| 01:17.03 | *** join/#edev djerome (~djerome@ip68-2-20-108.ph.ph.cox.net) |
| 02:01.21 | *** join/#edev ffio (~fire@unaffiliated/security) |
| 02:11.48 | *** join/#edev ffio (~fire@unaffiliated/security) |
| 02:21.24 | *** join/#edev JoeLlama (~snork@unaffiliated/joellama) |
| 03:00.57 | *** join/#edev roxfan (dunno@52.184-246-81.adsl-dyn.isp.belgacom.be) |
| 03:51.22 | *** join/#edev TheSeven (~quassel@rockbox/developer/TheSeven) |
| 04:18.47 | *** join/#edev Crofton (~balister@pool-96-240-166-228.ronkva.east.verizon.net) |
| 05:33.31 | *** join/#edev clio (~andrej@85.159.109.222) |
| 05:33.59 | *** join/#edev amarsman (~marsman@90-145-17-249.wxdsl.nl) |
| 05:48.55 | *** join/#edev mbuf (user@nat/redhat/x-jhccdnlkamooypxm) |
| 06:08.12 | *** join/#edev ffio (~fire@unaffiliated/security) |
| 07:15.12 | *** join/#edev ffio (~fire@unaffiliated/security) |
| 09:26.23 | *** join/#edev flo_lap (~fuchs@Maemo/community/contributor/florian) |
| 10:30.31 | Kasreyn | time to end support for an old processor.. ? this otherwise i686 class processor lacks CMOV which means the kernel won't boot (if compiled for ATOM with CONFIG_X86_GENERIC). ( VIA Samuel 2 (CentaurHauls C3-2 Nehemiah) ) |
| 10:32.19 | _av500_ | too bad :) |
| 10:32.46 | Kasreyn | the only way i'm getting 3.9.3 to boot is with CONFIG_M486.. not even CONFIG_MVIAC3_2 works.. |
| 10:33.28 | _av500_ | send a patch? |
| 10:34.05 | Kasreyn | CONFIG_MVIAC3_2: "Note, this kernel will not boot on older (pre model 9) C3s." |
| 10:34.31 | Kasreyn | model on this is 7 >( |
| 10:37.25 | Kasreyn | this hardware end of lifed a long time ago. it would be nice if it could work but i don't think it matters that much.. |
| 10:39.40 | Kasreyn | it only matters because i happen to have one (and use it) :) |
| 10:41.15 | Kasreyn | oh ok, that's what CONFIG_MCYRIXIII is ofr |
| 10:44.31 | Kasreyn | and.. this also has 3DNow! |
| 10:45.24 | Kasreyn | perfect. |
| 10:58.38 | *** join/#edev amarsman (~marsman@90-145-17-249.wxdsl.nl) |
| 12:04.55 | *** join/#edev g1powermac (~g1powerma@rrcs-74-87-150-6.west.biz.rr.com) |
| 12:04.56 | *** join/#edev g1powermac (~g1powerma@unaffiliated/g1powermac) |
| 12:39.02 | *** join/#edev Crofton (~balister@63.164.30.72) |
| 13:13.21 | *** join/#edev ffio_ (~fire@unaffiliated/security) |
| 13:23.19 | *** join/#edev Crofton (~balister@63.164.30.72) |
| 13:44.00 | *** join/#edev dgerlach (~a0274052@192.94.92.11) |
| 13:53.00 | *** join/#edev ccole (~cole@coledd.com) |
| 14:08.25 | *** join/#edev gustavoz (~gustavoz@host108.190-226-80.telecom.net.ar) |
| 14:42.42 | *** join/#edev vpopov (~happylife@dyn-32-201.fttbee.kis.ru) |
| 14:55.54 | *** join/#edev prpplague (~danders@adsl-68-88-72-115.dsl.rcsntx.swbell.net) |
| 15:32.44 | *** join/#edev Crofton (~balister@63.164.30.28) |
| 16:20.46 | *** join/#edev dgerlach (~a0274052@192.91.66.186) |
| 16:24.47 | iospace | i reserve the right to skullfuck anyone who uses magic numbers in production code |
| 16:33.12 | *** join/#edev gustavoz (~gustavoz@host108.190-226-80.telecom.net.ar) |
| 16:38.53 | SpeedEvil | What about pi? |
| 16:39.12 | iospace | #define PI_VAL 3.14 :P |
| 16:39.16 | iospace | unless it's pre-defined |
| 16:45.30 | MonMotha | I make exceptions for 0, 1, sometimes 2, and 8 (though CHAR_BIT is there for a reason) |
| 16:45.53 | MonMotha | otherwise yes, magic numbers are (probably, ymmv) "evil" |
| 16:46.32 | MonMotha | I only except 8 (allowing its use instead of CHAR_BIT) because CHAR_BIT is a bit long and unwieldy |
