* libguile/vm-engine.c (allocate-struct, struct-ref, struct-set!): New
instructions, to complement their "immediate" variants.
* module/language/cps/compile-bytecode.scm (compile-fun):
* module/system/vm/assembler.scm (system): Wire up the new instructions.
Allocating an instance of a class with a #:class or #:each-subclass slot
allocation should not re-initialize the class-allocated slot. In Guile
1.8, this worked by effectively doing a slot-bound? within
%initialize-object. In Guile 2.0 we instead initialize the slot when it
is allocated -- in compute-get-n-set.
* module/oop/goops.scm (compute-getters-n-setters): Don't set an
init-thunk for class-allocated slots.
(compute-get-n-set): Initialize class-allocated slots here, if an
init-thunk or init-value are present.
* test-suite/tests/goops.test ("#:each-subclass"): Add test.
* doc/body.texi: Reorder documentation, making jit_frame
and jit_tramp the lightning response to the need of
trampolines, continuations and tail call optimizations.
A pseudo code example of a factorial function was added.
Also added a section for description of the available
predicates.
* doc/fact.c: New file, implementing a simple example of
a translation of a trivial, recursive, tail call optimization
into lightning calls. This is the conversion to functional C
code of the example in doc/body.texi.
* doc/Makefile.am: Update for the next test case.
Fixes <http://bugs.gnu.org/19621>.
Before that, in 'cbip_fill_input', BUFFERED would be set to 0 when
reading from 'scm_getc' et al, because 'shortbuf' was being used. Thus,
we could eventually execute this line:
/* Copy the data back to the internal buffer. */
memcpy ((char *) c_port->read_pos, SCM_BYTEVECTOR_CONTENTS (bv),
c_octets);
But 'read_pos' would quickly point to the fields beyond 'shortbuf',
thereby leading to a corruption of the 'scm_t_port' itself.
* libguile/r6rs-ports.c (cbip_setvbuf): When READ_SIZE is 0, keep using
BV as the 'read_buf'.
(cbip_fill_input): Adjust assertion to accept 'read_buf_size = 1'.
* test-suite/tests/r6rs-ports.test ("7.2.7 Input Ports")["custom binary
input port unbuffered & 'get-string-all'", "custom binary input port
unbuffered UTF-8 & 'get-string-all'"]: New tests.
* include/lightning.h, lib/jit_aarch64.c,
lib/jit_alpha.c, lib/jit_arm-vfp.c, lib/jit_arm.c,
lib/jit_hppa.c, lib/jit_ia64.c, lib/jit_mips.c,
lib/jit_ppc.c, lib/jit_s390.c, lib/jit_sparc.c,
lib/jit_x86.c: Add the new jit_arg_register_p predicate.
The predicate is expected to be used to know if an
argument is in a register, what would need special
handling if code that can overwrite non callee save
registers is executed.
* check/carg.c: New test case to check consistency and
expected usage of jit_arg_register_p.
* check/Makefile.am: Update for new test case.
* include/lightning/jit_aarch64.h,
include/lightning/jit_alpha.h,
include/lightning/jit_arm.h,
include/lightning/jit_hppa.h,
include/lightning/jit_mips.h,
include/lightning/jit_ppc.h,
include/lightning/jit_s390.h,
include/lightning/jit_sparc.h,
include/lightning/jit_x86.h,
lib/jit_aarch64.c, lib/jit_alpha.c,
lib/jit_arm.c, lib/jit_hppa.c,
lib/jit_ia64.c, lib/jit_mips.c,
lib/jit_ppc.c, lib/jit_s390.c,
lib/jit_sparc.c, lib/jit_x86.c: Remove jit_arg_reg_p and
jit_arg_f_reg_p from a public header, and define it only
on port specific files where an integer offset is used
to qualify an argument identifier. Exported code expects
an opaque pointer (but of jit_node_t* type) to "qualify"
an argument identifier.
