Broly D-Link DIR-882 A1 AC2600 thread

Post new topic   Reply to topic    DD-WRT Forum Forum Index -> Ralink SoC based Hardware
Goto page 1, 2, 3, 4, 5  Next
Author Message
broly
DD-WRT Novice


Joined: 03 Apr 2010
Posts: 33
Location: edmonton

PostPosted: Mon Sep 18, 2017 4:53    Post subject: Broly D-Link DIR-882 A1 AC2600 thread Reply with quote
moved my big post down since the developers don't respect other people's time.

i may post the odd update here and there, but i feel it's fair to say i've hit the limit (arguably did around sept 25)

since the developer "community" has a strong desire to see the hindus' shitty-MIPS-clone, aka ARM, win, i've decided to say FUCK the LINUX project.** i know sir tim and bill would understand.

*[edit 20122017 17:54GMT]-- changed "fuck this project (dd-wrt)" to "fuck the LINUX project" because mr gottschall deserves more respect after he posted 4.14 for me pretty much. thank you again. i wouldn't have found the IRQ-MIPS-GIC issue otherwise!

this entire situation feels like powerpc trying to bury x86, but with IBM and INTEL's finances reversed. so fuck this pathetic show. now i know why padavan peaced it out years ago.

i should have known from the massive "updates" to the mips interrupt handler (IRQ_IPI,SMP_IPI&, and their forced inclusion) at the expense of proper hardirq support, that they would stop at nothing:
Quote:
11-09 10:37:17.108: genirq: Flags mismatch irq 23. 00000004 (wlan0) vs. 00000004 (Ralink_ESW)


yeah what a great fucking IRQ handler. HARDIRQs down the shitter so "they" (along with "their" token "techie") can feel GUHOODH (good) about themselves, while we're stuck in shitty softIRQ land.

thanks alot, MUTTI (merkel needa get MERCED imo)


[27122017 20:00GMT]Just for MITT, who thought his "recruits" could step


you wanna maul with the kid?
watch your step
you might fall trying to do what i did

(ask mitt's recruits who had a 6+year head start and unlimited support ROFFLES)

remove great passages from "ibm and the holocaust" because people should be able to get the idea just from this comment. if you haven't read it, i recommend you do. it's a fantastic book

Escape From Dullesville indeed, Ted (Rogers [Video]). Thanks for setting the bar high for all Canadians, Sir (William) Stephenson and Mr. Scott.
(last one because it's FUCKING HILARIOUS):
hahahhahahahahaha when will you guys just pack it in, man?
hold on, hold on... one more
"vite vay is muslim vay" amirite? *kneeslap* hahahahahhahaha
*(or vice-versa)

well, we finally learn the truth (Bankers and the search for a separate peace during World War II, by Jonathan Victor Marshall).
i want to be clear about a few things:
  • the area(s) of Psychiatry, Neurology, and their intersection(s) are deeply indebted to Wallenberg family members who agreed to fund the pioneering work (1,2, a sample of the typical, sophisticated, medical commentary in the field that resulted resulted from this work) of Dr. David Ingvar, Dr Niels Lassen and Dr Goran Franzen that focused on gleaning neuronal activity using haemodynamic properties of the brain. i would not be here if not for this work. as sir isaac said: if i have seen further, it is by standing on the shoulders of giants
  • i am NOT a communist, nor capitalist. some tend may associate certain beliefs (of mine) as "left" or "right", but such leanings have never influenced my beliefs. leave me out of this aimless armchair-philosophising, political tug-of-war; i have better things to do. (such as ensuring BAB's conduct is reprimanded. something he thought wouldn't happen)
  • my job was to ensure the parties who have (indirectly; typical Clan Campbell fashion) (ab)used the Bell Telephone Company assets ("The Telephone Plot") were exposed. vilifying other countries for television ratings that juice paper promises ("profits") are not the economics Smith spoke of.
    • stop blaming other countries for your own fuckups. i've had to sit on my ass with my colleagues' words of caution "be good boy, broly" echoing in my head for five fucking years because it seems this disaster has profound implications on everyone. 1-1427918

  • as Sir Robert (WALPOLE, WHO ELSE? not PEEL. that's for sure!) and Father/Son Pitt so-passionately yearned to say: the gig is up "Clan Campbell". no more deflecting of your role in all of this. you should be embarrassed for your conduct towards fellow clans over the past few centuries; forcing citizens to harbour terrorists on the soil they call HOME. shame on you.


the mutter of the meathead
"HUHUHUHUHUH [sped-up beavis and butthead laugh] THANKS YOUR MAJESTY HUHUHUHUHUH" MLSE when they've been lead to believe they've "successfully" dumped another billion-dollar bank bomb on HRH--i.e., converted their inflated trash debt to sovereign debt. haven't you fucking delinquent junkies exceeded your tab by now? fuck sakes


Last edited by broly on Tue Jun 19, 2018 2:07; edited 176 times in total
Sponsor
broly
DD-WRT Novice


Joined: 03 Apr 2010
Posts: 33
Location: edmonton

PostPosted: Wed Sep 20, 2017 22:05    Post subject: static g++/gcc/binutils is a success Reply with quote
here's a GNUBS'D-out System V (or "linux" if you take uname seriously) DD-WRT image with a native BASH shell (with Linux-PAM, DNSCRYPT, OPENDPI, BLOGIC's Hardware NAT for MT7621 devices, WPA3 Support, ENTWARE; don't think the first five are in BS' build). tools were built/based on a NeXTSTEP derivative with GLIBC.

currently kernel 4.14.187 and DD-WRT v43652
YO TP-LINK/LINKSYS/ASUS. hook up an A10/AC85P/EA7500etc, i'd love to finish/fill out my build with that extra 128MB (assuming the SPI can tolerate >16MB payload)
latest FW SE [special edition dedicated to John Crispin, aka blogic): factory-to-ddwrt.bin, dlink-dir882-a1-webflash.bin,
LOOTBAG*


*lootbag contains: glibc/bash/coreutils locales+tor0.4.2.5+PHP7.4.1+manpages+GNU Midnight Commander4.8.23+GNUMake4.2+sqlite3.30.1+JAVA (via JamVM)+GRoFF1.22.4+minidlna+VIm8.1(2-2123)+tcl8.6.9+Tk8.6.9+libX11-1.6.7*

directory tree of latest FW: mt_wifi: https://pastebin.com/xSjhxWwC LAST UPDATED 11102019 18:29GMT
(still-not-ready-for-primetime mt76: https://pastebin.com/ZJj7WZbP LAST UPDATED 10022020 01:52GMT)
bootlog of the latest FW: mt_wifi: https://pastebin.com/CR2Lgayj LAST UPDATED 11102019 18:29GMT

(still-not-ready-for-primetime mt76: https://pastebin.com/3vkfN2qd LAST UPDATED 10022020 01:52GMT)

older (mt_wifi, because mt76 is still p bad) versions
(4.14.170/42293) factory-to-ddwrt.bin, dlink-dir882-a1-webflash.bin
(4.14.175/42863) factory-to-ddwrt.bin, dlink-dir882-a1-webflash.bin
(4.14.176/42967) factory-to-ddwrt.bin, dlink-dir882-a1-webflash.bin
(4.14.178/43055) factory-to-ddwrt.bin, dlink-dir882-a1-webflash.bin
(4.14.183/43336) factory-to-ddwrt.bin, dlink-dir882-a1-webflash.bin
(4.14.184/43476) factory-to-ddwrt.bin, dlink-dir882-a1-webflash.bin

