Home
release-team@conference.openafs.org
Friday, February 2, 2018< ^ >
Room Configuration
Room Occupants

GMT+0
[13:59:41] Marcio Barbosa joins the room
[14:00:29] meffie joins the room
[14:00:51] <meffie> good morning/afternoon
[14:01:03] <kaduk@jabber.openafs.org/barnowl> greetings
[14:01:41] <Marcio Barbosa> good morning
[14:02:28] wiesand joins the room
[14:02:33] <wiesand> Hi
[14:02:58] <wiesand> So, shall we go ahead with the 1.6.22.2 release?
[14:03:43] <kaduk@jabber.openafs.org/barnowl> It seems so.
[14:04:21] <meffie> that for putting together 1.6.22.2, sorry i wasnt able to give it a try yesterday; was in meetings all day.
[14:05:29] <wiesand> Ok, let's do it. Ben could you please push the tag?
[14:06:10] <meffie> thanks.
[14:06:13] <wiesand> Bringing live the web change now.
[14:06:15] <kaduk@jabber.openafs.org/barnowl> To ssh://gerrit.openafs.org/openafs.git
* [new tag]             openafs-stable-1_6_22_2 -> openafs-stable-1_6_22_2
[14:06:58] <wiesand> And I prepared the announcement using you suggested description of EL7.4 getcwd(). Sending…
[14:07:42] <wiesand> Sent.
[14:08:14] <wiesand> Pestering the moderators in a moment should be all that's left to do. (the volumes were released earlier)
[14:09:58] <wiesand> Pestered
[14:10:36] <meffie> yay
[14:10:59] <kaduk@jabber.openafs.org/barnowl> yay
[14:11:08] <kaduk@jabber.openafs.org/barnowl> The release process is dead; long-live the release process
[14:11:18] <wiesand> I also did the pullups I promised a week ago today. Shame on me.
[14:11:27] <wiesand> Did a few more too though.
[14:11:56] <kaduk@jabber.openafs.org/barnowl> I saw those; thanks
[14:12:09] <kaduk@jabber.openafs.org/barnowl> Though for the freebsd param.h ones it is a little tricky
[14:12:11] <wiesand> Unfortunately the Windows builders are rather slow, thus I'v clogged the queues for the rest of the day I guess.
[14:12:18] <wiesand> Ah?
[14:12:46] <kaduk@jabber.openafs.org/barnowl> I guess that's only for the 1.8->1.6 jump, though -- we should still
have the fbsd_generic.h version in 1.8 as well as master
[14:13:56] <wiesand> Ok, I'll try to keep that in mind.
[14:15:18] <wiesand> Marcio, thanks for your pullups.
[14:15:21] <kaduk@jabber.openafs.org/barnowl> I tend to look at the diff from the previous freebsd version's header to
the one being added; there should just be a few added _ENV lines and the
different sysname+id
[14:15:40] <wiesand> In general, we should cherry-pick from 1.8 though.
[14:16:19] <Marcio Barbosa> oh
[14:18:19] <Marcio Barbosa> should i update the commit message?
[14:18:49] <wiesand> Let's see… some of those will need to be rebased anyway I think.
[14:19:55] <wiesand> While we had some amount of review on the ubik stack so far…Ben: I'd at least like to hear whether you consider it suitable for 1.6 at all.
[14:20:38] <kaduk@jabber.openafs.org/barnowl> I think the intent was to get it on 1.6, yes
[14:20:53] <wiesand> Ok, thanks.
[14:21:56] <wiesand> I'll start merging the trivial changes first anyway, to make things a bit more wieldy. That should happen soon.
[14:22:05] <kaduk@jabber.openafs.org/barnowl> Sounds good.
[14:22:29] <wiesand> And pre1 will be ~ a week later than I hoped last week. Sorry.
[14:23:00] <kaduk@jabber.openafs.org/barnowl> Things happen.
[14:23:26] <wiesand> Other than that, I think the (my) hot topic right now is the RHEL 7.5 client
[14:23:49] <meffie> yes, redhat.
[14:24:41] <kaduk@jabber.openafs.org/barnowl> I am so divorced from the red hat world that I don't even know if I can
legally get my hands on the OS image in question to look at it :-/
[14:24:57] <wiesand> I'll look into it a bit more, but I'm afraid this is beyond my skills to fix.
[14:25:50] <kaduk@jabber.openafs.org/barnowl> *nods*
[14:26:08] <kaduk@jabber.openafs.org/barnowl> Can you post the config.log entry for the has-iterate test?
