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>
Preparation for collapsing the two byte swaps, adjust_endianness and
handle_bswap, along the I/O path.
Target dependant attributes are conditionalized upon NEED_CPU_H.
Signed-off-by: Tony Nguyen <tony.nguyen@bt.com>
Acked-by: David Gibson <david@gibson.dropbear.id.au>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Acked-by: Cornelia Huck <cohuck@redhat.com>
Message-Id: <81d9cd7d7f5aaadfa772d6c48ecee834e9cf7882.1566466906.git.tony.nguyen@bt.com>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
This patch moves the define of target access alignment earlier from
target/foo/cpu.h to configure.
Suggested in Richard Henderson's reply to "[PATCH 1/4] tcg: TCGMemOp is now
accelerator independent MemOp"
Signed-off-by: Tony Nguyen <tony.nguyen@bt.com>
Message-Id: <11e818d38ebc40e986cfa62dd7d0afdc@tpw09926dag18e.domain1.systemhost.net>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: tony.nguyen@bt.com <tony.nguyen@bt.com>
migration/qemu-file.h neglects to include it even though it needs
ram_addr_t. Fix that. Drop a few superfluous inclusions elsewhere.
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Tested-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Message-Id: <20190812052359.30071-14-armbru@redhat.com>
Basically, the context could get the MachineState reference via call
chains or unrecommended qdev_get_machine() in !CONFIG_USER_ONLY mode.
A local variable of the same name would be introduced in the declaration
phase out of less effort OR replace it on the spot if it's only used
once in the context. No semantic changes.
Signed-off-by: Like Xu <like.xu@linux.intel.com>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Message-Id: <20190518205428.90532-4-like.xu@linux.intel.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Perform a per-element conditional move. This combination operation is
easier to implement on some host vector units than plain cmp+bitsel.
Omit the usual gvec interface, as this is intended to be used by
target-specific gvec expansion call-backs.
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
This operation performs d = (b & a) | (c & ~a), and is present
on a majority of host vector units. Include gvec expanders.
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Allow the backend to expand dup from memory directly, instead of
forcing the value into a temp first. This is especially important
if integer/vector register moves do not exist.
Note that officially tcg_out_dupm_vec is allowed to fail.
If it did, we could fix this up relatively easily:
VECE == 32/64:
Load the value into a vector register, then dup.
Both of these must work.
VECE == 8/16:
If the value happens to be at an offset such that an aligned
load would place the desired value in the least significant
end of the register, go ahead and load w/garbage in high bits.
Load the value w/INDEX_op_ld{8,16}_i32.
Attempt a move directly to vector reg, which may fail.
Store the value into the backing store for OTS.
Load the value into the vector reg w/TCG_TYPE_I32, which must work.
Duplicate from the vector reg into itself, which must work.
All of which is well and good, except that all supported
hosts can support dupm for all vece, so all of the failure
paths would be dead code and untestable.
Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
This case is similar to INDEX_op_mov_* in that we need to do
different things depending on the current location of the source.
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
---
v3: Added some commentary to the tcg_reg_alloc_* functions.
The i386 backend already has these functions, and the aarch64 backend
could easily split out one. Nothing is done with these functions yet,
but this will aid register allocation of INDEX_op_dup_vec in a later patch.
Adjust the aarch64 tcg_out_dupi_vec signature to match the new interface.
Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
PowerPC Altivec does not support direct moves between vector registers
and general registers. So when tcg_out_mov fails, we can use the
backing memory for the temporary to perform the move.
Acked-by: David Hildenbrand <david@redhat.com>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
This patch merely changes the interface, aborting on all failures,
of which there are currently none.
Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
Reviewed-by: David Hildenbrand <david@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Reviewed-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
The only fixed_reg is cpu_env, and it should not be modified
during any TB. Therefore code that tries to special-case moves
into a fixed_reg is dead. Remove it.
Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
Reviewed-by: David Hildenbrand <david@redhat.com>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
If the TB generates too much code, such that backend relocations
overflow, try again with a smaller TB. In support of this, move
relocation processing from a random place within tcg_out_op, in
the handling of branch opcodes, to a new function at the end of
tcg_gen_code.
This is not a complete solution, as there are additional relocs
generated for out-of-line ldst handling and constant pools.
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
If a TB generates too much code, try again with fewer insns.
