1
Fork 0
mirror of git://git.sv.gnu.org/emacs.git synced 2026-03-23 07:12:12 -07:00
Commit graph

2202 commits

Author SHA1 Message Date
Richard Brooksby
d462d4b36d Repairing superclass by taking a kind, which can later be checked.
Copied from Perforce
 Change: 190850
 ServerID: perforce.ravenbrook.com
2016-04-10 12:28:51 +01:00
Richard Brooksby
3bc19084f6 Fixing mustbea to not return a void *.
Copied from Perforce
 Change: 190849
 ServerID: perforce.ravenbrook.com
2016-04-08 20:29:33 +01:00
Richard Brooksby
a15e1139c1 Fixing nulls in class methods revealed by more thoroughly applying class checking functions.
Copied from Perforce
 Change: 190842
 ServerID: perforce.ravenbrook.com
2016-04-08 20:00:23 +01:00
Richard Brooksby
a56de59509 Generating type-safe functions for getting and setting the class of instances.
Fixing warning revealed by GCC that I wasn't actually checking classes after init.
Fixing abstract classes so that they pass their own checks!

Copied from Perforce
 Change: 190837
 ServerID: perforce.ravenbrook.com
2016-04-08 19:50:12 +01:00
Richard Brooksby
d2f191f0d2 Using common code for class getters and setters.
Copied from Perforce
 Change: 190836
 ServerID: perforce.ravenbrook.com
2016-04-08 18:33:34 +01:00
Richard Brooksby
5a0c818945 Explicitly making inststruct the prefix of instances, eliminating pointer punning of class pointers, and reducing likelihood of editing errors.
Copied from Perforce
 Change: 190830
 ServerID: perforce.ravenbrook.com
2016-04-08 17:51:21 +01:00
Richard Brooksby
a5ea15c8ff Eliminating issubclasspoly and replacing with compile-time constant test.
Copied from Perforce
 Change: 190829
 ServerID: perforce.ravenbrook.com
2016-04-08 17:51:21 +01:00
Richard Brooksby
260964fdac Replacing prime factor subclass test with array-based test, because the type ids were overflowing 32 bits.
Copied from Perforce
 Change: 190828
 ServerID: perforce.ravenbrook.com
2016-04-08 17:51:21 +01:00
Richard Brooksby
43b3f1a5cb Generating a static superclass mapping and eliminating other superclass macros.
Copied from Perforce
 Change: 190827
 ServerID: perforce.ravenbrook.com
2016-04-08 17:51:20 +01:00
Richard Brooksby
298addf910 Implementing isa, mustbea, couldbea, and reshaping the land classes to put the init, finish, and describe methods the right way in.
Copied from Perforce
 Change: 190826
 ServerID: perforce.ravenbrook.com
2016-04-08 17:51:20 +01:00
Richard Brooksby
1d28eaff07 Eliminating the concept of "alias classes" in favour of "kinds".
Copied from Perforce
 Change: 190825
 ServerID: perforce.ravenbrook.com
2016-04-08 17:51:20 +01:00
Richard Brooksby
2e191a810e Referring to classes by their base identifiers.
Copied from Perforce
 Change: 190823
 ServerID: perforce.ravenbrook.com
2016-04-08 17:51:20 +01:00
Richard Brooksby
09e73dd1da Renaming protocolinst to inst and protocolclass to instclass to make naming consistent with other classes.
Copied from Perforce
 Change: 190822
 ServerID: perforce.ravenbrook.com
2016-04-08 17:51:20 +01:00
Richard Brooksby
19adf3ec16 Adding an assert that can be used in expressions.
Copied from Perforce
 Change: 190821
 ServerID: perforce.ravenbrook.com
2016-04-08 17:51:20 +01:00
Richard Brooksby
e1b7e69a02 Adding class-based checking macros.
Copied from Perforce
 Change: 190820
 ServerID: perforce.ravenbrook.com
2016-04-08 17:51:20 +01:00
Richard Brooksby
32de0701e0 Removing always-zero offset from pool class structure.
Copied from Perforce
 Change: 190819
 ServerID: perforce.ravenbrook.com
2016-04-08 17:51:20 +01:00
Richard Brooksby
ed1782a479 Removing unused offset field in arena classes.
Copied from Perforce
 Change: 190818
 ServerID: perforce.ravenbrook.com
2016-04-08 17:51:19 +01:00
Richard Brooksby
50ad46051c Using prime factors to determine subclass relationships quickly.
Copied from Perforce
 Change: 190817
 ServerID: perforce.ravenbrook.com
2016-04-08 17:51:19 +01:00
Richard Brooksby
e041841a88 Moving common fields into the base class, starting with the class name.
Copied from Perforce
 Change: 190816
 ServerID: perforce.ravenbrook.com
2016-04-08 17:51:19 +01:00
Richard Brooksby
780e5991cb Removing never-used multiple inheritance speculation.
Copied from Perforce
 Change: 190815
 ServerID: perforce.ravenbrook.com
2016-04-08 17:51:19 +01:00
Richard Brooksby
847272c3b5 Eliminating structure copy by sharing class init method.
Copied from Perforce
 Change: 190814
 ServerID: perforce.ravenbrook.com
2016-04-08 17:51:19 +01:00
Richard Brooksby
93b423efdf Sharing definition of class declaration in class definition.
Copied from Perforce
 Change: 190813
 ServerID: perforce.ravenbrook.com
2016-04-08 17:51:19 +01:00
Richard Brooksby
c1abe0e9fd Abstracting class declarations and references to classes.
Copied from Perforce
 Change: 190812
 ServerID: perforce.ravenbrook.com
2016-04-08 17:51:19 +01:00
Richard Brooksby
2743081e86 Getting rid of "old style" class ensure functions.
Copied from Perforce
 Change: 190811
 ServerID: perforce.ravenbrook.com
2016-04-08 17:51:19 +01:00
Richard Brooksby
484d47205a Updating copyright date to current year.
Copied from Perforce
 Change: 190707
 ServerID: perforce.ravenbrook.com
2016-04-05 11:30:42 +01:00
Richard Brooksby
3ce48493c5 Updating .p4ignore files with miscellaneous entries collected from work with git.
Copied from Perforce
 Change: 190702
 ServerID: perforce.ravenbrook.com
2016-04-05 11:27:30 +01:00
Richard Brooksby
dcc027082e Fixing horrible intialization of segment class by poking before calling seginit.
Copied from Perforce
 Change: 190694
 ServerID: perforce.ravenbrook.com
2016-04-04 23:28:28 +01:00
Richard Brooksby
1199a63bab Linking .p4ignore to .gitignore to share configuration between perforce and git.
Copied from Perforce
 Change: 190667
 ServerID: perforce.ravenbrook.com
2016-04-04 19:20:26 +01:00
Richard Brooksby
50ecd3fb88 Adding new build directories to .p4ignore.
Copied from Perforce
 Change: 190643
 ServerID: perforce.ravenbrook.com
2016-04-02 15:54:24 +01:00
Richard Brooksby
49534e914f Protecting some segment accessor macros with rvalue.
Removing duplicate definition of SegPoolRing.

