Commit graph

472 commits

Author SHA1 Message Date
Richard Henderson f7e0c9e079
target/arm: Implement SVE FP Compare with Zero Group
Backports commit 4d2e2a03384a43c641e0cbca7ac79d7d0c50f666 from qemu
2018-07-03 03:49:52 -04:00
Richard Henderson 942f3c835e
target/arm: Implement SVE Floating Point Unary Operations - Unpredicated Group
Backports commit 3887c0388d39930ab419d4ae6e8ca5ea67a74ad5 from qemu
2018-07-03 03:44:40 -04:00
Richard Henderson f9f228efec
target/arm: Implement SVE FP Fast Reduction Group
Backports commit 23fbe79faa38cb4acc59f956a63feba3c2cc73ac from qemu
2018-07-03 03:41:00 -04:00
Richard Henderson cf3c7824ff
target/arm: Implement SVE Floating Point Multiply Indexed Group
Backports commit ca40a6e6e390eb1cad7ade881dc7c622793f9324 from qemu
2018-07-03 03:35:49 -04:00
Richard Henderson c718ef4243
target/arm: Implement SVE floating-point arithmetic with immediate
Backports commit cc48affe83fff4b2886c064265d7103dee5e4a14 from qemu
2018-07-03 03:24:46 -04:00
Richard Henderson db1d39ab4a
target/arm: Implement SVE floating-point compare vectors
Backports commit abfdefd5bd444b629d16dcefc2b60ac8da37e87d from qemu
2018-07-03 03:14:24 -04:00
Richard Henderson be65f60799
target/arm: Implement SVE scatter store vector immediate
Backports commit 408ecde97bd30f8ec13f831976d0a9a6535bb569 from qemu
2018-07-03 03:09:51 -04:00
Richard Henderson 892a9a66eb
target/arm: Implement SVE first-fault gather loads
Backports commit ed67eb7fa2a63b6709ec94397d833bc3686f7833 from qemu
2018-07-03 03:06:07 -04:00
Richard Henderson 1782f20dde
target/arm: Implement SVE gather loads
Backports commit 673e9fa6c29e030f4ab6ceae5d0f50bd36fe0ee0 from qemu
2018-07-03 02:58:18 -04:00
Richard Henderson 3a5d095277
target/arm: Implement SVE prefetches
Backports commit dec6cf6b43a1e3b18626852064d1e6e863c9b681 from qemu
2018-07-03 02:42:06 -04:00
Richard Henderson b78e283513
target/arm: Implement SVE scatter stores
Backports commit f6dbf62a7e3d00e9a1dcc7fe3e53b32c3ed93e24 from qemu
2018-07-03 02:40:17 -04:00
Richard Henderson d71cdfa41f
target/arm: Implement SVE store vector/predicate register
Backports commit 5047c204d0d4a0fff616a24963b2b45c7d9ba4c4 from qemu
2018-07-03 02:32:10 -04:00
Richard Henderson c497dc0a83
target/arm: Implement SVE load and broadcast element
Backports commit 684598640dc3b28f86ccc28cc9af50ba257f4cc8 from qemu
2018-07-03 02:27:00 -04:00
Richard Henderson 2caa99929e
target/arm: Implement SVE Floating Point Accumulating Reduction Group
Backports commit 7f9ddf64d5fe5bfaa91ae0ec52217d86f4d86452 from qemu
2018-07-03 02:21:41 -04:00
Richard Henderson 44d89b4cb1
target/arm: Implement SVE FP Multiply-Add Group
Backports commit 6ceabaad1101b0b33490b0fd4bed5b6445b0a34e from qemu
2018-07-03 02:17:16 -04:00
Richard Henderson d1c5d5b728
target/arm: Implement SVE floating-point arithmetic (predicated)
Backports commit ec3b87c28eb120b6575cc1ed7bfbfbf1b0060163 from qemu
2018-07-03 02:13:04 -04:00
Richard Henderson e7f53b6d1b
target/arm: Implement SVE integer convert to floating-point
Backports commit 8092c6a314c9625d8976fb6c5b6969f2a1105cd8 from qemu
2018-07-03 02:05:50 -04:00
Richard Henderson 8a99e1bad2
target/arm: Implement SVE load and broadcast quadword
Backports commit 05abe304be2987cb3576729a14dab96e9ccfaec9 from qemu
2018-07-03 02:01:23 -04:00
Richard Henderson 3a51861c05
target/arm: Implement SVE Memory Contiguous Store Group
Backports commit 1a039c7e58533d5f9431a2064771113d5961c964 from qemu
2018-07-03 01:57:52 -04:00
Richard Henderson 22a155a5d4
target/arm: Implement SVE Contiguous Load, first-fault and no-fault
Backports commit e2654d757598d6c06d1ceb25c62ddf7d63cac32f from qemu
2018-07-03 01:52:38 -04:00
Richard Henderson da1eeb226e
target/arm: Implement SVE Memory Contiguous Load Group
Backports commit c4e7c493da2fdd2c92a7958e592e38200af2f1ba from qemu
2018-07-03 01:40:18 -04:00
Peter Maydell 4abcf14b52
target/arm: Handle small regions in get_phys_addr_pmsav8()
Allow ARMv8M to handle small MPU and SAU region sizes, by making
get_phys_add_pmsav8() set the page size to the 1 if the MPU or
SAU region covers less than a TARGET_PAGE_SIZE.

