pixelbeat__

14 ms·

61.
by pixelbeat__ 5y ago
We used CPU intrinsics in version 9 of GNU coreutils, for the cksum utility. From the release notes: "cksum [-a crc] is now up to 4 times faster by using a slice by 8 algorithm, and at least 8 times faster where pclmul instructions are
62.
GNU coreutils 9 is released
119 points by pixelbeat__ 5y ago|17 comments
63.
by pixelbeat__ 7y ago
Agreed. Correct handling of SIGPIPE is quite subtle and often done incorrectly. I've some notes on SIGPIPE considerations at: http://www.pixelbeat.org/programming/sigpipe_handling.html
64.
by pixelbeat__ 7y ago
GNU coreutils has been very careful when adding new options. We're also careful to provide constructive feedback on why it might be best _not_ to add an option, and have summarized many of these discussions https://www.gnu.o
65.
by pixelbeat__ 7y ago
A couple of points with comparing coreutils, * recompiling with -march=native can give significant wins over more generic binaries provided by linux distros. * parallel processing helps with bigger files, and it's easy enough to levera
66.
by pixelbeat__ 7y ago
Some notes re coreutils and ssh sha256sum uses 128KiB blocks when reading, which may be more optimized than the openssl program. sha256sum supports using openssl libs as they are generally faster. This is enabled on Red Hat flavored distros
67.
by pixelbeat__ 7y ago
The GNU coreutils timeout command encapsulates a lot of these issues. It's surprisingly difficult to handle all the edge cases and races. https://www.maizure.org/projects/decoded-gnu-coreutils/timeo...
68.
by pixelbeat__ 8y ago
It takes a lot of focused work to keep them cohesive and performant https://www.gnu.org/software/coreutils/rejected_requests.htm...
69.
by pixelbeat__ 8y ago
Some tend to use debuggers for development details which I agree isn't great, as one tends to focus on details rather than overall structure. But for actual debugging, especially for code you're not familiar with, they can save hu
70.
by pixelbeat__ 8y ago
https://www.pixelbeat.org/cmdline.html#sets
71.
by pixelbeat__ 8y ago
Set operations are very useful. Here's a summary: http://https://www.pixelbeat.org/cmdline.html#sets
72.
by pixelbeat__ 8y ago
iKid
73.
by pixelbeat__ 8y ago
same in firefox nightly
74.
by pixelbeat__ 8y ago
Any practical program has the potential for bugs. Note one can build the coreutils like busybox with the --enable-single-binary configure option. In this mode all 100 or so utils are combined to a single 1MB binary. I would not consider t
75.
by pixelbeat__ 8y ago
Apparently Sir Michael Atiyah is presenting a proof of the Riemann hypothesis on Monday using "a radically new approach based on the work of von Neumann, Dirac, and Hirzebruch"
76.
A Possible Proof of the Riemann Hypothesis from Sir Michael Atiyah(twitter.com)
4 points by pixelbeat__ 8y ago|1 comments
77.
by pixelbeat__ 8y ago
It's interesting how people's expectations and attitudes develop. I remember landing at San Diego airport and asking how I could walk the relatively short distance into the city center. I just got looks of derision and directions
78.
by pixelbeat__ 8y ago
Canonical has about 500 employees, Red Hat about 12,000. It takes a certain scale to herd all the open source cats
79.
by pixelbeat__ 8y ago
Very nice guide, debugging a practical example My own debugging with gdb summary: http://www.pixelbeat.org/programming/debugger/
80.
by pixelbeat__ 8y ago
I've found int overflow one of the most pernicious issues in C programming, and have been maintaining notes (which reference this article) at: http://www.pixelbeat.org/programming/gcc/integer_overflow.ht...
81.
by pixelbeat__ 8y ago
This is one of the reasons GNU ls now quotes names with unsafe characters by default, so the common case of copy and pasting output from ls to commands will be safe, or if typing the argument, there is a visual indication of a potential iss
82.
by pixelbeat__ 8y ago
I've a summary of the operations at: http://www.pixelbeat.org/cmdline.html#sets Note comm output is a bit awkward to parse, so I use another coreutils `join` command to process already sorted data
83.
by pixelbeat__ 8y ago
Right. Even the UTC+7 forms are confusing. Best to use location based zones for input at least http://www.pixelbeat.org/docs/linux_timezones/
84.
by pixelbeat__ 8y ago
libguestfs > losetup > fdisk I.E. libguestfs is safer when dealing with images of unknown origin, and losetup has inbuilt support now for partition scanning. Here's an old fdisk/parted wrapper I used to use: http:/&#x2
85.
by pixelbeat__ 9y ago
aflfast is mentioned in my second link and was the most used fuzzing implementation to find bugs in coreutils. We've had a quick look at using oss-fuzz, which will need a bit of work since it's more suited to libraries rather than
86.
by pixelbeat__ 9y ago
I'm the GNU coreutils maintainer and have fuzzed them extensively. For example a recent bad bug in TZ handling was found using AFL and fixed by: http://git.sv.gnu.org/gitweb/?p=gnulib.git;a=commitdiff;h=94... We p
87.
by pixelbeat__ 9y ago
You can use strchr() for ascii, or strstr() otherwise. See utf8, searching section in http://www.pixelbeat.org/docs/utf8_programming.html
88.
by pixelbeat__ 9y ago
It's been getting slower over time with noticeable stalls. Performance is a feature
89.
by pixelbeat__ 9y ago
It's great to see continued maintenance of this important project, and it'll be interesting to see the results of a visual refresh. thunderbird has over 20 years lineage, and I've screenshots of the major stages at: http:&#x
90.
by pixelbeat__ 9y ago
Some thanks would definitely be appreciated at least to balance out the negative minority. As a maintainer of a project¹ that many use, I've had private messages encouraging me to commit suicide etc. Not ideal. ¹ https://www

More ›