| 16:46.49 | MonMotha | though if you think you need to be portable to a DSP, fraking use it |
| 16:48.38 | iospace | MonMotha: i usually make an exception for -1 as well |
| 16:49.22 | iospace | like if i wanna set an entire value to all ones, it's actually a useful trick to go uint8_t var = (unit8_t)(-1); |
| 16:49.25 | iospace | :P |
| 16:50.57 | MonMotha | iospace: I prefer ~1 for that |
| 16:51.08 | MonMotha | though you run into some funny type promotion things, sometimes |
| 16:51.20 | MonMotha | and er ~0 |
| 16:51.23 | iospace | :P |
| 16:51.33 | iospace | either way |
| 16:51.40 | MonMotha | but yes, same idea |
| 16:51.52 | MonMotha | basically, if it's a known C idiom, I'm OK with it having a magic number in it |
| 16:51.59 | MonMotha | e.g. (1 << BIT_POSITION) |
| 16:52.12 | iospace | yeah |
| 16:52.20 | iospace | hence why i'm ok with -1, 0, and 1 |
| 16:52.30 | iospace | (that was the rule at school too) |
| 16:53.07 | MonMotha | depending on your code, I may allow you to repeatedly mask using a literal power of 2 |
| 16:53.26 | MonMotha | cases where I find that totally acceptable is pretty rare, but I've seen it and been OK with it |
| 16:53.40 | MonMotha | usually when accompanied by suitable documentation and doing it another way would make it hard to read |
| 16:55.03 | iospace | yeah |
| 16:55.11 | iospace | but the code i'm looking at doesn't have the comments :< |
| 16:56.04 | MonMotha | ah, that's bad |
| 16:56.13 | iospace | it also has "if((Var = Op()) != Val) in places |
| 16:56.23 | iospace | HAVE YOU EVER HEARD OF READABILITY? |
| 16:56.24 | iospace | T_T |
| 16:56.37 | MonMotha | I am currently working in an "amusing" (not in a good way) codebase. The code style itself is fine - not my preference but fine. However, the structure is total spaghetti. |
| 16:56.44 | iospace | :P |
| 16:57.15 | MonMotha | it's a GIANT state machine, but the state transitions are scattered across 3 files, and nothing has a clear, single function. EVERYTHING has side effects. |
| 16:57.31 | MonMotha | oh, and it has an 8 page ISR |
| 16:57.37 | iospace | pats MonMotha |
| 16:57.38 | iospace | there there |
| 16:57.43 | MonMotha | heh |
| 16:58.13 | MonMotha | fortunately, the client is paying well (they got stuck with this code from another consultant), and I've been authorized to "do whatever I deem necessary" up to and including re-writing it |
| 16:58.24 | iospace | :D |
| 17:01.08 | MonMotha | so right now I'm just going through and reverse engineering an op spec out of the system |
| 17:01.13 | iospace | heh |
| 17:01.23 | MonMotha | I at least can compile the code, so I have traceability from function back to code. |
| 17:01.49 | MonMotha | it'll also be nice to actually have an op spec |
| 17:01.55 | SpeedEvil | "20% of the final budget was spent not on development, but on taking out a contract on the previous programmer' |
| 17:02.14 | MonMotha | darn near - I'm not sure how this got so bad |
| 17:02.42 | iospace | haha |
| 17:02.51 | MonMotha | the people I've talked to are of the impression that the previous consultant was more used to working on larger, higher level systems, but this really looks like somebody who was used to programming little 8051s and 1kB PICs got over their head on something a bit bigger |
| 17:03.04 | iospace | ah |
| 17:03.11 | iospace | yeah I work on UEFI for i7s |
| 17:03.22 | MonMotha | it definitely has the "Just keep adding code until it does what you want" vibe |
| 17:03.42 | iospace | ... see |