Copied from Perforce
 Change: 190642
 ServerID: perforce.ravenbrook.com
2016-04-02 15:53:57 +01:00
Richard Brooksby
29fd6ebdd0 Don't attempt to merge a non-white segment with a white one.
Copied from Perforce
 Change: 190634
 ServerID: perforce.ravenbrook.com
2016-04-02 12:07:46 +01:00
Richard Brooksby
fffb906f11 Fixing default behaviour for merging segments with differing summaries and therefore write barriers.
Copied from Perforce
 Change: 190629
 ServerID: perforce.ravenbrook.com
2016-04-02 09:18:58 +01:00
Richard Brooksby
e27b3ce6a8 Don't allow pools to whiten segments without condemning objects, so that a condemned size of zero implies no white segments, allowing quick trace destruction.
Copied from Perforce
 Change: 190607
 ServerID: perforce.ravenbrook.com
2016-03-31 23:41:46 +01:00
Richard Brooksby
4cb09a6d47 Cherry-picking low-risk fix for job003880. see <https://info.ravenbrook.com/mail/2016/03/29/20-23-20/0/>.
Copied from Perforce
 Change: 190579
 ServerID: perforce.ravenbrook.com
2016-03-31 12:30:00 +01:00
Richard Brooksby
0d7f8273fd Reducing over-ambitious random alignment increases in apss test.
Copied from Perforce
 Change: 190575
 ServerID: perforce.ravenbrook.com
2016-03-31 09:28:41 +01:00
Richard Brooksby
61dc9d8345 Cross-referencing comment about non-working debugging aps to job003995.
Copied from Perforce
 Change: 190477
 ServerID: perforce.ravenbrook.com
2016-03-27 21:17:53 +01:00
Richard Brooksby
7c1e988d2b Documenting and checking constraints on mps_key_align on manual pools.
Copied from Perforce
 Change: 190472
 ServerID: perforce.ravenbrook.com
2016-03-27 20:48:38 +01:00
Richard Brooksby
1d8377136a Correcting return type of zonesetofrange to a zoneset.
Copied from Perforce
 Change: 190395
 ServerID: perforce.ravenbrook.com
2016-03-24 21:59:39 +00:00
Richard Brooksby
ff35af0f15 Merging branch/2016-03-13/defer-write-barrier into master sources.
Copied from Perforce
 Change: 190380
 ServerID: perforce.ravenbrook.com
2016-03-24 12:07:20 +00:00
Richard Brooksby
681886a8ac Catch-up merge from master sources to branch/2016-03-11/shield-coalesce.
Copied from Perforce
 Change: 190373
 ServerID: perforce.ravenbrook.com
2016-03-24 11:08:24 +00:00
Richard Brooksby
397f3713f9 Fixing build errors on windows (w3i6mv).
tagtest was missing from makefiles.
Implicit conversion warning from Clock to double.

