Commit Graph

5225 Commits

Author SHA1 Message Date
Aleksandar Markovic
ed037799c7
target/mips: Amend MXU instruction opcodes
Amend MXU instruction opcodes. Pool04 is actually only instruction
OPC_MXU_S16MAD. Two cases within S16MAD are recognized by 1-bit
subfield 'aptn1'.

Backports commit eab0bdb07cbed1131be2d1f541059c7b96b05e32 from qemu
2018-11-11 05:58:35 -05:00
Craig Janeczek
8429d98b40
target/mips: Define a bit for MXU in insn_flags
Define a bit for MXU in insn_flags. This is the first non-MIPS
(third party) ASE supported in QEMU for MIPS, so it is placed in
the section "bits 56-63: vendor-specific ASEs".

Backports commit a031ac61619294ae473a78d1834e757fad8b59e5 from qemu
2018-11-11 05:52:18 -05:00
Craig Janeczek
58dc377890
target/mips: Introduce MXU registers
Define and initialize the 16 MXU registers - 15 general computational
register, and 1 control register). There is also a zero register, but
it does not have any corresponding variable.

Backports commit eb5559f67dc8dc12335dd996877bb6daaea32eb2 from qemu.
2018-11-11 05:50:52 -05:00
Aleksandar Markovic
a1ba04ae44
target/mips: Add two missing breaks for NM_LLWPE and NM_SCWPE decoder cases
Coverity found two fallthroughs that miss break statement. Fix them.

Backports commit 2431a422d325c1832d77dd64fa3135ec303b00de from qemu
2018-11-11 05:44:23 -05:00
Dimitrije Nikolic
948d49db81
target/mips: Implement emulation of nanoMIPS EVA instructions
Implement emulation of nanoMIPS EVA instructions. They are all
part of P.LS.E0 instruction pool, or one of its subpools.

Backports commit d046a9ea1b8877a570a8b12a2d0125ec59fe5b22 from qemu
2018-11-10 12:18:11 -05:00
Aleksandar Markovic
006f0a5873
target/mips: Add nanoMIPS CRC32 instruction pool
Backports commit ba1e81171fb761aea9a9a4ccadedf808e34eaae2 from qemu
2018-11-10 12:15:01 -05:00
Aleksandar Markovic
9f728678ec
target/mips: Fix decoding of ALIGN and DALIGN instructions
Opcode for ALIGN and DALIGN must be in fact ranges of opcodes, to
allow paremeter 'bp' to occupy two and three bits, respectively.

Backports commit 373ecd3823f949fd550ec49685299e287af5753e from qemu
2018-11-10 12:14:01 -05:00
Aleksandar Markovic
caa67a9d17
target/mips: Fix the title of translate.c
Replace MIPS32 with MIPS, since the file covers all generations
of MIPS architectures.

Backports commit ab99e0e44bc7b0e2e52d9083a673866b18470536 from qemu
2018-11-10 12:12:27 -05:00
Fredrik Noring
a93e15aad1
target/mips: Define the R5900 CPU
The primary purpose of this change is to support programs compiled by
GCC for the R5900 target and thereby run R5900 Linux distributions, for
example Gentoo.

GCC in version 7.3, by itself, by inspection of the GCC source code
and inspection of the generated machine code, for the R5900 target,
only emits two instructions that are specific to the R5900: the three-
operand MULT and MULTU. GCC and libc also emit certain MIPS III
instructions that are not part of the R5900 ISA. They are normally
trapped and emulated by the Linux kernel, and therefore need to be
treated accordingly by QEMU.

A program compiled by GCC is taken to mean source code compiled by GCC
under the restrictions above. One can, with the apparent limitations,
with a bit of effort obtain a fully functioning operating system such
as R5900 Gentoo. Strictly speaking, programs need not be compiled by
GCC to make use of this change.

Instructions and other facilities of the R5900 not implemented by this
change are intended to signal provisional exceptions. One such example
is the FPU that is not compliant with IEEE 754-1985 in system mode. It
is therefore provisionally disabled. In user space the FPU is trapped
and emulated by IEEE 754-1985 compliant software in the kernel, and
this is handled accordingly by QEMU. Another example is the 93
multimedia instructions specific to the R5900 that generate provisional
reserved instruction exception signals.

