Datasets:

showvikdbz

/

code-reviews

like 0

you can delete this main logic as the wrapper py handles it for you

Hi, it looks like you got the Chinese drivers. Please check if the solder bridge is on there. http://marlinfw.org/docs/hardware/tmc_drivers.html (search for fystec) If the two contacts are soldered together the drivers won't work with spi and will report rubbish like short to ground.

remove from here and target path below

tenant id must be nullable false

i guess users is a generic group so it is okay to assign goofy to this group

i agree with yamamoto here

netmask hardcoding is terrible idea i hope that this is temporary solution

this part should be deleted because attributeerror will be thrown

> prob and rand not once but 3 times per step what the fuck no

pdb should be removed

is this a libvirt only change

drive by dlog done makes sense this is totally useless for users

this cleanup probably does not belong here you leave similar garbage intact in other files

eqlx cli timeout is no longer being used and should probably be removed

the phonesubinfoproxy instance will check for dump permission

in fact the shadow charger mob has no spawn ingame wtf (fel steed is a mini boss so it's fine)

you know that had not even occurred to me but now i wants it p

Yes, I mean the VSCode changes being understood by typescript in ridiculous amount of time. It's definitely not only me, we had it several times during mob sessions, but recently I started to feel it even when I'm doing solo dev.

this comment is unnecessary after the refactor and should be removed

hmm okay fair point we should probably call it ffs then which is what linux uses for ffs although it looks terribly ugly

it is a tuple in real life so i guess it should be some tuple here

kwargs dict cluster id cluster id kwargs

Your post is basically information-free, this is barely a step up from "yo this sucks". Last people reporting performance issues solved them by reinstalling Keras / Theano / cuDNN. https://github.com/fchollet/keras/issues/1250

use pls self assertequalobjects

@PJB3005 ah shit, yeah

init pull request: i sleep when merged: real shit

Using an event dual purpose like this is ugly. It should be an event for strap instead.

What is it? Isn't it some MOFED hack?

check self fuel web assert pacemaker below will fail if you destroy self env d env nodes slaves

Oh damn, very good point. I'll look for another solution.

tls disabled should be removed now

use extra specs

please remove white spaces

i recently added gating test section at line please delete it and move test fuel agent section here

makes me think my previous question is in fact a dumb question

Darn, I did this :roll_eyes:

I prefer the lowercase letters. If I add ` m_ps_sel.dither = 1` inside the if, then I would need an else ` m_ps_sel.dither = 0` and kinda looks ugly.

Svalinn sucks by a ton to be a relic weapon

not that it matters but you need a space at the end here duplicated

Shit your approach after description

root helper utils get root helper

nit document the arg true is official build

Ah shit did you forget to set it back to its original brightness here

nit these two temporaries do not add much value

this explict call to rootwrap appears unique in the code base i am not sure which is right but things like cinder brick initiator linuxfc py do it differently so i am querying it here

again i happen to disagree with the naming here retry controllers should be named in a way that makes code that uses them nice and readable example of good name lf flow name retry retry alwaysrevert nice and says all of it lf flow name retry retry retrycounter looks confusing to me i would suggest some stupid but more clear name e g lf flow name retry retry times

isn't this a bit overkill? Runtime should be fuck all compared to other systems. Plus throttling this will actually have effects on player experience

sorry the result of me changing my workflow blindly using git add all git commit vs git gui

Oh shit, all of them? Can I attach the singulo directly to my person? That'd be neat.

ports synchronized common utils create synchronized decorator n hv driver

zopfli is used by update assets sh to compress the assets when they change

please remove white spaces

i was confused at first then I realized the parentheses oh fuck me

this should be defined on the command line upstream this is handled by handlellvmoptions cmake

oh ick this needs to die asap

leaving alone for now since it does not make much sense to propagate such a tristate up to the superclass

please convert these tabs into spaces

need to remove the white space around execute putils execute

Yuck! Maybe make that into a loop?

