This fixes a problem that occurs when compiled by gcc-10, as the code
is relying on undefined overflow behavior. This is fixed by replacing
compares between 32-bit and 64-bit results with compares that just use
the 64-bit results with a cast.
PR sim/27483
* simulator.c (set_flags_for_add32): Compare uresult against
itself. Compare sresult against itself.
All the scripts were using this implicitly already, so there's no real
change for them, but we want to call it explicitly as the CPP tool is
used to generate nltvals.def.
This file is quite large and is getting unmanageable. Split it apart
to follow aclocal best practices by putting one-macro-per-file. There
shouldn't be any real functional changes here as can be seen in the
configure script regens.
Rather than hand maintain m4 includes in various autotool files,
use AC_CONFIG_MACRO_DIRS to declare the relevant search paths.
This simplifies the code, makes it more robust, and cleans out
unused logic from configure.
These settings might have made sense in darker compiler times, but I
think they're largely obsolete now. Looking through the values that
get used in HDEFINES, it's quite limited, and configure itself should
handle them. If we still need something, we can leverage standard
autoconf macros instead, after we get a clear user report.
TDEFINES was never set anywhere and was always empty, so prune that.
Since we require C11 now, we can assume many headers exist, and
clean up all of the conditional includes. It's not like any of
this code actually accounted for the headers not existing, just
whether we could include them.
The strings.h cleanup is a little nuanced: it isn't in C11, but
every use of it in the codebase will include strings.h only if
string.h doesn't exist. Since we now assume the C11 string.h
exists, we'll never include strings.h, so we can delete it.
We've had this off for a long time because the sim code was way too
full of warnings for it to be feasible. However, I've cleaned things
up significantly from when this was first merged, and we can start to
turn this around.
Change the macro to enable -Werror by default, and allow ports to opt
out. New ports will get it automatically (and we can push back on
them if they try to turn it off).
Also turn it off for the few ports that still hit warnings for me.
All the rest will get the new default, and we'll wait for feedback
if/when new issues come up.
With GDB requiring a C++11 compiler now, this hopefully shouldn't
be a big deal. It's been 10 years since C11 came out, so should
be plenty of time to upgrade.
This will allow us to start cleaning up random header logic and
many of our non-standard custom types.
A comment in the implementation of blr says:
/* The pseudo code in the spec says we update LR before fetching.
the value from the rn. */
With 'rn' being the register holding the destination address.
This may have been true at one point, but the ISA manual now clearly
shows the destination register being read before the link register is
written.
This commit updates the implementation of blr to match.
sim/aarch64/ChangeLog:
PR sim/25318
* simulator.c (blr): Read destination register before calling
aarch64_save_LR.
Change-Id: Icb1c556064e3d9c807ac28440475caa205ab1064
This commit:
commit ef9866970c
Date: Thu Mar 28 06:40:30 2019 +0900
sim/common: convert sim-arange to use sim-inline
Broke the simulator build for aarch64 - some required macros are no
longer included where needed, fixed in this commit.
sim/aarch64/ChangeLog:
* cpustate.c: Add 'libiberty.h' include.
* interp.c: Add 'sim-assert.h' include.
sim/aarch64/
* simulator.c (vec_load): Add M argument. Rewrite to iterate over
registers based on structure size.
(LD4, LD3, LD2, LD1_2, LD1_3, LD1_4): Pass new arg to vec_load.
(LD1_1): Replace with call to vec_load.
(vec_store): Add new M argument. Rewrite to iterate over registers
based on structure size.
(ST4, ST3, ST2, ST1_2, ST1_3, ST1_4): Pass new arg to vec_store.
(ST1_1): Replace with call to vec_store.
sim/testsuite/sim/aarch64/
* fcvtz.s, fstur.s, ldn_single.s, ldnr.s, mla.s, mls.s, uzp.s: Align
data.
* sumulh.s: Delete unnecessary data alignment.
* stn_single.s: Align data. Fix unaligned ldr insns. Adjust cmp
arguments to match change.
* ldn_multiple.s, stn_multiple.s: New.
sim/aarch64/
* simulator.c (set_flags_for_add32): Cast result to uint32_t in carry
flag check.
sim/testsuite/sim/aarch64/
* adds.s: Add checks for values -2 and 1, where C is not set.
sim/aarch64/
* simulator.c (mul64hi): Shift carry left by 32.
(smulh): Change signum to negate. If negate, invert result, and add
carry bit if low part of multiply result is zero.
sim/testsuite/sim/aarch64/
* sumov.s: Correct compare test values.
* sumulh.s: New.
sim/aarch64/
* simulator.c (do_vec_ADDV): Mov val declaration inside each case,
with type set to input type size.
(do_vec_xtl): Change bias from 3 to 4 for byte case.
sim/testsuite/sim/aarch64/
* bit.s: Change cmp immediates to account for addv bug fix.
* cmtst.s, ldn_single.s, stn_single.s: Likewise.
