1 2017-08-26 00:01:10	0|bitcoin-git|[13bitcoin] 15danra opened pull request #11151: Fix header guards using reserved identifiers (06master...06fix/header-guards) 02https://github.com/bitcoin/bitcoin/pull/11151
 2 2017-08-26 00:22:05	0|jimpo|I know you guys are busy with the 0.15 release, but I'd really appreciate a review on either 11116 or 11113.
 3 2017-08-26 00:27:27	0|gmaxwell|jimpo: Thanks!
 4 2017-08-26 01:41:32	0|bitcoin-git|[13bitcoin] 15RHavar opened pull request #11152: Increase DEFAULT_ANCESTOR_LIMIT and DEFAULT_DESCENDANT_LIMIT to 100 (06master...06limits) 02https://github.com/bitcoin/bitcoin/pull/11152
 5 2017-08-26 02:33:13	0|bitcoin-git|[13bitcoin] 15RHavar closed pull request #11152: Increase DEFAULT_ANCESTOR_LIMIT and DEFAULT_DESCENDANT_LIMIT to 100 (06master...06limits) 02https://github.com/bitcoin/bitcoin/pull/11152
 6 2017-08-26 02:35:35	0|jtimon|ping https://github.com/bitcoin/bitcoin/pull/10669
 7 2017-08-26 02:39:00	0|jtimon|I know nobody cares about this and that's why it's taken ages, but I think it's good even if I didn't care about signed blocks...
 8 2017-08-26 02:47:01	0|meshcollider|I'll review in a second :)
 9 2017-08-26 02:54:26	0|meshcollider|lgtm
10 2017-08-26 07:12:25	0|bitcoin-git|[13bitcoin] 15MeshCollider opened pull request #11153: Add meshcollider Gitian PGP key (06master...06201708_add_meshcollider_gpg) 02https://github.com/bitcoin/bitcoin/pull/11153
11 2017-08-26 10:21:32	0|bitcoin-git|[13bitcoin] 15danra opened pull request #11155: Trivial: Documentation fixes for CVectorWriter ctors (06master...06patch-8) 02https://github.com/bitcoin/bitcoin/pull/11155
12 2017-08-26 10:29:30	0|bitcoin-git|[13bitcoin] 15danra opened pull request #11156: Fix memory leaks in qt/guiutil.cpp (06master...06fix/qt-guiutil-memory-leaks) 02https://github.com/bitcoin/bitcoin/pull/11156
13 2017-08-26 11:00:05	0|bitcoin-git|13bitcoin/06master 14bc70ab5 15Dan Raviv: Fix header guards using reserved identifiers...
14 2017-08-26 11:00:05	0|bitcoin-git|[13bitcoin] 15laanwj pushed 2 new commits to 06master: 02https://github.com/bitcoin/bitcoin/compare/07c92b98e2a0...7fd49d01dc2e
15 2017-08-26 11:00:06	0|bitcoin-git|13bitcoin/06master 147fd49d0 15Wladimir J. van der Laan: Merge #11151: Fix header guards using reserved identifiers...
16 2017-08-26 11:00:45	0|bitcoin-git|[13bitcoin] 15laanwj closed pull request #11151: Fix header guards using reserved identifiers (06master...06fix/header-guards) 02https://github.com/bitcoin/bitcoin/pull/11151
17 2017-08-26 12:12:36	0|bitcoin-git|[13bitcoin] 15danra opened pull request #11160: Tests: Improve versionbits_computeblockversion test code consistency and fix… (06master...06patch-9) 02https://github.com/bitcoin/bitcoin/pull/11160
18 2017-08-26 13:38:38	0|bitcoin-git|[13bitcoin] 15danra opened pull request #11161: Remove redundant explicitly defined CFeeRate copy ctor. (06master...06fix/redundant-CFeeRate-copy-ctor) 02https://github.com/bitcoin/bitcoin/pull/11161
19 2017-08-26 16:34:37	0|bitcoin-git|[13bitcoin] 15danra opened pull request #11163: Remove redundant explicitly defined CTxMemPoolEntry copy ctor (06master...06fix/redundant-CTxMemPoolEntry-copy-ctor) 02https://github.com/bitcoin/bitcoin/pull/11163
20 2017-08-26 17:53:26	0|luke-jr|weird, I've got an apparent memory leak in custom bitcoind code, and valgrind fails to see it at all :/
21 2017-08-26 17:54:40	0|luke-jr|valgrind says 142 MB memory used total (incl reachable), yet ps shows 12.5 GB memory use, and if I run outside valgrind I rapidly hit swap
22 2017-08-26 18:03:37	0|gmaxwell|there are allocations happening via things other than malloc (and new) so valgrind can't see them.