LAUNDRY LIST/INFO

  • [02072020 21:30GMT] (43547++) i figured out the problem with upnp. it uses http headers which causes httpd to return a failed authorisation. i am trying to see if there's a simple fix.
  • [21042020 16:38GMT] (43055++) apparently pppoe works now. so if you have w/ pppoe, power cycle the modem and router.
  • [16042020 19:32GMT] (42912++) glibc is now 2.31, only because of a bug that a dude from redhat helped me fix (got a mainline patch for binutils! how cool is that?. but that's only because nick was super cool. anyways that's the only notable change.
  • [UPDATED INSTRUCTIONS 30042020 19:55GMT] (42422++) i forgot to mention that i've added region capabilities. if you set the nvram variable "region" to US, CA, DE, GB, IM, IE, RU, JP and country of the dog eaters with the stinkiest defecation (KR, freaks). it will use the appropriate band settings for those countries. sorry for not telling you guys this earlier.
  • [17022020 23:24GMT] (42422++) BACK ON MT_WIFI. mt76 is just NOT THERE YET. it's a shame some people have dedicated their lives to wasting my time and perpetuating false notions of mt76's functionality. i don't know what they get out of this behaviour. it's extremely primitive, just like their inter"cultural" conduct. any praise of the mt76 driver in terms of the mt7615 should be taken as anti-chinese, because there is no excuse for why it should be used over the proprietary driver, if not for this bias
  • [01122019 22:00GMT] just an FYI, iwpriv's TxPower command is a PERCENTAGE. i think the best way to tweak txpower is using the manual command. if people could test it out and report back i think all ralink users would be grateful (and help restore this brand's status to "the best"). here is the format

    the manual command is "iwpriv <radio> set TxPwrManualSet=<input>" where i believe input takes five arguments separated by colons (i.e. <arg1>:<arg2>:<arg3>:<arg4>:<arg5>) where each arg is TWO digits (i.e. 14 character string in TOTAL) and takes the following arguments:

    Quote:

    <arg1> (manual mode on or off): 00 (manual mode off), or 01 (manual mode on)
    <arg2> (band width): 00 (CCK), 01 (OFDM), 02 (HT), 03 (VHT)
    <arg3> (frequency, only 20 or 40 for HT): 20 (MHz), 40 (MHz), 80 (MHz, mode 03 only), 160 (MHz, mode 03 only)
    <arg4> (manual power rate):
    for CCK (00): 0: 1 Mbps ; 1: 2 Mbps; 2: 5 Mbps; 3: 11 Mbps
    for OFDM (01): 0: 6 Mbps ; 1: 9 Mbps; 2: 12 Mbps; 3: 18 Mbps; 4: 24 Mbps; 5: 36 Mbps; 6: 48 Mbps; 7: 54 Mbps
    for HT20/HT40 (02): HT20: MCS 0 ~ 7; HT40: MCS 0 ~ 8
    for VHT20/VHT40/VHT80/VHT160 (03): MCS - 0 ~ 9 for VHT20 ; MCS - 0 for VHT40/VHT80/VHT160
    <arg5>: actual value for power (in dB, i think).
    -----------------------------------------------------------------------------
    SetTxPwrManualCtrl: Set wrong parameters
    CCK (0) only supports rate setting: 0 ~ 3 and no supports different BW
    OFDM (1) only supports rate setting: 0 ~ 7 and no supports different BW
    HT20 (2) only supports rate setting: MCS 0 ~ 7
    HT40 (2) only supports rate setting: MCS 0 ~ 8
    VHT20 (3) only supports rate setting: 0 ~ 9
    VHT40 (3) only supports rate setting: 0
    -----------------------------------------------------------------------------


  • [06112019 01:48GMT] (41474+) i've updated samba3. i am trying to cross compile samba4 but it's an absolute chore for osx; talk about operating system discrimination. what a bunch of fuckboys.
  • [20062019 03:08GMT] i really feel like, this build is as good as when i had dd-wrt on my WNDR3700v1 (AR7161). that was the goal. it seems it has been realised. wow

  • entware/using the SES/WPS buttons

  • the SES (left) button now disables the 2.4G radio, and the WPS (right) button disables the 5G radio. enjoy this fully-functional killer firmware. so good that people are trying to create false rumours about being unable to flash lol.
  • i symlinked /usr/local/ to /opt/. so if you mount a usb drive on /opt/, it should have its contents appear in /usr/local. i find this highly desirable. also, enjoy the binutils/gcc suite for this platform. i have tested it and it can compile perl from source. must be pretty, pretty good Wink
  • to use ENTWARE, do the following:
    Quote:
    mkdir /mnt/main #make main mount point directory
    mount /dev/sda1 /mnt/main #mount USB drive to main mount directory
    opt-opkg-upd.sh #run the opkg/entware installer script

    you now have opkg, and can install packages willy-nilly (tested myself! ENJOY! THIS IS WAY BETTER THAN I ANTICIPATED)
  • additional information about RT2880 (MT7615x2) configuration/manipulation

    • each radio now uses one configuration file (i.e. CONFIG_MULTI_PROFILE=n produces AC3467 at 5GHz)
    • VAPs can have distinct wireless channels from the main ssid. To set them, use wl{0,1}_channel=ssid1chan;vap1chan;vap2chan;vap3chan;...;vapNchan where the ordering of the N VAPs' channels (not sure if the limit N is 4, 8 or 16 different channels for VAPS on a single radio) should correspond to the order they are listed in the DD-WRT interface (or nvram settings). I think it's possible to have more VAPs than distinct channels for the respective radio. This feature should really let the mt7615 shine (deal with the reboots. don't create VAPs willy nilly and expect it in 10 seconds k, sometimes it takes 25-30seconds more)
    • users can instantiate the "wl{0,1}_dbdc={0,1}" nvram variable to enable DBDC mode for whichever radio. AC3467 is currently only possible without DBDC mode from my cursory assessment. still a work in progress.
    • PRO TIP #1: switching from BS to my build, or vice-versa, CAN (but SHOULD not [from my experience flashing back and forth for debugging purposes]) corrupt your nvram because we use different NVRAM drivers. BACK UP your settings and restore them after first-boot.

  • additional information about SNMPD

    • snmpd now linked to openssl and supports many more encryption types compared to before.
    • snmpd_conf is now leveraged by services. you can include additional arguments you'd like in snmpd.conf using this nvram variable, which is present in the httpd GUI when enabling the service. enabling encryption requires multiple lines, which can be entered as follows (thx ArchWiki) <angle brackets for command line-equivanet>
      Quote:
      <nvram set snmpd_conf=">rouser *desired username for snmp crypto*<hit enter>
      createUser *desired username for snmp crypto* SHA *desired SHA password* AES *desired AES password*"<hit enter>


  • additional information regarding NVRAM usage:

    • unlike BS's build, my build adheres to the manufacturer's NVRAM partitioning specification, thereby enabling usage of the full 88KB NVRAM spread across the (manufacturer's) five partition scheme. this approach yields 24KB more than BS's "generic" nvram driver that does not use the manufacturer's partitions AT ALL, instead opting to allocate+create a distinct 64KB NVRAM partition. choice is up to YOU, the (YO)USER
    • by default
      Quote:
      nvram show

      will display all nvram banks. i have added the option to show only ONE bank, specified either by the bank name or number (first number in square brackets, second number is the size [in bytes]) as follows:
      Quote:
      nvram show {RT2860_NVRAM [0;32768],RTDEV_NVRAM [1;8192], WIFI3_NVRAM [2;8192], CERT_NVRAM [3;4096], CONFIG2_NVRAM [4;32768]}

    • by default, {RTDEV,WIFI3,CERT}_NVRAM are used as a last-resort (i.e. only if {CONFIG,CONFIG2}_NVRAM are nearly full). the "almost full" heuristic used by the NVRAM is "(nvram_size-nvram_used) / nvram_size < .05"--i.e., less than 5% free.