Works on PC. This is almost certainly one of those wacky Adreno stencil/discard driver bugs. Which GPU does your device have again?

you're shitting me i forgot to remove it AGAIN

ah i had tried that but got various build errors regarding libprotobuf mutator so was not sure whether that cl is supposed to build at all here s the error fwiw it s possible i am doing something stupid i am not too familiar with platform ec cr e b f mnissler toroa trunk src platform ec board cr make j buildfuzztests build host cr fuzz build host host command fuzz make build host host command fuzz host command fuzz exe is up to date make obj mnt host source src platform ec build host cr fuzz cryptoc support unaligned config upto sha y c mnt host source src third party cryptoc make mnt host source src platform ec build host cr fuzz cryptoc libcryptoc a is up to date cxx rw fuzz cr fuzz o fuzz cr fuzz cc fatal error src libfuzzer libfuzzer macro h file not found include src libfuzzer libfuzzer macro h error generated make makefile rules build host cr fuzz rw fuzz cr fuzz o error make makefile rules host cr fuzz error

nit these migth be the first check in the routine i think we will hit this of cases

Lord this is disgusting, can you add a TODO for me to fix it.

why create a function instead of just using readlink script dir readlink f dirname

all of those if s would be better as if flags whatever

oh shit sorry this is left over, ill remove it

ok so it was very obvious still dumb people like me might wonder what the format is

unrelated change these few lines are just dumb

this seems rather dangerous to me actually if we start expecting that the display name that we give to the instance is in dns and we do not fail the operation if we cannot satisfy that then users have the potential to do insecure things without realizing for example nova boot foo ssh u myuser foo provider com accept host key y ssh password for myuser foo provider com permission denied ssh password for myuser foo provider com and thus i have just given my password to someone who guessed the hostname that i wanted unless we expose the actual hostname back to the nova user somehow we cannot really blame someone for committing the above offense imho which is pretty bad

Version 5.0.5 still broken ... ridiculous

after a while i have understand what you whant to do

about to wrap this PR up, gonna set an engi/atmos access requirement on multitool menus and call it a day(seriously nobody wants an assistant breaking EVERY. DAMN. AIR. VENT.)

i think we should be consistent and use the check create complete instead of eventlet sleep

Sucks that we got cornered into doing `.Result`, would it be feasible to make this `TryGetGeneratedOutputAsync`?

this should be moved to iscsideployprivatemethodstestcase and delete iscsideployvalidateparameterstestcase

Farken Travis...I give up this is ridiculous.

Never hit, would look ugly as hell even if it did.

>pretty compelling system That's a pretty funny way of spelling fucking awful.

Actually no, it's still WTF! Because it's suppose to be the hash of the block not hash of the header.

Wth are you talking about @SergeGautherie ?

this seems like dirty hack extensions shall be deterministic

Fuck off. No one actually shoots that far.

silly me possibly something we want our linter to catch

get the status of slave builders to a master

This is absolutely horrific bug = you create a workflow from a history and most of your parameters are lost, so the workflow is unusuable

how embarrassing since I wrote the darn tool lol

return here is useless

I see. In that case I would categorize this as, we had a breaking change that we communicated poorly. Which sucks - I'm sorry. We are working on doing this better with gathering all breaking changes into one message, and releasing a bit less often (monthly).

i think ubsan works for static executables but we do not support ubsan in sanitize target sgtm

this should work as well return value hostname foo host

i think you should mention about your puppet glance change in commit message

wtf is this can you please get rid of it

As I said, I left that there to plug-in your hack. Or if really really any you don't want it, we'd still have to warn user they are screwed below that version.

i think you mistakenly dropped the compute from taking the midonet data so this will not be passed here

I believe I dealt with the merge confict, but LMK if I screwed it up!

Ehm.. dude wtf be careful and remove the previous step also and change the number of steps ...

do NOT use data holy FUCK data is B A D use the dedicated "how many times have I ticked" var holy FUCK BALLS SHIT AAAAAAAA

Actually, skip this suggestion. It looks crap :P

Do you wanna tidy that up, that looks weird now. Thanks Clang :P