| 00:06.54 | *** join/#maemo-ssu RedW (~redw@89-73-179-171.dynamic.chello.pl) |
| 00:29.51 | *** join/#maemo-ssu rd_ (~rd@p4FD1DE37.dip0.t-ipconnect.de) |
| 00:31.19 | xes | freemangordon: well done! |
| 08:27.00 | *** join/#maemo-ssu arcean (~arcean@apn-46-76-96-172.dynamic.gprs.plus.pl) |
| 08:46.34 | *** join/#maemo-ssu RzR (~RzR@82.236.136.171) |
| 09:02.06 | *** join/#maemo-ssu rd_ (~rd@p4FD1DE37.dip0.t-ipconnect.de) |
| 09:37.45 | freemangordon | merlin1991: I think it is relatively safe o include openssl in the next -testing |
| 09:37.49 | freemangordon | *to |
| 09:39.33 | freemangordon | if there is something broken, I doubt we'll catch it while the package is in -devel. so far I tested everything I can think about and there seems to be no problem - https, email (imap tls), ssh, supl. |
| 09:40.28 | bencoh | maybe we should nm all the binaries and grep for affected symbols |
| 09:40.40 | bencoh | might help catching potential issues |
| 09:41.30 | bencoh | but the main problem here is things will probably "work" but we might end up with invalid reads/writes at some point |
| 09:48.34 | freemangordon | bencoh: symbols should be there, it is supposed to be the same ABI |
| 09:49.02 | bencoh | yeah but some structures differ |
| 09:49.18 | freemangordon | but you won;t catch that with nm |
| 09:49.20 | bencoh | some of which might be instantiated from the app and passed to functions |
| 09:49.38 | bencoh | no, but you can grep on the functions which are passed those struct/struct* |
| 09:49.47 | freemangordon | I doubt there is a transparent structures to be used in applications |
| 09:49.53 | freemangordon | *are |
| 09:50.09 | freemangordon | if an application uses such, then I'll call it buggy |
| 09:50.42 | freemangordon | even if it malloc's it |
| 09:50.54 | bencoh | there are structures defined in headers |
| 09:51.05 | freemangordon | bencoh: but feel free to do ^^^ if you think it is possible |
| 09:51.10 | freemangordon | so? |
| 09:51.46 | bencoh | they're not private/opaque :) |
| 09:51.53 | bencoh | so, I dont know |
| 09:52.51 | bencoh | hm, headers look bloated in openssl :D |
| 09:52.56 | freemangordon | :nod: |
| 09:55.39 | kerio | let's just move to libressl |
| 09:55.51 | kerio | we only need to recompile everything |
| 09:56.07 | bencoh | :/ |
| 09:56.42 | bencoh | there are things we cant rebuild :( |
| 09:57.19 | kerio | :( |
| 10:00.53 | bencoh | most most datatype changes reported in http://upstream.rosalinux.ru/versions/openssl.html say "99 symbols affected" (basically every symbole, more or less) |
| 10:02.21 | kerio | wut |
| 10:03.21 | bencoh | that means nm+grep wont work |
| 10:03.42 | freemangordon | bencoh: look at the correct version, 0.9.8 that is |
| 10:05.36 | kerio | yeah i mean |
| 10:05.49 | kerio | we're going 0.9.8n to 0.9.8zf |
| 10:06.06 | freemangordon | :nod: |
| 10:06.44 | *** join/#maemo-ssu Pali (~pali@Maemo/community/contributor/Pali) |
| 10:07.27 | bencoh | freemangordon: yeah that's what I'm checking :) |
| 10:07.56 | bencoh | looks like za reverts datatype issues introduced in y |
| 10:08.16 | bencoh | so we're left with issues introduced in p |
