[PATCH v2] Huge performance boost for recursion (cp, du, find, ls, rm, mv)

this project up. If the old/slow and new/fast code should be chosen with a compile-time config option then I'm happy to do that as well.From 69312a6928c188ac8be3c714db3a53724b85dd09 Mon Sep 17 00:00:00 2001 From: Jody Bruchon Date: Wed, 10 Apr 2024 18:08:00 -0400 Subject: [PATCH v2] Huge performance

Re: [PATCH] Huge performance boost for recursion (cp, du, find, ls, rm, , mv)

. I left the old code commented instead of taking it out because there is a good chance the maintainers will want to see this "bloat" gated behind a config option. On 2024-04-11 1:47 AM, Rolf Eike Beer wrote: On Mittwoch, 10. April 2024 21:36:06 CEST Jody Bruchon wrote: This patc

[PATCH] Huge performance boost for recursion (cp, du, find, ls, rm, , mv)

-by: Jody Bruchon From 638521733744e1292ba2701c638983a2124761b2 Mon Sep 17 00:00:00 2001 From: Jody Bruchon Date: Wed, 10 Apr 2024 15:10:26 -0400 Subject: [PATCH] Huge performance boost for recursion (cp, du, find, ls, rm, mv) This patch uses pre-calculated name lengths to massively speed up

[PATCH] Style guide: add "Required_argument" example to applet_longopts

From 48532ee17f752a9bbf4dbf6cb1d7662004c41756 Mon Sep 17 00:00:00 2001 From: Jody Bruchon Date: Wed, 31 Mar 2021 11:30:31 -0400 Subject: [PATCH] Style guide: add "Required_argument" example to applet_longopts Signed-off-by: Jody Bruchon --- docs/style-guide.txt | 5 +++-- 1 file

[PATCH] Style guide: add "Required_argument" example to applet_longopts

From 48532ee17f752a9bbf4dbf6cb1d7662004c41756 Mon Sep 17 00:00:00 2001 From: Jody Bruchon Date: Wed, 31 Mar 2021 11:30:31 -0400 Subject: [PATCH] Style guide: add "Required_argument" example to applet_longopts Signed-off-by: Jody Bruchon --- docs/style-guide.txt | 5 +++-- 1 file

Re: Searching the lists

There is a Google custom search bar immediately below the text. Scroll down more and enjoy! On July 12, 2020 5:14:43 AM EDT, Baruch Burstein wrote: >Hi, > >How do I search the lists? When I go to >https://www.busybox.net/lists.html it >just says to search the lists, but doesn't say how to do

Re: [PATCH 3/4] lsscsi: code shrink and refactor

Allowing erroneous pages to render is not inherently bad, especially with standards that can change in the future. Look at earlier HTML (< 4) vs XHTML. If a page is written in XHTML and is read by a browser that doesn't understand XHTML, the browser still attempting to render it by ignoring any

Re: [PATCH] libbb/last_char_is: rewrite for smaller and faster code

This is what I love about coding. We'll play with optimizing some of the simplest and least important things on a whim. May the winds of hackerdom ever be in your favor. On July 3, 2020 3:34:23 AM EDT, Bernd Petrovitsch wrote: >On Fri, 2020-07-03 at 08:12 +0200, Tito wrote: >[...] >> Improved

Re: [PATCH] libbb/last_char_is: rewrite for smaller and faster code

On July 2, 2020 3:26:29 PM EDT, Tito wrote: >hi, >just for fun I add my version: > >char* FAST_FUNC last_char_is(const char *s, int c) { > while (s && *s && *(s + 1)) s++; > return (c == *s) ? (char *)s : NULL; >} That's pretty small, but it performs three comparisons for each

[PATCH] libbb.h: change prototype for last_char_is() to match new code

From de10e585d933793cb309a03c88faac17cdedb370 Mon Sep 17 00:00:00 2001 From: Jody Bruchon Date: Wed, 1 Jul 2020 18:03:32 -0400 Subject: [PATCH] libbb.h: change prototype for last_char_is() to match new code ---  include/libbb.h | 2 +-  1 file changed, 1 insertion(+), 1 deletion(-) diff --git