This patch, and the code review/simplification done during
it also corrected some bugs:
o Inconsistent jit_arg_d value of double argument after 3
integer arguments in arm for jit_functions; tested, C
functions were being properly called.
o Inconsistent use of getarg_{f,d} and putarg*_{f,d} on
s390 (32-bit) that happened to not have a proper test
case, as it would only happen for jit functions, and
tested, called C functions had proper arguments.
o Corrected a "last minute" correction that did not go
to the committed version, and would not compile on hppa,
due to bad _jit_putargi_d prototype definition.
* include/lightning.h, lib/jit_aarch64.c,
lib/jit_alpha.c, lib/jit_arm.c, lib/jit_hppa.c,
lib/jit_ia64.c, lib/jit_mips.c, lib/jit_ppc.c,
lib/jit_s390.c, lib/jit_sparc.c, lib/jit_x86.c:
Implement jit_putarg*. It works as a mix of jit_getarg*
and jit_pusharg*, in the way that the first argument is
a register or immediate, and the second is a pointer
returned by jit_arg*. The use of the interface is to change
values of arguments to the current jit function.
* check/put.ok, check/put.tst: New test cases exercising
the new jit_putarg* interface.
* check/Makefile.am, check/lightning.c: Update for the
new test case and interface.
* module/ice-9/command-line.scm (shell-usage): Add 'newline' call when
FMT is true.
(compile-shell-switches)[error]: Prepend "error: " and append "~%" to
FMT.
Use a lower-case message.
This yields a 50% improvement on the "narrow string" benchmark of
'hash.bm', 41% on "wide string", and 76% on "long string".
* libguile/hash.c (scm_i_string_hash): Rewrite to avoid
'scm_i_string_ref' calls.
This function has been unused internally for some time and is undocumented.
* libguile/hash.c (scm_string_hash): Wrap if #if SCM_ENABLE_DEPRECATED
== 1.
* libguile/hash.h (scm_string_hash): Likewise, and replace SCM_API with
SCM_DEPRECATED.
This shows a 19% improvement on the "string without escapes"
micro-benchmark of 'write.bm', and 12% on "string with escapes".
* libguile/print.c (iprin1) <scm_tc7_string>: Replace 'scm_i_string_ref'
loop with a call to 'write_string'.
(display_character): Adjust description of return value in comment.
(write_string): New function.
* include/lightning.h, include/lightning/jit_private.h,
include/lightning/jit_s390x.h, lib/jit_disasm.c,
lib/jit_s390x-cpu.c, lib/jit_s390x-fpu.c, lib/jit_s390x-sz.c,
lib/jit_s390x.c, lib/jit_size.c, lib/lightning.c:
Add support for generating jit for s390 32 bit. This change
also removed %f15 from the list of temporaries fpr registers;
it was not being used, but if were, it would corrupt the
stack frame because the spill address would overwrite grp
offsets.
* include/lightning.h: Split jit_htonr in the new 3 interfaces
jit_htonr_us, jit_htonr_ui and jit_htonr_ul, the later only
available on 64 bit. The plain/untyped jit_htonr macro call
maps to the wordsize one.
* lib/jit_aarch64-cpu.c, lib/jit_aarch64-sz.c, lib/jit_aarch64.c,
lib/jit_alpha-cpu.c, lib/jit_alpha-sz.c, lib/jit_alpha.c,
lib/jit_arm-cpu.c, lib/jit_arm-sz.c, lib/jit_arm.c,
lib/jit_hppa-cpu.c, lib/jit_hppa-sz.c, lib/jit_hppa.c,
lib/jit_ia64-cpu.c, lib/jit_ia64-sz.c, lib/jit_ia64.c,
lib/jit_mips-cpu.c, lib/jit_mips-sz.c, lib/jit_mips.c,
lib/jit_ppc-cpu.c, lib/jit_ppc-sz.c, lib/jit_ppc.c,
lib/jit_s390x-cpu.c, lib/jit_s390x-sz.c, lib/jit_s390x.c,
lib/jit_sparc-cpu.c, lib/jit_sparc-sz.c, lib/jit_sparc.c,
lib/jit_x86-cpu.c, lib/jit_x86-sz.c, lib/jit_x86.c:
Update backends for the new jit_htonr*.