      • RT2860_NVRAM does not use the heuristic criteria above; instead there is a hard cap of 1000 variables to ensure each variable can accommodate a reasonably-large string.
      • use RALINK's nvram_set if you want to manually set variables to a desired bank; services will have no problem reading/using them.

    • CERT_NVRAM holds the sshd_host key (can't hold much more. 4096 capacity. thought i'd make use of it for the certificate)

  • supported languages are (as of 35771): traditional chinese, german, hungarian, italian, latvian, polish, russian, portuguese, and english.
  • setuserpasswd/httpd's passwd change mechanism uses crypt_r, where i generate the (necessary) 2-byte salt via arc4random_buf--i.e., it is not generated by PAM, but is "PAM-esque". if you want a "pam password", use the "passwd" utility via sshd/telnet


THANK YOUS


----
interlude:
truly grateful to be here, and not live in fear of being persecuted for being an independent thinker. don't know if i'd get that opportunity in any other country.

thank you for taking an immeasurable leap of faith in trying to teach us "The Way" over 200 years ago, Lord/Marquess Cornwallis
this is for you, SIR (doctor) MacKay (and of course: clan MacKAY), Sir TIM (you guys should know who this is), Clan(s) Cunningham Ross (ofc) and MANY others. thank you so much for taking a chance on us.), and of course, HRH HM!!
----

  • What's Different in your version of DD-WRT compared to the rest?

    • first and foremost: gLIBC 2.31. we disclose briefly disclose noteworthy distinctions, so readers can compare and contrast.

      • bash 5.0.16(2) [bash/bashbug] latest patches applied.
      • util-linux 2.35.1 [libmount/libuuid/liblkid/libfdisk{.so.1.x.x}, cfdisk, mount, umount, findmnt, login, pivot_root, hwclock, swapon, swapoff, agetty, rfkill]
      • libBSD ncurses 6.2 [patched up to 13022020] [libncurses++tw,libncursestw/libsmartcolstw/libpanel/libformtw] (tw suffix meanes "threaded+wide").compiled with reentrant,pthread-eintr,color-fgbgb,ext-colors, and the screen driver (among other things).
      • glib20 2.55.0 (using wide+threaded curses,util-linux mount, and libpcre-jit 8.41 for internal PCRE) [gmodule/gthread/gobject/gio/libiconv/gettext/gnuintl]
      • net-tools 1.60n [ifconfig]
      • SysV syslogd/klogd 1.5.0 because the busybox versions weren't working the way i wanted.
      • BSD(Apple) telnetd [telnetd] uses PAM authentication. (busybox's telnetd can't accommodate PAM-based logins)
      • GNU 'which' 2.21 (c'man, who WOULDN'T want GNU which?!)
      • coreutils 8.32 [all programs in single binary](threads+iconv+gnuintl)
      • iproute2, iproute2-ss170905 [ip,tc]
      • iptables 1.6.2 [iptables, ip6tables]
      • ebtables 2.0.10-4 [ebtables]
      • libpcap 1.8.1 [libpcap.so]
      • libdnet 1.10/1.11 [libdnet.so.2.43.2 {unofficial}/libdnet.1.0.1{official}] (1.10 used by common linux distros; 1.11 is OG. kept both to cement DD-WRT's OG status)
      • libgudev-233 [libgudev-1.0.so.0.2.0] (gnome and SysV's seemingly-eternal, yet recently-disturbed matrimony needed this*)
      • procps 3.2.8 [libproc-3.2.8.so/proc/ps/sysctl] again, their inclusion is justified by the added functionality
      • kmod-2.6 [kmod+{modprobe,depmod,insmod,rmmod,lsmod,modinfo} symlinks] (added functionality over busybox justifies inclusion)
      • pci-utils 3.5.5 [libpci.so.3.5.5] (built with kmod for an immersive experienced)
      • e2fsprogs 1.44.0 [libe2p.so/libext2fs.so, mke2fs, tune2fs]
      • GNUfindutils 4.7.0 [xargs,find] (iconv+intl+pthreads)
      • GNUawk 5.1.0 [gawk] (libsigsegv+iconv+pthread+gnuintl'd)
      • GNUbison 3.5.4 [bison] (iconv+intl+pthreads)
      • GNUsed 4.8 [sed] (iconv+intl+pthreads)
      • GNUscreen 4.6.2[/url] [screen] (working on optimising usage of screen. would like it to get called at login time; ~/.screenrc points to /usr/local/home/.screenrc). thank you maarten and jussi
      • cap 2.25 [libcap.so] (capabilities library)
      • grep 3.4 [grep] (pcre+iconv+threads+intl)
      • attr 1.43.4 [libattr.so] (attributes library)
      • acl 2.2.52 [libacl.so] (access control lists library)
      • Linux-PAM 1.3.1 [/lib/security, /etc/pam.d]
      • shadow 4.5 [user{add,mod,del},group{add,mod,del}] (user-friendly utilities [when we get root] to manage users/groups/privs)
      • sysfsutils v2.1.0 [libsysfs.so] (kong inspired this addition. happy i imitated his style [love u kong;)])
      • file 5.32 [libmagic.so,file] (still limited definitions. add some symlinks in /usr/local/share/misc/magic to increase capabilities!)
        and of course...
      • libqm(FUCKING)i 1.20.2 [libqmi-glib.so.5.2.0,qmicli,qmiproxy,qmi-firmware-update] (because it's DD-WRT and LEGIT NETWORKS people get PRIORITY. not noobs)
      • libmbim 1.16.2 [libmbim-glib.so.4.2.0, mbimcli,mbim-proxy,mbim-network]
      • RALINK RT2800_APPS from 2017 SDK (all built on gLIBC; functional unless otherwise stated.):

        • bandsteering plus (bndstng2) due to popular demand. i suspect the VAPpers love it. to use bndstng2, you must have two SSIDs with the same name, on different bands, on a given radio (iirc).
        • flash (lets you read flash memory in linux; saving you from having to hit up uboot)
        • rtdot1x (previously known as rt2860apd) is the new 80211x client for ralink radios. should work if you set the radius variables. i have verified that rtdot1x communicates with the MT7615 linux driver correctly.
        • nvram utilities [nvram_daemon] (for those wanting to return to RALINK's factory firmware [and thus, the factory NVRAM]), call nvram_daemon before flashing the stock firmware.

    • Why should I use this version of DD-WRT as opposed to others (assuming availability)?

      • simple answer: don't.
      • i'm neither looking to attract casuals, nor infringe upon territories that my contemporaries have managed so well, for so long.

        • it should be clear that my "target audience" deviates significantly from my contemporaries (if posting serial-only fw files [not deliberately, mind you] didn't signify this already).
        • i hope that, in the near future, when i get the hang of the bare minimums required for basic network interface configuration, this will change. but again: not looking to infringe upon my contemporaries' territories, so it's not a priority


    • So you don't care if I use it, but you're posting like you want me to, what the fuck motivated this undertaking?

      • simple answer: a lot of my interest and involvement in dd-wrt is at the operating systems-level, which is the area of expertise of individuals who influenced me ("bill", and sir tim). i felt building the kernel from scratch for an embedded networking device is the "new frontier" since each radio can be looked at as a CPU, as it must shuttle its data separately from our activities.

