aboutsummaryrefslogtreecommitdiff
path: root/tcg
AgeCommit message (Collapse)AuthorFilesLines
2020-06-02tcg: Implement gvec support for rotate by scalarRichard Henderson6-0/+32
No host backend support yet, but the interfaces for rotls are in place. Only implement left-rotate for now, as the only known use of vector rotate by scalar is s390x, so any right-rotate would be unused and untestable. Reviewed-by: Alex Bennée <alex.bennee@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
2020-06-02tcg: Remove expansion to shift by vector from do_shiftsRichard Henderson1-24/+11
We do not reflect this expansion in tcg_can_emit_vecop_list, so it is unused and unusable. However, we actually perform the same expansion in do_gvec_shifts, so it is also unneeded. Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
2020-06-02tcg: Implement gvec support for rotate by vectorRichard Henderson7-1/+141
No host backend support yet, but the interfaces for rotlv and rotrv are in place. Reviewed-by: Alex Bennée <alex.bennee@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- v3: Drop the generic expansion from rot to shift; we can do better for each backend, and then this code becomes unused.
2020-06-02tcg: Implement gvec support for rotate by immediateRichard Henderson7-1/+87
No host backend support yet, but the interfaces for rotli are in place. Canonicalize immediate rotate to the left, based on a survey of architectures, but provide both left and right shift interfaces to the translators. Reviewed-by: Alex Bennée <alex.bennee@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
2020-05-15disas: include an optional note for the start of disassemblyAlex Bennée1-2/+2
This will become useful shortly for providing more information about output assembly inline. While there fix up the indenting and code formatting in disas(). Signed-off-by: Alex Bennée <alex.bennee@linaro.org> Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com> Message-Id: <20200513175134.19619-9-alex.bennee@linaro.org>
2020-05-06tcg: Fix integral argument type to tcg_gen_rot[rl]i_i{32,64}Richard Henderson1-8/+8
For the benefit of compatibility of function pointer types, we have standardized on int32_t and int64_t as the integral argument to tcg expanders. We converted most of them in 474b2e8f0f7, but missed the rotates. Reviewed-by: Alex Bennée <alex.bennee@linaro.org> Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
2020-05-06tcg: Add load_dest parameter to GVecGen2Richard Henderson1-13/+32
We have this same parameter for GVecGen2i, GVecGen3, and GVecGen3i. This will make some SVE2 insns easier to parameterize. Reviewed-by: Alex Bennée <alex.bennee@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
2020-05-06tcg: Improve vector tail clearingRichard Henderson1-20/+64
Better handling of non-power-of-2 tails as seen with Arm 8-byte vector operations. Reviewed-by: Alex Bennée <alex.bennee@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
2020-05-06tcg: Remove tcg_gen_gvec_dup{8,16,32,64}iRichard Henderson1-28/+0
These interfaces are now unused. Reviewed-by: LIU Zhiwei <zhiwei_liu@c-sky.com> Reviewed-by: David Hildenbrand <david@redhat.com> Reviewed-by: Alex Bennée <alex.bennee@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
2020-05-06tcg: Use tcg_gen_gvec_dup_imm in logical simplificationsRichard Henderson1-4/+4
Replace the outgoing interface. Reviewed-by: LIU Zhiwei <zhiwei_liu@c-sky.com> Reviewed-by: Alex Bennée <alex.bennee@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
2020-05-06tcg: Add tcg_gen_gvec_dup_immRichard Henderson1-0/+7
Add a version of tcg_gen_dup_* that takes both immediate and a vector element size operand. This will replace the set of tcg_gen_gvec_dup{8,16,32,64}i functions that encode the element size within the function name. Reviewed-by: LIU Zhiwei <zhiwei_liu@c-sky.com> Reviewed-by: David Hildenbrand <david@redhat.com> Reviewed-by: Alex Bennée <alex.bennee@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
2020-04-12tcg/mips: mips sync* encode errorlixinyu1-5/+5
