| 11 |
Date: |
Date: |
| 12 |
Tag: <post-commit CVS tag> |
Tag: <post-commit CVS tag> |
| 13 |
Description: |
Description: |
| 14 |
|
|
| 15 |
|
---------------------------------------------------------------------- |
| 16 |
|
Name: Matthias Blume |
| 17 |
|
Date: 2001/03/22 15:10:00 EST |
| 18 |
|
Tag: blume-20010322-rawmem-parcm |
| 19 |
|
Description: |
| 20 |
|
|
| 21 |
|
1. All "raw memory access" primitives for the new FFI are implemented now |
| 22 |
|
(at least on the x86). |
| 23 |
|
2. Some further cleanup of CM's parallel make mechanism. |
| 24 |
|
|
| 25 |
|
---------------------------------------------------------------------- |
| 26 |
|
Name: Matthias Blume |
| 27 |
|
Date: 2001/03/19 17:53:00 EST |
| 28 |
|
Tag: blume-20010319-parallel |
| 29 |
|
Description: |
| 30 |
|
|
| 31 |
|
Parallel make (using compile servers) now works again. |
| 32 |
|
|
| 33 |
|
To this end, CM.stabilize and CMB.make have been modified to work in |
| 34 |
|
two passes when compile servers are attached: |
| 35 |
|
1. Compile everything, do not perform stabilization; this pass |
| 36 |
|
uses compile servers |
| 37 |
|
2. Stabilize everything; this pass does not use compile servers |
| 38 |
|
If there are no compile servers, the two passes are combined into one |
| 39 |
|
(as before). Splitting the passes increases the inherent parallelism |
| 40 |
|
in the dependency graph because the entire graph including all |
| 41 |
|
libraries is available at the same time. This, in turn, improves |
| 42 |
|
server utilization. The downside is that the master process will |
| 43 |
|
have to do some extra work after compilation is done (because for |
| 44 |
|
technical reasons it must re-read all the binfiles during stabilization). |
| 45 |
|
|
| 46 |
|
---------------------------------------------------------------------- |
| 47 |
|
Name: Matthias Blume |
| 48 |
|
Date: 2001/03/16 12:22:00 EST |
| 49 |
|
Tag: blume-20010316-bootfiles |
| 50 |
|
Description: |
| 51 |
|
|
| 52 |
|
Created a new set of bootfiles (for your automatic installation convenience). |
| 53 |
|
|
| 54 |
|
---------------------------------------------------------------------- |
| 55 |
|
Name: Matthias Blume |
| 56 |
|
Date: 2001/03/16 11:00:00 EST |
| 57 |
|
Tag: blume-20010316-MLTREE-fixup |
| 58 |
|
Description: |
| 59 |
|
|
| 60 |
|
This is a minor fixup for an (untagged) earlier commit by Allen. |
| 61 |
|
(A file was missing). |
| 62 |
|
|
| 63 |
|
---------------------------------------------------------------------- |
| 64 |
|
Name: Allen Leung |
| 65 |
|
Date: Mon Mar 5 18:54:57 EST 2001 |
| 66 |
|
Tag: leunga-20010305-cut-support |
| 67 |
|
|
| 68 |
|
1. New support for alternative control-flow in MLTREE. |
| 69 |
|
Currently we support |
| 70 |
|
|
| 71 |
|
FLOW_TO(CALL ...., [k1,...,kn]) |
| 72 |
|
|
| 73 |
|
This is needed for 'cuts to' in C-- and try/handle-like constructs |
| 74 |
|
in Moby |
| 75 |
|
|
| 76 |
|
New assembler flag "asm-show-cutsto" to turn on control-flow debugging. |
| 77 |
|
|
| 78 |
|
2. Register Allocator |
| 79 |
|
|
| 80 |
|
Changes in interface [from Fermin, John] |
| 81 |
|
|
| 82 |
|
3. Alpha 8-bit SLL support [Fermin] |
| 83 |
|
|
| 84 |
|
4. All architectures |
| 85 |
|
|
| 86 |
|
A new module (ClusterExpandCopies) for expanding parallel copies. |
| 87 |
|
|
| 88 |
|
---------------------------------------------------------------------- |
| 89 |
|
Name: Allen Leung |
| 90 |
|
Date: 2001/02/27 23:07:00 EST |
| 91 |
|
Tag: leunga-20010227-minor-stuff |
| 92 |
|
|
| 93 |
|
1. Alpha bug fix for CMOVNE |
| 94 |
|
2. Handle mltree COND(..,FCMP ...,...) |
| 95 |
|
3. Bug fix in simplifier |
| 96 |
|
|
| 97 |
|
---------------------------------------------------------------------- |
| 98 |
|
Name: Matthias Blume |
| 99 |
|
Date: 2001/01/30 17:50:00 EST |
| 100 |
|
Tag: blume-20010130-sync |
| 101 |
|
Description: |
| 102 |
|
|
| 103 |
|
This is just a minor update to sync my devel branch with the main brach. |
| 104 |
|
The only visible change is the addition of some README files. |
| 105 |
|
|
| 106 |
|
---------------------------------------------------------------------- |
| 107 |
|
Name: Matthias Blume |
| 108 |
|
Date: 2001/01/12 23:30:00 JST |