[PATCH] libbb/last_char_is: rewrite for smaller and faster code

From 10389ff81b4ee5a7db293cf5d6205d8944e51901 Mon Sep 17 00:00:00 2001 From: Jody Bruchon Date: Wed, 1 Jul 2020 17:55:04 -0400 Subject: [PATCH] libbb/last_char_is: rewrite for smaller and faster code Signed-off-by: Jody Bruchon ---  libbb/last_char_is.c | 14 +-  1 file changed, 5

Re: [PATCH 4/4] last_char_is: code shrink

nal implementation to simplify reading: char* FAST_FUNC last_char_is(const char *s, int c) {     if (s) {         char *index = strrchr(s, c);         if (index && *(index + 1) == '\0')             return index;     }     return NULL; } On Tue, 30 Jun 2020 at 10:59, Jody Bruchon <mailto

Re: [PATCH 4/4] last_char_is: code shrink

strlen scans the string; strrchr also scans the string. On 2020-07-01 10:23, Martin Lewis wrote: Hi, I'm not sure what I'm missing, from a quick glance at strrchr (glibc's) it seems that it scans the string only once? Thank you, Martin On Tue, 30 Jun 2020 at 01:34, Denys Vlasenko

Re: Non-responsive maintainer?

Understood. Be well. On June 4, 2020 9:45:39 PM EDT, Eli Schwartz wrote: >On 6/4/20 9:01 PM, Jody Bruchon wrote: >> If things hang for too long, I'd be willing to fork the project and >> maintain it so it can keep on chugging. I'm not as good as Denys >> though. Can't pr

Re: Non-responsive maintainer?

If things hang for too long, I'd be willing to fork the project and maintain it so it can keep on chugging. I'm not as good as Denys though. Can't promise my mailing list replies will be bottom-posted though ;-) On June 4, 2020 12:42:42 PM EDT, Xabier Oneca -- xOneca wrote: >> I'm new to

Re: Sime makelike wrapper scripts support

of program that very clearly do not belong in BusyBox for philosophical and scope reasons, even if they cannot technically be added without much effort. I don't see much else to discuss on this subject. I applaud the effort, but I believe it is misguided. - Jody Bruchon -- Sent from my Android ph

Re: rustybox, an experimental fork written entirely in Rust

, a thing to enjoy playing with but not to be taken too seriously as a replacement program for its parent. Let them have their fun! Let's all be excellent to each other. A rising tide lifts all boats. - Jody Bruchon On 11/16/2019 7:01 AM, Laurent Bercot wrote:  The busybox mailing-list has

Re: chmod issue

What kind of filesystem is it on? Is that filesystem mounted noexec or read-only? Is SELinux enabled? Is the file you're changing in a layer of an overlay filesystem? I noticed the time didn't change on the file either. On 6/14/2019 8:54 AM, JB Data31 wrote: Hi, I am a busybox user

Re: [PATCH] losetup: Add partition scanning option

Thank you very much for sending in a patch for this. It has been on my list of pain points for BusyBox for a while now since I need to 'losetup -fP disk_image' for some things I do semi-regularly, but you've just saved me the trouble. I'm looking forward to this patch being accepted! On

Re: [v2] jdupes 1.11.1 port to BusyBox

On 12/19/2018 8:16 AM, Tito wrote: On 11/12/18 18:29, Jody Bruchon wrote: This updated version of the jdupes duplicate file finder port to BusyBox adds a whole host of feature exclusions that allow further reduction of the applet's size. Nearly all of the final actions performed by jdupes

Re: [v2] jdupes 1.11.1 port to BusyBox

On 12/19/2018 7:10 AM, Xabier Oneca -- xOneca wrote: Hi, The redundant unknown option text is there so the big 'case' has a default block. It should never be reached, but it's there for safety just in case. I suppose the fprintf() could be changed out to the new unsupported_option()

Re: [v2] jdupes 1.11.1 port to BusyBox

On 12/19/2018 5:02 AM, Xabier Oneca -- xOneca wrote: Hello Jody, Why are you treating disabled options with the message 'Option %s is disabled in this build', when you already provide similar message for unknown options ('Sorry, using '-%c' is not supported.')? Won't save much, but I see them a

