bdc5faad93
using if-zero in even more places.
2012-07-16 19:38:49 -07:00
1a72249983
math: using if-zero in more places.
2012-06-18 14:32:39 -07:00
d37851fb5d
cpu.x86.32.bootstrap: oops, meant bootstrap-cell
2011-12-19 20:55:07 -08:00
d533bbb88c
cpu.x86.32.bootstrap: update unwind-native-frames
...
The arguments passed from C are in a different position now.
2011-12-19 18:42:14 -08:00
d6b520734d
cpu.x86.64: define jit-load-vm as a no-op
...
This was the signal handler prolog code can be shared completely between x86.32 and x86.64.
2011-12-14 11:33:05 -08:00
9078282855
oops, handler still need to jit-load-vm on x86-32
2011-12-14 09:56:50 -08:00
af989f533a
cpu.x86.*: clean up signal handler prolog code
...
Also bump the signal handler frame sizes up again since it looks like we shrank them too much.
2011-12-14 09:56:50 -08:00
6dd5973bdd
cpu.x86: reduce compiled frames by 2 cells too
2011-12-14 09:56:49 -08:00
1c44a4e8ae
simplify function prologs
2011-12-14 09:56:49 -08:00
5f0102fb7c
cpu.*.bootstrap: return address for (set-context)
...
Instead of pushing zero as the fake return address for a set-context call, push an address that points back into the primitive so that callstack walking can match the frame back to the subprimitive.
2011-12-14 09:56:46 -08:00
8981fccbe8
vm: frame size for signal handler code blocks
2011-12-14 09:56:46 -08:00
9186a1bdf3
vm: store stack frame size in code blocks
...
Change modify-code-heap primitive so it takes a sixth element in each array for the frame size.
2011-12-14 09:56:45 -08:00
819e1602be
cpu.x86: enable POPCNT with a flag
...
If bootstrapped with '-enable-popcnt' and the host CPU supports it, enable the POPCNT-based fixnum-bit-count intrinsic. Fixes #436 .
2011-12-12 16:43:49 -08:00
eee483654a
use radix literals
2011-11-23 19:03:40 -08:00
62e665bb88
compiler: emit safepoints in optimized prologs
2011-11-23 11:11:25 -08:00
ce205e1e3a
code beautification
2011-11-13 16:10:26 -08:00
2c4c04f973
compiler: add intrinsic for PMOVMSKB/MOVMSKP[SD]
...
Combined with a fast bit-count this will let us rice byte-counting.
2011-11-13 16:10:20 -08:00
4984bde393
cpu.*.bootstrap: clear faulting flag directly
...
No reason to go through a safepoint in unwind-native-frames really.
2011-11-12 13:02:57 -08:00
beb0f48da9
vm: more defense against multi-faulting
...
* Clear faulting_p from a safepoint rather than inside general_error, because jumping into unwind-native-frames could blow up.
* Handle multiple faults from fatal_error by breakpointing. Is there anything else we can safely do at that point?
* Verify memory protection faults in the top half of the signal handlers because signal dispatch could fault. Treat memory faults during gc or fep as fatal errors.
* Add a function factor_vm::abort() that restores the default SIGABRT handler and ::abort()s. Use it from fatal_error() so we get useful context from gdb and so the user gets feedback from the system crash reporter that Factor blew up and didn't just disappear.
* In factorbug(), don't proceed with .s .r .c if it would be unsafe to do so.
* Don't pile on signals if we've already called fatal_error().
2011-11-12 13:02:57 -08:00
5b43301eb6
vm: strip out call-counting profiler
...
This makes the separate "code" and "entry_point" fields in word and quotation redundant, so also remove them to reclaim an additional cell per word and quotation object, which should help with #318 .
2011-11-10 16:01:07 -08:00
c78b4d7f60
cpu.x86.*.bootstrap: save flags in signal-handlers
2011-11-04 11:56:01 -07:00
db1a47ffc1
cpu.x86.bootstrap: cell -> bootstrap-cell
2011-11-04 11:54:40 -07:00
a944e41e79
cpu.x86.assembler: PUSHF, POPF instructions
2011-11-04 11:23:11 -07:00
d4f24f44c3
cpu.x86.64.bootstrap: save all the regs on signal
...
Appears to fix another sporadic crash from mason on x86-64. There's no reason to cut corners here, really.
2011-11-04 10:01:56 -07:00
a28192859a
cpu.x86.bootstrap: don't safepoint in a safepoint
...
