Logo Created by the Dynamic-Duo at RoboDesign

[ LibTomCrypt ] [ LibTomMath ] [ LibTomPoly ] [ LibTomNet ]
[ LibTomFloat ] [ TomsFastMath ] [ LTC Book ]

[ News ] [ Features ] [ Changes ] [ Download ] [ Mirror1 ] [ Mirror2 ]

---

Welcome

Welcome to the LibTomCrypt home here at libtomcrypt.com. LibTomCrypt is a public domain open source cryptographic toolkit written in portable C. The library supports a variety of cryptographic primitives including ciphers, hashes, random number generators, authentication schemes, and public key algorithms. (see more)

This project, like all LibTom projects, is public domain and free for all purposes.

Send patches, bug reports, errata, etc, to me.

My C.V. is here and makes for good readin'

Jump to the news.

Project Fundraising

Yeah, everyone is doing fundraising this time of year. My goals are a bit more lax but just as important. I'm planning on my first expense being to Toorcon in March 2007 for the amount of $1500 USD. I haven't decided about going to Shmooocon but that's likely out as it stands. I'm definitely going to Toorcon (to present and hand out freebies). So the goal is simply to break even on my smaller budget of $2500 USD. This of course, doesn't include any hardware or other costs (like domain names, artwork, etc). I'm hoping to raise at least $1000 by March 2007 and the remainder by September.

So if you work for a company which uses the library, consider picking up a copy (or three). All I need to do is sell 20 copies of the hard cover copy, or 40 copies of the soft cover copies to make my March 2007 goals (or some combination of the two).

In the event you don't want to kill a tree but still want to contribute to the LibTom Projects, contact me via e-mail about sending in a donation via cheque (or money order). And please remember, the projects are still public domain, and I will still be donating to the various causes regardless of the outcome. This endeavour is just an attempt to balance my books sanely.