Re: [v2] jdupes 1.11.1 port to BusyBox

It's been a week since I sent this in. Does anyone have any questions, comments, thoughts or opinions? On 12/11/2018 12:29 PM, Jody Bruchon wrote: This updated version of the jdupes duplicate file finder port to BusyBox adds a whole host of feature exclusions that allow further reduction

[v2] jdupes 1.11.1 port to BusyBox

n and I've noted that in the help text. /* jdupes (C) 2015-2018 Jody Bruchon Derived from fdupes (C) 1999-2018 Adrian Lopez Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to

Re: inode usage

Do you have an alias for 'df' by some chance? On 12/11/2018 11:10 AM, David Henderson wrote: So I was checking with the 'df' from coreutils on another computer and in doing so I stumbled upon something bizarre. Going back to my main computer (that only has the BB 'df'): # which df /bin/df #

jdupes 1.11.1 port to BusyBox

3648 1020201   f9129 busybox_with_jdupes 1005355    6755    3600 1015710   f7f9e busybox_with_jdupes_minimal  991308    6731    3216 1001255   f4727 busybox_without_jdupes Signed-off-by: Jody Bruchon /* jdupes (C) 2015-2018 Jody Bruchon Derived from fdupes (C) 1999-2018 Adrian Lopez Permission is

Re: Question: Finding libraries that are required for busybox??

that are required for busybox?? To: busybox@busybox.net From: Eli Schwartz Date sent: Mon, 25 Jun 2018 15:30:21 -0400 On 06/25/2018 03:20 PM, Jody Bruchon wrote: By the way, you can also grab the relocations directly from the binary with this horrific expression (replace the path as needed

Re: Question: Finding libraries that are required for busybox??

expression (replace the path as needed): readelf -d /bin/busybox | sed 's/.*\[\([^]]*\)]/\1/' Scary but effective. -Jody Bruchon On June 25, 2018 3:01:56 PM EDT, "Michael D. Setzer II" wrote: >Ran into a problem when trying to clean up libraries that were not >needed >within

Re: bug in vi's undo

buffer becomes // zero-length, a newline is added back, which requires this to compensate. Can any of you[1] think other situations where this would break? Thanks, Kartik [1] Particularly Jody Bruchon, original author of commit a8d6f9be in Apr 2014. Thanks in advance, Jody. I am the author o

Re: Cannot download busybox with busybox

On 8/29/2017 7:58 AM, Denys Vlasenko wrote: I'll cc this to bbox mailing list to humiliate this guy. On Tue, Aug 29, 2017 at 1:21 PM, wrote: I'm trying to update busybox binary on busybox-based system. When i am trying to download busybox, i have this error # wget

Re: [PATCH] ash: support platforms that don't have '%m' printf specifier

On 2017-07-27 12:16, Xabier Oneca -- xOneca wrote: Hello: 2017-07-27 17:48 GMT+02:00 Ron Yorston : Kang-Che Sung wrote: dietlibc supports %m only if WANT_ERROR_PRINTF is defined in when building, and its commented out by default. I don't know how to check the macro

Re: [PATCH] ash: avoid GLIBC'ism %m

On 2017-07-19 14:34, Markus Gothe wrote: Look over here http://pubs.opengroup.org/onlinepubs/9699919799/functions/scanf.html instead: The %m in BusyBox is a printf specifier, not a scanf specifier. How is this link relevant to printf("%m")? ___

Re: [PATCH] ash: avoid GLIBC'ism %m

https://www.gnu.org/software/gnulib/manual/html_node/fprintf.html -Jody Bruchon ___ busybox mailing list busybox@busybox.net http://lists.busybox.net/mailman/listinfo/busybox

[PATCH] Add help text for 'uniq -i'

Looks like I missed the help text in the last patch. This time I've attached it in a way that shouldn't mangle the tabs. From 653b0410685813d223acfdd62cc989e1cd8a555e Mon Sep 17 00:00:00 2001 From: Jody Bruchon <j...@jodybruchon.com> Date: Thu, 30 Mar 2017 12:24:56 -0400 Subject: [PATCH] Ad