        • with our data appetite seemingly growing, the CPU "brick wall" (3.2ish GHZ) going unaddressed, i felt the parallelisation/multiprocessing environment in a networks context would bear fruit at the theoretical (Computing Science) level.
        • there are interesting operating systems paradigms that arise, and i felt the current releases for embedded devices didn't share the same perspective (exception of the Beast known as Padavan; i'd have used his ralink_gpio if it worked with the DIR882, but no LEDs -_-)
        • lastly, i've been a DD-WRT "leech" for over 13 years, and looking at my contemporaries' work (eko/slayer/kong), i was both motivated and inspired to not be a moocher. as an example:

          • my source of inspiration for the recent, yet stripped-down, iptables/iproute2 were their 10 year old makefiles. without them, these programs are approximately five-fold larger. as such, i wanted to "step to the plate" and contribute to their great work, as my work is not possible without their decade-plus dedication to this area. they were doing great work while i was picking my ass, content with being mediocre.

        • in other words, my work is more of a "thank you" to the DD-WRT crew for their great work (evidenced by the stripped down makefiles/custom sources-where-necessary to facilitate reduced size) that many of us enjoy for free (for the most part).




* "IT'S THE GLOBAL ECKANAMY" <cue gesture:>



------
BONUS/HINT(s) (formerly titled "Bulletin Board") <- this "rant", while still mostly true, is retained since it contains a hint on how to fix the webGUI (for whoever is bold enough to build DD-WRT from source... there are only a few, but history has shown it is those who matter the most...)
i was being overtly generous to BS when i removed my original comment blaming the ugly webgui on the lack of repugly (it has nothing to do with my build configuration). the facts are that others have had this *EXACT* issue. i can't fault BS for not being forthcoming about the deliberate omissions for a working WEBGUI, as he has put a lot of effort into this project and has yet to receive adequate compensation proportional to this effort (something i think he was promised).

  • my take on this complicated situation is this: while the webgui was originally from Linksys, BS has put significant effort into the WEBGUI such that it has become synonymous with DD-WRT. as such, until he gets compensated (he has not, trust me) i will not get upset and start calling DD-WRT "buggy crap" because everything i've had to do was within the bounds of a properly-trained computing science undergraduate with expertise in Operating Systems. LONG STORY SHORT: none of us will get a working WEBGUI from the files in the SVN until BS is compensated. end of story.

    • in order to to shoot down the incorrect claims of "everything is in the source tree", and "all you need to do is uglify the *.js" files, i've attached a pretty decent set of compressed js files. if you compare them to kong's (inspect the webgui on one of his builds), you'll see it's extremely close (not identical obviously). however it's much uglier than the default set of files: (secret hint: BS was right that it's "all on SVN", but if you find the other hint in my "rant", you'll figure it out soon enough [sooner than it took me, at least {you're welcome}]).
      this (should) shoot down the idea that all that is missing is an uglifier.(secret knowledge only for those who spend the time reading my "blabber" and want a functional webGUI; two words: revision 7400)

      • further, if we are to assume all that is missing is uglification of the JS, i point readers to the fact that the compressed form of window_effects.js is the same as the uncompressed, further complicating any claim that uglification was the solution since the full set of uncompressed files are necessary
      • as i've hopefully belaboured throughout: i'm not "upset" or mad that they're withholding information to produce a working DD-WRT WEBGUI, but i would appreciate some honesty. unfortunately they're going to stick to their lie (just like ol' BAB and MLSE [don't forget the samuelis of broadCRAP]). at least be straight about it, jeez. your only critics are the losers who think everything in life should be free, and their expectations are bullshit. stop trying to meet them "on paper" but dodge them in reality, it doesn't look good.

    • lastly, do not believe the claims of "success", or you will be disappointed (i only saw those for the first time a few days ago, when i began thoroughly researching this problem lol). you can/will get a northstar/ARMv6+-based webgui working, but that's about it.

      • the alleged successes (save tsynik, hanskloss and like two other people) were all around the time i began my foray into this area, which certain "powers that (used to) be" used as an opportunity to show their recruits were worthy. however, as has been demonstrated time and time again, these recruits' tendencies to deliberately skew perceptions has been counterproductive (for nearly two centuries). in this case, anyone who is truly familiar with the source tree will know many of their alleged "solutions" are simply not true. such mistruths can only be learned when a post such as this respects/appreciates the sophistication of httpd (it really is amazing).



_________________
"But of this I can assure you that there is not a movement of any body of Men however small whether on Horse-back or on foot, nor an operation or March of any description nor any Service in the field that is not formed upon some mathematical principle, and in the performance of which the knowledge and practical application of the mathematicks will be found not only useful but necessary. The application of the Mathematicks to Gunnery, Fortification, Tactics, the survey and knowledge of formal Castrametation [setting up camp] etc. cannot be acquired without study"
"Only by adopting an infinitely small unit for observation, the differential in history otherwise known as human homogeneity, and perfecting the art of integration (the adding up of infinitesimals) can we have any hope of determining the laws of history"


Last edited by broly on Sat Jul 04, 2020 20:00; edited 1161 times in total
shadowjig
DD-WRT Novice


Joined: 23 Sep 2017
Posts: 3

PostPosted: Thu Oct 05, 2017 10:30    Post subject: Re: static g++/gcc/binutils is a success Reply with quote
My wifi keeps dropping on the 882. Any tips on how to troubleshoot it?

broly wrote:
greetings everyone.

even though we are locked out of superuser land because of "the losers" i've mentioned before, here is a log of a successful compilation (small program sure) on the MT7621 platform (mipsel-none-linux-gnu target) using gcc.

Quote:

sh-4.4# gcc -v
Using built-in specs.
COLLECT_GCC=gcc
COLLECT_LTO_WRAPPER=/mnt/root/bin/../libexec/gcc/mipsel-none-linux-gnu/7.2.0/lto-wrapper
Target: mipsel-none-linux-gnu
Configured with: ../gcc7/configure --prefix=/tools --libdir=/tools/lib --build=x86_64-cross-darwin14 --host=mipsel-none-linux-gnu --target=mipsel-none-linux-gnu <SNIPPED BY BROLY>
Thread model: posix
gcc version 7.2.0 (GCC)
sh-4.4# g++ -v
Using built-in specs.
COLLECT_GCC=g++
COLLECT_LTO_WRAPPER=/mnt/root/bin/../libexec/gcc/mipsel-none-linux-gnu/7.2.0/lto-wrapper
Target: mipsel-none-linux-gnu
Configured with: ../gcc7/configure --prefix=/tools --libdir=/tools/lib --build=x86_64-cross-darwin14 --host=mipsel-none-linux-gnu --target=mipsel-none-linux-gnu <SNIPPED BY BROLY>
Thread model: posix
gcc version 7.2.0 (GCC)
sh-4.4# cat t
tmp/ tools/
sh-4.4# cat /mnt/root/bin/test.c
#include <stdlib.h>

int main() {


printf("Hi\n");
}

sh-4.4# pwd
/
sh-4.4# cd /mnt/root/bin
sh-4.4# rm a.out
sh-4.4# gcc test.c
test.c:1:10: fatal error: stdlib.h: No such file or directory
#include <stdlib.h>
^~~~~~~~~~
compilation terminated.
sh-4.4# gcc -I/mnt/root/include test.c
test.c: In function 'main':
test.c:6:2: warning: implicit declaration of function 'printf' [-Wimplicit-function-declaration]
printf("Hi\n");
^~~~~~
test.c:6:2: warning: incompatible implicit declaration of built-in function 'printf'
test.c:6:2: note: include '<stdio.h>' or provide a declaration of 'printf'
sh-4.4# ./a.out
Hi
sh-4.4# ar
Usage: ar [emulation options] [-]{dmpqrstx}[abcDfilMNoPsSTuvV] [--plugin <name>] [member-name] [count] archive-file file...