We choose to use a size of 1 because it makes no difference to
the core code, and avoids having to track both the base and
limit for SAU and MPU and then convert into an artificially
restricted "page size" that the core code will then ignore.

Since the core TCG code can't handle execution from small
MPU regions, we strip the exec permission from them so that
any execution attempts will cause an MPU exception, rather
than allowing it to end up with a cpu_abort() in
get_page_addr_code().

(The previous code's intention was to make any small page be
treated as having no permissions, but unfortunately errors
in the implementation meant that it didn't behave that way.
It's possible that some binaries using small regions were
accidentally working with our old behaviour and won't now.)

We also retain an existing bug, where we ignored the possibility
that the SAU region might not cover the entire page, in the
case of executable regions. This is necessary because some
currently-working guest code images rely on being able to
execute from addresses which are covered by a page-sized
MPU region but a smaller SAU region. We can remove this
workaround if we ever support execution from small regions.

Backports commit 720424359917887c926a33d248131fbff84c9c28 from qemu
2018-07-03 00:55:37 -04:00
Peter Maydell 5e8e8b9bf8
target/arm: Set page (region) size in get_phys_addr_pmsav7()
We want to handle small MPU region sizes for ARMv7M. To do this,
make get_phys_addr_pmsav7() set the page size to the region
size if it is less that TARGET_PAGE_SIZE, rather than working
only in TARGET_PAGE_SIZE chunks.

Since the core TCG code con't handle execution from small
MPU regions, we strip the exec permission from them so that
any execution attempts will cause an MPU exception, rather
than allowing it to end up with a cpu_abort() in
get_page_addr_code().

(The previous code's intention was to make any small page be
treated as having no permissions, but unfortunately errors
in the implementation meant that it didn't behave that way.
It's possible that some binaries using small regions were
accidentally working with our old behaviour and won't now.)

Backports commit e5e40999b5e03567ef654546e3d448431643f8f3 from qemu
2018-07-03 00:49:51 -04:00
Julia Suvorova f28514178a
target/arm: Strict alignment for ARMv6-M and ARMv8-M Baseline
Unlike ARMv7-M, ARMv6-M and ARMv8-M Baseline only supports naturally
aligned memory accesses for load/store instructions.

Backports commit 2aeba0d007d33efa12a6339bb140aa634e0d52eb from qemu
2018-06-29 14:15:33 -05:00
Julia Suvorova 65d0e176c3
target/arm: Introduce ARM_FEATURE_M_MAIN
This feature is intended to distinguish ARMv8-M variants: Baseline and
Mainline. ARMv7-M compatibility requires the Main Extension. ARMv6-M
compatibility is provided by all ARMv8-M implementations.