OPC_SYNC_WMB, OPC_SYNC_MB, OPC_SYNC_ACQUIRE, OPC_SYNC_RELEASE and OPC_SYNC_RMB have wrong encode. According to the mips manual, their encode should be 'OPC_SYNC | 0x?? << 6' rather than 'OPC_SYNC | 0x?? << 5'. Wrong encode can lead illegal instruction errors. These instructions often appear with multi-threaded simulation. Fixes: 6f0b99104a3 ("tcg/mips: Add support for fence") Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Reviewed-by: Aleksandar Markovic <aleksandar.qemu.devel@gmail.com> Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org> Signed-off-by: lixinyu <precinct@mail.ustc.edu.cn> Message-Id: <20200411124612.12560-1-precinct@mail.ustc.edu.cn> Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
2020-04-07tcg/i386: Fix %r12 guest_base initializationRichard Henderson1-1/+1
When %gs cannot be used, we use register offset addressing. This path is almost never used, so it was clearly not tested. Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Reviewed-by: Alex Bennée <alex.bennee@linaro.org> Tested-by: Alex Bennée <alex.bennee@linaro.org> Message-Id: <20200406174803.8192-1-richard.henderson@linaro.org> Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
2020-03-30tcg/i386: Fix INDEX_op_dup2_vecRichard Henderson1-3/+7
We were only constructing the 64-bit element, and not replicating the 64-bit element across the rest of the vector. Cc: qemu-stable@nongnu.org Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
2020-03-17tcg/i386: Bound shift count expanding sari_vecRichard Henderson1-3/+6
A given RISU testcase for SVE can produce tcg-op-vec.c:511: do_shifti: Assertion `i >= 0 && i < (8 << vece)' failed. because expand_vec_sari gave a shift count of 32 to a MO_32 vector shift. In 44f1441dbe1, we changed from direct expansion of vector opcodes to re-use of the tcg expanders. So while the comment correctly notes that the hw will handle such a shift count, we now have to take our own sanity checks into account. Which is easy in this particular case. Fixes: 44f1441dbe1 Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
2020-02-28tcg/arm: Expand epilogue inlineRichard Henderson1-10/+2
It is, after all, just two instructions. Profiling on a cortex-a15, using -d nochain to increase the number of exit_tb that are executed, shows a minor improvement of 0.5%. Signed-off-by: Richard Henderson <rth@twiddle.net>
2020-02-28tcg/arm: Split out tcg_out_epilogueRichard Henderson1-8/+11
We will shortly use this function from tcg_out_op as well. Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com> Signed-off-by: Richard Henderson <rth@twiddle.net>
2020-02-25tcg: save vaddr temp for plugin usageAlex Bennée1-3/+20
While do_gen_mem_cb does copy (via extu_tl_i64) vaddr into a new temp this won't help if the vaddr temp gets clobbered by the actual load/store op. To avoid this clobbering we explicitly copy vaddr before the op to ensure it is live my the time we do the instrumentation. Suggested-by: Richard Henderson <richard.henderson@linaro.org> Signed-off-by: Alex Bennée <alex.bennee@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Reviewed-by: Emilio G. Cota <cota@braap.org> Cc: qemu-stable@nongnu.org Message-Id: <20200225124710.14152-18-alex.bennee@linaro.org>
2020-02-12tcg: Add tcg_gen_gvec_5_ptrRichard Henderson1-0/+32
Extend the vector generator infrastructure to handle 5 vector arguments. Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com> Reviewed-by: Alex Bennée <alex.bennee@linaro.org> Reviewed-by: Taylor Simpson <tsimpson@quicinc.com> Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
2020-01-15tcg: Move TCG headers to include/tcg/Philippe Mathieu-Daudé6-3516/+0
Reviewed-by: Paolo Bonzini <pbonzini@redhat.com> Reviewed-by: Alistair Francis <alistair.francis@wdc.com> Reviewed-by: Stefan Weil <sw@weilnetz.de> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com> Message-Id: <20200101112303.20724-4-philmd@redhat.com> Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
2020-01-15tcg: Search includes in the parent source directoryPhilippe Mathieu-Daudé8-14/+14
All the *.inc.c files included by tcg/$TARGET/tcg-target.inc.c are in tcg/, their parent directory. To simplify the preprocessor search path, include the relative parent path: '..'. Patch created mechanically by running: $ for x in tcg-pool.inc.c tcg-ldst.inc.c; do \ sed -i "s,#include \"$x\",#include \"../$x\"," \ $(git grep -l "#include \"$x\""); \ done Acked-by: David Gibson <david@gibson.dropbear.id.au> (ppc parts) Reviewed-by: Paolo Bonzini <pbonzini@redhat.com> Reviewed-by: Alistair Francis <alistair.francis@wdc.com> Reviewed-by: Stefan Weil <sw@weilnetz.de> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com> Message-Id: <20200101112303.20724-3-philmd@redhat.com> Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