* xtl.s: New.
sim/aarch64/
* simulator.c (do_vec_bit): Change loop limits from 16 and 8 to 4 and
2. Move test_false if inside loop. Fix logic for computing result
stored to vd.
sim/testsuite/sim/aarch64
* bit.s: New.
sim/aarch64/
* simulator.c: (LDn_STn_SINGLE_LANE_AND_SIZE): New.
(do_vec_LDn_single, do_vec_STn_single): New.
(do_vec_LDnR): Add and set new nregs var. Replace switch on nregs with
loop over nregs using new var n. Add n times size to address in loop.
Add n to vd in loop.
(do_vec_load_store): Add comment for instruction bit 24. New var
single to hold instruction bit 24. Add new code to use single. Move
ldnr support inside single if statements. Fix ldnr register counts
inside post if statement. Change HALT_NYI calls to HALT_UNALLOC.
sim/testsuite/sim/aarch64/
* ldn_single.s: New.
* ldnr.s: New.
* stn_single.s: New.
sim/aarch64/
* simulator.c (do_vec_ADDV): Call aarch64_set_vec_u64 instead of
aarch64_set_reg_u64. In case 2, call HALT_UNALLOC if not full. In
case 3, call HALT_UNALLOC unconditionally.
(do_vec_XTN): Delete shifts. In case 2, change index from i + 4 to
i + 2. Delete if on bias, change index to i + bias * X.
sim/testsuite/sim/aarch64/
* addv.s: New.
* xtn.s: New.
sim/aarch64/
* cpustate.c: Include math.h.
(aarch64_set_FP_float): Use signbit to check for signed zero.
(aarch64_set_FP_double): Likewise.
* simulator.c (do_vec_MOV_immediate, case 0x8): Add missing break.
(do_vec_mul): In all DO_VEC_WIDENING_MUL calls, make second and fourth
args same size as third arg.
(fmaxnm): Use isnan instead of fpclassify.
(fminnm, dmaxnm, dminnm): Likewise.
(do_vec_MLS): Reverse order of subtraction operands.
(dexSimpleFPCondSelect): Call aarch64_get_FP_double or
aarch64_get_FP_float to get source register contents.
(UINT_MIN, ULONG_MIN, FLOAT_UINT_MAX, FLOAT_UINT_MIN,
DOUBLE_UINT_MAX, DOUBLE_UINT_MIN, FLOAT_ULONG_MAX, FLOAT_ULONG_MIN,
DOUBLE_ULONG_MAX, DOUBLE_ULONG_MIN): New.
(do_fcvtzu): Use ULONG instead of LONG, and UINT instead of INT in
raise_exception calls.
sim/testsuite/sim/aarch64/
* fcsel.s: New.
* fcvtz.s: New.
* fminnm.s: New.
* mls.s: New.
* mul.s: New.
sim/aarch64
* simulator.c (NEG, POS): Move before set_flags_for_add64.
(set_flags_for_add64): Replace with a modified copy of
set_flags_for_sub64.
sim/testsuite/sim/aarch64
* testutils.inc (pass): Move .Lpass to start.
(fail): Move .Lfail to start. Return 1 instead of 0.
(start): Moved .Lpass and .Lfail to here.
* adds.s: New.
* fstur.s: New.
* tbnz.s: New.
The common sim tracing code already handles loading and tracking of
symbols from the target program so that it can show symbol info in
trace/disassembly calls. Once we touch up the trace code and add a
few API callbacks, ports don't need to do loading and searching of
symbol tables themselves anymore.
* simulator.c (aarch64_step): Revert pervious delta.
(aarch64_run): Call sim_events_tick after each
instruction is simulated, and if necessary call
sim_events_process.
* simulator.h: Revert previous delta.
* interp.c (sim_create_inferior): Allow for being called with a
NULL abfd parameter. If a bfd is provided, initialise the sim
with that start address.
* simulator.c (HALT_NYI): Just print out the numeric value of the
instruction when not tracing.
(aarch64_step): Change from static to global.
* simulator.h: Add a prototype for aarch64_step().
The major reason this header was needed, bfd_default_set_arch_mach,
has now moved to bfd.h.
gdb/
* amd64-darwin-tdep.c: Don't include libbfd.h.
* i386-darwin-tdep.c: Likewise.
* rs6000-nat.c: Likewise.
* rs6000-tdep.c: Likewise.
sim/aarch64/
* memory.c: Don't include libbfd.h.
sim/rl78/
* load.c: Don't include libbfd.h.
(rl78_load): Don't use private iovec seek or read.
sim/rx/
* load.c: Don't include libbfd.h.
(rx_load): Don't use private iovec seek or read.
* cpustate.h: Include config.h.
(union GRegisterValue): Add WORDS_BIGENDIAN check. For big endian code
use anonymous structs to align members.
* simulator.c (aarch64_step): Use sim_core_read_buffer and
endian_le2h_4 to read instruction from pc.
* cpustate.c (aarch64_set_reg_s32): New function.
(aarch64_set_reg_u32): New function.
(aarch64_get_FP_half): Place half precision value into the correct
slot of the union.