[14:26:15] <kaduk@jabber.openafs.org/barnowl> (You probably did already, I guess)
[14:26:30] <wiesand> No I didn't
[14:26:56] <meffie> mvita should be able to help late next week
[14:27:57] <wiesand> It looks completely unsuspicious…
[14:28:42] <wiesand> configure:12229: checking for iterate in struct file_operations
make -C /usr/src/kernels/3.10.0-830.el7.x86_64 M=/tmp/2/openafs-1.6.22.2/conftest.dir modules KBUILD_VERBOSE=1
make: Entering directory `/usr/src/kernels/3.10.0-830.el7.x86_64'
test -e include/generated/autoconf.h -a -e include/config/auto.conf || (                \
echo >&2;                                                       \
echo >&2 "  ERROR: Kernel configuration is invalid.";           \
echo >&2 "         include/generated/autoconf.h or include/config/auto.conf are missing.";\
echo >&2 "         Run 'make oldconfig && make prepare' on kernel src to fix it.";      \
echo >&2 ;                                                      \
/bin/false)
mkdir -p /tmp/2/openafs-1.6.22.2/conftest.dir/.tmp_versions ; rm -f /tmp/2/openafs-1.6.22.2/conftest.dir/.tmp_versions/*
make -f scripts/Makefile.build obj=/tmp/2/openafs-1.6.22.2/conftest.dir
  gcc -Wp,-MD,/tmp/2/openafs-1.6.22.2/conftest.dir/.conftest.o.d  -nostdinc -isystem /usr/lib/gcc/x86_64-redhat-linux/4.8.5/include -I./arch/x86/include -Iarch/x86/include/generated  -Iinclude -I./arch/x86/include/uapi -Iarch/x86/include/generated/uapi -I./include/uapi -Iinclude/generated/uapi -include ./include/linux/kconfig.h -D__KERNEL__ -Wall -Wundef -Wstrict-prototypes -Wno-trigraphs -fno-strict-aliasing -fno-common -Werror-implicit-function-declaration -Wno-format-security -fno-delete-null-pointer-checks -std=gnu89 -O2 -m64 -mno-mmx -mno-sse -mpreferred-stack-boundary=3 -mtune=generic -mno-red-zone -mcmodel=kernel -funit-at-a-time -maccumulate-outgoing-args -Wframe-larger-than=2048 -DCONFIG_AS_CFI=1 -DCONFIG_AS_CFI_SIGNAL_FRAME=1 -DCONFIG_AS_CFI_SECTIONS=1 -DCONFIG_AS_FXSAVEQ=1 -DCONFIG_AS_AVX=1 -DCONFIG_AS_AVX2=1 -DCONFIG_AS_AVX512=1 -DCONFIG_AS_SHA1_NI=1 -DCONFIG_AS_SHA256_NI=1 -pipe -Wno-sign-compare -fno-asynchronous-unwind-tables -mno-sse -mno-mmx -mno-sse2 -mno-3dnow -mno-avx -Wframe-larger-than=2048 -fstack-protector-strong -Wno-unused-but-set-variable -fno-omit-frame-pointer -fno-optimize-sibling-calls -g -pg -mfentry -DCC_USING_FENTRY -fno-inline-functions-called-once -Wdeclaration-after-statement -Wno-pointer-sign -fno-strict-overflow -fconserve-stack -DCC_HAVE_ASM_GOTO  -DMODULE  -D"KBUILD_STR(s)=#s" -D"KBUILD_BASENAME=KBUILD_STR(conftest)"  -D"KBUILD_MODNAME=KBUILD_STR(conftest)" -c -o /tmp/2/openafs-1.6.22.2/conftest.dir/.tmp_conftest.o /tmp/2/openafs-1.6.22.2/conftest.dir/conftest.c
  if [ "-pg" = "-pg" ]; then if [ /tmp/2/openafs-1.6.22.2/conftest.dir/conftest.o != "scripts/mod/empty.o" ]; then ./scripts/recordmcount  "/tmp/2/openafs-1.6.22.2/conftest.dir/conftest.o"; fi; fi;
(cat /dev/null;   echo kernel//tmp/2/openafs-1.6.22.2/conftest.dir/conftest.ko;) > /tmp/2/openafs-1.6.22.2/conftest.dir/modules.order
make -f ./scripts/Makefile.modpost
  find /tmp/2/openafs-1.6.22.2/conftest.dir/.tmp_versions -name '*.mod' | xargs -r grep -h '\.ko$' | sort -u | sed 's/\.ko$/.o/' | scripts/mod/modpost -m -a -i ./Module.symvers -I /tmp/2/openafs-1.6.22.2/conftest.dir/Module.symvers  -o /tmp/2/openafs-1.6.22.2/conftest.dir/Module.symvers  -w -s -T -
