118 entries in 0.544s
bvt: diana_coman: ty for spotting this. i will regring vpatches p.1 and p.2; i wanted to make the vpatch p.1 name the same in manifest and
file system, but did the wrong thing there just editing the line from previous vpatch in vpatch p.2.
diana_coman: and given the two-class
system those are effectively priorities: at every ask-opportunity, the Requester will choose first object request and only second
file request (those really are the ONLY two types of questions the client may ask the server)
a111: Logged on 2019-05-01 07:14 mp_en_viaje: wherein is a text
file : "for linux users, please download the driver for your
system from our website".
mp_en_viaje: wherein is a text
file : "for linux users, please download the driver for your
system from our website".
☟︎ phf: trinque: i'm going to play with some link combinations, but perhaps it would be worthwhile to at least check if the named
file exists on the
system after failed genesis production.
BingoBoingo: I suspect
file system corruption is more likely. This may mean hardware failure or it could mean an accumulation of small errors piling up during those power cycles.
a111: Logged on 2018-10-24 19:57 phf:
http://btcbase.org/log/2018-10-23#1865503 << i threw your patch on btcbase, it looks good, though i'm not sure i agree with the decision to put temp
file in /tmp. the point of putting it in same hierarchy as press, was to avoid the whole cross-
file-
system issue
phf:
http://btcbase.org/log/2018-10-23#1865503 << i threw your patch on btcbase, it looks good, though i'm not sure i agree with the decision to put temp
file in /tmp. the point of putting it in same hierarchy as press, was to avoid the whole cross-
file-
system issue
☝︎☟︎ a111: Logged on 2018-10-16 11:44 mircea_popescu:
http://btcbase.org/log/2018-10-15#1863111 << thinking about this in the interim, it is EVIDENTLY the case. "files" as in ~the fucking thing REFERENCED~ in the "
file"
system names aren't even first class abstractions! the drive's a BYTE device! there's no fucking files ANYWHERE except in the imagination.
mircea_popescu: what passes for a "
file system" currently is this attempt to force a byte-device (sometimes with some half-baked geographical allocation bolted on) through a half-baked database model. wtf is a "FAT" or a "journal" besides poorly implemented mysql/postgres dben ?
a111: Logged on 2018-10-15 19:24 mircea_popescu: s a
file system AT ALL, and the "all things are files" is just a thin wraper on a turd sausage made out of "we have no data model beyond ram and our disk doesn't actually work".
mircea_popescu:
http://btcbase.org/log/2018-10-15#1863111 << thinking about this in the interim, it is EVIDENTLY the case. "files" as in ~the fucking thing REFERENCED~ in the "
file"
system names aren't even first class abstractions! the drive's a BYTE device! there's no fucking files ANYWHERE except in the imagination.
☝︎☟︎ mircea_popescu: so the situation may actually be that no computer
system currently deployed actually has a
file system.
mircea_popescu: s a
file system AT ALL, and the "all things are files" is just a thin wraper on a turd sausage made out of "we have no data model beyond ram and our disk doesn't actually work".
☟︎ a111: Logged on 2018-10-15 10:26 ave1: btw, gnat specific;
System.OS_Lib has 'Create_New_File' (it also has a temp
file generator, but I cannot recommend, uses digits). bvt's implemention looks the way to go (Although the string allocations to talk to C should be removed)
ave1: btw, gnat specific;
System.OS_Lib has 'Create_New_File' (it also has a temp
file generator, but I cannot recommend, uses digits). bvt's implemention looks the way to go (Although the string allocations to talk to C should be removed)
☟︎ bvt: and i'm don't know yet if files created this way can be mv'ed to
file system a111: Logged on 2018-09-22 18:46 phf: asciilifeform: that i can see, but i'm not sure why it's picking up
system wide includes. there's an isystem there that points to correct musl include tree, that has the sys/types.h
file in it
phf: asciilifeform: that i can see, but i'm not sure why it's picking up
system wide includes. there's an isystem there that points to correct musl include tree, that has the sys/types.h
file in it
☟︎ ave1: Also no UNIX sockets yet, I was reading the documentation and came accross how linux now supports "abstract" unix sockets which have no equivalent on the
file system. Pretty big WTF all over, it's implemented by having a string start with a 0 (zero) byte.