| 109 |
|
Tag: blume-20010112-bootfiles |
| 110 |
|
Description: |
| 111 |
|
|
| 112 |
|
Made a new set of bootfiles that goes with the current state of the |
| 113 |
|
repository. |
| 114 |
|
|
| 115 |
|
---------------------------------------------------------------------- |
| 116 |
|
Name: Matthias Blume |
| 117 |
|
Date: 2001/01/12 21:20:00 JST |
| 118 |
|
Tag: blume-20010112-sync |
| 119 |
|
Description: |
| 120 |
|
|
| 121 |
|
I am just flushing out some minor changes that had accumulated in |
| 122 |
|
my private branch in order to sync with the main tree. (This is |
| 123 |
|
mainly because I had CVS trouble when trying to merge _into_ my |
| 124 |
|
private branch.) |
| 125 |
|
|
| 126 |
|
Most people should be completely unaffected by this. |
| 127 |
|
|
| 128 |
|
---------------------------------------------------------------------- |
| 129 |
|
Name: Allen Leung |
| 130 |
|
Date: Thu Jan 11 21:03:00 EST 2001 |
| 131 |
|
Tag: leunga-20010111-labexp=mltree |
| 132 |
|
Description: |
| 133 |
|
|
| 134 |
|
1. Removed the type LabelExp and replace it by MLTree. |
| 135 |
|
2. Rewritten mltree-simplify with the pattern matcher tool. |
| 136 |
|
3. There were some bugs in alpha code generator which would break |
| 137 |
|
64-bit code generation. |
| 138 |
|
4. Redo the tools to generate code with the |
| 139 |
|
5. The CM files in MLRISC (and in src/system/smlnj/MLRISC) |
| 140 |
|
are now generated by perl scripts. |
| 141 |
|
|
| 142 |
|
---------------------------------------------------------------------- |
| 143 |
|
Name: Matthias Blume |
| 144 |
|
Date: 2001/01/10 21:55:00 JST |
| 145 |
|
Tag: blume-20010110-rcc |
| 146 |
|
Description: |
| 147 |
|
|
| 148 |
|
The RCC stuff now seems to work (but only on the x86). |
| 149 |
|
This required hacking of the c-calls interface (and -implementation) in |
| 150 |
|
MLRISC. |
| 151 |
|
|
| 152 |
|
Normal compiler users should be unaffected. |
| 153 |
|
|
| 154 |
|
---------------------------------------------------------------------- |
| 155 |
|
Name: Matthias Blume |
| 156 |
|
Date: 2001/01/09 01:20:00 JST |
| 157 |
|
Tag: blume-20010109-rcc |
| 158 |
|
Description: |
| 159 |
|
|
| 160 |
|
This is a fairly big patch, flushing out a large number of pending |
| 161 |
|
changes that I made to my development copy over the last couple of days. |
| 162 |
|
|
| 163 |
|
Of practical relevance at this moment is a workaround for a pickling |
| 164 |
|
bug that Allen ran into the other day. The cause of the bug itself is |
| 165 |
|
still unknown and it might be hard to fix it properly, but the |
| 166 |
|
workaround has some merits of its own (namely somewhat reducing pickling |
| 167 |
|
overhead for certain libraries). Therefore, I think this solution should |
| 168 |
|
be satisfactory at this time. |
| 169 |
|
|
| 170 |
|
The rest of the changes (i.e., the vast majority) has to do with my |
| 171 |
|
ongoing efforts of providing direct support for C function calls from |
| 172 |
|
ML. At the moment there is a new primop "RAW_CCALL", typing magic |
| 173 |
|
in types/cproto.sml (invoked from FLINT/trans/translate.sml), a new |
| 174 |
|
case in the FLINT CPS datatype (RCC), changes to cps/convert.sml to |
| 175 |
|
translate uses of RAW_CCALL into RCC, and changes to mlriscGen.sml to |
| 176 |
|
handle RCC. |
| 177 |
|
|
| 178 |
|
The last part (the changes to mlriscGen.sml) are still known to be |
| 179 |
|
wrong on the x86 and not implemented on all other architectures. But |
| 180 |
|
the infrastructure is in place. I had to change a few functor |
| 181 |
|
signatures in the backend to be able to route the CCalls interface |
| 182 |
|
from MLRISC there, and I had to specialize the mltree type (on the |
| 183 |
|
x86) to include the necessary extensions. (The extensions themselves |
| 184 |
|
were already there and redy to go in MLRISC/x86). |
| 185 |
|
|
| 186 |
|
Everything should be very happy as soon as someone helps me with |
| 187 |
|
mlriscGen.sml... |
| 188 |
|
|
| 189 |
|
In any case, nothing of this should matter to anyone as long as the |
| 190 |
|
new primop is not being used (which is going to be the case unless you |
| 191 |
|
find it where I hid it :). The rest of the compiler is completely |
