|
wait this is a quote not edit snipe! noooooooooooooooooo
|
#
?
Oct 20, 2019 15:54
|
|
|
|
| # ? Apr 29, 2024 11:05 |
|
|
chasing down bugs like a moron keeps me employed, op
|
|
#
?
Oct 20, 2019 16:09
|
|
|
you shouldn’t really need an unoptimized executable for debugging, obviously a lot of things about debug info will degrade under optimization but you can generally still piece together what‘s happening, and being able to step and inspect memory is a lifesaver. the mgmt is right, there’s no excuse for not having a debugserver implementation on your device
|
|
#
?
Oct 20, 2019 16:16
|
|
|
admittedly i have only tried step-debugging a release build on hardware once, but it didn't go well either maybe if I get some spare time I'll try again and get that worked out 
|
|
#
?
Oct 20, 2019 16:18
|
|
|
i feel like i had a breakthrough the other day, instead of looping through a code structure that segfaults, you can just let it fail and then reverse until it's back in the code you wrote.
|
|
#
?
Oct 20, 2019 16:22
|
|
|
Ciaphas posted:realpost: tldr, but suffice to say MTP is seemingly impossible to get working, Hardware took the UART out of the "newest" revision, whatever module provides ActiveSync doesn't work reliably for debugging via IP, and even if by some miracle it DOES all work, the machine is too gutless to run a debug executable at anything other than unreasonable treacle-like speeds. we're on vs2005 for that so when it does work it at least works well, but yeesh what a fuckin farce to get there and god help me if what i need to test is buried behind slow-running actions right, activesync not mtp. it’s been a long time since I’ve winced. my experience with it has been pretty good debugging-wise, but I suppose it depends on the BSP. golang? enjoy your giant binaries.
|
|
#
?
Oct 20, 2019 16:51
|
|
|
i;m drunk and thinking about debugging, op
|
|
#
?
Oct 20, 2019 17:04
|
|
|
the race condition doesn't happen anymore when you attach a debugger
|
|
#
?
Oct 20, 2019 17:31
|
|
|
suffix posted:the race condition doesn't happen anymore when you attach a debugger these are the worst. someone in the TP thread told me about intel inspector and that thing is really drat slick at finding these and memory problems Ciaphas fucked around with this message at 17:43 on Oct 20, 2019 |
|
#
?
Oct 20, 2019 17:38
|
|
|
The Management posted:right, activesync not mtp. it’s been a long time since I’ve winced. my experience with it has been pretty good debugging-wise, but I suppose it depends on the BSP. i've been warned lots about golang but tbh it looks overall a lot less foot-shooty than C++ has been in my career so I'm all for it (no generics is a fuckin' grumble tho)
|
|
#
?
Oct 20, 2019 17:40
|
|
|
also it's not printf it's NKDbgPrintfW, god 
|
|
#
?
Oct 20, 2019 17:42
|
|
|
hifi posted:i feel like i had a breakthrough the other day, instead of looping through a code structure that segfaults, you can just let it fail and then reverse until it's back in the code you wrote. using a debugger is like solving a maze by starting from the end. break when an error happens, figure out how you got there instead of figuring out all of the possible ways to get there
|
|
#
?
Oct 20, 2019 17:47
|
|
|
suffix posted:the race condition doesn't happen anymore when you attach a debugger debuggers are fairly innocuous for things like that. you generally add some breakpoints at the start and then run. printfs, on the other hand, can have massive implications on timing as well as changing code generation to hide some bugs.
|
|
#
?
Oct 20, 2019 17:49
|
|
|
also prints to the same log often literally synchronize threads that doesn’t actually eliminate races, but it does make them much harder to find because they don’t reproduce unless specific segments between logging statements are concurrent as opposed to the much longer sequences between the actual useful synchronization points
|
|
#
?
Oct 20, 2019 18:20
|
|
|
jimmyjams posted:what is code code is like a program only useless
|
|
#
?
Oct 21, 2019 11:21
|
|
|
I use the debugger as much as I can but there are a lot of situations where the debugger isn't practical at all like if a bug is caused by the interactions of several clients in a multiplayer game your only way to figure it out is logs and perhaps a frame by frame replay of a video capture of the bug (possibly with debug prints/draw enabled)
|
|
#
?
Oct 21, 2019 14:14
|
|
|
i learned to use print() in my intro to python course on datacamp, are you telling me datacamp dont know what theyre talking about?
|
|
#
?