| 10:09.14 | bencoh | which might be solved just by keeping the old PQ_64BIT behaviour |
| 10:09.28 | bencoh | I wouldnt be surprised if debian did that for us |
| 10:10.57 | kerio | i like how openssl has shit like "is_probably_safari" |
| 10:11.02 | bencoh | :)) |
| 10:11.33 | bencoh | kerio: then what about PQ_64BIT_IS_BIGNUM (bignum was 20bytes ... 160bits) ? ;p |
| 10:12.41 | freemangordon | bencoh: could you check the patches? |
| 10:13.27 | freemangordon | https://github.com/community-ssu/openssl/tree/0.9.8zf-1%2Bmaemo1%2B0m5%2B0cssu0/debian/patches |
| 10:15.38 | freemangordon | hmm, debian has patches up to 0.9.8o :( |
| 10:15.50 | freemangordon | I guess it is because of the p breaking ABI |
| 10:15.51 | bencoh | is this patch what I think ? https://github.com/community-ssu/openssl/blob/master/debian/patches/12_valgrind.patch |
| 10:16.01 | bencoh | reminds me of an (in)famous debian patch |
| 10:16.18 | bencoh | I think someone had a look at it before, but ... |
| 10:16.35 | bencoh | freemangordon: hm probably yeah |
| 10:17.41 | bencoh | they keep backporting changes in squeeze-lts |
| 10:17.51 | bencoh | why not use squeeze-lts as upstream for openssl ? |
| 10:17.52 | freemangordon | http://anonscm.debian.org/viewvc/pkg-openssl/openssl/tags/0.9.8o-4squeeze15/debian/patches/ |
| 10:18.19 | freemangordon | because the newest patch there is 9 months old |
| 10:18.30 | bencoh | https://packages.qa.debian.org/o/openssl/news/20150320T214900Z.html |
| 10:19.01 | freemangordon | hmm, where did you get that from? |
| 10:19.12 | freemangordon | I mean - where is the source repo? |
| 10:19.27 | bencoh | dunno |
| 10:19.36 | bencoh | I didnt get it - yet |
| 10:20.36 | bencoh | and ... google cant find the packages |
| 10:21.08 | bencoh | http://ftp.de.debian.org/debian/pool/main/o/openssl/openssl_0.9.8o-4squeeze20.debian.tar.gz |
| 10:21.14 | bencoh | google hasnt indexed it |
| 10:21.19 | bencoh | (yet) |
| 10:23.40 | freemangordon | we'd rather fix the ABI breakage (if any) |
| 10:23.56 | bencoh | the valgrind.patch is still in debian, so I guess (hope ? :D) it's not the infamous 2008ish debianssl bg |
| 10:24.00 | bencoh | bug |
| 10:24.43 | bencoh | freemangordon: there our best bet is probably defining PQ_64BIT_IS_BIGNUM to 1 and PQ_64BIT_IS_INTEGER to 0 |
| 10:24.52 | bencoh | (and maybe the other two PQ_64BIT constants) |
| 10:27.56 | freemangordon | bencoh: look at pqueue/pq_compat.h |
| 10:29.00 | bencoh | yeah |
| 10:29.09 | freemangordon | and it seems defined in SB |
| 10:29.23 | freemangordon | ~/workspace/cssu/openssl/git/openssl-0.9.8zf] > grep -r THIRTY * |
| 10:29.26 | bencoh | oh ? |
| 10:29.44 | freemangordon | ... |
| 10:29.46 | freemangordon | crypto/opensslconf.h:#define THIRTY_TWO_BIT |
| 10:29.50 | bencoh | :D |
| 10:30.13 | bencoh | nice, we might be safe then |
| 10:30.18 | freemangordon | I'll put some garbage in that header, to make sure this is really defained |
| 10:31.17 | bencoh | hm we need to make sure it's used everywhere |
| 10:35.38 | freemangordon | ../../include/openssl/pq_compat.h:113:2: error: #error "# if (defined(THIRTY_TWO_BIT) && !defined(BN_LLONG)) || defined(SIXTEEN_BIT) || defined(EIGHT_BIT) *IS NOT* defined, we are busted :(" |