  gcc -Wp,-MD,/tmp/2/openafs-1.6.22.2/conftest.dir/.conftest.mod.o.d  -nostdinc -isystem /usr/lib/gcc/x86_64-redhat-linux/4.8.5/include -I./arch/x86/include -Iarch/x86/include/generated  -Iinclude -I./arch/x86/include/uapi -Iarch/x86/include/generated/uapi -I./include/uapi -Iinclude/generated/uapi -include ./include/linux/kconfig.h -D__KERNEL__ -Wall -Wundef -Wstrict-prototypes -Wno-trigraphs -fno-strict-aliasing -fno-common -Werror-implicit-function-declaration -Wno-format-security -fno-delete-null-pointer-checks -std=gnu89 -O2 -m64 -mno-mmx -mno-sse -mpreferred-stack-boundary=3 -mtune=generic -mno-red-zone -mcmodel=kernel -funit-at-a-time -maccumulate-outgoing-args -Wframe-larger-than=2048 -DCONFIG_AS_CFI=1 -DCONFIG_AS_CFI_SIGNAL_FRAME=1 -DCONFIG_AS_CFI_SECTIONS=1 -DCONFIG_AS_FXSAVEQ=1 -DCONFIG_AS_AVX=1 -DCONFIG_AS_AVX2=1 -DCONFIG_AS_AVX512=1 -DCONFIG_AS_SHA1_NI=1 -DCONFIG_AS_SHA256_NI=1 -pipe -Wno-sign-compare -fno-asynchronous-unwind-tables -mno-sse -mno-mmx -mno-sse2 -mno-3dnow -mno-avx -Wframe-larger-than=2048 -fstack-protector-strong -Wno-unused-but-set-variable -fno-omit-frame-pointer -fno-optimize-sibling-calls -g -pg -mfentry -DCC_USING_FENTRY -fno-inline-functions-called-once -Wdeclaration-after-statement -Wno-pointer-sign -fno-strict-overflow -fconserve-stack -DCC_HAVE_ASM_GOTO  -D"KBUILD_STR(s)=#s" -D"KBUILD_BASENAME=KBUILD_STR(conftest.mod)"  -D"KBUILD_MODNAME=KBUILD_STR(conftest)" -DMODULE  -c -o /tmp/2/openafs-1.6.22.2/conftest.dir/conftest.mod.o /tmp/2/openafs-1.6.22.2/conftest.dir/conftest.mod.c
  ld -r -m elf_x86_64 -T ./scripts/module-common.lds --build-id  -o /tmp/2/openafs-1.6.22.2/conftest.dir/conftest.ko /tmp/2/openafs-1.6.22.2/conftest.dir/conftest.o /tmp/2/openafs-1.6.22.2/conftest.dir/conftest.mod.o
make: Leaving directory `/usr/src/kernels/3.10.0-830.el7.x86_64'
configure:12329: result: yes
[14:29:46] <kaduk@jabber.openafs.org/barnowl> And you said in mail there was a RH_KBI_COMPAT (mumble) conditional
around the field in the header?  Which way does the conditional go?
[14:30:10] <wiesand> KABI_EXTEND
[14:30:13] <kaduk@jabber.openafs.org/barnowl> It's possible they left the field in so as to not change structure
offsets but it is now unused in favor of some other mechanism
[14:31:00] <wiesand> But it seems that's not a conditional. Just a macro to allow adding to interfaces without changing the KABI hash…
[14:31:14] <wiesand> hang on
[14:32:02] <wiesand> * RH_KABI_EXTEND - simple macro for adding a new element to a struct while
*                  preserving the kabi agreement (by wrapping with GENKSYMS).
[14:34:22] <wiesand> There are some more changes that may well have an impact on us, like renaming vfs_readdir to dir_iterate
[14:35:18] <kaduk@jabber.openafs.org/barnowl> Okay.  Well, we should probably not try to drill into it in the middle
of the meeting.
(After, maybe.)
[14:35:49] <wiesand> (after, I'll start enjoying my weekend ;-)
[14:36:21] <kaduk@jabber.openafs.org/barnowl> The hidden risks of the new meeting slot...
[14:37:23] <wiesand> Well it's not really urgent… 7.5 GA will take at least 2 months I think.
[14:37:35] <meffie> ok, that's good news.
[14:38:01] <wiesand> (maybe I shouldn't have said that ;-)
[14:38:06] <meffie> ha!
[14:38:40] <wiesand> So that was about my agenda for today.