One of the benefits of running a Linux distribution under QEMU is that
programs can be compiled with a native compiler, where the host and
target are the same, as opposed to a cross-compiler, where they are
not the same. This is especially important in cases where the target
hardware does not have the resources to run a native compiler.

Problems with cross-compilation are often related to host and target
differences in integer sizes, pointer sizes, endianness, machine code,
ABI, etc. Sometimes cross-compilation is not even supported by the
build script for a given package. One effective way to avoid those
problems is to replace the cross-compiler with a native compiler. This
change of compilation methods does not resolve the inherent problems
with cross-compilation.

The native compiler naturally replaces the cross-compiler, because one
typically uses one or the other, and preferably the native compiler
when the circumstances admit this. The native compiler is also a good
test case for the R5900 QEMU user mode. Additionally, Gentoo is well-
known for compiling and installing its packages from sources.

This change has been tested with Gentoo compiled for R5900, including
native compilation of several packages under QEMU.

Backports commit ed4f49ba9bb56ebca6987b1083255daf6c89b5de from qemu.
2018-11-10 12:11:11 -05:00
Fredrik Noring
e9a4475a35
target/mips: Make R5900 DMULT[U], DDIV[U], LL[D] and SC[D] user only
The Linux kernel traps certain reserved instruction exceptions to
emulate the corresponding instructions. QEMU plays the role of the
kernel in user mode, so those traps are emulated by accepting the
instructions.

This change adds the function check_insn_opc_user_only to signal a
reserved instruction exception for flagged CPUs in QEMU system mode.

The MIPS III instructions DMULT[U], DDIV[U], LL[D] and SC[D] are not
implemented in R5900 hardware. They are trapped and emulated by the
Linux kernel and, accordingly, therefore QEMU user only instructions.

Backports commit 96631327be14c4f54cc31f873c278d9ffedd1e00 from qemu
2018-11-10 12:10:01 -05:00
Fredrik Noring
cb7b097204
target/mips: Support R5900 MOVN, MOVZ and PREF instructions from MIPS IV
The R5900 is taken to be MIPS III with certain modifications. From
MIPS IV it implements the instructions MOVN, MOVZ and PREF.

Backports commit 5601e6217d90ed322b4b9a6d68e8db607db91842 from qemu
2018-11-10 12:08:14 -05:00
Fredrik Noring
296d8bd729
target/mips: Support R5900 DIV1 and DIVU1 instructions
Backports commit be9c42c90d162100a63111ddeb8bfe50be897873 from qemu
2018-11-10 12:07:24 -05:00
Fredrik Noring
b42c8e80b8
target/mips: Support R5900 MFLO1, MTLO1, MFHI1 and MTHI1 instructions
Backports commit 8d927f7cb4b3ec06d6a7cf71221fd6a48e9a8fb5 from qemu
2018-11-10 12:06:20 -05:00
Fredrik Noring
3914d65fb9
target/mips: Support R5900 three-operand MULT1 and MULTU1 instructions
Add support for MULT1 and MULTU1 instructions.

Backports commit 06de726b2d4da185dfec9d06b5f1032059ad3554 from qemu
2018-11-10 12:04:28 -05:00
Fredrik Noring
91cd7c20e5
target/mips: Support R5900 three-operand MULT and MULTU instructions
The three-operand MULT and MULTU are the only R5900-specific
instructions emitted by GCC 7.3. The R5900 also implements the three-
operand MADD and MADDU instructions, but they are omitted in QEMU for
now since they are absent in programs compiled by current GCC versions.

Likewise, the R5900-specific pipeline 1 instruction variants MULT1,
MULTU1, DIV1, DIVU1, MADD1, MADDU1, MFHI1, MFLO1, MTHI1 and MTLO1
are omitted here as well.