Other News

• December 22nd, 2006. Got more orders for the LTC manuals. Thanks!!! Ever closer to the March 2007 goal! BTW check out the prototype site here. It will be going live as libtom.org in a week or two.
• December 19th, 2006. Hard cover LTC manuals are available at the store front. $50 of each hard cover goes to the LibTom Projects. w00t. And thanks to the person who bought a (for-profit) soft-cover LTC manual on the 2nd day it was out. Coolies!
• December 18th, 2006. Muahahahaha, I got neato faster ECC by making TFM a single compile unit. Check this out. TFM 0.11 will support this by using "NEED_SPEED=1 make" which binds the entire lib to one .c file and compiles it. Sure it's big, but it's also crazy fast.
• December 16th, 2006. Release of LTC 1.16. Also the LTC book/manual is available for purchase. The $40 copy is the profit copy which funds my random LibTom projects (see the LTC book page for more info), my take is $25 USD per for-profit book sold. I make $0 on the "at-cost" copies.
• December 11th, 2006. Ok, so let's recap. Wiki posts article about me, I vote AFD, so far I'm losing. Ok, fine, let's make the article more interesting at least. Also, this weekend was a writeoff, expect LTC 1.16 ETA a week or so late. Like I said earlier, the code is frozen, the manual is what I have to work on. And please, don't write about Tom without first asking in the future. I keep my outlet of insanity limited to libtomcrypt.com for a reason. kthxbye.
• December 8th, 2006. Nearly finished the first pass of the manual, going to add examples on the weekend.
• December 6th, 2006. Worked on a couple extra chapters last night, going to spend the weekend adding examples to the manual to beef it up a bit. Code is frozen at this point, only bug fixes. Still in good shape for the 13th. Bad news for the people (of which there are 7) that I promised LTC manuals to. Turns out my "overspending" went a bit too far. Since I promised Toorcon the stipends before I did the LTC manuals that takes precedence. Most likely I won't be able to do the manuals until late January, or sometime in February. Sorry, but hey what you expect, I'm just one dude doing crazy things!
• December 4th, 2006. Added a lot of fixes to the ECC/RSA/DSA code to clean up the error handling (w.r.t. heap errors). Merged in a patchset from CRI that clears up some warnings when using MSVC. I've also cleaned up the ECC/DSA chapters of the manual (getting ever closer to ready to print), and finally I've added Shamir's Trick testing to the testprof suite (monte carlo style). LTC 1.16 is still set for the 13^th, and I highly recommend getting it if you use LTC.
• November 32nd, 2006. ... err December 2nd, 2006, added Shamir's trick to the ECC lib, 34 to 40% faster ECC verifications. w00t. Also fixed a handful of memory leaks (that would only arise when you ran out of heap, kinda a double whammy). Update: I added the shamir trick thingy to the Fixed Point code. ~14-19% faster. In short, on my 3.42GHz Core 2 Duo I can do about 4800 ECC-192 EC-DSA verifies per second. Not bad for a 2.4GHz core mid-range core.
• November 30th, 2006. Bad news, seems I went a bit over budget with some recent spendings. I won't have the freebie copies of the LTC manuals out to the people who asked for them until sometime towards the end of January. On the plus side, the test printing does look sweet.
• November 30th, 2006. Got my test print of the LTC manual (soft cover). It looks awesome. Still waiting on the hard cover.
• November 29th, 2006. Found a few bugs in LTC. Foremost is the undefined behaviour in ecc_sign_hash() when mp_init_multi() fails. Fixed. Runner up is that (imho) pkcs_1_pss_decode() should not return CRYPT_OK if the padding is invalid (note: it will catch invalid signatures, so it's not a security problem, this is a style thingy).
• November 29th, 2006. Finished beta code for OLPC (the code for the curious). Even includes a mini-heap library which I may clean up and add as a side lib to LTC for folk who want to avoid using libc in their embedded work. Also found some stray memcmp/etc in LTC. Cleaned that up. w00t.
• November 28th, 2006. Busy busy, planning on LTC 1.16 around Dec 13th. Will have fixed GCM (apply the patch in the meantime), EC-IES support and accelerated EC-DSA for those using MECC_FP. I also plan on releasing the LTC book on that day. In my near future, I have shmooocon slides to write tonight, then OLPC code over the next two days due. Crazy crazy. In essence, I'm not slated to have free time until the 14th (that is if I want to get this book put together and LTC 1.16 out the door). All worth it imho. Just means I have to party harder.
• November 26th, 2006. One of my RAID-5 drives died today :-( so I'll be spending the day backing up. Likely I'll move from 3x250 to 2x320 and just RAID-1 (easier and fewer drives). Fortunately I keep copies of the LT CVS on various random drives. Update: I've bought a pair of Seagate 320GB SATA drives (RAID-1) and am in the process of creating the array. No data loss fortunately. Sadly this did cost me a precious weekend day. I'll have to catch up during the week... arrg (lesson for the kids: use RAID arrays in your personal boxes)
• November 25th, 2006. Added UTF-8 support to the ASN1 code. Yes, that's how I spent my Saturday evening ... :-( at least I had good techno in the background.
• November 25th, 2006. Wrote the first half of the code for OLPC's signed BIOS code. Tomorrow I work on my Shmoocon presentation. Next week (during the week) I coordinate with OLPC to get the 2nd half done. After that I go back to the LTC manual and after that I go towards LTC upgrades (like UTF-8 support in ASN1). Note: The release of the LTC manual and LTC 1.16 will be synchronized so that the manual refers to the "latest" code.
• November 23rd, 2006. Got my E6600 running at 3465MHz (385x9). Built LTC 1.16 (wip) in 9.4 seconds with "make -j3". A time rivaled only by an Opteron 285 workstation with two dual core processors and "make -j5" (which got ~8.2 seconds for LTC 1.14).
