| 14 |
|
|
| 15 |
---------------------------------------------------------------------- |
---------------------------------------------------------------------- |
| 16 |
Name: Matthias Blume |
Name: Matthias Blume |
| 17 |
|
Date: 2002/02/08 15:00:00 EST |
| 18 |
|
Tag: blume-20020208-uniquepid |
| 19 |
|
Description: |
| 20 |
|
|
| 21 |
|
0. Version number has been bumped to 110.38.1. NEW BOOTFILES!!! |
| 22 |
|
|
| 23 |
|
1. The installer (config/install.sh) has gotten smarter: |
| 24 |
|
|
| 25 |
|
- Configuration options are a bit easier to specify now |
| 26 |
|
(in config/targets). |
| 27 |
|
- Bug in recognizing .tar.bz2 files fixed. |
| 28 |
|
- Installer automatically resolves dependencies between |
| 29 |
|
configuration options (e.g., if you ask for eXene, you will |
| 30 |
|
also get cml -- regardless whether you asked for it or not). |
| 31 |
|
- Installer can run in "quieter mode" by setting the environment |
| 32 |
|
variable INSTALL_QUIETLY to "true". "Quieter" does not mean |
| 33 |
|
"completely silent", though. |
| 34 |
|
- Build HashCons library as part of smlnj-lib. |
| 35 |
|
|
| 36 |
|
2. A new scheme for assigning persistent identifiers to compilation |
| 37 |
|
units (and, by extension, to types etc.) has been put into place. |
| 38 |
|
This fixes a long-standing bug where types and even dynamic values |
| 39 |
|
can get internally confused, thereby compromising type safety |
| 40 |
|
(abstraction) and dynamic correctness. See |
| 41 |
|
|
| 42 |
|
http://cm.bell-labs.com/cm/cs/who/blume/pid-confusion.tgz |
| 43 |
|
|
| 44 |
|
for an example of how things could go wrong until now. |
| 45 |
|
|
| 46 |
|
The downside of the new scheme is that pids are not quite as |
| 47 |
|
persistent as they used to be: CM will generate a fresh pid |
| 48 |
|
for every compilation unit that it thinks it sees for the first |
| 49 |
|
time. That means that if you compile starting from a clean, fresh |
| 50 |
|
source tree at two different times, you end up with different |
| 51 |
|
binaries. |
| 52 |
|
|
| 53 |
|
Cutoff recompilation, however, has not been compromised because |
| 54 |
|
CM keeps pid information in special caches between runs. |
| 55 |
|
|
| 56 |
|
---------------------------------------------------------------------- |
| 57 |
|
Name: Lal George |
| 58 |
|
Date: 2002/02/07 15:34:13 EST 2002 |
| 59 |
|
Tag: <none> |
| 60 |
|
Description: |
| 61 |
|
|
| 62 |
|
Compilers that generate assembly code may produce global labels |
| 63 |
|
whose value is resolved at link time. The various peephole optimization |
| 64 |
|
modules did not take this in account. |
| 65 |
|
|
| 66 |
|
TODO. The Labels.addrOf function should really return an option |
| 67 |
|
type so that clients are forced to deal with this issue, rather |
| 68 |
|
than an exception being raised. |
| 69 |
|
|
| 70 |
|
---------------------------------------------------------------------- |
| 71 |
|
Name: Lal George |
| 72 |
|
Date: 2002/02/06 13:55:02 EST |
| 73 |
|
Tag: george-20020206-ra-breakup |
| 74 |
|
Description: |
| 75 |
|
|
| 76 |
|
1. A bug fix from Allen. |
| 77 |
|
|
| 78 |
|
A typo causes extra fstp %st(0)'s to be generated at compensation |
| 79 |
|
edges, which might cause stack underflow traps at runtime. This |
| 80 |
|
occurs in fft where there are extraneous fstps right before the 'into' |
| 81 |
|
trap instruction (in this case they are harmless since none of the |
| 82 |
|
integers overflow.) |
| 83 |
|
|
| 84 |
|
2. Pulled out various utility modules that were embedded in the modules |