the more verbose output (ar/as/readelf/objdump/gcc/g++ --help commands) is below.
https://pastebin.com/vKyTpSGe

truly grateful to be here, and not live in fear of being persecuted for being an independent thinker. don't know if i'd get that opportunity in any other country.

this is for you, SIR (doctor) MacKay, Sir TIM (you guys should know who this is), and of course, HRH HM!!
BrainSlayer
Site Admin


Joined: 06 Jun 2006
Posts: 6959
Location: Dresden, Germany

PostPosted: Fri Nov 16, 2018 17:46    Post subject: Re: static g++/gcc/binutils is a success Reply with quote
broly wrote:
here's a GNUBS'D-out System V (or "linux" if you take uname seriously) DD-WRT image with a native BASH shell (with **working** resetbutton, Linux-PAM, DNSCRYPT, BLOGIC's Hardware NAT for MT7621 devices, and ENTWARE; don't think the first four are in BS' build). tools were built/based on a NeXTSTEP derivative with GLIBC.

currently kernel 4.14.81 and DD-WRT v37715
YO TP-LINK. hook up an A10, i'd love to finish/fill out my build with that extra 128MB (assuming the SPI can tolerate >16MB payload)
latest FW SE [special edition dedicated to John Crispin, aka blogic): factory-to-ddwrt.bin, dlink-dir882-a1-webflash.bin
directory tree of latest FW: https://pastebin.com/anfyEqiV LAST UPDATED 13112018 01:42GMT
bootlog of the latest FW: https://pastebin.com/EFgXicqG LAST UPDATED 12092018 21:57GMT
Linux 4.14.x kernel .config file
older FWs:
37398 factory-to-ddwrt, webflash
37441 factory-to-ddwrt, webflash
37488 factory-to-ddwrt, webflash
37618 factory-to-ddwrt, webflash
37634 factory-to-ddwrt, webflash

here are some additional applications that wouldn't fit on the ramdisk (install instructions are in the download msg):

  • glibc/bash/coreutils locales+tor0.3.4.8+PHP7.2.11+GNU Midnight Commander4.8.11+JAVA (via JamVM)+GRoFF1.22.3+minidlna+VIm8.8+tcl8.6.8+bitchx-git in a nice, tidy LOOTBAG (pam samba policy [pam_smbpass.so] included).
  • (Bell Labs-badgeworthy) NATIVE (mipsel-thread-multi) SHARED BSD perl 5.28.0 (prove me wrong when i say this is the best perl available on any embedded platform (c89std,sysvipc,decimal float,quadmath,manpages,etc).
  • NATIVE SHARED Python 2.7.14 [python] (hashlib [openSSL integration], Tk/Tcl, curses, curses_panel, libffi ctypes,nis,etc). here's what it DOESN'T have:_bsddb,_sqlite3,bsddb185,bz2,dbm,gdbm,readline,sunaudiodev
  • **SHARED+STATIC GCC 8.2.0/BINUTILS2.31.1 (with ld.gold+C11 threads support [libc2.28])** (C++/C/OBJC support, should be fully loaded) i think i've met the goals that i set out a few years ago, so i'll share this sucker. enjoy.
    Quote:
    root@DD-WRT:~# gcc -v
    Using built-in specs.
    COLLECT_GCC=gcc
    COLLECT_LTO_WRAPPER=/tmp/mnt/main/usr/local/bin/../libexec/gcc/mipsel-none-linux-gnu/8.2.0/lto-wrapper
    Target: mipsel-none-linux-gnu
    Configured with: ../gcc-8.2.0/configure --build=i386-apple-darwin14 --host=mipsel-none-linux-gnu --enable-languages=c,c++,objc --with-system-zlib --enable-install-libiberty --enable-frame-pointer --enable-libffi --enable-decimal-float --enable-libquadmath --enable-libgomp --enable-libbacktrace --disable-multilib --with-native-system-header-dir=/include --cache-file=config.cache
    Thread model: posix
    gcc version 8.2.0 (GCC)
    root@DD-WRT:~# ld -V
    GNU ld (GNU Binutils) 2.31.1
    Supported emulations:
    elf32ltsmip
    elf32btsmip
    elf32ltsmipn32
    elf64ltsmip
    elf32btsmipn32
    elf64btsmip

  • [13112018 0136GMT] links have been refreshed, toolchain is now using libc-2.28.
  • ** NOTE **: in case it is not obvious, much of the software compiled below is for mipsel-linux (and not a specific router), meaning it should work on ANY glibc-based MIPS32(el) platform.
  • everything else you need is probably available on Padavan's entware. i have only included the perl/python since i am confident they're as good (maybe better) than the entware offerings, but i feel the rest of entware is probably as-good as what was tediously posted over multiple links.


LAUNDRY LIST/INFO

  • [09112018 22:45GMT] (37634++) the toolchain's peculiar linker error has been fixed. users would typically experience
    Quote:

    /tmp/mnt/main/usr/local/bin/../lib/gcc/mipsel-none-linux-gnu/8.2.0/../../../../mipsel-none-linux-gnu/bin/ld: /tmp/mnt/main/usr/local/bin/../lib/gcc/mipsel-none-linux-gnu/8.2.0/../../../crt1.o: in function `__start':
    /Volumes/xtoolshit/glibc-2.28/csu/../sysdeps/mips/start.S:99: undefined reference to `__libc_csu_init'
    /tmp/mnt/main/usr/local/bin/../lib/gcc/mipsel-none-linux-gnu/8.2.0/../../../../mipsel-none-linux-gnu/bin/ld: /Volumes/xtoolshit/glibc-2.28/csu/../sysdeps/mips/start.S:100: undefined reference to `__libc_csu_fini'

    when compiling executables. turns out libc_nonshared.a was corrupt. enjoy the seamless experience.
  • to use ENTWARE, do the following:
    Quote:
    mkdir /mnt/main #make main mount point directory
    mount /dev/sda1 /mnt/main #mount USB drive to main mount directory
    opt-opkg-upd.sh #run the opkg/entware installer script

    you now have opkg, and can install packages willy-nilly (tested myself! ENJOY! THIS IS WAY BETTER THAN I ANTICIPATED)
  • additional information about RT2880 (MT7615x2) configuration/manipulation

    • each radio now uses one configuration file (i.e. CONFIG_MULTI_PROFILE=n produces AC3467 at 5GHz)
    • VAPs can have distinct wireless channels from the main ssid. To set them, use wl{0,1}_channel=ssid1chan;vap1chan;vap2chan;vap3chan;...;vapNchan where the ordering of the N VAPs' channels (not sure if the limit N is 4, 8 or 16 different channels for VAPS on a single radio) should correspond to the order they are listed in the DD-WRT interface (or nvram settings). I think it's possible to have more VAPs than distinct channels for the respective radio. This feature should really let the mt7615 shine (deal with the reboots. don't create VAPs willy nilly and expect it in 10 seconds k, sometimes it takes 25-30seconds more)
    • users can instantiate the "wl{0,1}_dbdc={0,1}" nvram variable to enable DBDC mode for whichever radio. AC3467 is currently only possible without DBDC mode from my cursory assessment. still a work in progress.
    • PRO TIP #1: switching from BS to my build, or vice-versa, CAN (but SHOULD not [from my experience flashing back and forth for debugging purposes]) corrupt your nvram because we use different NVRAM drivers. BACK UP your settings and restore them after first-boot.