| 10:35.38 | freemangordon | make[3]: *** [pqueue.o] Error 1 |
| 10:35.40 | bencoh | okay, that's the only place PQ_64BIT is defined, so we should be safe |
| 10:35.46 | bencoh | oh ? |
| 10:36.48 | bencoh | hmm right, we're 32b but we have long long support on arm |
| 10:38.19 | freemangordon | yeah |
| 10:38.37 | freemangordon | that one should be fixed. I guess |
| 10:38.54 | freemangordon | however, I gtg, will look at it later |
| 10:39.10 | bencoh | that means we give up on long long optim ... well, if those didnt exist before 0.9.8p, we wont miss anything |
| 10:41.28 | freemangordon | well, we can try to wrap |
| 10:42.32 | freemangordon | those are just a couple of functions anyway |
| 10:43.00 | bencoh | yeah |
| 10:43.27 | bencoh | funny thing is ... I cant find those changes in openssl git (git log between tags) :/ |
| 10:43.35 | bencoh | dunno what they did there |
| 10:44.01 | freemangordon | check on github, they have "file history" |
| 10:44.18 | freemangordon | https://github.com/openssl/openssl |
| 10:44.29 | bencoh | ? |
| 10:45.23 | freemangordon | https://github.com/openssl/openssl/commits/OpenSSL_0_9_8zf/crypto/pqueue/pqueue.h |
| 10:45.59 | freemangordon | https://github.com/openssl/openssl/commit/e2ac4732cd4f0efc3dff3261d19aed0ced2c4dde |
| 10:46.38 | freemangordon | hmm, wait, isn;t that pq shit internal? |
| 10:46.54 | bencoh | no, it's reflected in public structures |
| 10:47.21 | bencoh | dtls1_/ssl3_ stuff |
| 10:47.38 | freemangordon | hmm, yeah |
| 10:47.40 | freemangordon | https://github.com/openssl/openssl/commit/e2ac4732cd4f0efc3dff3261d19aed0ced2c4dde#diff-bfa5a0561be6e4126229a7fd57bc24e0R299 |
| 10:47.46 | bencoh | ssl3_record_st has seq_num for instance |
| 10:47.59 | bencoh | right ;) |
| 10:48.13 | freemangordon | ok, we can wrap it |
| 10:48.38 | freemangordon | or maybe not |
| 10:48.49 | bencoh | :] |
| 10:49.34 | freemangordon | wait... do we really need to fix that? |
| 10:49.48 | bencoh | dunno |
| 10:49.52 | freemangordon | the new structure is smaller than the old one |
| 10:50.14 | freemangordon | merlin1991: Pali: ^^^ |
| 10:52.31 | freemangordon | maybe someone should ask on #openssl |
| 10:52.42 | freemangordon | or whateer the channel is |
| 10:52.55 | *** join/#maemo-ssu amiconn (~amiconn@rockbox/developer/amiconn) |
| 10:53.14 | freemangordon | is afk |
| 10:53.16 | Pali | I really do not know :-( |
| 10:53.38 | freemangordon | Pali: noone knows, that is why we investigate :) |
| 10:58.16 | freemangordon | bencoh: could you check what sizeof(those 2 structs) gives with old and with new openssl? |
| 10:58.52 | bencoh | I'm checking diffs between (current cssu) openssl-0.9.8n and zf, and ... |
| 10:58.56 | bencoh | diff -w -u crypto/pqueue/pq_compat.h ../openssl-cssu/openssl-0.9.8zf/crypto/pqueue/pq_compat.h |
| 10:58.59 | bencoh | returns nothing |
| 10:59.24 | bencoh | I feel like I missed something at some point |
| 10:59.50 | bencoh | or upstream.rosalinux.ru says shit |
| 11:01.13 | freemangordon | keep it mind that it might say shit, becuse of the missing arch specific stuff |
| 11:01.23 | bencoh | true |