Backports commit 21e8e8b230af38b6bd8c953fa5f31e4a5a128e1c from qemu
2018-11-10 11:57:54 -05:00
Fredrik Noring
7da9860cf5
target/mips: Add a placeholder for R5900 MMI3 instruction subclass
Add a placeholder for MMI3 subclass.

Backports commit ec1944fc8eb862d48a0c74671105ce13acbc9c9e from qemu
2018-11-10 11:54:48 -05:00
Fredrik Noring
13d02e9aad
target/mips: Add a placeholder for R5900 MMI2 instruction subclass
Backports commit 6c1e48d38a2d26cca2d2b00f331a4ac7dbfae3ca from qemu
2018-11-10 11:52:47 -05:00
Fredrik Noring
63d9f0c6b7
target/mips: Add a placeholder for R5900 MMI1 instruction subclass
Backports commit 7a803ca23a6d6ac0d1d34469cd3f010a66cd381a from qemu
2018-11-10 11:51:47 -05:00
Fredrik Noring
685ccc52e7
target/mips: Add a placeholder for R5900 MMI0 instruction subclass
Add a placeholder for MMI0 subclass.

Backports commit 88eafe0b28c8e88ed6d38eb27dbf482f9c885745 from qemu
2018-11-10 11:50:48 -05:00
Fredrik Noring
e71d43fd8d
target/mips: Add a placeholder for R5900 MMI instruction class
Add a placeholder for MMI class. This is the main palceholder for
MMI ASE.

backports commit 71b8a6b3f3d89283b094e04962c077aa01401438 from qemu
2018-11-10 11:49:19 -05:00
Fredrik Noring
496fe3ba14
target/mips: Add a placeholder for R5900 LQ
Add a placeholder for LQ instruction.

Backports commit f08099ad7a4dd32e12a3f9d4b4b04c32d7522b03 from qemu
2018-11-10 11:41:10 -05:00
Fredrik Noring
cfc33249ce
target/mips: Add a placeholder for R5900 SQ, handle user mode RDHWR
Add placeholder for SQ instruction, handle RDHWR.

Backports commit bb41e74b66a8879ba5c23db145039faa27df5766 from qemu
2018-11-10 11:40:18 -05:00
Fredrik Noring
32a644a4a1
target/mips: Bunch of opcodes
Globs a bunch of MIPS commits together from qemu.
2018-11-10 11:39:20 -05:00
Fredrik Noring
5a6b86a346
target/mips: Define R5900 MMI class, and LQ and SQ opcode constants
Define MMI class, LQ, and SQ R5900 opdoces.

Backports commit f99c0d6da3424dd48ad1f0345464f63515949be6 from qemu
2018-11-10 11:32:27 -05:00
Fredrik Noring
fa90149289
target/mips: Add R5900 Multimedia Instruction overview note
Add a comment on R5900 MMI ASE (short overview).

Backports commit 497f072b159effc4b19d9629e4818e6185be5776 from qemu
2018-11-10 11:30:57 -05:00
Fredrik Noring
93fb8273c8
target/mips: Define R5900 ISA, MMI ASE, and R5900 CPU preprocessor constants
The R5900 implements the 64-bit MIPS III instruction set except
DMULT, DMULTU, DDIV, DDIVU, LL, SC, LLD and SCD. The MIPS IV
instructions MOVN, MOVZ and PREF are implemented. It has the
R5900-specific three-operand instructions MADD, MADDU, MULT and
MULTU as well as pipeline 1 versions MULT1, MULTU1, DIV1, DIVU1,
MADD1, MADDU1, MFHI1, MFLO1, MTHI1 and MTLO1. A set of 93 128-bit
multimedia instructions specific to the R5900 is also implemented.

The Toshiba TX System RISC TX79 Core Architecture manual:

https://wiki.qemu.org/File:C790.pdf

describes the C790 processor that is a follow-up to the R5900. There
are a few notable differences in that the R5900 FPU

- is not IEEE 754-1985 compliant,
- does not implement double format, and
- its machine code is nonstandard.