23 2017-08-26 18:03:55	0|bitcoin-git|[13bitcoin] 15danra closed pull request #11163: Remove redundant explicitly defined CTxMemPoolEntry copy ctor (06master...06fix/redundant-CTxMemPoolEntry-copy-ctor) 02https://github.com/bitcoin/bitcoin/pull/11163
24 2017-08-26 18:04:42	0|gmaxwell|are you sure it's allocations and not mmaped files
25 2017-08-26 18:06:25	0|luke-jr|gmaxwell: well, things get real slow when I begin swapping
26 2017-08-26 18:06:39	0|luke-jr|I'd expect mmap can't do that
27 2017-08-26 18:10:39	0|luke-jr|thankfully, the actual code being executed is relatively small, I guess I can just start commenting out parts to see where the culprit is
28 2017-08-26 18:10:52	0|luke-jr|(the wallet rescan logic specifically)
29 2017-08-26 18:13:39	0|luke-jr|curiously, valgrind did report: crypto/common.h:48:22: runtime error: load of misaligned address 0x000046682d26 for type 'uint32_t', which requires 4 byte alignment
30 2017-08-26 18:14:05	0|luke-jr|and it looks like our SHA2 code does such access without checking alignment. wonder why it works.
31 2017-08-26 18:14:30	0|luke-jr|(but this is an older codebase, maybe already changed)
32 2017-08-26 18:17:06	0|gmaxwell|that should be fixed.  alignment isn't required on x86, so it works even though its undefined behavior.
33 2017-08-26 18:19:05	0|sipa|luke-jr: which function is that?
34 2017-08-26 18:19:18	0|luke-jr|SSSE3 definitely segfaulted when not aligned on my Haswell CPU, but that's another specific case :p
35 2017-08-26 18:19:27	0|sipa|last line of WriteLE32 ?
36 2017-08-26 18:19:28	0|luke-jr|sipa: ReadBE32
37 2017-08-26 18:20:07	0|sipa|what version is that?
38 2017-08-26 18:20:41	0|luke-jr|0.13.2-based
39 2017-08-26 18:20:47	0|luke-jr|I think this is a close stack http://codepad.org/oA6M1Oef
40 2017-08-26 18:21:11	0|sipa|fixed in #9512
41 2017-08-26 18:22:35	0|gmaxwell|luke-jr: yes, unaligned aligned simd reads can crash on x86.
42 2017-08-26 18:23:12	0|luke-jr|gmaxwell: the fun part of that was that glibc did it for memcpy XD
43 2017-08-26 18:24:14	0|gmaxwell|wtf. memcpy has no alignment requirements.
44 2017-08-26 18:26:53	0|sipa|gmaxwell: but memcpy may be implemented by casting pointers to int pointers and then assigning
45 2017-08-26 18:27:00	0|sipa|(i assume that's what luke meant)
46 2017-08-26 18:27:30	0|luke-jr|right
47 2017-08-26 19:05:01	0|bitcoin-git|[13bitcoin] 15danra opened pull request #11164: Fix boost headers included as user instead of system headers (06master...06fix/boost-system-includes) 02https://github.com/bitcoin/bitcoin/pull/11164
48 2017-08-26 22:18:44	0|bitcoin-git|[13bitcoin] 15sipa opened pull request #11167: Full BIP173 (Bech32) support (06master...06201708_bech32) 02https://github.com/bitcoin/bitcoin/pull/11167
49 2017-08-26 23:11:00	0|luke-jr|oh. it seems -fsanitize=undefined causes the memory leak -.-
50 2017-08-26 23:11:23	0|gmaxwell|that isn't a memory leak.
51 2017-08-26 23:11:45	0|gmaxwell|msan needs scads of virtual address space and a fair amount of ram.
52 2017-08-26 23:11:55	0|luke-jr|gmaxwell: over 20 GB?
53 2017-08-26 23:12:16	0|luke-jr|also, this is undefined sanitizer, not msan <.<
54 2017-08-26 23:12:24	0|gmaxwell|same reason.
55 2017-08-26 23:12:37	0|gmaxwell|I've seen it get up to a petabyte of virt.
56 2017-08-26 23:12:50	0|luke-jr|it wasn't merely virt
57 2017-08-26 23:12:55	0|gmaxwell|(oh actually no, .1 PB)
58 2017-08-26 23:13:04	0|gmaxwell|yes, it uses a lot more RES too...
59 2017-08-26 23:13:53	0|luke-jr|kinda makes it useless :/
60 2017-08-26 23:15:04	0|gmaxwell|have more ram...
61 2017-08-26 23:15:56	0|luke-jr|UBSAN claims "The checks have small runtime cost…"
62 2017-08-26 23:17:20	0|gmaxwell|hm. I was confused above, I thought you were using asan. I'm not aware of undefined using a lot of ram.
63 2017-08-26 23:19:38	0|gmaxwell|jonasschnelli: I just had someone on IRC that had their GUI not displaying after upgrading to 0.14.2 ... this fixed it https://github.com/bitcoin/bitcoin/issues/7869#issuecomment-209265754  is there some underlying bug we need to fix?
64 2017-08-26 23:28:56	0|Lightsword|do we need to bump the OSX SDK version? I think it’s on 10.11 which doesn’t seem to have getentropy