diana_coman: logs.minigame.biz/2018-07-21.log.html#t18:54:49 -> for future reference: this required cflags set because it turned out that configure failed to look in the correct place for the .h
file; gentoo's own gcc used "--with-
system-zlib" and got away with it; ave1's gcc did not
a111: Logged on 2018-06-27 18:03 phf: PeterL: the approach that we've been taking with legacy C code is pulling out autotools, and replacing with a single #ifdef/.. configuration header. outside of linux/bsd code is probably not going to work, and the approach is to look at the configuration header
file (
http://btcbase.org/patches/vdiff_sha_static/tree/vtools/src/system.h#L145 in case of vtools) and patch it for your
system a111: Logged on 2014-09-03 11:56 mircea_popescu: By default, systemd saves core dumps to the journal, instead of the
file system. Core dumps must be explicitly queried using coredumpctl4. Besides going against all reason, it also creates complications in multi-user environments (good luck running gdb on your program's core dump if it's dumped to the journal and you don't have root access)
phf: asciilifeform: is it possible to combine musl and libc on a same gentoo
system (gentoo insists i use something they call crossdev, i haven't yet looked into it further) (i'm using your aarch64 gentoo root
file system)?
phf: but where this subthread started "_outside of linux/bsd_ code is probably not going to work, and the approach is to look at the configuration header
file and patch it _for your system_
phf: PeterL: the approach that we've been taking with legacy C code is pulling out autotools, and replacing with a single #ifdef/.. configuration header. outside of linux/bsd code is probably not going to work, and the approach is to look at the configuration header
file (
http://btcbase.org/patches/vdiff_sha_static/tree/vtools/src/system.h#L145 in case of vtools) and patch it for your
system ☟︎ phf: asciilifeform: you'd be amused by the latest emacs release, "Limited form of concurrency with Lisp threads" "Emacs now uses double buffering to reduce flicker on the X Window
System" "Flymake has been completely redesigned" "TRAMP has a new connection method for Google Drive" "A systemd user unit
file is provided". it's almost like a self-parody
mircea_popescu: hey, you get a free db with the
file system, i never was arsed to do any better on my reports. not that i'm proposing my laziness as the model for anyone else nor that i think the bikeshedding discussion needs to continue, now extented into a theoretical comparison of the merits and dismerits of postgress and implicit-fs-db.
mircea_popescu: bwahahaha "bug in npm changes permissions on /
file system destroys productions linux or unix boxes. “sudo npm” will chown “/“."
ben_vulpes: shinohai:
system is complaining that the
file dun exist when it quite clearly does, gonna hafta look into this but not atm, thanks for chiming in
PeterL: the problem I see with the current
system is that you can make a change in one
file which relies on a change in behavior of a function defined in a different
file. You end up with two "sister" patches, but the second one is invalid without the first.
PeterL:
http://btcbase.org/log/2018-01-06#1765616 << I dun see why patches have to change much? Thinking of the
system proposed by mircea_popescu you would have one line change in a "patch version"
file, the rest of the patch would be identical to what we have now
☝︎ ben_vulpes: also, single-
file vpatches drives the
system towards whole sourcetree hashing, otherwise they'll definitionally never depend on one another
phf: fwiw bulk of these tools have been written through the 90s and what was worthwhile from orcland was published that way then. until silent takeover by latex & 1.8gb tex installations happened and none of these tricks work anymore (because the necessary hooks are so deep within the layer of cruft it's near impossible to get to them, and one way they did it is through standard
file system lay out, that requires a chain of compilation stages to move files
phf: well, autoconf is not just scripts. it's also compatibility shims, which is a bit tricky in case of a differ, since, unlike mpi, it has a lot of
file system interaction code, that might or might not be portable. anyway, we'll see
mircea_popescu: anywya, this
system'd be purrfect : if hash unchanged, "this is THE SAME
file by a different name (or path, same thing" ; if hash changed "this is DIFFERENT
FILE by same name"
mircea_popescu: geting hard numbers on
file system performance, something everyone involved in data storage for the past 30 years has been vehehehery deliberately hiding, is certainly better use of time than, for instance, attempting to reason with djb jzw & all.
phf: it does an equivalent of patch, but without calling out to c programs and without the result (or intermediate steps) touching the
file system at any point
erlehmann: every idiot who just takes an uploaded
file and converts it using ffmpeg is just a 4 line text
file away from me filling whatever storage the idiot has on the converter
system phf: writing out each one of those concerns separately will teach you how to do it (or whether it can even be done, like with "atomic
file system")
thestringpuller: "RAND_poll seeds the random number generator using a
system-specific entropy source, which is /dev/urandom on UNIX-like operating systems" << so openssl default is PRNG??? RE: "The urandom device may lack sufficient entropy for your needs, and you might want to reseed it immediately from /dev/random. On Unix and other operating systems that provide the block device, you can use RAND_load_file to load directly from /dev/random."
a111: Logged on 2016-07-18 18:08 asciilifeform: i know of no
file system that would not choke.