Fixes: https://bugs.launchpad.net/bugs/1824853
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
This will let backends implement the double-word shift operation.
Reviewed-by: David Hildenbrand <david@redhat.com>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
dump_exec_info() takes an fprintf()-like callback and a FILE * to pass
to it.
Its only caller hmp_info_jit() passes monitor_fprintf() and the
current monitor cast to FILE *. monitor_fprintf() casts it right
back, and is otherwise identical to monitor_printf(). The
type-punning is ugly.
Drop the callback, and call qemu_printf() instead.
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Message-Id: <20190417191805.28198-5-armbru@redhat.com>
dump_opcount_info() takes an fprintf()-like callback and a FILE * to
pass to it.
Its only caller hmp_info_opcount() passes monitor_fprintf() and the
current monitor cast to FILE *. monitor_fprintf() casts it right
back, and is otherwise identical to monitor_printf(). The
type-punning is ugly.
Drop the callback, and call qemu_printf() instead.
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Message-Id: <20190417191805.28198-4-armbru@redhat.com>
Currently, a jump to a label that is not defined anywhere will
be emitted not be relocated. This results in a jump to a random
jump target. With tcg debugging, print a diagnostic to the -d op
file and abort.
This could help debug or detect errors like
c2d9644e6d ("target/arm: Fix crash on conditional instruction in an IT block")
Reported-by: Roman Kapl <code@rkapl.cz>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Free the argument register only after we have verified that the
temporary is not already in that register. This case is likely
now that we are back propagating the preferred register.
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
With these preferences, we can arrange for function call arguments to
be computed into the proper registers instead of requiring extra moves.
Reviewed-by: Emilio G. Cota <cota@braap.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Use this to notice the opcodes that exit the TB, which implies
that local temps are really dead and need not be synced.
Previously we so marked the true end of the TB, but that was
immediately overwritten by the la_bb_end invoked by any
TCG_OPF_BB_END opcode, like exit_tb.
Reviewed-by: Emilio G. Cota <cota@braap.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
No need for a "tcg_" prefix for a static function; we already
have another "la_" prefix for indicating liveness analysis.
Pass in nb_globals and nb_temps, as we will already have them
in registers for other loops within the parent function.
Reviewed-by: Emilio G. Cota <cota@braap.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
There are two blocks of the form
if (foo) {
stuff1;
goto bar;
} else {
baz:
stuff2;
}
which have unnecessary and confusing indentation.
Remove the else and unindent stuff2.
Reviewed-by: Emilio G. Cota <cota@braap.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Try harder to honor the output_pref. When we're forced to allocate
a second register for the input, it does not need to use the input
constraint; that will be honored by the register we allocate for the
output and a move is already required.
Reviewed-by: Emilio G. Cota <cota@braap.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Allocate storage for, but do not yet fill in, per-opcode
preferences for the output operands. Pass it in to the
register allocation routines for output operands.
Reviewed-by: Emilio G. Cota <cota@braap.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
This new argument will aid register allocation by indicating how
the temporary will be used in future. If the preference cannot
be satisfied, fall back to the constraints of the current insn.
Short circuit the preference when it cannot be satisfied or if
it does not further constrain the operation.
With an eye toward optimizing function call sequences, optimize
for the preferred_reg set containing a single register.
For the moment, all users pass 0 for preference.
Reviewed-by: Emilio G. Cota <cota@braap.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Delete trivially dead code that follows unconditional branches and
noreturn helpers. These can occur either via optimization or via
the structure of a target's translator following an exception.
Reviewed-by: Emilio G. Cota <cota@braap.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Increment when adding branches, and decrement when removing them.
Reviewed-by: Emilio G. Cota <cota@braap.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
It's unused since 75e8b9b7aa.
Signed-off-by: Emilio G. Cota <cota@braap.org>
Message-Id: <20181209193749.12277-9-cota@braap.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
This will move the assert for success from within (subroutines of)
patch_reloc into the callers. It will also let new code do something
different when a relocation is out of range.
For the moment, all backends are trivially converted to return true.
Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
When we implemented per-vCPU TCG contexts, we forgot to also
distribute the tcg_time counter, which has remained as a global
accessed without any serialization, leading to potentially missed
counts.