Backports commit 6f692818a7b53630702d25a709cd61282fd139ad from qemu
2018-11-10 11:29:25 -05:00
Richard Henderson
e9fe7d3c8d
target/arm: Only flush tlb if ASID changes
Since QEMU does not implement ASIDs, changes to the ASID must flush the
tlb. However, if the ASID does not change there is no reason to flush.

In testing a boot of the Ubuntu installer to the first menu, this reduces
the number of flushes by 30%, or nearly 600k instances.

Backports commit 93f379b0c43617b1361f742f261479eaed4959cb from qemu
2018-11-10 11:26:24 -05:00
Richard Henderson
15a6b3f771
target/arm: Remove writefn from TTBR0_EL3
The EL3 version of this register does not include an ASID,
and so the tlb_flush performed by vmsa_ttbr_write is not needed.

Backports commit f478847f1ee0df9397f561025ab2f687fd923571 from qemu
2018-11-10 11:25:39 -05:00
Richard Henderson
1bcba0737e
target/arm: Reorg NEON VLD/VST single element to one lane
Instead of shifts and masks, use direct loads and stores from
the neon register file.

Backports commit 2d6ac920837f558be214ad2ddd28cad7f3b15e5c from qemu
2018-11-10 11:24:37 -05:00
Richard Henderson
37103f1bc4
target/arm: Promote consecutive memory ops for aa32
For a sequence of loads or stores from a single register,
little-endian operations can be promoted to an 8-byte op.
This can reduce the number of operations by a factor of 8.

Backports commit e23f12b3a252352b575908ca7b94587acd004641 from qemu
2018-11-10 11:19:15 -05:00
Richard Henderson
1cab7a41ac
target/arm: Reorg NEON VLD/VST all elements
Instead of shifts and masks, use direct loads and stores from the neon
register file. Mirror the iteration structure of the ARM pseudocode
more closely. Correct the parameters of the VLD2 A2 insn.

Note that this includes a bugfix for handling of the insn
"VLD2 (multiple 2-element structures)" -- we were using an
incorrect stride value.

Backports commit ac55d00709e78cd39dfa298dcaac7aecb58762e8 from qemu
2018-11-10 11:18:45 -05:00
Richard Henderson
a2239b9f5b
target/arm: Use gvec for NEON VLD all lanes
Backports commit 7377c2c97e20e64ed9b481eb2d9b9084bfd5b7e9 from qemu
2018-11-10 11:08:29 -05:00
Richard Henderson
985acb9cde
target/arm: Use gvec for NEON_3R_VTST_VCEQ, NEON_3R_VCGT, NEON_3R_VCGE
Move cmtst_op expanders from translate-a64.c.

Backports commit ea580fa312674c1ba82a8b137caf42b0609ce3e3 from qemu
2018-11-10 11:03:42 -05:00
Richard Henderson
5d9c0e52bf
target/arm: Use gvec for NEON_3R_VML
Move mla_op and mls_op expanders from translate-a64.c.

Backports commit 4a7832b095b9ce97a815749a13516f5cfb3c5dd4 from qemu
2018-11-10 10:58:44 -05:00
Richard Henderson
79bbb7c730
target/arm: Use gvec for VSRI, VSLI
Move shi_op and sli_op expanders from translate-a64.c.

Backports commit f3cd8218d1d3e534877ce3f3cb61c6757d10f9df from qemu
2018-11-10 10:53:28 -05:00
Lioncash
edb36c7505
target/arm: Use gvec for VSRA 2018-11-10 10:32:29 -05:00
Richard Henderson
b5877f1dfb
target/arm: Use gvec for VSHR, VSHL
Backports commit 1dc8425e551be1371d657e94367f37130cd7aede from qemu
2018-11-10 10:14:31 -05:00
Lioncash
7790ca1ccb
target/arm: Use gvec for NEON_3R_VMUL 2018-11-10 10:11:10 -05:00
Richard Henderson
dfdc6bc05c
target/arm: Use gvec for NEON_2RM_VMN, NEON_2RM_VNEG
Backports commit 4bf940bebad273e4b3534ae3f83f2c9d1191d3a2 from qemu
2018-11-10 10:09:38 -05:00
Richard Henderson
7b4b5ac249
target/arm: Use gvec for NEON_3R_VADD_VSUB insns
Backports commit e4717ae02dd0c2e544a07302c1ed473775209aba from qemu
2018-11-10 10:08:23 -05:00
Richard Henderson
0965b9513a
target/arm: Use gvec for NEON_3R_LOGIC insns
Move expanders for VBSL, VBIT, and VBIF from translate-a64.c.