(aarch64_set_FP_half): Likewise.
* cpustate.h: Add prototypes for aarch64_set_reg_s32 and
aarch64_set_reg_u32.
* memory.c (FETCH_FUNC): Cast the read value to the access type
before converting it to the return type. Rename to FETCH_FUNC64.
(FETCH_FUNC32): New macro. Duplicates FETCH_FUNC64 but for 32-bit
accesses. Use for 32-bit memory access functions.
* simulator.c (ldrsb_wb): Use sign extension not zero extension.
(ldrb_scale_ext, ldrsh32_abs, ldrsh32_wb): Likewise.
(ldrsh32_scale_ext, ldrsh_abs, ldrsh64_wb): Likewise.
(ldrsh_scale_ext, ldrsw_abs): Likewise.
(ldrh32_abs): Store 32 bit value not 64-bits.
(ldrh32_wb, ldrh32_scale_ext): Likewise.
(do_vec_MOV_immediate): Fix computation of val.
(do_vec_MVNI): Likewise.
(DO_VEC_WIDENING_MUL): New macro.
(do_vec_mull): Use new macro.
(do_vec_mul): Use new macro.
(do_vec_MLA): Read values before writing.
(do_vec_xtl): Likewise.
(do_vec_SSHL): Select correct shift value.
(do_vec_USHL): Likewise.
(do_scalar_UCVTF): New function.
(do_scalar_vec): Call new function.
(store_pair_u64): Treat reads of SP as reads of XZR.
* cpustate.c: Remove space after asterisk in function parameters.
* decode.h (greg): Delete unused function.
(vreg, shift, extension, scaling, writeback, condcode): Likewise.
* simulator.c: Use INSTR macro in more places.
(HALT_NYI): Use sim_io_eprintf in place of fprintf.
Remove extraneous whitespace.
* cpustate.c (aarch64_get_FP_half): New function. Read a vector
register as a half precision floating point number.
(aarch64_set_FP_half): New function. Similar, but for setting
a half precision register.
(aarch64_get_thread_id): New function. Returns the value of the
CPU's TPIDR register.
(aarch64_get_FPCR): New function. Returns the value of the CPU's
floating point control register.
(aarch64_set_FPCR): New function. Set the value of the CPU's FPCR
register.
* cpustate.h: Add prototypes for new functions.
* sim-main.h (struct _sim_cpu): Add FPCR and tpidr fields.
* memory.c: Use unaligned core access functions for all memory
reads and writes.
* simulator.c (HALT_NYI): Generate an error message if tracing
will not tell the user why the simulator is halting.
(HALT_UNREACHABLE): Delete. Delete (unneeded) uses of the macro.
(INSTR): New time-saver macro.
(fldrb_abs): New function. Loads an 8-bit value using a scaled
offset.
(fldrh_abs): New function. Likewise for 16-bit values.
(do_vec_SSHL): Allow for negative shift values.
(do_vec_USHL): Likewise.
(do_vec_SHL): Correct computation of shift amount.
(do_vec_SSHR_USHR): Correct decision of signed vs unsigned
shifts and computation of shift value.
(clz): New function. Counts leading zero bits.
(do_vec_CLZ): New function. Implements CLZ (vector).
(do_vec_MOV_element): Call do_vec_CLZ.
(dexSimpleFPCondCompare): Implement.
(do_FCVT_half_to_single): New function. Implements one of the
FCVT operations.
(do_FCVT_half_to_double): New function. Likewise.
(do_FCVT_single_to_half): New function. Likewise.
(do_FCVT_double_to_half): New function. Likewise.
(dexSimpleFPDataProc1Source): Call new FCVT functions.
(do_scalar_SHL): Handle negative shifts.
(do_scalar_shift): Handle SSHR.
(do_scalar_USHL): New function.
(do_double_add): Simplify to just performing a double precision
add operation. Move remaining code into...
(do_scalar_vec): ... New function.
(dexLoadUnsignedImmediate): Call new fldrb_abs and fldrh_abs
functions.
(system_get): Add support for TPIDR, CTR, FPCR, FPSR and CPSR
registers.
(system_set): New function.
(do_MSR_immediate): New function. Stub for now.
(do_MSR_reg): New function. Likewise. Partially implements MSR
instruction.
(do_SYS): New function. Stub for now,
(dexSystem): Call new functions.
* simulator.c (set_flags_for_sub32): Correct type of signbit.
(CondCompare): Swap interpretation of bit 30.
(DO_ADDP): Delete macro.
(do_vec_ADDP): Copy source registers before starting to update
destination register.
(do_vec_FADDP): Likewise.
(do_vec_load_store): Fix computation of sizeof_operation.
(rbit64): Fix type of constant.
(aarch64_step): When displaying insn value, display all 32 bits.
Rather than stuffing the command line with a bunch of -D flags, start
moving things to config.h which is managed by autoheader. This makes
the makefile a bit simpler and the build output tighter, and it makes
the migration to automake easier as there are fewer vars to juggle.
We'll want to move the other options out too, but it'll take more work.