[PATCH] uniq: add -i option to ignore case

From 5ebbe05516c6735df219e8c0069cc4330ae9184b Mon Sep 17 00:00:00 2001 From: Jody Bruchon <j...@jodybruchon.com> Date: Thu, 30 Mar 2017 11:09:21 -0400 Subject: [PATCH] uniq: add -i option to ignore case Signed-off-by: Jody Bruchon <j...@jodybruchon.com> --- coreutils/uniq.c | 7 +++

Re: mounting errors

On 2017-02-24 08:49, David Henderson wrote: Thanks for the tip Rob! So it looks like my only resolution to this problem is the dmesg silencing? The only issue I see with that is that perhaps something I need to see won't get shown. Setting the level higher will restrict immediate console

Re: mounting errors

On 2017-02-22 08:43, David Henderson wrote: Good morning all! Yes I am not specifying the type as ext4 explicitly, so it is being probed. The partitions mount and work correctly, I just keep getting these error messages any time they are mounted (e.g. bootup, attaching external drives, etc).

Re: mounting errors

On 2017-02-22 2:28 AM, Mattias Schlenker wrote: Am 21.02.2017 um 22:09 schrieb David Henderson: EXT4-fs (sde2): couldn't mount as ext3 due to feature incompatibilities EXT4-fs (sde2): couldn't mount as ext2 due to feature incompatibilities Are you in control of the kernel configuration? In

Re: mounting errors

On 2017-02-21 4:09 PM, David Henderson wrote: Good afternoon all! I am using the BB mount command to mount several partitions, but when the command is executed I get: EXT4-fs (sde2): couldn't mount as ext3 due to feature incompatibilities EXT4-fs (sde2): couldn't mount as ext2 due to feature

Re: Please revert "ash: make dot command search current directory first, as bash does."

On 2017-01-10 3:42 AM, Kang-Che Sung wrote: Please revert commit 8ad78e1ec7b2e873953f9f476fb63b5893526c39 "ash: make dot command search current directory first, as bash does." This dot command behavior neither follows bash's behavior nor POSIX's. It is also a security risk to have the dot

Re: [PATCH] fdisk: Add support for GPT partition type codes

On 11/27/2016 12:11 AM, Denys Vlasenko wrote: > I looked at these "type codes of GPT partitions" and they seem to be > a figment of imagination of "GPT fdisk" people: they are not stored on disk, > fdisk authors just assigned them to some GUIDs. > > How do you find them useful? I use gptfdisk

[PATCH] Re: Possible Vulnerability in httpd.c

eported. I haven't tested it but it should be functionally identical and close the most obvious sprintf security holes I found on a cursory examination. Hope this helps. -Jody Bruchon >From a6d75d30f172e50e9c46efc9d1d6aba62dd4908b Mon Sep 17 00:00:00 2001 From: Jody Bruchon <j...@jodybrucho

Re: # CONFIG_BUSYBOX is not set: breakage on master

On 2016-08-27 08:18, Tito wrote: >> $ ./busybox >> busybox: applet not found The help text and applet list should not be shown and no error thrown as this is the expected behavior so it should be: >> $ ./busybox >> $ echo $? >> 0 This makes sense; return 0 if the busybox

Re: Was examples/var_service useful for anyone?

On 2016-07-10 14:18, Denys Vlasenko wrote: Hi guys, Please speak up if you ever looked into examples/var_service/* and found it useful in any way. I believe it was very useful for me when I was trying to understand how runsvdir and friends worked for an embedded init system. It has been a

Re: whois doesn't work

On 2016-07-04 11:34, walter harms wrote: Am 04.07.2016 17:22, schrieb Jody Bruchon: On 2016-07-04 11:18, Vito Mulè wrote: Can you strace it and post the output? I can confirm this problem has been going on for a while; one day it just stopped working. Here's an strace: $ strace busybox

Re: two new generated files to be ignored?