| 85 |
|
of the register allocator. I need these modules for other purposes, but |
| 86 |
|
they are not complete enough to put into a library (just yet). |
| 87 |
|
---------------------------------------------------------------------- |
| 88 |
|
Name: Matthias Blume |
| 89 |
|
Date: 2002/01/31 16:05:00 EST |
| 90 |
|
Tag: blume-20020131-sparc-ccalls |
| 91 |
|
Description: |
| 92 |
|
|
| 93 |
|
1. C-calls on Sparc needlessly allocated a huge chunk (96 bytes) |
| 94 |
|
of extra stack space by mistake. Fixed. |
| 95 |
|
|
| 96 |
|
2. Bug in logic of handling of command-line options in ml-nlffigen fixed. |
| 97 |
|
|
| 98 |
|
---------------------------------------------------------------------- |
| 99 |
|
Name: Allen Leung |
| 100 |
|
Date: 2002/01/30 |
| 101 |
|
Tag: leunga-20020130-nowhere-bug-fix |
| 102 |
|
Description: |
| 103 |
|
|
| 104 |
|
MLRISC bug fixes: |
| 105 |
|
1. Fixed a bindings computation bug in the 'nowhere' program generator tool. |
| 106 |
|
2. MachineInt.fromString was negating its value. |
| 107 |
|
|
| 108 |
|
---------------------------------------------------------------------- |
| 109 |
|
Name: Matthias Blume |
| 110 |
|
Date: 2002/01/29 |
| 111 |
|
Tag: blume-20020129-INSTALL |
| 112 |
|
Description: |
| 113 |
|
|
| 114 |
|
- Added somewhat detailed installation instructions (file INSTALL). |
| 115 |
|
- Fixed curl-detection bug in config/install.sh. |
| 116 |
|
- It is now possible to select the URL getter using the URLGETTER |
| 117 |
|
environment variable: |
| 118 |
|
|
| 119 |
|
not set / "unknown" --> automatic detection (script tries wget, |
| 120 |
|
curl, and lynx) |
| 121 |
|
"wget" / "curl" / "lynx" --> use the specified program (script "knows" |
| 122 |
|
how to properly invoke them) |
| 123 |
|
other --> use $URLGETTER directly, it must take |
| 124 |
|
precisely two command-line arguments |
| 125 |
|
(source URL and destination file name) |
| 126 |
|
|
| 127 |
|
---------------------------------------------------------------------- |
| 128 |
|
Name: Matthias Blume |
| 129 |
|
Date: 2002/01/28 |
| 130 |
|
Tag: blume-20020128-sparc-ccalls |
| 131 |
|
Description: |
| 132 |
|
|
| 133 |
|
- Fixed problem with calculation of "used" registers in sparc-c-calls. |
| 134 |
|
- Make use of the allocParam argument in sparc-c-calls. |
| 135 |
|
|
| 136 |
|
---------------------------------------------------------------------- |
| 137 |
|
Name: Matthias Blume |
| 138 |
|
Date: 2002/01/28 |
| 139 |
|
Tag: blume-20020128-allocParam |
| 140 |
|
Description: |
| 141 |
|
|
| 142 |
|
John Reppy: Changes c-calls API to accept client-callback for |
| 143 |
|
allocating extra stack space. |
| 144 |
|
me: Corresponding changes to mlriscGen (using a dummy argument that |
| 145 |
|
does not change the current behavior). |
| 146 |
|
|
| 147 |
|
---------------------------------------------------------------------- |
| 148 |
|
Name: Matthias Blume |
| 149 |
Date: 2002/01/28 12:00:00 |
Date: 2002/01/28 12:00:00 |
| 150 |
Tag: Release_110_38 |
Tag: Release_110_38 |
| 151 |
Description: |
Description: |
| 705 |
|
|
| 706 |
---------------------------------------------------------------------- |
---------------------------------------------------------------------- |
| 707 |
Name: Matthias Blume |
Name: Matthias Blume |
|
>>>>>>> 1.169 |
|
| 708 |
Date: 2001/09/18 15:35:00 EDT |
Date: 2001/09/18 15:35:00 EDT |
| 709 |
Tag: blume-20010918-readme11036 |
Tag: blume-20010918-readme11036 |
| 710 |
Description: |
Description: |
Parent Directory
| 
Revision Log
| 
Patch