2020-01-15tcg: Search includes from the project root source directoryPhilippe Mathieu-Daudé10-18/+18
We currently search both the root and the tcg/ directories for tcg files: $ git grep '#include "tcg/' | wc -l 28 $ git grep '#include "tcg[^/]' | wc -l 94 To simplify the preprocessor search path, unify by expliciting the tcg/ directory. Patch created mechanically by running: $ for x in \ tcg.h tcg-mo.h tcg-op.h tcg-opc.h \ tcg-op-gvec.h tcg-gvec-desc.h; do \ sed -i "s,#include \"$x\",#include \"tcg/$x\"," \ $(git grep -l "#include \"$x\""); \ done Acked-by: David Gibson <david@gibson.dropbear.id.au> (ppc parts) Reviewed-by: Paolo Bonzini <pbonzini@redhat.com> Reviewed-by: Alistair Francis <alistair.francis@wdc.com> Reviewed-by: Stefan Weil <sw@weilnetz.de> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com> Message-Id: <20200101112303.20724-2-philmd@redhat.com> Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
2020-01-15cputlb: Rename helper_ret_ld*_cmmu to cpu_ld*_codeRichard Henderson1-29/+0
There are no uses of the *_cmmu names other than the bare wrapping within the *_code inlines. Therefore rename the functions so we can drop the inlines. Use abi_ptr instead of target_ulong in preparation for user-only; the two types are identical for softmmu. Tested-by: Philippe Mathieu-Daudé <philmd@redhat.com> Reviewed-by: Aleksandar Markovic <amarkovic@wavecomp.com> Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
2019-12-18Add use of RCU for qemu_logfile.Robert Foley1-2/+10
This now allows changing the logfile while logging is active, and also solves the issue of a seg fault while changing the logfile. Any read access to the qemu_logfile handle will use the rcu_read_lock()/unlock() around the use of the handle. To fetch the handle we will use atomic_rcu_read(). We also in many cases do a check for validity of the logfile handle before using it to deal with the case where the file is closed and set to NULL. The cases where we write to the qemu_logfile will use atomic_rcu_set(). Writers will also use call_rcu() with a newly added qemu_logfile_free function for freeing/closing when readers have finished. Signed-off-by: Robert Foley <robert.foley@linaro.org> Reviewed-by: Alex Bennée <alex.bennee@linaro.org> Signed-off-by: Alex Bennée <alex.bennee@linaro.org> Message-Id: <20191118211528.3221-6-robert.foley@linaro.org>
2019-12-18qemu_log_lock/unlock now preserves the qemu_logfile handle.Robert Foley1-8/+8
qemu_log_lock() now returns a handle and qemu_log_unlock() receives a handle to unlock. This allows for changing the handle during logging and ensures the lock() and unlock() are for the same file. Also in target/tilegx/translate.c removed the qemu_log_lock()/unlock() calls (and the log("\n")), since the translator can longjmp out of the loop if it attempts to translate an instruction in an inaccessible page. Signed-off-by: Robert Foley <robert.foley@linaro.org> Reviewed-by: Alex Bennée <alex.bennee@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Signed-off-by: Alex Bennée <alex.bennee@linaro.org> Message-Id: <20191118211528.3221-5-robert.foley@linaro.org>
2019-11-11tcg/LICENSE: Remove out of date claim about TCG subdirectory licensingPeter Maydell1-3/+0
Since 2008 the tcg/LICENSE file has not changed: it claims that everything under tcg/ is BSD-licensed. This is not true and hasn't been true for years: in 2013 we accepted the tcg/aarch64 target code under a GPLv2-or-later license statement. We also have generic vector optimisation code under the LGPL2.1-or-later, and the TCI backend is GPLv2-or-later. Further, many of the files are not BSD licensed but MIT licensed. We don't really consider the tcg subdirectory to be a distinct part of QEMU anyway. Remove the LICENSE file, since claiming false information about the license of the code is confusing. Update the main project LICENSE file also to be clearer about the licenses used by TCG. Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Message-Id: <20191025155848.17362-5-peter.maydell@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