Oct 21, 2019 15:17
|
|
|
Boiled Water posted:current employer doesn't want us writing log files. "Everything must go through AWS CloudWatch, so says the SOP!" you can do both also cloud watch is pretty clunky
|
|
#
?
Oct 21, 2019 15:18
|
|
|
Boiled Water posted:current employer doesn't want us writing log files. "Everything must go through AWS CloudWatch, so says the SOP!" if it makes you feel any better, cloudwatch probably won't be helpful even when it is configured
|
|
#
?
Oct 21, 2019 16:59
|
|
|
you can use either / or \ as path separators // and \\ are escape sequences that let you include a / or a \ in a filename or directory name
|
|
#
?
Oct 21, 2019 17:35
|
|
|
current debugging environment: DOS 2.1 DEBUG.COM on a 128-KB IBM PCjr 
|
|
#
?
Oct 21, 2019 17:47
|
|
|
one of the last teapot threads before he was permabanned was where he asserted that needing to use a debugger on your code was a sign of weakness and being a bad programmer.
|
|
#
?
Oct 21, 2019 17:52
|
|
|
something like "You should know what your code is doing, you wrote it. You should not need a debugger if you've done your job"
|
|
#
?
Oct 21, 2019 17:53
|
|
|
i saw teapot irl on BART once. I herded my children behind me, they were so young, I didn't want them trolled.
|
|
#
?
Oct 21, 2019 17:54
|
|
|
rotor posted:one of the last teapot threads before he was permabanned was where he asserted that needing to use a debugger on your code was a sign of weakness and being a bad programmer. makes sense.
|
|
#
?
Oct 21, 2019 18:10
|
|
|
rotor posted:one of the last teapot threads before he was permabanned was where he asserted that needing to use a debugger on your code was a sign of weakness and being a bad programmer. technically correct; all programs and programmers are bad
|
|
#
?
Oct 21, 2019 19:00
|
|
|
printf("%i",printf(size_of(int))); am I doing this right
|
|
#
?
Oct 21, 2019 19:12
|
|
|
so what if i want to print things other than 'f' though? what am i supposed to do then? that's why javascript is better, you can just console.log any kind of character or string.
|
|
#
?
Oct 21, 2019 19:14
|
|
|
rotor posted:so what if i want to print things other than 'f' though? what am i supposed to do then? that's why javascript is better, you can just console.log any kind of character or string. thats why i moved to node.js , to be closer to the metal
|
|
#
?
Oct 21, 2019 21:13
|
|
|
printf(“%s:%u\n”, __FUNCTION__, __LINE__); repeated a thousand times in your file
|
|
#
?
Oct 21, 2019 21:15
|
|
|
ANAmal.net posted:if it makes you feel any better, cloudwatch probably won't be helpful even when it is configured it's amazingly bad. We have it set up for logging output of choice select ECS jobs but you can't actually find the logs if you close the tab or an hour passes and you're forced to re-auth.
|
|
#
?
Oct 21, 2019 21:36
|
|
|
Zlodo posted:you can use either / or \ as path separators i meant to post this in the yosos thread btw, it isn't my secret printf based debugging techinque
|
|
#
?
Oct 21, 2019 22:02
|
|
|
Zlodo posted:i meant to post this in the yosos thread btw, it isn't my secret printf based debugging techinque jeez, post much?
|
|
#
?
Oct 21, 2019 22:20
|
|
|
josh2112 posted:current debugging environment: DOS 2.1 DEBUG.COM on a 128-KB IBM PCjr I remember using debug.com to create small executables by copying hexadecimal numbers from a book 
|
|
#
?
Oct 21, 2019 22:45
|
|
|
Ciaphas posted:technically correct; all programs and programmers are bad 
|
|
#
?
Oct 23, 2019 01:33
|
|
|
Ciaphas posted:technically correct; all programs and programmers are bad mine are all good, don't know what you're talking about.
|
|
#
?
Oct 23, 2019 01:37
|
|
|
akadajet posted:mine are all good, don't know what you're talking about. ah, a non-programmer
|
|
#
?
Oct 23, 2019 01:40
|
|
|
what is a program
|
|
#
?
Oct 23, 2019 02:38
|
|
|
jimmyjams posted:what is a program no much, what's program with you?
|
|
#
?
Oct 23, 2019 18:20
|
|
|
|
| # ? Apr 29, 2024 11:05 |
|
|
hahahahaha
|
|
#
?
Oct 23, 2019 18:21
|
|