On 2016-04-23 16:56, Cristian Ionescu-Idbohrn wrote: $ git status On branch master Your branch is up-to-date with 'origin/master'. Untracked files: (use "git add ..." to include in what will be committed) include/common_bufsiz.h include/common_bufsiz.h.method nothing added

Re: [PATCH] Compress common substrings in applet names to save space

On 2016-04-20 08:03, Denys Vlasenko wrote: > On Wed, Apr 20, 2016 at 12:55 PM, Guillermo Rodriguez Garcia > wrote: When strings are abbreviated the new find_applet_by_name() is about 34% slower than currently. (All comparisons are for the default

Re: why are init's arguments wiped ?

On February 1, 2016 9:28:58 AM EST, Sean Mathews wrote: >Maybe write a simple C init wrapper and have it call the real init >process. Write a shell script that stores its arguments in a particular text file i.e. 'echo "$@" > /var/log/init_cmdline'. Have that shell script

Re: why are init's arguments wiped ?

On 2016-01-28 12:02, Nicolas CARRIER wrote: Why are init's arguments wipe ? I had the hope to pass init some arguments, then read them by looking into /proc/1/cmdline, but no luck... I don't know the answer to the question about init's arguments being erased, but assuming you're on Linux, you

Re: [PATCH] umount: always use umount2 syscall with specified flags

coding practice (being clever is almost always a future maintainability nightmare) and should not be accepted into BusyBox without being rewritten in a maintainable way, including detailed comments explaining why and how it does what it does. Jody Bruchon On October 13, 2015 6:23:46 AM EST, walter

Re: [PATCH] umount: always use umount2 syscall with specified flags

On 2015-10-13 08:42, Denys Vlasenko wrote: > On Tue, Oct 13, 2015 at 1:00 PM, Jody Bruchon <j...@jodybruchon.com> wrote: >> I agree. There is no good reason for doing this at all. It looks like someone has intentionally obfuscated the code. If it's for throwing an e

Re: [PATCH] umount: always use umount2 syscall with specified flags

On October 13, 2015 11:02:45 AM EST, Denys Vlasenko wrote: is obvious you did not bother reading the code before coming up >with criticism. I apologize and withdraw my statements. I will say that a comment explaining why that typedef exists would still be beneficial.

[OT] Re: xfs_fsr on Buffalo Linkstation

l be quite slow if xfs_fsr is running while you're trying to use the device for its intended purpose. Read the XFS FAQ to get more information on this topic. The XFS people have covered all of these questions very comprehensively. Jody Bruchon On October 12, 2015 11:21:39 AM EST, Mik

Re: [PATCH] su: support denying accounts with blank password

no reason to argue with making such a feature available. Jody Bruchon ___ busybox mailing list busybox@busybox.net http://lists.busybox.net/mailman/listinfo/busybox

Re: AW: [RFC/PATCH v2 3/5] libbb: add ends_with() function

I did not see the email you've quoted on the mailing list. Interesting. I shall respond to it now. On 2015-08-25 03:00, dietmar.schind...@manroland-web.com wrote: Von: Tito Gesendet: Montag, 24. August 2015 21:29 Hi Jody, so you pass the buck to the silly caller to guess what is consistent,

Re: [RFC/PATCH v2 3/5] libbb: add ends_with() function

On 2015-08-21 17:26, Tito wrote: what sense would it make to check if str\0 ends_with \0 if all strings do end with \0 ? In this case, I view this as a problem of handling all possible valid inputs in a robust manner. It can also make the code more efficient since \0 is just another byte and

Re: [PATCH] fdisk: fix overflow in GPT size math

On July 15, 2015 5:27:52 PM EDT, Gregory Fong gregory.0...@gmail.com wrote: Promote the first arg to unsigned long long to avoid this. Why not uint64_t? ___ busybox mailing list busybox@busybox.net http://lists.busybox.net/mailman/listinfo/busybox

Re: How to use read in Busybox v1.1.1 ?

On 5/13/2015 7:45 AM, Mike Yates wrote: read -p Enter name: $name I have found that the $ is the issue - Busybox will only accept a variable name without it, while Bash accepts either, or the null variable. My understanding is that BusyBox does not attempt to support Bashisms beyond a certain

Re: How to use read in Busybox v1.1.1 ?