2019-11-11tcg/ppc/tcg-target.opc.h: Add copyright/licensePeter Maydell1-0/+20
Add the copyright/license boilerplate for tcg/i386/tcg-target.opc.h. This file has had only two commits, 4b06c216826b7e4 and d9897efa1fd3174ec, both by a Linaro engineer. The license is MIT, since that's what the rest of tcg/ppc/ is. Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Message-Id: <20191025155848.17362-4-peter.maydell@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
2019-11-11tcg/i386/tcg-target.opc.h: Add copyright/licensePeter Maydell1-3/+25
Add the copyright/license boilerplate for tcg/i386/tcg-target.opc.h. This file has had only one commit, 770c2fc7bb70804a, by a Linaro engineer. The license is MIT, since that's what the rest of tcg/i386/ is. Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Message-Id: <20191025155848.17362-3-peter.maydell@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
2019-11-11tcg/aarch64/tcg-target.opc.h: Add copyright/licensePeter Maydell1-3/+12
Add the copyright/license boilerplate for target/aarch64/tcg-target.opc.h. This file has only had two commits: 14e4c1e2355473ccb29 and 79525dfd08262d8, both by the same Linaro engineer. The license is GPL-2-or-later, since that's what the rest of tcg/aarch64 uses. Reviewed-by: Alex Bennée <alex.bennee@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Message-Id: <20191025155848.17362-2-peter.maydell@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
2019-10-30Merge remote-tracking branch ↵Peter Maydell5-9/+93
'remotes/stsquad/tags/pull-tcg-plugins-281019-4' into staging TCG Plugins initial implementation - use --enable-plugins @ configure - low impact introspection (-plugin empty.so to measure overhead) - plugins cannot alter guest state - example plugins included in source tree (tests/plugins) - -d plugin to enable plugin output in logs - check-tcg runs extra tests when plugins enabled - documentation in docs/devel/plugins.rst # gpg: Signature made Mon 28 Oct 2019 15:13:23 GMT # gpg: using RSA key 6685AE99E75167BCAFC8DF35FBD0DB095A9E2A44 # gpg: Good signature from "Alex Bennée (Master Work Key) <alex.bennee@linaro.org>" [full] # Primary key fingerprint: 6685 AE99 E751 67BC AFC8 DF35 FBD0 DB09 5A9E 2A44 * remotes/stsquad/tags/pull-tcg-plugins-281019-4: (57 commits) travis.yml: enable linux-gcc-debug-tcg cache MAINTAINERS: add me for the TCG plugins code scripts/checkpatch.pl: don't complain about (foo, /* empty */) .travis.yml: add --enable-plugins tests include/exec: wrap cpu_ldst.h in CONFIG_TCG accel/stubs: reduce headers from tcg-stub tests/plugin: add hotpages to analyse memory access patterns tests/plugin: add instruction execution breakdown tests/plugin: add a hotblocks plugin tests/tcg: enable plugin testing tests/tcg: drop test-i386-fprem from TESTS when not SLOW tests/tcg: move "virtual" tests to EXTRA_TESTS tests/tcg: set QEMU_OPTS for all cris runs tests/tcg/Makefile.target: fix path to config-host.mak tests/plugin: add sample plugins linux-user: support -plugin option vl: support -plugin option plugin: add qemu_plugin_outs helper plugin: add qemu_plugin_insn_disas helper plugin: expand the plugin_init function to include an info block ... Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
2019-10-28cputlb: ensure _cmmu helper functions follow the naming standardAlex Bennée1-6/+14
We document this in docs/devel/load-stores.rst so lets follow it. The 32 bit and 64 bit access functions have historically not included the sign so we leave those as is. We also introduce some signed helpers which are used for loading immediate values in the translator. Fixes: 282dffc8 Signed-off-by: Alex Bennée <alex.bennee@linaro.org> Message-Id: <20191021150910.23216-1-alex.bennee@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
2019-10-28tcg: let plugins instrument virtual memory accessesEmilio G. Cota2-9/+32
To capture all memory accesses we need hook into all the various helper functions that are involved in memory operations as well as the injected inline helper calls. A later commit will allow us to resolve the actual guest HW addresses by replaying the lookup. Signed-off-by: Emilio G. Cota <cota@braap.org> [AJB: drop haddr handling, just deal in vaddr] Signed-off-by: Alex Bennée <alex.bennee@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