| 17:03.44 | MonMotha | tons of page-long copy/paste with 1-2 line changes, etc. |
| 17:03.47 | iospace | that's not doing it right D: |
| 17:03.51 | MonMotha | no, it's not |
| 17:04.50 | MonMotha | that said, isn't UEFI just a mess by definition? :) |
| 17:05.21 | MonMotha | it's like the Win32 and ACPI people got together and had a red-headed stepchild |
| 17:05.57 | MonMotha | just different enough that you can deny it's not Win32 or ACPI, but it still has all the cruft |
| 17:05.58 | iospace | ha |
| 17:06.03 | iospace | hahahahacpinotbeingamess |
| 17:06.05 | iospace | hahahaahahahah |
| 17:06.36 | MonMotha | oh, no...ACPI is definitely a mess |
| 17:06.50 | MonMotha | I remember trying to figure out which random 4 letter combination of letters switched the video mux on my laptop |
| 17:06.53 | MonMotha | (it's ATPX, IIRC) |
| 17:06.57 | iospace | actually my perspective on things is rather limited, I haven't worked on anything else in terms of BIOS and such :P |
| 17:07.06 | iospace | only just graduated :P |
| 17:08.56 | *** join/#edev Crofton (~balister@63.164.30.94) |
| 17:10.35 | MonMotha | I assure you that while trying to find out which one it was, I came up with several other 4 letter combinations |
| 17:10.53 | MonMotha | many of which you've probably heard |
| 17:11.22 | iospace | heh |
| 17:28.16 | *** join/#edev dgerlach (~a0274052@192.91.66.186) |
| 17:29.43 | *** join/#edev florian (~fuchs@Maemo/community/contributor/florian) |
| 17:37.17 | prpplague | CosmicPenguin: ping |
| 17:45.01 | *** join/#edev bsdfox (~Bob@c-71-197-82-145.hsd1.ca.comcast.net) |
| 17:45.02 | *** join/#edev bsdfox (~Bob@unaffiliated/bsdfox) |
| 18:04.48 | landley | iospace: I went a step worse. You'll love line 1.117 of http://landley.net/hg/toybox/rev/ce0a9368b82f#l1.90 |
| 18:06.25 | landley | As described in http://lists.landley.net/pipermail/toybox-landley.net/2013-May/000994.html |
| 18:06.32 | iospace | eh |
| 18:06.34 | iospace | shrugs |
| 18:07.00 | *** join/#edev dgerlach (~a0274052@192.91.66.186) |
| 18:07.04 | landley | Not just magic number, but doing math in ioctl space. |
| 18:07.12 | iospace | heh |
| 18:08.10 | MonMotha | landley: that's not terrible |
| 18:08.20 | MonMotha | landley: to me that falls under the "idiom" category |
| 18:08.49 | MonMotha | though I might have done (t->off & 0xFF00) == 0x8900 |
| 18:09.16 | MonMotha | now, the 0x89 part - that's a magic number, yes |
| 18:09.23 | landley | MonMotha: yeah, it gets promoted to int either way. |
| 18:09.43 | landley | MonMotha it's actually a documented magic number, but you have to dig. man ioctl_list I think. |
| 18:09.51 | MonMotha | right, and to me, an AND is a more obvious mask than an OR |
| 18:09.52 | *** join/#edev g1powermac (~g1powerma@host-063-131-200-114.dhcp.fewpb.net) |
| 18:09.53 | *** join/#edev g1powermac (~g1powerma@unaffiliated/g1powermac) |
| 18:10.23 | MonMotha | landley: hum is there no #define or similar for it? |
| 18:10.40 | landley | MonMotha: there are _IOW macros, or some such. |
| 18:11.46 | landley | I've been elaborately explaining my cleanups on the toybox list. The second link was my explanation of why I made those changes. |
| 18:12.31 | landley | (That's also why I've checked in over a dozen changes to this file already and am still maybe 2/3 of the way through: I'm doing it in baby steps so I can explain my reasoning in hopes other people will learn how to turn a 1500 line unreadable file into a 500 line readable file.) |