| 192 |
|
unaffected. |
| 193 |
|
|
| 194 |
|
---------------------------------------------------------------------- |
| 195 |
|
Name: Matthias Blume |
| 196 |
|
Date: 2001/01/05 00:30:00 JST |
| 197 |
|
Tag: blume-20010105-primops |
| 198 |
|
Description: |
| 199 |
|
|
| 200 |
|
Added some experimental support for work that I am doing right now. |
| 201 |
|
These changes mostly concern added primops, but there is also a new |
| 202 |
|
experimental C library in the runtime system (but currently not enabled |
| 203 |
|
anywhere except on Linux/X86). |
| 204 |
|
|
| 205 |
|
In the course of adding primops (and playing with them), I discovered that |
| 206 |
|
Zhong's INL_PRIM hack (no type info for certain primops) was, in fact, badly |
| 207 |
|
broken. (Zhong was very right he labeled this stuff as "major gross hack".) |
| 208 |
|
To recover, I made type information in INL_PRIM mandatory and changed |
| 209 |
|
prim.sml as well as built-in.sml accordingly. The InLine structure now |
| 210 |
|
has complete, correct type information (i.e., no bottom types). |
| 211 |
|
|
| 212 |
|
Since all these changes mean that we need new binfiles, I also bumped the |
| 213 |
|
version number to 110.32.1. |
| 214 |
|
|
| 215 |
|
---------------------------------------------------------------------- |
| 216 |
|
Name: Matthias Blume |
| 217 |
|
Date: 2000/12/30 22:10:00 JST |
| 218 |
|
Tag: blume-20001230-various |
| 219 |
|
Description: |
| 220 |
|
|
| 221 |
|
Added proxy libraries for MLRISC and let MLRISC libraries refer |
| 222 |
|
to each other using path anchors. (See CM manual for explanation.) |
| 223 |
|
|
| 224 |
|
Updated CM documentation. |
| 225 |
|
|
| 226 |
|
Fixed some bugs in CM. |
| 227 |
|
|
| 228 |
|
Implemented "proxy" libraries (= syntactic sugar for CM). |
| 229 |
|
|
| 230 |
|
Added "-quiet" option to makeml and changed runtime system accordingly. |
| 231 |
|
|
| 232 |
|
Added cleanup handler for exportML to reset timers and compiler stats. |
| 233 |
|
|
| 234 |
|
---------------------------------------------------------------------- |
| 235 |
|
Name: Lal George |
| 236 |
|
Date: 2000/12/22 22:22:58 EST 2000 |
| 237 |
|
Tag: Release_110_32 |
| 238 |
|
Description: |
| 239 |
|
|
| 240 |
|
Infinite precision used throughout MLRISC. |
| 241 |
|
see MLRISC/mltree/machine-int.sig |
| 242 |
|
|
| 243 |
|
---------------------------------------------------------------------- |
| 244 |
|
Name: Matthias Blume |
| 245 |
|
Date: 2000/12/22 23:16:00 JST |
| 246 |
|
Tag: blume-20001222-warn |
| 247 |
|
Description: |
| 248 |
|
|
| 249 |
|
Corrected wording and formatting of some CM warning message which I |
| 250 |
|
broke in my previous patch. |
| 251 |
|
|
| 252 |
|
---------------------------------------------------------------------- |
| 253 |
|
Name: Matthias Blume |
| 254 |
|
Date: 2000/12/22 21:20:00 JST |
| 255 |
|
Tag: blume-20001222-anchorenv |
| 256 |
|
Description: |
| 257 |
|
|
| 258 |
|
Fixed CM's handling of anchor environments in connection with CMB.make. |
| 259 |
|
|
| 260 |
|
---------------------------------------------------------------------- |
| 261 |
|
Name: Matthias Blume |
| 262 |
|
Date: 2000/12/22 13:15:00 JST |
| 263 |
|
Tag: blume-20001222-cleanup |
| 264 |
|
Description: |
| 265 |
|
|
| 266 |
|
Removed src/cm/ffi which does not (and did not) belong here. |
| 267 |
|
|
| 268 |
|
---------------------------------------------------------------------- |
| 269 |
|
Name: Matthias Blume |
| 270 |
|
Date: 2000/12/21 23:55:00 JST |
| 271 |
|
Tag: blume-20001221-exn |
| 272 |
|
Description: |
| 273 |
|
|
| 274 |
|
Probably most important: CM no longer silently swallows all exceptions |
| 275 |
|
in the compiler. |
| 276 |
|
Plus: some other minor CM changes. For example, CM now reports some |
| 277 |
|
sizes for generated binfiles (code, data, envpickle, lambdapickle). |
| 278 |
|
|
| 279 |
---------------------------------------------------------------------- |
---------------------------------------------------------------------- |
| 280 |
Name: Matthias Blume |
Name: Matthias Blume |
| 281 |
Date: 2000/12/15 00:01:05 JST |
Date: 2000/12/15 00:01:05 JST |
Parent Directory
| 
Revision Log
| 
Patch