Framedragger: asciilifeform: on what does it run again? nfs as in network
file system, or sth else?
a111: Logged on 2016-07-18 18:08 asciilifeform: i know of no
file system that would not choke.
Framedragger: each comment saved as separate
file, to be removed etc using normal
system tools; any modifcation under tree triggers static content regeneration
phf: also it doesn't re-parse txt files from
file system on each page hit...
mod6: boy oh boy. looks like all you can hope for with Ada and issuing
system commands is to redirect the output to a
file, and the read the
file.
ascii_butugychag: my point was that there is not a special knob, you don't need it, your unix
file system is the knob
assbot: Logged on 16-01-2016 22:31:04; mod6: the idea of V is a versioning
system based upon patches that include SHA512 hashes of the
file before and after the given patch is applied -- and checks the given signatures of the wot entities who have signed off on the patch.
mod6: the idea of V is a versioning
system based upon patches that include SHA512 hashes of the
file before and after the given patch is applied -- and checks the given signatures of the wot entities who have signed off on the patch.
☟︎ mircea_popescu: ok that's not it, does the
file it wants actually exist on your
system ? can you find it ?
pete_dushenski: "According to an analysis [PDF] by Duo Security, a bundled plugin reinstalls the root CA
file if it is removed. First, you must delete Dell.Foundation.Agent.Plugins.eDell.dll from your
system (search for it) and then remove the eDellRoot root CA certificate."
shinohai: !s /rotor/TEST2/ourlibs/lib/libboost_system.a: error adding symbols:
File format not recognized
phf: "The inclusion of OS X default applications in the list [some list of SIP locations --phf] means that they inherit the same
file system protections as any other SIP location—if you want to delete Mail.app because you only use your third-party mail application, you’re out of luck."
phf: "The end result is that in El Cap, root is no longer an account with effectively unlimited access to either the
file system or to memory and running processes. SIP places kernel-level checks on root’s privilege that can only be bypassed by the kernel itself."
ascii_field: r loads a lot of DLLs from the executable directory first, so by copying the vpndownloader.exe
file from Program Files to a temporary directory and dropping an appropriately named DLL you can get code execution as
SYSTEM.'
mats: how does anybody meaningfully build on e.g. 'msdos' when it has only one operating mode, is purposed for a 'single-task', and doesn't blow up when some level of the abstraction (that by any reasonable person could be reasoned as multi-task if it is to do anything useful, e.g, graphics, networking,
file system access!!1) has a bug in it?
kakobrekla: J_EDGAR_HOOVER is the main security guardian of the
system, called upon at process start,
file open, user log on, etc. < lmao
gernika: mod6 Built v0.5.4-TEST2 with rotor but can't run it on the
system I built it on because: "-bash: ./bitcoind: cannot execute binary
file: Exec format error." This is on gentoo built from stage3-i486-20150728.tar.bz2
ascii_field: 'In monolithic operating systems, a driver can write to any word of memory and thus accidentally trash user programs. In MINIX 3, when a user expects data from, for example, the
file system, it builds a descriptor telling who has access and at what addresses. It then passes an index to this descriptor to the
file system, which may pass it to a driver. The
file system or driver then asks the kernel to write via the
phf: in my experience it's cheaper to literally go a log
file and reconstruct data manually the one time your
system crash, then introduce uknowable redundancies that tend to increase complexity and ultimately result in the crash, because doesn't fit in head
copypaste: The
system also would not handle concurrency well, due to the fact that it comes up with the ID by doing ls, then finding the biggest numbered text
file in _g and adding one to it.
ascii_field: but operator should select path on
file system and give, e.g., destination pubkey for a ciphergram, as operand on commandline.
ascii_field: '"I suppose the idea is that everything will be in the downloaded
file, so nothing depends on the local libraries on the target
system. Unfortunately with Linux, and I think anything else using GLIBC, this still isn't quite true. There's this "libnss" (name service switch, some people seem to call it network security
system) which provides functions for accessing various databases for authentication, network information,
assbot: Logged on 10-10-2014 17:08:55; asciilifeform: (easiest solution: forget
file system in the 1st place, just use raw block device. and then you don't even need msdos for anything.)