Safepointing in the signal-handler primitive causes a crash if the safepoint is guarded and the signal being handled was raised by a safepoint. Put our own RET in the subprimitive to bail out before the standard epilog gets written out. This tentatively appears to fix #349 .
2011-11-03 20:19:38 -07:00
1155da01c9
cpu.x86.32.bootstrap: typo
2011-11-03 13:07:13 -07:00
7ac94bd09f
vm: always generate safepoints in jit
...
Even if there's no stack frame we still need to safepoint before leaving the function. Fixes #332 .
2011-11-03 11:57:43 -07:00
75cba3106a
remove BSD,solaris,etc. misc platform support code
2011-11-02 17:13:51 -07:00
a3979cd123
fix platform dependency in leaf-signal-handler
2011-11-02 13:23:12 -07:00
0ee07f1f2a
cpu.x86.bootstrap: simplify leaf-signal-handler
...
We can use RET (stack-frame-size - cell) instead of doing stack gymnastics.
2011-11-02 13:23:08 -07:00
be12f2b27d
Make "foo.private" require load foo instead.
...
Move require from vocabs.loader to vocabs. Update everything.
Fixes #201 .
2011-11-02 11:50:03 -07:00
09088ec3ce
cleanups from code review
2011-10-27 21:18:23 -07:00
871ea10cb0
cpu.x86.64: also must save ctx-reg
2011-10-27 21:18:21 -07:00
66ffd0f138
cpu.x86.32: update for resumable signal handlers
2011-10-27 21:18:19 -07:00
9c7731e4c9
vm: suppress signal-handler stack frame generation
...
This is a hack; see #295
2011-10-27 21:18:19 -07:00
1eddaf34f1
vm: resumably handle signals from leaf procedures
2011-10-27 21:18:18 -07:00
cb6f175ef8
vm: dispatch signal handlers through subprimitive
...
We also need to save C ABI volatile registers before calling the signal handler in order to be able to reliably resume. Add signal-handler and leaf-signal-handler subprimitives to preserve volatile registers before invoking the signal handler C function.
2011-10-27 21:18:18 -07:00
d41c3f2709
cpu.x86.(32,64).bootstrap: get rid of PUSH malarky
...
In the function prologue, move the stack pointer once after the stack frame is set up instead of PUSHing the stack frame piece by piece.
2011-10-27 21:14:50 -07:00
cf45e0d581
compiler: add ##safepoint IR insn
2011-10-27 21:14:46 -07:00
6e237abb5e
cpu.x86.assembler: add MOVABS instruction
...
MOVABS generates the one-byte opcode version of MOV eAX, [absolute] or MOV [absolute], eAX, which saves a byte per safepoint on x86-32.
2011-10-27 21:14:45 -07:00
09d0fdaf8d
cpu.x86.bootstrap: add safepoint to jit-epilog
2011-10-27 21:14:45 -07:00
cb41cc99d8
Fix 32bit platform bootstrap
2011-10-24 16:24:01 -07:00
14a2511d57
Found a few more places to use lookup-c-type
2011-10-24 15:58:40 -07:00
7bfbe17861
Rename c-type to lookup-c-type. Fixes #230 .
2011-10-24 14:37:24 -07:00
b1b7b24ad5
Add RT_INLINE_CACHE_MISS relocation type to avoid frequent dlsym lookups when compiling PICs. Fixes #123
2011-10-03 00:26:21 -07:00
8336d898a3
Rework non-optimizing compiler backend to use compiler.codegen.relocation
2011-10-03 00:26:21 -07:00
83f602c1da
Split up compiler.codegen.fixup into compiler.codegen.{gc-maps,labels,relocation}
2011-10-03 00:25:47 -07:00
94f3ca3b8c
Remove all non-core uses of (scan-token)
...
Add scan-datum
Add scan-number
Document more words
Fixes #225 .
2011-10-02 12:00:08 -07:00
17e0fbe501
Remove Windows CE from core/ basis/ and build-support/
...
Rename the winnt singleton to windows in core/ basis/ extra/
Rename boot images winnt -> windows
Fixes #84 .
2011-09-18 23:19:06 -05:00
238fea7b7e
cpu.ppc: add 'not loaded' tags so that load-all doesn't break an x86 image
2011-08-24 23:52:02 -07:00
John Benediktsson
Joe Groff
Doug Coleman
Slava Pestov