| 11:01.34 | bencoh | it might just be that |
| 11:01.46 | freemangordon | that is why the real test should be comparing the sizeof structs, just to be sure |
| 11:02.06 | freemangordon | but I don;t have time now |
| 11:02.13 | freemangordon | buy, bbl |
| 11:02.21 | bencoh | should I just compile a oneliner against the headers ? |
| 11:28.48 | *** join/#maemo-ssu M4rtinK (~M4rtinK@mail.melf.eu) |
| 11:41.44 | *** join/#maemo-ssu futpib (~futpib@176.104.210.0) |
| 11:55.33 | bencoh | I should have disabled the checks in debian/rules ... it's taking ages |
| 12:02.04 | *** join/#maemo-ssu trx (ns-team@devbin/founder/trx) |
| 12:10.44 | bencoh | freemangordon: http://pastebin.notk.org/pastebin.php?show=m6cc01d7d |
| 12:10.47 | bencoh | looks good |
| 12:17.23 | *** join/#maemo-ssu rd_ (~rd@p4FD1DE37.dip0.t-ipconnect.de) |
| 12:34.59 | Pali | why is git read-tree so sloooow? |
| 12:35.30 | Pali | probably it removes all inodes information from .git/index... |
| 13:00.58 | *** join/#maemo-ssu rd_ (~rd@p4FD1DE37.dip0.t-ipconnect.de) |
| 13:03.29 | bencoh | http://pastebin.notk.org/pastebin.php?show=m597f644e |
| 13:22.54 | Pali | only two structs? |
| 13:23.05 | Pali | this is not too much |
| 13:38.03 | *** join/#maemo-ssu arcean (~arcean@apn-37-248-53-180.dynamic.gprs.plus.pl) |
| 14:08.36 | bencoh | some structs might be missing in this test, but at least it covers the pq64 thing, and a few others |
| 15:07.42 | freemangordon | bencoh: yes, looks good |
| 15:08.39 | freemangordon | except that x509_cinf_st thingie |
| 15:08.59 | bencoh | yeah, dunno where that comes from |
| 15:12.27 | freemangordon | bencoh: does headers differ? |
| 15:12.44 | freemangordon | it is crypto/x509/x509.h |
| 15:15.07 | bencoh | hmm right, looks like they added ASN1_ENCODING enc |
| 15:15.13 | bencoh | couldnt see it in git log :) |
| 15:16.09 | freemangordon | bencoh: https://github.com/openssl/openssl/commit/6cb5746b65a53207bc794121a0c78603bff8ea30 |
| 15:16.19 | bencoh | and hm_fragment_st has unsigned char *reassembly |
| 15:18.19 | freemangordon | bencoh: anyway X509_CINF structure should not be allocated with malloc, there is X509_CINF *X509_CINF_new(); function |
| 15:18.37 | bencoh | hmm, so we should be fine |
| 15:19.33 | bencoh | hm_fragment doesnt seem to be used in any public function (?!) |
| 15:22.12 | freemangordon | I don't see it used at all, weird |
| 15:22.44 | freemangordon | oh |
| 15:22.52 | freemangordon | it is hw_fragment |
| 15:23.27 | freemangordon | yep, seems like internal struct |
| 15:23.59 | freemangordon | bencoh: seems the ABI is the same |
| 15:24.22 | bencoh | hm_* not hw_* |
| 15:24.33 | freemangordon | yeah, a typo |
| 15:24.37 | bencoh | but ... yeah, everything's fine in the end |
| 15:24.37 | freemangordon | :) |
| 15:24.45 | bencoh | I guess I just worried too much :) |
| 15:25.05 | freemangordon | no, it is ok, better be on the safe side with such a critical library |
| 16:29.52 | *** join/#maemo-ssu rd_ (~rd@p4FD1DE37.dip0.t-ipconnect.de) |
| 16:31.58 | *** join/#maemo-ssu xes_ (~xes@unaffiliated/xes) |
| 16:46.42 | *** join/#maemo-ssu LauRoman (~LauRoman@5-14-6-29.residential.rdsnet.ro) |