• November 23rd, 2006. Update: The shop (ShopRBC) has agreed to move my box to the front of the list so I can have it back tonight. They rock. Thanks guys! BTW this is where I get all my LibTom desktops from. Me Core 2 Duo box is in the shop, upgrading to an E6600 (2.4Ghz 4MB L2) so I can give my FSB a rest (bump it down to ~380MHz from 420MHz). I've got a CD-R copy of the LT CVS so I'll work on my laptop until it gets back to me. :-) (oh yeah and new images on the ltcbook page)
• November 22nd, 2006. Added the LTC book page to the website. Read it.
• November 21st, 2006. Worked some on the manual, made corrections and additions to the hash, MAC, and RSA chapters. Even after adding the new ECC and build documentation I'll make an additional half dozen passes looking for errors and missing stuff. Looks good for a Dec 1st printing. Also got a few quotes for the back cover from Greg Rose and Paul Kocher (testimonials on a free product? Sure why not!)
• November 21st, 2006. Got some private samplings printed in hardcover from lulu.com. Needless to say, the quality is great. I'm going to step it up and try to get manual edited during the week and work on OLPC on the weekend. I'll try to have the LTC manual for the first week of December, then I'll offer it with the warning that i haven't performed a test print yet. Once I get my copies I'll give the nod for those wanting to wait to make sure the book is setup correctly. In theory by mid December people could order printed copies.
• November 20th, 2006. Brian Gladman has offered vectors for GCM which I'm going to add to LTC. OLPC work this weekend and next, then I resume working on the LTC manual. I plan to offer @cost soft/hard copies after I get all the kinks out (~January). I may offer >cost copies to raise money for my 2007 endeavours. We'll see. I'd appreciate feedback (privately) from people concerning whether they're even interested in buying a non-@cost copy. Please contact me!
• November 19th, 2006. Brian Gladman pointed out a recent change to how I handled IVs in GCM was wrong, I reverted the code (patch), and it complies against his code now. The GCM code is frozen now. Current plans are to work on the OLPC code over the next two weeks. The goal is to deliver beta code to them for Dec 1st. Which is when I'll resume work on LTC 1.16 and the manual. This means likely the completed manual will not be up until the new year. Please apply the GCM patch in the mean time :-) (fortunately, it's just a bug and not a security problem)
• November 17th, 2006. LTC 1.15 is up. Keep in mind the manual is INCOMPLETE for this release and will be fixed for v1.16. Hint: stay away from F9-MAC since it's not documented...
• November 16th, 2006. Ok, to pull LTC 1.15 out on time I've decided to delay the manuals by a week or so. The printed manuals will reflect v1.16. This way I can at least get the code out on time. I have OLPC code due soon, so most likely printed LTC manuals won't materialize until the end of the year. Sorry folks. I'll try my best to speed things up. Even after the manual is done I need a sample print before I can order more. So there is a 2 week lead time on all of this...
• November 15th, 2006. Bad news. First, I may have to curtail my conference list a bit. Sadly I can't count. Still will be at Toorcon, and in theory I'll try for Shmoocon at least. Second bit. I may be a day or two late with LTC 1.15. The code is ready but the manual is not. I'll try my best but so far this week is a mess. On the plus side I have my UK tour planned (mostly). Will be in England for April 1 - 14, inside that trip I'm visiting Paris, Budapest (briefly), and Arad. :-) (maybe that's why I don't have conference money...)
• November 14th, 2006. Christophe Devine found a bug in anubis on the Tru64 platform (stupid upper bits hehehe). Fixed. LTC 1.15 builds and runs on the Tru64 platform. w00t. Should work on the other UNIX platforms too...
• November 13th, 2006. Found a bug in LTM 0.39 (line 83 of bn_fast_s_mp_mul_high_digs.c) thanks to Valgrind. I will do a quicky release of LTM 0.40 on Friday along with LTC 1.15 (found a Yarrow bug with Valgrind as well... w00t)
• November 13th, 2006. Working on the manual ... about 25% through it. Gonna call it quits for the night. Code is frozen at this point for the 1.15 release.
• November 13th, 2006. Began working on LTC 1.15 again (don't ask me about the weekend...). Plan to have everything together for this Friday. Please ask me if you want a pre-release to test out. BTW to clear up, I am still planning to allow people to buy the manuals. I just will sell them at cost. If you want to help me out in my cause just email me and we can sort something out. Still at least 2-3 weeks away from having the books up for sale. Also a prelim list of conferences I plan to attend (with freebies) is : Codecon, Shmoocon, Crypto, CHES (doubtful), and Toorcon. I have something in store for David and Toorcon :-)
• November 9th, 2006. I've got KASUMI/F9 working fine [so far]. I've cleaned up Andreas' PKCS #1 v1.5 code to actually comply with the spec (bad Andreas bad!). BTW, "openssl rsautl -sign" doesn't follow the spec either ... (it doesn't encapsulate the signature in the SEQUENCE, it just pads with 0xFF). Release is still on schedule for the 17th of November. w00t. Still haven't updated the manual .. arrg .. so much work...
• November 8th, 2006. I've added KASUMI and F9-MAC to the lib. Can't find test vectors for EITHER anywhere. If anyone has official vectors please send tom. Update: Found Kasumi vectors on cryptonessie and F9-MAC complies with 3GPP vectors. (caveat: F9-MAC requires the caller perform the padding).
• The Eighth of November, 2006. I've finished my pass through the manual. I have code to write (adding F9-MAC and more callbacks, maybe even KASUMI to round off the 3GPP side) then it's all glory. Why didn't anyone tell me how bad the manual was? My god, grammar much? Lets just say there is a fair number of fixes to make this half decent. And since I'm mostly just winging it, that's really like a quarter decent.