2019-10-28plugin-gen: add module for TCG-related codeEmilio G. Cota4-0/+56
We first inject empty instrumentation from translator_loop. After translation, we go through the plugins to see what they want to register for, filling in the empty instrumentation. If if turns out that some instrumentation remains unused, we remove it. This approach supports the following features: - Inlining TCG code for simple operations. Note that we do not export TCG ops to plugins. Instead, we give them a C API to insert inlined ops. So far we only support adding an immediate to a u64, e.g. to count events. - "Direct" callbacks. These are callbacks that do not go via a helper. Instead, the helper is defined at run-time, so that the plugin code is directly called from TCG. This makes direct callbacks as efficient as possible; they are therefore used for very frequent events, e.g. memory callbacks. - Passing the host address to memory callbacks. Most of this is implemented in a later patch though. - Instrumentation of memory accesses performed from helpers. See the corresponding comment, as well as a later patch. Signed-off-by: Emilio G. Cota <cota@braap.org> [AJB: add alloc_tcg_plugin_context, use glib, rm hwaddr] Signed-off-by: Alex Bennée <alex.bennee@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
2019-10-28tcg: add tcg_gen_st_ptrEmilio G. Cota1-0/+5
Will gain a user soon. Signed-off-by: Emilio G. Cota <cota@braap.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
2019-10-28trace: add mmu_index to mem_infoAlex Bennée1-4/+4
We are going to re-use mem_info later for plugins and will need to track the mmu_idx for softmmu code. Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
2019-10-28cputlb: ensure _cmmu helper functions follow the naming standardAlex Bennée1-6/+14
We document this in docs/devel/load-stores.rst so lets follow it. The 32 bit and 64 bit access functions have historically not included the sign so we leave those as is. We also introduce some signed helpers which are used for loading immediate values in the translator. Fixes: 282dffc8 Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Signed-off-by: Alex Bennée <alex.bennee@linaro.org> Message-Id: <20191021150910.23216-1-alex.bennee@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
2019-10-28tci: Add implementation for INDEX_op_ld16u_i64Stefan Weil1-0/+15
This fixes "make check-tcg" on a Debian x86_64 host. Signed-off-by: Stefan Weil <sw@weilnetz.de> Tested-by: Thomas Huth <thuth@redhat.com> Message-Id: <20190410194838.10123-1-sw@weilnetz.de> Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
2019-10-14tcg/ppc: Update vector support for v3.00 dup/dupiRichard Henderson1-2/+26
These new instructions are conditional on MSR.VEC for TX=1, so we can consider these Altivec instructions. Tested-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk> Reviewed-by: Aleksandar Markovic <amarkovic@wavecomp.com> Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
2019-10-14tcg/ppc: Update vector support for v3.00 load/storeRichard Henderson1-9/+38
These new instructions are a mix of those like LXSD that are only conditional only on MSR.VEC and those like LXV that are conditional on MSR.VEC for TX=1. Thus, in the end, we can consider all of these as Altivec instructions. Tested-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk> Reviewed-by: Aleksandar Markovic <amarkovic@wavecomp.com> Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
2019-10-14tcg/ppc: Update vector support for v3.00 AltivecRichard Henderson2-1/+24
These new instructions are conditional only on MSR.VEC and are thus part of the Altivec instruction set, and not VSX. This includes negation and compare not equal. Tested-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk> Reviewed-by: Aleksandar Markovic <amarkovic@wavecomp.com> Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
2019-10-14tcg/ppc: Update vector support for v2.07 FPRichard Henderson1-6/+26
These new instructions are conditional on MSR.FP when TX=0 and MSR.VEC when TX=1. Since we only care about the Altivec registers, and force TX=1, we can consider these to be Altivec instructions. Since Altivec is true for any use of vector types, we only need test have_isa_2_07. This includes moves to and from the integer registers. Tested-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk> Reviewed-by: Aleksandar Markovic <amarkovic@wavecomp.com> Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
