` instruction in executed in HS-mode or M-mode,
193
+
control returns to VS-mode or VU-mode (as selected by `mstatus.SPP`) when
194
+
`hstatus.SPV` is 1 and to HS-mode or U-mode otherwise.
220
195
type: RW
221
196
reset_value: UNDEFINED_LEGAL
222
197
GVA:
223
198
location: 6
224
199
long_name: Guest Virtual Address
225
-
description:
226
-
- id: csr-hstatus-gva-behavior
227
-
normative: false
228
-
text: |
229
-
Written by hardware whenever a trap is taken into HS-mode:
200
+
description: |
201
+
Written by hardware whenever a trap is taken into HS-mode:
230
202
231
-
* Writes 1 when a trap causes a guest virtual address to be written into `stval` (`Breakpoint`, `* Address Misaligned`, `* Access Fault`, `* Page Fault`, or `* Guest-Page Fault`).
232
-
* Writes 0 otherwise
203
+
* Writes 1 when a trap causes a guest virtual address to be written into `stval` (`Breakpoint`, `* Address Misaligned`, `* Access Fault`, `* Page Fault`, or `* Guest-Page Fault`).
204
+
* Writes 0 otherwise
233
205
234
-
Does not affect execution.
206
+
Does not affect execution.
235
207
type: RW
236
208
reset_value: UNDEFINED_LEGAL
237
209
VSBE:
238
210
location: 5
239
211
long_name: VS-mode Big Endian
240
212
description:
241
-
- id: csr-hstatus-vgein-behavior
242
-
normative: false
243
-
text: |
213
+
- text: |
244
214
Controls the endianness of data VS-mode (0 = little, 1 = big).
245
215
Instructions are always little endian, regardless of the data setting.
246
216
247
-
- id: csr-hstatus-vgein-little-endian
248
-
normative: false
249
-
text: |
217
+
- text: |
250
218
Since the CPU does not support big endian in VS-mode, this is hardwired to 0.
251
219
when(): return VS_MODE_ENDIANNESS == "little";
252
220
253
-
- id: csr-hstatus-vgein-big-endian
254
-
normative: false
255
-
text: |
221
+
- text: |
256
222
Since the CPU does not support little endian in VS-mode, this is hardwired to 1.
0 commit comments