[14:39:31] <wiesand> On to 1.8? Master?
[14:39:48] <kaduk@jabber.openafs.org/barnowl> I am afraid that I have to start off with some bad news:
[14:40:12] <wiesand> ok…
[14:40:26] <kaduk@jabber.openafs.org/barnowl> I have been selected for a two-year term as one of the IETF Security
Area Directors, which is described as being at least a 20 hour/week
commitment, with no upper bound on the amount of time that can be
involved.
[14:40:48] <meffie> congradulations
[14:40:51] <kaduk@jabber.openafs.org/barnowl> So, there will be more things fighting for my time and things might go
slower for openafs on my end.
[14:40:58] <wiesand> yes, congratulations
[14:41:21] <kaduk@jabber.openafs.org/barnowl> Thanks :)
[14:42:20] <wiesand> It's already relieving that you didn't say you'd quit or take a complete sabbatical
[14:42:30] <meffie> i agree.
[14:42:37] <kaduk@jabber.openafs.org/barnowl> Hah!
[14:43:07] <meffie> and maybe this is good motivation to recruit more people :)
[14:43:13] <kaduk@jabber.openafs.org/barnowl> Anyway, on the 1.8 front, Stephan and Mike have pulled up a bunch of
things in gerrit, so thank you
[14:43:31] <kaduk@jabber.openafs.org/barnowl> IIRC, it includes all the stuff mentioned two weeks ago, but I did not
explictily double-check.
[14:44:04] <wiesand> I think CellServDB update is new, but…
[14:44:35] <wiesand> NB I put it on top of the redhat changes (path conflict…) so you'll have to merge those first
[14:44:58] <kaduk@jabber.openafs.org/barnowl> Good to know
[14:45:46] <wiesand> The rfc3961.h think is new too I believe
[14:46:08] <kaduk@jabber.openafs.org/barnowl> Yeah, but both are clearly useful/needed
[14:46:39] <wiesand> And yes, the rest was discussed last week.
[14:47:27] <kaduk@jabber.openafs.org/barnowl> So, assuming it all gets reviewed and checks out, is it time for a pre5?
[14:47:41] <wiesand> It seems so
[14:48:08] <kaduk@jabber.openafs.org/barnowl> I think we heard of the RHEL issues with pre4, but 1.6.x also has those,
and it is unclear if we should wait for a fix, since RHEL (and linux in
general) can introduce new breakage at arbitrary times.
[14:48:54] <meffie> i should mention, andrew has been talking to dhowells re the rx packet thing (rt 133446)
[14:48:55] <kaduk@jabber.openafs.org/barnowl> Some traffic on IRC reminded me of 11629, an oldie but maybe goodie
[14:49:14] <wiesand> Unless it's clearly a regression, I wouldn't delay pre5 for it
[14:50:05] <wiesand> Ah, that one.. (11629)
[14:50:35] <meffie> oh, 11629 yes. hrmph.
[14:51:00] <kaduk@jabber.openafs.org/barnowl> It is arguably too late in the release process for it to go in.
And it's a pretty cleanly separtaed new feature that could go in a 1.8.1
(or a 1.6.x even), so it doesn't really feel right for me to try to
force it into 1.8.0.
[14:51:14] <meffie> agreed
[14:51:41] <kaduk@jabber.openafs.org/barnowl> But it is kind of a shame that our defaults are so bad
[14:51:48] <meffie> yes
[14:52:05] <kaduk@jabber.openafs.org/barnowl> Is anyone set up to try and do some testing on a scheme like this?
[14:52:10] <meffie> to bad we didnt just implement -xxxlarge :)
[14:52:23] <wiesand> +1 on (could go in 1.8.1 or even 1.6.x)
[14:53:46] <wiesand> Ben: I'd be, but I won't promise more than I can for the next week….
[14:54:12] <kaduk@jabber.openafs.org/barnowl> Understood.
[14:54:25] <kaduk@jabber.openafs.org/barnowl> And since we're agreed on 1.8.1, it can wait a few weeks.
[14:55:04] <kaduk@jabber.openafs.org/barnowl> Anyone want to put together a NEWS update and version-number bump for pre5?
[14:55:12] <meffie> i will.
[14:55:20] <wiesand> applause
[14:55:26] <kaduk@jabber.openafs.org/barnowl> Thanks!
[14:56:24] <wiesand> What will happen after the 1.8.0 release? Shall I take over responsibility for the 1.8.x branch then?
[14:56:44] <kaduk@jabber.openafs.org/barnowl> I was going to ask you in email if you would be willing to do that.
But since you asked here...