Backports commit cc2ae7c9de14efd72c6205825eb7cd980ac09c11 from qemu
2018-06-29 14:14:18 -05:00
Edgar E. Iglesias dfc6bd6a0b
target-arm: Add the Cortex-R5F
Add the Cortex-R5F with the optional FPU enabled.

Backports commit ebac5458c7517ed7b8ee06eb90beacc7472b295d from qemu
2018-06-29 13:53:31 -05:00
Julia Suvorova 38747fc125
target/arm: Minor cleanup for ARMv6-M 32-bit instructions
The arrays were made static, "if" was simplified because V7M and V8M
define V6 feature.

Backports commit 8297cb13e407db8a96cc7ed6b6a6c318a150759a from qemu
2018-06-29 13:50:30 -05:00
Julia Suvorova f447a6f668
target/arm: Allow ARMv6-M Thumb2 instructions
ARMv6-M supports 6 Thumb2 instructions. This patch checks for these
instructions and allows their execution.
Like Thumb2 cores, ARMv6-M always interprets BL instruction as 32-bit.

This patch is required for future Cortex-M0 support.

Backports commit 14120108f87b3f9e1beacdf0a6096e464e62bb65 from qemu
2018-06-15 14:12:20 -04:00
Richard Henderson d81cc5f5cd
target/arm: Implement SVE Floating Point Arithmetic - Unpredicated Group
Backports commit 29b80469dc51ae4064e9ef9223967882d2610523 from qemu
2018-06-15 14:10:16 -04:00
Richard Henderson e56db70fee
target/arm: Implement SVE Integer Wide Immediate - Unpredicated Group
Backports commit 6e6a157d683058d86a224cf128320d67b0915365 from qemu
2018-06-15 14:05:11 -04:00
Richard Henderson b5352c6ba1
target/arm: Implement FDUP/DUP
Backports commit ed49196125360c037d7f23c1c315a85cc234e72d from qemu
2018-06-15 13:56:03 -04:00
Richard Henderson 496bb35c97
target/arm: Implement SVE Integer Compare - Scalars Group
Backports commit caf1cefc72be98497e0907d0e07f4327fc641e96 from qemu
2018-06-15 13:54:05 -04:00
Richard Henderson bc55b3e570
target/arm: Implement SVE Predicate Count Group
Backports commit 9ee3a611de28b8d0862fa687215b04b5aad20747 from qemu
2018-06-15 13:49:58 -04:00
Richard Henderson bb930f35b0
target/arm: Implement SVE Partition Break Group
Backports commit 35da316f5e847292ffbe7b6d16cd3988043dfe22 from qemu
2018-06-15 13:42:35 -04:00
Richard Henderson ade246e87b
target/arm: Implement SVE Integer Compare - Immediate Group
Backports commit 38cadeba0daf0f16cf2aeaa5b2752b26fb0676c5 from qemu
2018-06-15 13:35:40 -04:00
Richard Henderson 2969a38d61
target/arm: Implement SVE Integer Compare - Vectors Group
Backports commit 757f9cff1b63895bfd6fc8d66a6e52d7c40baa7b from qemu
2018-06-15 13:29:15 -04:00
Richard Henderson 7211d415a4
target/arm: Implement SVE Select Vectors Group
Backports commit d3fe4a29d754dee73cbf3cb7584db222981179ac from qemu
2018-06-15 13:17:47 -04:00
Richard Henderson 7698c1634e
target/arm: Implement SVE vector splice (predicated)
Backports commit b48ff24098c72f86e187e6abb7e9ca4de40a7fb4 from qemu
2018-06-15 13:14:33 -04:00
Richard Henderson 7d930e8515
target/arm: Implement SVE reverse within elements
Backports commit dae8fb9019d2aa6ccb151a19871df40de6c98e29 from qemu
2018-06-15 13:12:14 -04:00
Richard Henderson 53d151de58
target/arm: Implement SVE copy to vector (predicated)
Backports commit 792a557847697235037fea30eaaacb9b45b4c9e5 from qemu
2018-06-15 13:03:36 -04:00
Richard Henderson 0bb2fdd752
target/arm: Implement SVE conditionally broadcast/extract element
Backports commit ef23cb726dc32375bc2fca7ac3e9f34816f6ee13 from qemu
2018-06-15 13:01:40 -04:00
Richard Henderson 8ba3bde59b
target/arm: Implement SVE compress active elements
Backports commit 3ca879aeb3412bc2be35d01a7bedf5fada960b5d from qemu
2018-06-15 12:52:19 -04:00
Richard Henderson d9ed221567
target/arm: Implement SVE Permute - Interleaving Group
Backports commit 234b48e9c68759aea78ff5a1e49c2ba806cd1d83 from qemu
2018-06-15 12:49:42 -04:00
Richard Henderson 3722ab310b
target/arm: Implement SVE Permute - Predicates Group
Backports commit d731d8cb3c74258669211f065c918353eb7b8f4a from qemu
2018-06-15 12:44:50 -04:00
Richard Henderson c57ff23c56
target/arm: Implement SVE Permute - Unpredicated Group
Backports commit 30562ab716bcec0bf718b47b5268949856b17604 from qemu
2018-06-15 12:37:56 -04:00
Richard Henderson 4dc2b5ea79
target/arm: Extend vec_reg_offset to larger sizes
Rearrange the arithmetic so that we are agnostic about the total size
of the vector and the size of the element. This will allow us to index
up to the 32nd byte and with 16-byte elements.