| 18:13.30 | MonMotha | and this pretty much sums up why "Lines of code per day" is such an awful metric |
| 18:13.43 | MonMotha | either that or you interpret it as "lower is better" (including negative numbers) |
| 18:14.58 | *** join/#edev Crofton (~balister@63.164.30.94) |
| 18:17.01 | landley | While I was editing The Art of Unix Programming, Ken Thompson's review of the first 8 chapters came in the mail. |
| 18:17.21 | landley | I got to see his comment "One of my most productive days was throwing away 1000 lines of code." in the margin of the maunscript in his own writing. (Red pen.) |
| 18:17.41 | landley | It was my email .sig for a while. :) |
| 18:37.01 | *** join/#edev calculu5 (~calculus@gentoo/user/calculus) |
| 18:46.54 | SpeedEvil | :) |
| 19:27.23 | *** join/#edev JoeLlama (~snork@unaffiliated/joellama) |
| 20:06.52 | *** join/#edev Anaphaxeton (~george@unaffiliated/anaphaxeton) |
| 20:15.49 | *** join/#edev contempt (contempt@unaffiliated/contempt) |
| 20:39.58 | *** join/#edev vpopov (~happylife@dyn-32-201.fttbee.kis.ru) |
| 20:42.09 | billtraynor | landley: you edited The Art of Unix Programming? |
| 20:42.57 | landley | billtraynor: yup. |
| 20:43.08 | landley | For a definition of "edit" that goes "it was 9 chapters when I started, 20 when I stopped". |
| 20:43.38 | landley | Paragraph 2: http://www.catb.org/esr/writings/taoup/html/pr01s06.html |
| 20:44.41 | landley | This was before Eric's libertarian leanings got hijacked by Koch industries and he went into full climate change denialism, singing the praises of the book "the bell curve" mode. |
| 20:45.21 | billtraynor | heh |
| 20:45.42 | landley | (I crashed on the couch in his basement for 3 months editing TAOUP in 2003. Our falling out was... 2010?) |
| 20:45.55 | landley | Pity. I miss my friend. Not hugely fond of the guy who's replaced him. |
| 20:46.33 | billtraynor | i was on a plane with him once and thought about talking to him, but then thought better of it |
| 20:46.41 | landley | Anyway, somewhere between 1/3 and 1/2 of TAOUP is material I contributed, but I made Eric type it up because it was his book and needed a consistent narrative style. |
| 20:46.41 | billtraynor | it was fairly recent |
| 20:47.18 | landley | billtraynor: I met him at Atlanta Linux Symposium in 2009 and was impressed by the fact he knew the words to rather a lot of Tom Lehrer songs. |
| 20:47.25 | landley | I mean 1999. |
| 20:47.27 | landley | I'm old. |
| 20:48.11 | landley | I sent him a copy of the movie "1776" (yes, with the musical number about the triangle trade), because he was Ben Franklin, Linus was Thomas Jefferson, and Stallman was John Adams. |
| 20:48.34 | billtraynor | is also old |
| 20:48.38 | landley | (I also sent one to Stallman, who did not appreciate the comparison.) |
| 20:49.16 | billtraynor | bbiab, spawn need feeding |
| 21:13.28 | *** join/#edev _ffio_ (~fire@unaffiliated/security) |
| 21:36.39 | *** join/#edev solomint (~solomint0@unaffiliated/solomint) |
| 21:37.06 | *** part/#edev solomint (~solomint0@unaffiliated/solomint) |
| 21:40.59 | *** join/#edev dgerlach (~a0274052@192.91.66.186) |
| 21:49.05 | *** join/#edev g1powermac (~g1powerma@rrcs-74-87-150-6.west.biz.rr.com) |
| 21:49.06 | *** join/#edev g1powermac (~g1powerma@unaffiliated/g1powermac) |
| 21:51.26 | *** join/#edev dijenerate (~dijenerat@65.48.188.233) |
| 22:32.51 | *** join/#edev Crofton (~balister@pool-96-240-166-228.ronkva.east.verizon.net) |
| 23:40.50 | *** join/#edev contempt (contempt@unaffiliated/contempt) |