  • additional information about SNMPD

    • snmpd now linked to openssl and supports many more encryption types compared to before.
    • snmpd_conf is now leveraged by services. you can include additional arguments you'd like in snmpd.conf using this nvram variable, which is present in the httpd GUI when enabling the service. enabling encryption requires multiple lines, which can be entered as follows (thx ArchWiki) <angle brackets for command line-equivanet>
      Quote:
      <nvram set snmpd_conf=">rouser *desired username for snmp crypto*<hit enter>
      createUser *desired username for snmp crypto* SHA *desired SHA password* AES *desired AES password*"<hit enter>


  • additional information regarding NVRAM usage:

    • unlike BS's build, my build adheres to the manufacturer's NVRAM partitioning specification, thereby enabling usage of the full 88KB NVRAM spread across the (manufacturer's) five partition scheme. this approach yields 24KB more than BS's "generic" nvram driver that does not use the manufacturer's partitions AT ALL, instead opting to allocate+create a distinct 64KB NVRAM partition. choice is up to YOU, the (YO)USER
    • by default
      Quote:
      nvram show

      will display all nvram banks. i have added the option to show only ONE bank, specified either by the bank name or number (first number in square brackets, second number is the size [in bytes]) as follows:
      Quote:
      nvram show {RT2860_NVRAM [0;32768],RTDEV_NVRAM [1;8192], WIFI3_NVRAM [2;8192], CERT_NVRAM [3;4096], CONFIG2_NVRAM [4;32768]}

    • by default, {RTDEV,WIFI3,CERT}_NVRAM are used as a last-resort (i.e. only if {CONFIG,CONFIG2}_NVRAM are nearly full). the "almost full" heuristic used by the NVRAM is "(nvram_size-nvram_used) / nvram_size < .05"--i.e., less than 5% free.

      • RT2860_NVRAM does not use the heuristic criteria above; instead there is a hard cap of 1000 variables to ensure each variable can accommodate a reasonably-large string.
      • use RALINK's nvram_set if you want to manually set variables to a desired bank; services will have no problem reading/using them.

    • CERT_NVRAM holds the sshd_host key (can't hold much more. 4096 capacity. thought i'd make use of it for the certificate)

  • supported languages are (as of 35771): traditional chinese, german, hungarian, italian, latvian, polish, russian, portuguese, and english.
  • setuserpasswd/httpd's passwd change mechanism uses crypt_r, where i generate the (necessary) 2-byte salt via arc4random_buf--i.e., it is not generated by PAM, but is "PAM-esque". if you want a "pam password", use the "passwd" utility via sshd/telnet


THANK YOUS


----
interlude:
truly grateful to be here, and not live in fear of being persecuted for being an independent thinker. don't know if i'd get that opportunity in any other country.

thank you for taking an immeasurable leap of faith in trying to teach us "The Way" over 200 years ago, Lord/Marquess Cornwallis
this is for you, SIR (doctor) MacKay (and of course: clan MacKAY), Sir TIM (you guys should know who this is), Clan(s) Cunningham Ross (ofc) and MANY others. thank you so much for taking a chance on us.), and of course, HRH HM!!
----

  • What's Different in your version of DD-WRT compared to the rest?

    • first and foremost: gLIBC 2.28. we disclose briefly disclose noteworthy distinctions, so readers can compare and contrast.

      • bash 5.0 beta [bash/bashbug] again, if i find it's unstable i will revert back to 4.4.19.
      • util-linux 2.33 [libmount/libuuid/liblkid/libfdisk{.so.1.x.x}, cfdisk, mount, umount, findmnt, login, pivot_root, losetup, runuser, hwclock, ctrlaltdelete, swapon, swapoff, agetty, rfkill]
      • libBSD ncurses 6.1 [patched up to 27102018] [libncurses++tw,libncursestw/libsmartcolstw/libpanel/libformtw] (tw suffix meanes "threaded+wide").compiled with reentrant,pthread-eintr,color-fgbgb,ext-colors, and the screen driver (among other things).
      • glib20 2.55.0 (using wide+threaded curses,util-linux mount, and libpcre-jit 8.41 for internal PCRE) [gmodule/gthread/gobject/gio/libiconv/gettext/gnuintl]
      • net-tools 1.60n [ifconfig]
      • SysV syslogd/klogd 1.5.0 because the busybox versions weren't working the way i wanted.
      • BSD(Apple) telnetd [telnetd] uses PAM authentication. (busybox's telnetd can't accommodate PAM-based logins)
      • GNU 'which' 2.21 (c'man, who WOULDN'T want GNU which?!)
      • coreutils 8.30 [all programs in single binary](threads+iconv+gnuintl)
      • ex-vi 050325 [ex,vi] (paying homage to the SysV/BSD guys. much better than bbawx's vi, but may need some tweaks)
      • iproute2, iproute2-ss170905 [ip,tc]
      • iptables 1.6.2 [iptables, ip6tables]
      • ebtables 2.0.10-4 [ebtables]
      • libpcap 1.8.1 [libpcap.so]
      • libdnet 1.10/1.11 [libdnet.so.2.43.2 {unofficial}/libdnet.1.0.1{official}] (1.10 used by common linux distros; 1.11 is OG. kept both to cement DD-WRT's OG status)
      • libgudev-232 [libgudev-1.0.so.0.2.0] (gnome and SysV's seemingly-eternal, yet recently-disturbed matrimony needed this*)
      • procps 3.2.8 [libproc-3.2.8.so/proc/ps/sysctl] again, their inclusion is justified by the added functionality
      • kmod-2.5 [kmod+{modprobe,depmod,insmod,rmmod,lsmod,modinfo} symlinks] (added functionality over busybox justifies inclusion)
      • pci-utils 3.5.5 [libpci.so.3.5.5] (built with kmod for an immersive experienced)
      • e2fsprogs 1.44.0 [libe2p.so/libext2fs.so, mke2fs, tune2fs]
      • autoconf 2.69 [autoconf,autoreconf,autoheader,aclocal] (see below)
      • automake 1.15 [aclocal automake] (including these is an easy one: with the new setup, mounting a foreign perl install like the one i gave above is easy. if you have internet, getting the automake package is also easy. so you should be good to go quite fast.)
      • GNUmake 4.2 [make] (iconv+intl+pthreads)
      • GNUfindutils 4.4.2 [xargs,find] (iconv+intl+pthreads)
      • GNUawk 4.2.1 [gawk] (libsigsegv+iconv+pthread+gnuintl'd)
      • GNUbison 3.2.1 [bison] (iconv+intl+pthreads)
      • GNUsed 4.5 [sed] (iconv+intl+pthreads)
      • GNUscreen 4.6.2[/url] [screen] (working on optimising usage of screen. would like it to get called at login time; ~/.screenrc points to /usr/local/home/.screenrc). thank you maarten and jussi
      • cap 2.25 [libcap.so] (capabilities library)
      • grep 3.1 [grep] (pcre+iconv+threads+intl)
      • attr 1.43.4 [libattr.so] (attributes library)
      • acl 2.2.52 [libacl.so] (access control lists library)
      • Linux-PAM 1.3.1 [/lib/security, /etc/pam.d]
      • shadow 4.5 [user{add,mod,del},group{add,mod,del}] (user-friendly utilities [when we get root] to manage users/groups/privs)
      • sysfsutils v2.1.0 [libsysfs.so] (kong inspired this addition. happy i imitated his style [love u kong;)])
      • file 5.32 [libmagic.so,file] (still limited definitions. add some symlinks in /usr/local/share/misc/magic to increase capabilities!)
        and of course...
      • libqm(FUCKING)i 1.20.0 [libqmi-glib.so.5.2.0,qmicli,qmiproxy,qmi-firmware-update] (because it's DD-WRT and LEGIT NETWORKS people get PRIORITY. not noobs)
      • libmbim 1.16.0 [libmbim-glib.so.4.2.0, mbimcli,mbim-proxy,mbim-network]
      • RALINK RT2800_APPS from 2017 SDK (all built on gLIBC; functional unless otherwise stated.):

        • bandsteering plus (bndstng2) (37497++) due to popular demand, has been re-added. i suspect the VAPpers love it. to use bndstng2, you must have two SSIDs with the same name, on different bands, on a given radio (iirc).
        • flash (lets you read flash memory in linux; saving you from having to hit up uboot)
        • rtdot1x (previously known as rt2860apd) is the new 80211x client for ralink radios. should work if you set the radius variables. i have verified that rtdot1x communicates with the MT7615 linux driver correctly.
        • nvram utilities [nvram_daemon/ralink_init] (for those wanting to return to RALINK's factory firmware [and thus, the factory NVRAM]).