* check/lightning.c, lib/jit_names.c, lib/lightning.c:
Update for the new jit_htonr* interfaces.
* check/Makefile.am: Update for new test cases.
* check/hton.ok, check/hton.tst: New test cases.
* include/lightning/jit_private.h, include/lightning/jit_x86.h,
lib/jit_disasm.c, lib/jit_x86-cpu.c, lib/jit_x86-sse.c,
lib/jit_x86-sz.c, lib/jit_x86-x87.c, lib/jit_x86.c,
size.c: Implement support for the x32 abi. Built and
tested on Gentoo default/linux/amd64/13.0/x32 profile.
* lib/jit_arm.c: Call __clear_cache for every page.
This should only be required for older boards or
toolchain setup, but has been reported to be required
for lightning at some point.
Fixes <http://bugs.gnu.org/19354>.
Reported by Linas Vepstas <linasvepstas@gmail.com>.
* module/system/vm/assembler.scm (write-sources): Intern the filename
only if it's a string. (For sockets, the filename is a symbol).
* libguile/memoize.c (memoize): Fix meta on subsequent case-lambda
clauses.
* module/ice-9/eval.scm (primitive-eval): Rewrite to compile expressions
to thunks, to avoid runtime dispatch cost.
* libguile/expand.c (convert_assignment): Handle creation of the default
lambda-case body here.
* libguile/eval.c (eval):
* module/ice-9/eval.scm (primitive-eval):
* libguile/memoize.h:
* libguile/memoize.c (MAKMEMO_BOX_REF, MAKMEMO_BOX_SET):
(MAKMEMO_TOP_BOX, MAKMEMO_MOD_BOX): Refactor all global var resolution
to go through "resolve". Add "box-ref" and "box-set!". Rename
memoize-variable-access! to %resolve-variable, and don't be
destructive.
* libguile/memoize.c (FULL_ARITY): Serialize "ninits" and the unbound
value instead of the init list.
(memoize): Adapt to FULL_ARITY changes. Remove LETREC case.
(unmemoize): Adapt to memoized code change.
* libguile/eval.c (BOOT_CLOSURE_PARSE_FULL): Adapt to parse ninits and
unbound instead of inits.
(eval): Lexical-ref can no longer raise an error.
(prepare_boot_closure_env_for_apply): Adapt to inits change.
* module/ice-9/eval.scm (primitive-eval): Adapt to ninits/unbound
change.
* libguile/expand.c (expand_named_let): Fix lambda-case creation to make
lists for opt and inits.
This is a followup to 5fac1a7.
* meta/Makefile.am (substitute): Add GUILE_* and LIBGUILE_* substitutions,
and $(dependency_substitutions).
(dependency_substitutions): New variable.
Reported in <http://bugs.gnu.org/18520>
by David Kastrup <dak@gnu.org>.
* doc/ref/api-io.texi (Random Access): Clarify the unit of the 'offset'
argument to 'seek'.
* configure.ac: Remove meta/guile-2.0.pc and
meta/guile-2.0-uninstalled.pc from 'AC_CONFIG_FILES'.
* meta/Makefile.am (substitute): New variable.
(guile-2.0.pc, guile-2.0-uninstalled.pc): New targets.
(guile-config, guild): Use $(substitute) instead of duplicated sed
script.
(CLEANFILES): Add the .pc files.
* meta/guile-2.0.pc.in (bindir, guild, guile): New variables.
* doc/ref/libguile-parallel.texi (Parallel Installations): Document the
'guild' and 'guile' pkg-config variables.