[14:57:24] <kaduk@jabber.openafs.org/barnowl> I would find it very helpful if you could take over the new stable
branch, yes.  But you should not feel pressured to do so.
[14:58:00] <wiesand> It would seem quite natural to me.
[14:58:18] <meffie> wonderful news!
[14:58:24] <kaduk@jabber.openafs.org/barnowl> Excellent; thank you!
[14:59:21] <wiesand> So, unless anyone else objects, fine.
[14:59:36] <kaduk@jabber.openafs.org/barnowl> On to master, then?
[14:59:47] <meffie> ok
[15:00:18] <kaduk@jabber.openafs.org/barnowl> It seems someone has gotten interested in the aarch64 stuff again, so I
expect we can get that in, now that master is unfrozen for 1.8
[15:00:39] <meffie> hooray!
[15:02:25] <meffie> shall i rebase them? i work on getting an aarch64 buildbot.
[15:02:52] <kaduk@jabber.openafs.org/barnowl> Ian already did rebase them
[15:03:41] <meffie> oh, great!
[15:04:13] <kaduk@jabber.openafs.org/barnowl> Though it looks like I left some comments on 11937 that are not yet
addressed
[15:04:23] <meffie> ok, i'll see about a nightly buildbot (its too slow for a gerrit builder)
[15:05:01] <kaduk@jabber.openafs.org/barnowl> *nods*
[15:05:12] <kaduk@jabber.openafs.org/barnowl> Do you want to look at 11937 as well?
[15:05:38] <meffie> ok, thanks for the reminder on 11937. i'll ask andrew to look
[15:05:42] <kaduk@jabber.openafs.org/barnowl> Okay
[15:06:01] <kaduk@jabber.openafs.org/barnowl> Ian also brought up the existence of the 'git review' add-on command.
[15:06:39] <kaduk@jabber.openafs.org/barnowl> I'm okay with adding the configuration for it to the tree if (n > 2?) people are
actively using it, but I don't know if anyone (other than Ian) is.
[15:08:23] <wiesand> Would make a nice gesture towards a new contributor…
[15:08:24] <meffie> i believe those patches are in use
[15:09:43] <kaduk@jabber.openafs.org/barnowl> So that's a +1?  Or maybe a +0.5?
[15:11:10] <wiesand> +0.5 from me
[15:11:38] <meffie> sorry need to run off to another meeting.
[15:12:07] <kaduk@jabber.openafs.org/barnowl> Let's see, other master stuff...
solaris vcache locking stuff that I've been meaning to look at for too
long
new rev of the ctf stuff, thanks Marcio -- I think we're approaching
ready on that one
Anything interesting to say about 12879 (vos zap -force + empty
directories)?
[15:12:11] <kaduk@jabber.openafs.org/barnowl> Okay, thanks Mike
[15:13:03] <wiesand> Mike said last week: btw, the vos zap stuff is not too urgent in my opinion. i dont think those should block 1.8.0
[15:13:22] <Marcio Barbosa> yeah
[15:13:30] <Marcio Barbosa> we have a workaround
[15:13:44] <Marcio Barbosa> we have to run vos zap -force twice to remove the left overs
[15:14:17] <Marcio Barbosa> left overs == empty directories
[15:14:29] <kaduk@jabber.openafs.org/barnowl> Ah.
[15:14:40] <wiesand> When you need to vos zap at all, you probably are having more serious troubles than that…
[15:14:56] <kaduk@jabber.openafs.org/barnowl> Pretty much...
[15:15:09] <Marcio Barbosa> specially with -force
[15:16:57] <wiesand> Is there more to discuss today?
[15:18:36] <kaduk@jabber.openafs.org/barnowl> I don't think there's anything else much.
I didn't have any time to touch rxgk.
[15:18:47] <kaduk@jabber.openafs.org/barnowl> Maybe I should take Andrew up on his offer to do some of the cleanup
sweeps
[15:19:11] <wiesand> If he'd do that…
[15:19:40] <wiesand> I'll call it a day then. Let's adjourn.
[15:19:56] <wiesand> Thanks a lot Ben & Marcio (&Mike)!
[15:20:14] wiesand leaves the room
[15:20:14] <kaduk@jabber.openafs.org/barnowl> Thanks everyone!
[17:45:04] meffie leaves the room
[17:57:16] meffie joins the room
[21:18:48] Marcio Barbosa leaves the room
[21:18:48] Marcio Barbosa joins the room
[21:28:57] Marcio Barbosa leaves the room
[22:24:43] meffie leaves the room
Powered by ejabberd Powered by Erlang Valid XHTML 1.0 Transitional Valid CSS!