    • Why should I use this version of DD-WRT as opposed to others (assuming availability)?

      • simple answer: don't.
      • i'm neither looking to attract casuals, nor infringe upon territories that my contemporaries have managed so well, for so long.

        • it should be clear that my "target audience" deviates significantly from my contemporaries (if posting serial-only fw files [not deliberately, mind you] didn't signify this already).
        • i hope that, in the near future, when i get the hang of the bare minimums required for basic network interface configuration, this will change. but again: not looking to infringe upon my contemporaries' territories, so it's not a priority


    • So you don't care if I use it, but you're posting like you want me to, what the fuck motivated this undertaking?

      • simple answer: a lot of my interest and involvement in dd-wrt is at the operating systems-level, which is the area of expertise of individuals who influenced me ("bill", and sir tim). i felt building the kernel from scratch for an embedded networking device is the "new frontier" since each radio can be looked at as a CPU, as it must shuttle its data separately from our activities.

        • with our data appetite seemingly growing, the CPU "brick wall" (3.2ish GHZ) going unaddressed, i felt the parallelisation/multiprocessing environment in a networks context would bear fruit at the theoretical (Computing Science) level.
        • there are interesting operating systems paradigms that arise, and i felt the current releases for embedded devices didn't share the same perspective (exception of the Beast known as Padavan; i'd have used his ralink_gpio if it worked with the DIR882, but no LEDs -_-)
        • lastly, i've been a DD-WRT "leech" for over 13 years, and looking at my contemporaries' work (eko/slayer/kong), i was both motivated and inspired to not be a moocher. as an example:

          • my source of inspiration for the recent, yet stripped-down, iptables/iproute2 were their 10 year old makefiles. without them, these programs are approximately five-fold larger. as such, i wanted to "step to the plate" and contribute to their great work, as my work is not possible without their decade-plus dedication to this area. they were doing great work while i was picking my ass, content with being mediocre.

        • in other words, my work is more of a "thank you" to the DD-WRT crew for their great work (evidenced by the stripped down makefiles/custom sources-where-necessary to facilitate reduced size) that many of us enjoy for free (for the most part).




* "IT'S THE GLOBAL ECKANAMY" <cue gesture:>



------
BONUS/HINT(s) (formerly titled "Bulletin Board") <- this "rant", while still mostly true, is retained since it contains a hint on how to fix the webGUI (for whoever is bold enough to build DD-WRT from source... there are only a few, but history has shown it is those who matter the most...)
i was being overtly generous to BS when i removed my original comment blaming the ugly webgui on the lack of repugly (it has nothing to do with my build configuration). the facts are that others have had this *EXACT* issue. i can't fault BS for not being forthcoming about the deliberate omissions for a working WEBGUI, as he has put a lot of effort into this project and has yet to receive adequate compensation proportional to this effort (something i think he was promised).

  • my take on this complicated situation is this: while the webgui was originally from Linksys, BS has put significant effort into the WEBGUI such that it has become synonymous with DD-WRT. as such, until he gets compensated (he has not, trust me) i will not get upset and start calling DD-WRT "buggy crap" because everything i've had to do was within the bounds of a properly-trained computing science undergraduate with expertise in Operating Systems. LONG STORY SHORT: none of us will get a working WEBGUI from the files in the SVN until BS is compensated. end of story.

    • in order to to shoot down the incorrect claims of "everything is in the source tree", and "all you need to do is uglify the *.js" files, i've attached a pretty decent set of compressed js files. if you compare them to kong's (inspect the webgui on one of his builds), you'll see it's extremely close (not identical obviously). however it's much uglier than the default set of files: (secret hint: BS was right that it's "all on SVN", but if you find the other hint in my "rant", you'll figure it out soon enough [sooner than it took me, at least {you're welcome}]).
      this (should) shoot down the idea that all that is missing is an uglifier.(secret knowledge only for those who spend the time reading my "blabber" and want a functional webGUI; two words: revision 7400)

      • further, if we are to assume all that is missing is uglification of the JS, i point readers to the fact that the compressed form of window_effects.js is the same as the uncompressed, further complicating any claim that uglification was the solution since the full set of uncompressed files are necessary
      • as i've hopefully belaboured throughout: i'm not "upset" or mad that they're withholding information to produce a working DD-WRT WEBGUI, but i would appreciate some honesty. unfortunately they're going to stick to their lie (just like ol' BAB and MLSE [don't forget the samuelis of broadCRAP]). at least be straight about it, jeez. your only critics are the losers who think everything in life should be free, and their expectations are bullshit. stop trying to meet them "on paper" but dodge them in reality, it doesn't look good.

    • lastly, do not believe the claims of "success", or you will be disappointed (i only saw those for the first time a few days ago, when i began thoroughly researching this problem lol). you can/will get a northstar/ARMv6+-based webgui working, but that's about it.

      • the alleged successes (save tsynik, hanskloss and like two other people) were all around the time i began my foray into this area, which certain "powers that (used to) be" used as an opportunity to show their recruits were worthy. however, as has been demonstrated time and time again, these recruits' tendencies to deliberately skew perceptions has been counterproductive (for nearly two centuries). in this case, anyone who is truly familiar with the source tree will know many of their alleged "solutions" are simply not true. such mistruths can only be learned when a post such as this respects/appreciates the sophistication of httpd (it really is amazing).




its really hard to follow up that drug or alcohol influence post as it seems, but i want to bring up one issue. the repugly which is nothing more than

for i in $1/*.js ; do uglifyjs --compress --mangle -- $i > $i.copy; cp $i.copy $i; rm $i.copy; done


uglifyjs can be found on github. but its nothing you need to use. its just to make everything small. its not required to get a working webinterface. the asp and js files can be used as is without mangeling

_________________
"So you tried to use the computer and it started smoking? Sounds like a Mac to me.." - Louis Rossmann https://www.youtube.com/watch?v=eL_5YDRWqGE&t=60s
SoLoR
DD-WRT Novice


Joined: 21 Jan 2019
Posts: 31

PostPosted: Sun Apr 21, 2019 21:13    Post subject: Re: static g++/gcc/binutils is a success Reply with quote
broly wrote:
currently kernel 4.14.111 and DD-WRT v39514


Thank for this build! Trying it now on dir878, lets see how is stability... got a reboot with offical r39296 after 6 and a half days :/ also not sure how interested you are with fixing minor bugs... for example you cant set custom ntp server, however client is obviously working since it sets time from pool.ntp.org, also you cant disable ssh (option does not save) and syslogd does not start with remote loging enabled, i would suspect this is because on offical builds syslog remote loging starts syslogd with -R <ip> option, but on this build syslogd complains this is invalid option.
rotarum
DD-WRT Novice


Joined: 06 Mar 2007
Posts: 18

PostPosted: Sat Jun 01, 2019 21:38    Post subject: Reply with quote
Set to Wide HT40.

But still the Clients connect with HT20 only (65mbps)

Build #39296

Encryption set to WPA2 PSK with AES... what can it be??