Fix it by distributing the field over the TCG contexts, embedding
it into TCGProfile with a field called "cpu_exec_time", which is more
descriptive than "tcg_time". Add a function to query this value
directly, and for completeness, fill in the field in
tcg_profile_snapshot, even though its callers do not use it.
Signed-off-by: Emilio G. Cota <cota@braap.org>
Message-Id: <20181010144853.13005-5-cota@braap.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
We forgot to initialize n in commit 15fa08f845 ("tcg: Dynamically
allocate TCGOps", 2017-12-29).
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Signed-off-by: Emilio G. Cota <cota@braap.org>
Message-Id: <20181010144853.13005-3-cota@braap.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Also, assert that we don't overflow any of two different offsets into
the TB. Both unwind and goto_tb both record a uint16_t for later use.
This fixes an arm-softmmu test case utilizing NEON in which there is
a TB generated that runs to 7800 opcodes, and compiles to 96k on an
x86_64 host. This overflows the 16-bit offset in which we record the
goto_tb reset offset. Because of that overflow, we install a jump
destination that goes to neverland. Boom.
With this reduced op count, the same TB compiles to about 48k for
aarch64, ppc64le, and x86_64 hosts, and neither assertion fires.
Cc: qemu-stable@nongnu.org
Reported-by: "Jason A. Donenfeld" <Jason@zx2c4.com>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Thereby making it per-TCGContext. Once we remove tb_lock, this will
avoid an atomic increment every time a TB is invalidated.
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
Signed-off-by: Emilio G. Cota <cota@braap.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
This paves the way for enabling scalable parallel generation of TCG code.
Instead of tracking TBs with a single binary search tree (BST), use a
BST for each TCG region, protecting it with a lock. This is as scalable
as it gets, since each TCG thread operates on a separate region.
The core of this change is the introduction of struct tcg_region_tree,
which contains a pointer to a GTree and an associated lock to serialize
accesses to it. We then allocate an array of tcg_region_tree's, adding
the appropriate padding to avoid false sharing based on
qemu_dcache_linesize.
Given a tc_ptr, we first find the corresponding region_tree. This
is done by special-casing the first and last regions first, since they
might be of size != region.size; otherwise we just divide the offset
by region.stride. I was worried about this division (several dozen
cycles of latency), but profiling shows that this is not a fast path.
Note that region.stride is not required to be a power of two; it
is only required to be a multiple of the host's page size.
Note that with this design we can also provide consistent snapshots
about all region trees at once; for instance, tcg_tb_foreach
acquires/releases all region_tree locks before/after iterating over them.
For this reason we now drop tb_lock in dump_exec_info().
As an alternative I considered implementing a concurrent BST, but this
can be tricky to get right, offers no consistent snapshots of the BST,
and performance and scalability-wise I don't think it could ever beat
having separate GTrees, given that our workload is insert-mostly (all
concurrent BST designs I've seen focus, understandably, on making
lookups fast, which comes at the expense of convoluted, non-wait-free
insertions/removals).
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
Signed-off-by: Emilio G. Cota <cota@braap.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
In 6001f7729e we partially attempt to address the branch
displacement overflow caused by 15fa08f845.
However, gcc/testsuite/gcc.target/aarch64/advsimd-intrinsics/vqtbX.c
is a testcase that contains a TB so large as to overflow anyway.
The limit here of 8000 ops produces a maximum output TB size of
24112 bytes on a ppc64le host with that test case. This is still
much less than the maximum forward branch distance of 32764 bytes.
Cc: qemu-stable@nongnu.org
Fixes: 15fa08f845 ("tcg: Dynamically allocate TCGOps")
Reviewed-by: Laurent Vivier <laurent@vivier.eu>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
ppc64 uses a BC instruction to call the tcg_out_qemu_ld/st
slow path. BC instruction uses a relative address encoded
on 14 bits.
The slow path functions are added at the end of the generated
instructions buffer, in the reverse order of the callers.
So more we have slow path functions more the distance between
the caller (BC) and the function increases.
This patch changes the behavior to generate the functions in
the same order of the callers.
Cc: qemu-stable@nongnu.org
Fixes: 15fa08f845 ("tcg: Dynamically allocate TCGOps")
Signed-off-by: Laurent Vivier <lvivier@redhat.com>
Message-Id: <20180429235840.16659-1-lvivier@redhat.com>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