Backports commit eabcd6faa90461e0b7463f4ebe75b8d050487c9c from qemu
2018-11-10 10:06:13 -05:00
Richard Henderson
9f767248a2
target/arm: Use gvec for NEON VMOV, VMVN, VBIC & VORR (immediate)
Backports commit 246fa4aca95e213fba10c8222dbc6bd0a9a2a8d4 from qemu
2018-11-10 09:56:30 -05:00
Richard Henderson
c1251a19e1
target/arm: Use gvec for NEON VDUP
Also introduces neon_element_offset to find the env offset
of a specific element within a neon register.

Backports commit 32f91fb71f4c32113ec8c2af5f74f14abe6c7162 from qemu
2018-11-10 09:51:40 -05:00
Richard Henderson
3d5f040608
target/arm: Mark some arrays const
Backports commit 308e5636152594daa4c5597b1188d44d7266db04 from qemu
2018-11-10 09:49:25 -05:00
Richard Henderson
931b49fb06
target/arm: Promote consecutive memory ops for aa64
For a sequence of loads or stores from a single register,
little-endian operations can be promoted to an 8-byte op.
This can reduce the number of operations by a factor of 8.

Backports commit 87f9a7f0c8d5122c36743885158782c2348a6d21 from qemu
2018-11-10 09:46:04 -05:00
Richard Henderson
e6707b900c
target/arm: Use tcg_gen_gvec_dup_i64 for LD[1-4]R
Backports commit 10e0b33c676b4e8ac80d5929980f4fa6be617c5a from qemu
2018-11-10 09:41:14 -05:00
Richard Henderson
74aba4ba51
target/arm: Don't call tcg_clear_temp_count
This is done generically in translator_loop.

Backports commit 7108e255c2d95b44c9dfee8075d0d6fb391281a8 from qemu
2018-11-10 09:40:06 -05:00
Richard Henderson
9bbc412c66
target/arm: Hoist address increment for vector memory ops
This can reduce the number of opcodes required for certain
complex forms of load-multiple (e.g. ld4.16b).

Backports commit a7d8143aed2268f147cc1abfebc848ed6282a313 from qemu
2018-11-10 09:39:20 -05:00
Peter Maydell
d60fe610bb
target/arm: Report correct syndrome for FP/SIMD traps to Hyp mode
For traps of FP/SIMD instructions to AArch32 Hyp mode, the syndrome
provided in HSR has more information than is reported to AArch64.
Specifically, there are extra fields TA and coproc which indicate
whether the trapped instruction was FP or SIMD. Add this extra
information to the syndromes we construct, and mask it out when
taking the exception to AArch64.

Backports commit 4be42f4013fa1a9df47b48aae5148767bed8e80c from qemu
2018-11-10 09:36:41 -05:00
Peter Maydell
075bac4d57
target/arm: Get IL bit correct for v7 syndrome values
For the v7 version of the Arm architecture, the IL bit in
syndrome register values where the field is not valid was
defined to be UNK/SBZP. In v8 this is RES1, which is what
QEMU currently implements. Handle the desired v7 behaviour
by squashing the IL bit for the affected cases:
* EC == EC_UNCATEGORIZED
* prefetch aborts
* data aborts where ISV is 0

(The fourth case listed in the v8 Arm ARM DDI 0487C.a in
section G7.2.70, "illegal state exception", can't happen
on a v7 CPU.)

This deals with a corner case noted in a comment.

Backports commit 2ed08180db096ea5e44573529b85e09b1ed10b08 from qemu
2018-11-10 09:29:13 -05:00