Backports commit 66f2dbd783d0b6172043e3679171421b2d0bac11 from qemu
2018-06-15 12:23:35 -04:00
Philippe Mathieu-Daudé 5992214b3b
target/arm: Add trailing '\n' to qemu_log() calls
Backports commit 0221c8fdf2321137b34bd4dbb2557d0725739822 from qemu
2018-06-08 19:24:38 -04:00
Richard Henderson 10e2b13650
tcg: Pass tb and index to tcg_gen_exit_tb separately
Do the cast to uintptr_t within the helper, so that the compiler
can type check the pointer argument. We can also do some more
sanity checking of the index argument.

Backports commit 07ea28b41830f946de3841b0ac61a3413679feb9 from qemu
2018-06-07 11:56:32 -04:00
Richard Henderson 533a3f6a6c
tcg: Fix helper function vs host abi for float16
Depending on the host abi, float16, aka uint16_t, values are
passed and returned either zero-extended in the host register
or with garbage at the top of the host register.

The tcg code generator has so far been assuming garbage, as that
matches the x86 abi, but this is incorrect for other host abis.
Further, target/arm has so far been assuming zero-extended results,
so that it may store the 16-bit value into a 32-bit slot with the
high 16-bits already clear.

Rectify both problems by mapping "f16" in the helper definition
to uint32_t instead of (a typedef for) uint16_t. This forces
the host compiler to assume garbage in the upper 16 bits on input
and to zero-extend the result on output.

Backports commit 6c2be133a7478e443c99757b833d0f265c48e0a6 from qemu
2018-06-02 10:10:12 -04:00
Peter Maydell 0f0b2e0bd8
target/arm: Honour FPCR.FZ in FRECPX
The FRECPX instructions should (like most other floating point operations)
honour the FPCR.FZ bit which specifies whether input denormals should
be flushed to zero (or FZ16 for the half-precision version).
We forgot to implement this, which doesn't affect the results (since
the calculation doesn't actually care about the mantissa bits) but did
mean we were failing to set the FPSR.IDC bit.

Backports commit 2cfbf36ec07f7cac1aabb3b86f1c95c8a55424ba from qemu
2018-06-02 10:02:57 -04:00