Copied from Perforce
 Change: 190372
 ServerID: perforce.ravenbrook.com
2016-03-24 11:04:58 +00:00
Richard Brooksby
4e0557e20e Improving shieldflushentries to allow for delayed protection changes on shieldlower.
Copied from Perforce
 Change: 190363
 ServerID: perforce.ravenbrook.com
2016-03-23 19:03:00 +00:00
Richard Brooksby
f2864ea179 Improving commentary: the mutator is suspended, not the arena.
Copied from Perforce
 Change: 190352
 ServerID: perforce.ravenbrook.com
2016-03-23 17:55:22 +00:00
Richard Brooksby
25bfc7b131 Responding to review by nb <https://info.ravenbrook.com/mail/2016/03/22/17-19-03/0/>.
Copied from Perforce
 Change: 190349
 ServerID: perforce.ravenbrook.com
2016-03-23 17:31:59 +00:00
Richard Brooksby
bcd30514d9 Forgot shieldhold/shieldrelease in tracecondemnzones. caught by hot builds of amcssth.
Copied from Perforce
 Change: 190344
 ServerID: perforce.ravenbrook.com
2016-03-23 15:48:58 +00:00
Richard Brooksby
33cb1b0596 Improving commentary in response to nb <https://info.ravenbrook.com/mail/2016/03/22/17-19-03/0/>.
Copied from Perforce
 Change: 190339
 ServerID: perforce.ravenbrook.com
2016-03-23 14:37:54 +00:00
Richard Brooksby
8bfd101b34 Responding to review by gdr <https://info.ravenbrook.com/mail/2016/03/21/21-05-49/0/>.
Copied from Perforce
 Change: 190336
 ServerID: perforce.ravenbrook.com
2016-03-23 14:23:14 +00:00
Richard Brooksby
405aed6b34 Bracketing out unused stringequal and tidying up stringlength.
Copied from Perforce
 Change: 190333
 ServerID: perforce.ravenbrook.com
2016-03-23 13:54:22 +00:00
Richard Brooksby
e711121e82 Moving quicksort's workspace off the stack to limit mps stack usage.
Copied from Perforce
 Change: 190332
 ServerID: perforce.ravenbrook.com
2016-03-23 13:47:55 +00:00
Richard Brooksby
88affc312f Added discussion of when it might be good to resume the mutator early in design.mps.shield.improv.resume.
Copied from Perforce
 Change: 190329
 ServerID: perforce.ravenbrook.com
2016-03-23 13:11:19 +00:00