2019-10-14tcg/ppc: Update vector support for v2.07 VSXRichard Henderson1-0/+11
These new instructions are conditional only on MSR.VSX and are thus part of the VSX instruction set, and not Altivec. This includes double-word loads and stores. Tested-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk> Reviewed-by: Aleksandar Markovic <amarkovic@wavecomp.com> Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
2019-10-14tcg/ppc: Update vector support for v2.07 AltivecRichard Henderson2-22/+67
These new instructions are conditional only on MSR.VEC and are thus part of the Altivec instruction set, and not VSX. This includes lots of double-word arithmetic and a few extra logical operations. Tested-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk> Reviewed-by: Aleksandar Markovic <amarkovic@wavecomp.com> Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
2019-10-14tcg/ppc: Update vector support for VSXRichard Henderson2-6/+51
The VSX instruction set instructions include double-word loads and stores, double-word load and splat, double-word permute, and bit select. All of which require multiple operations in the Altivec instruction set. Because the VSX registers map %vsr32 to %vr0, and we have no current intention or need to use vector registers outside %vr0-%vr19, force on the {ax,bx,cx,tx} bits within the added VSX insns so that we don't have to otherwise modify the VR[TABC] macros. Tested-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk> Reviewed-by: Aleksandar Markovic <amarkovic@wavecomp.com> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Signed-off-by: Aleksandar Markovic <amarkovic@wavecomp.com>
2019-10-14tcg/ppc: Enable Altivec detectionRichard Henderson1-0/+4
Now that we have implemented the required tcg operations, we can enable detection of host vector support. Tested-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk> (PPC32) Reviewed-by: Aleksandar Markovic <amarkovic@wavecomp.com> Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
2019-10-14tcg/ppc: Support vector dup2Richard Henderson1-0/+9
This is only used for 32-bit hosts. Tested-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk> Reviewed-by: Aleksandar Markovic <amarkovic@wavecomp.com> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Signed-off-by: Aleksandar Markovic <amarkovic@wavecomp.com>
2019-10-14tcg/ppc: Support vector multiplyRichard Henderson3-2/+121
For Altivec, this is always an expansion. Tested-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk> Reviewed-by: Aleksandar Markovic <amarkovic@wavecomp.com> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Signed-off-by: Aleksandar Markovic <amarkovic@wavecomp.com>
2019-10-14tcg/ppc: Support vector shift by immediateRichard Henderson2-3/+57
For Altivec, this is done via vector shift by vector, and loading the immediate into a register. Tested-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk> Reviewed-by: Aleksandar Markovic <amarkovic@wavecomp.com> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Signed-off-by: Aleksandar Markovic <amarkovic@wavecomp.com>
2019-10-14tcg/ppc: Add support for vector saturated add/subtractRichard Henderson2-1/+37
Add support for vector saturated add/subtract using Altivec instructions: VADDSBS, VADDSHS, VADDSWS, VADDUBS, VADDUHS, VADDUWS, and VSUBSBS, VSUBSHS, VSUBSWS, VSUBUBS, VSUBUHS, VSUBUWS. Tested-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk> Reviewed-by: Aleksandar Markovic <amarkovic@wavecomp.com> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Signed-off-by: Aleksandar Markovic <amarkovic@wavecomp.com>
2019-10-14tcg/ppc: Add support for vector add/subtractRichard Henderson1-0/+20
Add support for vector add/subtract using Altivec instructions: VADDUBM, VADDUHM, VADDUWM, VSUBUBM, VSUBUHM, VSUBUWM. Tested-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk> Reviewed-by: Aleksandar Markovic <amarkovic@wavecomp.com> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Signed-off-by: Aleksandar Markovic <amarkovic@wavecomp.com>
generated by cgit v1.1 at 2024-10-01 12:25:23 +0000