| 17:13.32 | kerio | freemangordon: do i yolo upgrade |
| 17:14.27 | kerio | are there some reported breakages |
| 17:15.12 | freemangordon | kerio: nothing broken so far |
| 17:15.16 | kerio | alright i found a packaging problem already |
| 17:15.38 | kerio | openssl doesn't depend on libssl0.9.8 >= 0.9.8zf |
| 17:15.38 | freemangordon | I even managed to connect to FB with QWebView |
| 17:15.57 | freemangordon | sure, why it should depend on that particular varion? |
| 17:16.02 | freemangordon | *version |
| 17:16.05 | kerio | because... it's openssl 0.9.8zf |
| 17:16.14 | freemangordon | but, ABI is 0.9.8 |
| 17:16.22 | kerio | k |
| 17:16.24 | freemangordon | not 0.9.8zf |
| 17:17.33 | kerio | freemangordon: there's a new busybox-power apparently |
| 17:18.18 | freemangordon | kerio: I have only 2 hands, still |
| 17:18.28 | freemangordon | and 24hours per day |
| 17:18.43 | freemangordon | deffinitely, openssl is of higher priority |
| 17:18.46 | kerio | freemangordon: http://www.blogcdn.com/www.engadget.com/media/2008/05/5-29-08-carpet-keyboard.jpg |
| 17:19.05 | freemangordon | those are legs, not hands |
| 17:21.16 | kerio | mh, who uploaded libc6 +cssu3? |
| 17:22.08 | freemangordon | merlin1991, afaik |
| 17:22.33 | kerio | why was a non-thumb libc6 in your repo |
| 17:23.33 | merlin1991 | kerio: +cssu3 is in -devel not in -thumb |
| 17:23.41 | kerio | +cssu2 is in -thumb |
| 18:09.21 | freemangordon | kerio: because thumb-compiled seemed to bring ptoblems |
| 18:09.31 | kerio | ;-; |
| 18:09.51 | freemangordon | so I increased the version, ARM-compiled it and put it in the repo |
| 18:11.00 | *** join/#maemo-ssu LauRoman (~LauRoman@5-14-6-29.residential.rdsnet.ro) |
| 18:13.24 | kerio | k |
| 18:32.01 | *** join/#maemo-ssu sparetire_ (~sparetire@unaffiliated/sparetire) |
| 18:39.17 | *** join/#maemo-ssu rd_ (~rd@p4FD1DE37.dip0.t-ipconnect.de) |
| 18:41.05 | *** join/#maemo-ssu arcean (~arcean@apn-95-41-246-32.dynamic.gprs.plus.pl) |
| 18:43.45 | freemangordon | Pali: did you backout linux-n900 on github? |
| 18:43.52 | freemangordon | *backup |
| 18:47.26 | *** join/#maemo-ssu futpib (~futpib@176.104.210.0) |
| 18:53.35 | Pali | freemangordon: week ago... https://github.com/pali/linux-n900 |
| 18:55.35 | freemangordon | Pali: ok |
| 18:56.47 | freemangordon | kerio: http://maemo.merlin1991.at/cssu/community-thumb/pool/free/b/busybox-power/ |
| 18:56.54 | kerio | yey |
| 19:11.19 | *** join/#maemo-ssu freemangordon (~ivo@46.249.74.23) |
| 19:25.49 | *** join/#maemo-ssu M4rtinK (~M4rtinK@ip-37-188-232-242.eurotel.cz) |
| 19:38.27 | *** join/#maemo-ssu M4rtinK (~M4rtinK@ip-37-188-232-242.eurotel.cz) |
| 20:14.42 | *** join/#maemo-ssu arcean (~arcean@apn-95-41-84-22.dynamic.gprs.plus.pl) |
| 20:23.12 | *** join/#maemo-ssu M4rtinK (~M4rtinK@ip-37-188-234-132.eurotel.cz) |
| 20:40.58 | *** join/#maemo-ssu futpib (~futpib@176.104.210.0) |
| 21:04.55 | *** join/#maemo-ssu futpib (~futpib@176.104.210.0) |
| 21:10.25 | *** join/#maemo-ssu futpib_ (~futpib@176.104.210.0) |
| 21:21.36 | *** join/#maemo-ssu dafox (~dafox@ip565ed369.direct-adsl.nl) |
| 22:52.30 | *** join/#maemo-ssu xes (~xes@unaffiliated/xes) |