EDIT:

Upgrade to the latest build from broly (first page)

DD-WRT v4.14-r39800M CRISPIN (05/14/19)


HT40 Mode and VHT160 same time working perfect!

Beside this the LED's show correctly too Smile
v43
DD-WRT Novice


Joined: 25 Aug 2018
Posts: 2

PostPosted: Mon Jun 03, 2019 2:35    Post subject: broly 39800 and 39924 rev firmwares Reply with quote
Hey all, I tried the broly 39800 and 39924 dd-wrt firmware for the dir-882 a1.

My issue is I could not setup openvpn client through this build of dd-wrt. I ensured I followed the full instructions from the vpn provider. I triple checked the user name and password, also that it is currently valid vpn account, which it is, and this is what my log said, see below, and claimed it was an authentication error. I did not have an issue with Brainslayer's beta in this regard. I have also set up other dd-wrt routers successfully using the openvpn instructions.

Otherwise this firmware build seems truly awesome from what I have seen and heard. Any help/input would be greatly appreciated. Thank you in advance. I am a newbie to networking so the answer might be something obvious.


Jun 2 21:42:05 DD-WRT httpd[1540]: pam_unix(httpd:auth): check pass; user unknown
Jun 2 21:42:05 DD-WRT httpd[1540]: pam_unix(httpd:auth): authentication failure; logname= uid=0 euid=0 tty= ruser= rhost=
Jun 2 21:42:12 DD-WRT httpd[1540]: pam_unix(httpd:auth): check pass; user unknown
Jun 2 21:42:12 DD-WRT httpd[1540]: pam_unix(httpd:auth): authentication failure; logname= uid=0 euid=0 tty= ruser= rhost=
Jun 2 21:42:19 DD-WRT httpd[1540]: pam_unix(httpd:auth): check pass; user unknown
Jun 2 21:42:19 DD-WRT httpd[1540]: pam_unix(httpd:auth): authentication failure; logname= uid=0 euid=0 tty= ruser= rhost=
Jun 2 21:42:27 DD-WRT httpd[1540]: pam_unix(httpd:auth): check pass; user unknown
Jun 2 21:42:27 DD-WRT httpd[1540]: pam_unix(httpd:auth): authentication failure; logname= uid=0 euid=0 tty= ruser= rhost=
Jun 2 21:42:35 DD-WRT httpd[1540]: pam_unix(httpd:auth): check pass; user unknown
Jun 2 21:42:35 DD-WRT httpd[1540]: pam_unix(httpd:auth): authentication failure; logname= uid=0 euid=0 tty= ruser= rhost=
Jun 2 21:42:36 DD-WRT httpd[1540]: pam_unix(httpd:auth): check pass; user unknown
Jun 2 21:42:41 DD-WRT start_single_service: openvpn : OpenVPN daemon (Client) successfully stopped
Jun 2 21:42:41 DD-WRT start_single_service: pptpd : daemon successfully stopped
Jun 2 21:42:43 DD-WRT start_single_service: openvpn : OpenVPN daemon (Client) hanging, send SIGKILL
Jun 2 21:42:43 DD-WRT httpd[1540]: pam_unix(httpd:auth): check pass; user unknown
Jun 2 21:42:43 DD-WRT start_single_service: openvpn : OpenVPN daemon (Client) starting/restarting...
jefbuan
DD-WRT Novice


Joined: 18 May 2011
Posts: 23

PostPosted: Tue Jun 04, 2019 7:30    Post subject: Reply with quote
rotarum wrote:
Set to Wide HT40.

But still the Clients connect with HT20 only (65mbps)

Build #39296

Encryption set to WPA2 PSK with AES... what can it be??



EDIT:

Upgrade to the latest build from broly (first page)

DD-WRT v4.14-r39800M CRISPIN (05/14/19)


HT40 Mode and VHT160 same time working perfect!

Beside this the LED's show correctly too Smile


I did broly lstest build but have you manage to enable NAS ??

edit: upgrade to broly's latest v39957 , I can mount my usb HD drive..Big thanks..
ashishchandra
DD-WRT Novice


Joined: 17 Oct 2018
Posts: 14

PostPosted: Sun Jun 09, 2019 17:27    Post subject: Reply with quote
I did broly lstest build but have you manage to enable NAS ??

edit: upgrade to broly's latest v39957 , I can mount my usb HD drive..Big thanks..[/quote]


Is it stable ?
jefbuan
DD-WRT Novice


Joined: 18 May 2011
Posts: 23

PostPosted: Mon Jun 10, 2019 9:23    Post subject: Reply with quote
yup very much stable no reboots at all..NAS and Entware also works great.Thanks to broly for bothering his PM .Butstill I dont have write permission on my NTFS USB drive , ext4 format no problems,,other than that it's stable
ashishchandra
DD-WRT Novice


Joined: 17 Oct 2018
Posts: 14

PostPosted: Tue Jun 11, 2019 21:24    Post subject: Reply with quote
jefbuan wrote:
yup very much stable no reboots at all..NAS and Entware also works great.Thanks to broly for bothering his PM .Butstill I dont have write permission on my NTFS USB drive , ext4 format no problems,,other than that it's stable


Hey, thanks for your response.

Process to revet to stock is also the same right ?
Iangh
DD-WRT Novice


Joined: 22 Apr 2019
Posts: 26

PostPosted: Tue Jun 11, 2019 23:22    Post subject: Reply with quote
Tried Broly's v39957. Faster speeds on 2.4Ghz, went from 50Mbps to 60Mbps sat next to the router. On 5Ghz actually saw a rate of 490Mbps in network properties of win10. Before, highest was 390Mbps. Did notice a lot more errors, nothing to be excited about but not zero as per standard dd-wrt. Has the same problem as before: couldn't assign static addresses. I could put one in and save and apply and it would revert to blank. Workaround was to install backup from standard, which captured the static/MACs but when I went to add one via the gui they all disappeared.

On a previous Broly it didn't capture the DDNS but this version did.

Anybody else using Broly's having an issue with static addresses?
jefbuan
DD-WRT Novice


Joined: 18 May 2011
Posts: 23

PostPosted: Wed Jun 12, 2019 10:04    Post subject: Reply with quote
ashishchandra wrote:
jefbuan wrote:
yup very much stable no reboots at all..NAS and Entware also works great.Thanks to broly for bothering his PM .Butstill I dont have write permission on my NTFS USB drive , ext4 format no problems,,other than that it's stable


Hey, thanks for your response.

Process to revet to stock is also the same right ?


yup same..if you tried his version , I think you will not mind reverting back to orig firware..
casualtester
DD-WRT User


Joined: 13 Dec 2015
Posts: 304

PostPosted: Tue Jun 18, 2019 0:32    Post subject: Reply with quote
Hello broly,
Would it be possible to use your firmware in a DIR-878?

Under the impression that the basic difference between the DIR-882 and DIR-878 was that the 878 was missing a USB port...
SoLoR
DD-WRT Novice


Joined: 21 Jan 2019
Posts: 31

PostPosted: Tue Jun 18, 2019 10:40    Post subject: Reply with quote
casualtester wrote:
Hello broly,
Would it be possible to use your firmware in a DIR-878?

Under the impression that the basic difference between the DIR-882 and DIR-878 was that the 878 was missing a USB port...


yes it possible im using it since april and it works...
Goto page 1, 2, 3, 4, 5  Next Display posts from previous:    Page 1 of 5
Post new topic   Reply to topic    DD-WRT Forum Forum Index -> Ralink SoC based Hardware All times are GMT

Navigation

 
Jump to:  
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum
You cannot attach files in this forum
You cannot download files in this forum