On 05/13/2015 11:27 AM, Michael Conrad wrote: My bash behaves like busybox. As far as I know it has never been possible to read $foo and get the value placed into $foo (unless $foo=foo) I confirmed that bash on one of my servers does allow use of 'read $foo' instead of 'read foo'. It seems like

Re: How to use read in Busybox v1.1.1 ?

On 05/13/2015 1:00 PM, Michael Conrad wrote: while read $foo; do echo $foo; done Does it start writing to an unintended variable on the second iteration? My bash appears to resolve $foo to nothing and puts the response in $REPLY. What's strange is that it is now NOT letting me use 'read

BB tar aborts while GNU tar continues to unpack

Today I encountered an invalid magic error using BusyBox 1.23.1 tar on x86_64 while unpacking a 16 GB archive full of database files. It got about halfway through and gave up. GNU tar unpacked it without any issues. It is possible that my error is caused by an ancient tar 1.12 without large

Re: [PATCH] build system: allow users to select the optimization level

On 4/15/2015 6:15 AM, Bartosz Golaszewski wrote: Allow users to select the compiler optimization level by adding appropriate config options. I would suggest dropping -O0 and -O1 and replacing them with -Og. This optimization is equal to -O2 but turns off *only* optimizations that cause

[OT] Re: Help Please

. If you need more assistance, try asking questions on a board centered on Linux users helping each other, such as LinuxQuestions.org. -Jody Bruchon ___ busybox mailing list busybox@busybox.net http://lists.busybox.net/mailman/listinfo/busybox

Re: [PATCH] truncate: new applet

On March 3, 2015 12:02:55 PM EST, Isaac Dunham ibid...@gmail.com wrote: Was this copy-pasted from coreutils/dd.c? The bloat could be reduced by making it shared between dd and truncate; Also, is there any real reason for the D suffix to exist in BusyBox at all? It seems unnecessary. My

Re: Proposition: use a hashtable instead of bsearch to locate applets

On June 1, 2014 6:28:00 PM EDT, Denys Vlasenko vda.li...@googlemail.com wrote: Hmm. awk.c already has hashtable implementation. A growable one at that. Perhaps it could be generalized a bit and pulled into libbb? -Jody ___ busybox mailing list

Re: [PATCH] Add verbose switch to cp, mv and rm commands

annoyance for me, and I would love to see this feature added. Jody Bruchon ___ busybox mailing list busybox@busybox.net http://lists.busybox.net/mailman/listinfo/busybox

Re: Undo for vi

') *p++ = '\0'; if (*q) colon(q); } while (p); free(initial_cmds[n]); initial_cmds[n] = NULL; n++; } } #endif -snip- Thanks in advance! -Jody Bruchon

Re: Undo for vi

On 3/18/2014 10:51 AM, Jody Bruchon wrote: Out of an interest in seeing this feature, I'm looking at vi.c I've started implementing the undo function; it looks to be easier than I expected now that I have a feel for the way the buffer is handled and should add very little actual code (though

Re: [PATCH] 'u' (Undo) command support for vi

On 3/18/2014 6:15 PM, Jody Bruchon wrote: As promised earlier, I have written undo support for BusyBox 'vi'. I forgot to add my sign-off to this patch; please tack it on if committed to Git. Signed-off-by: Jody Bruchon j...@jodybruchon.com

Re: [PATCH v2] 'u' (Undo) command support for vi

On 3/18/2014 6:15 PM, Jody Bruchon wrote: This code is in a works for me state but needs testing and refinement. D'oh, I didn't test building without the undo feature enabled and missed an #if/#endif. Apologies. Revised patch attached. Also, there is an unused parameter warning if undo

[PATCH] fix fdisk GPT size counting math errors

BusyBox fdisk GPT support does not multiply the total sector count by the sector size in bytes before calculating total disk and partition sizes. This patch fixes the math to display correct sizes. Regards, Jody Lee Bruchon --- a/util-linux/fdisk_gpt.c2013-05-11 19:30:45.0 -0400