• November 7th, 2006. I've changed my mind about the LTC manuals. I'll print copies for the people I said I would give copies to, but I won't offer it for sale. The updated material will be part of the source archive where you can free load and read it.
• November 7th, 2006. Added SECP 112, 128 and 160 bit curves (the 'r1' variants) to the ECC side. Added ANSI X9.63 import/export of public keys (section 4.3.6) as well.
• November 7th, 2006. Edited most of the LTC manual (first pass) going to make another pass shortly. Also organizing with the OLPC folk to work on their BIOS security. Busy Busy.
• November 6th, 2006. Got my draft copy of the LTC manual today. Looks nice. Clear print, properly bound, packed snugly. I'm going to spend the next week editing the manual and the following week typing up the changes. In theory, I could offer advance copies (for those willing to order at the same time I order my first print copies) as soon as 2 weeks from now.
• November 5th, 2006. Remember, remember, the fifth of november, The Gunpowder Treason and Plot, I know of no reason why Gunpowder Treason, Should ever be forgot. Oh yeah, I'm also putting pressure on my contacts at Intel to score me a non-commercial ICC 9.1 license :-) If anyone from Intel is reading this help the projects out y0!
• November 4th, 2006. Made the CTR mode RFC 3686 compliant (increment then encrypt) by adding the LTC_CTR_RFC3686 mode flag to ctr_start(). Also fixes to SOBER/RC4 for valgrind users, F8 chaining speed ups, and a bunch of other fixes. ETA on v1.15 is about 2 weeks.
• November 3rd, 2006. Added the Korean SEED cipher to the list of supported ciphers. Along with XCBC-MAC this release will have a bit more support for newer TLS specs... :-)
• November 3rd, 2006. Seems someone is joe-jobing me again. This time about the 2nd book. If you see posts with weird headers going all over the net about the 2nd book that would be this joe-jobber. I honestly don't get these people. I just ignore people I don't like, I don't spend my waking hours toiling over ways to make their lives complicated... I guess that's what separates us Adults from the children.
• November 2nd, 2006. I've added XCBC-MAC mode to the list of MAC algorithms supported. RFC 3566 compliant y0.
• November 2nd, 2006. I got (and paid for) the LTC manual covers. Here is a low-res copy of the front cover. Ain't it spiffy? I should have the edits finished in ~2 weeks and then a sample printing. If all goes to plan the book should be offered in ~4 weeks. More details to follow.
• November 1st, 2006. In theory I should get the draft print of the LTC manual on Monday. It will also be a good chance to check out what lulu.com quality looks like (hopefully really good).
• November 1st, 2006. RoboDesign is putting together the cover for the LTC manual. Just want to say they do awesome work. I'll be hosting a low res scan of it as soon as I pay for the kit (they still have to make the back cover first).
• November 1st, 2006. I've reformatted the LTC manual as 6x9 already. Just waiting on my print of the existing manual to hand edit and then update. Just so we're clear, the manual is still going to be part of the public domain archive. I also plan to release TFM v0.10 today which has a few bug fixes and the new AVR32 support (requires GCC 4.0.2 or higher).
• October 29th, 2006. I've setup the LTC manual on lulu.com (private only) to get a sample printing. I have plans to clean up, add to and publish the manual as a softcover 6x9 rated for cost+$15 (estimated cost around ~25USD plus shipping and taxes). The manual will be part of the source archive of course. The goal here is to fund the projects a bit and make my outgoing budget a bit smaller. Update: The book will be available in both soft and hard cover 6x9 prints (~200 pages). The soft will cost approximately $23.54 USD, and the hard will cost approximately $33.02 USD (both excluding taxes and shipping, includes $15 author royalties).
• October 26th, 2006. I review the last three chapters tonight, book hits the printers tomorrow. Also check out the new featured project above.
• October 25th, 2006. Atmel is preparing a new GCC build for me to try out. Hopefully that will let me test LTC. Did I mention Atmel rocks? w00t.
• October 24th, 2006. So far I've seen the first three chapters of the book all laid out and proper. Some formattting issues that I hope Syngress fixes before printing. Heavens knows they're working hard, but this is why my next book will be in latex. :-). In theory, we're on track to hand the book to the printer on Friday. If they get my corrections in the Quark'ed chapters the book will be a bit cleaner.
• October 21st, 2006. While waiting for the copy edits ... I ported TFM to that AVR32 board that Atmel had sent me. Testing it right now against LTM and it seems to be working. Took a bit to sort out the montgomery macros but other than that it was straightforward. TFM 0.10 will support the AVR32 processors. So far LTC won't run in it. I get an ICE (compiler error) during the build, and even with optimizations off it fails test vectors. Doesn't help that Atmel uses GCC 4.0.2 as their development suite ... booo!!!
• October 21st, 2006. Submitted entire book to Syngress. Now to review the copy edit and pray all turns out well. Just a heads up, along with my bignum math book, you may want to pick up this book on ECC. I avoid duplicating their effort in the last chapter (on PK algorithms), but give enough (I think) guidance to help out. Book goes to the printer this Friday, which means that in about 3-4 weeks it will show up in peoples hands.
• October 18th, 2006. 40% finished ch9 of the book [then I have reviews to do before printing NEXT WEEK!!]. got the AVR32 board. Unfortunately, the tools it comes with can't build LTC and I haven't even tried TFM yet. I'm going to push Atmel to get patches against GCC 4.1.x out so I can try that compiler. Right now the board is just pretty looking :-)
• October 15th, 2006. Moved back to Ottawa. I have two weeks to finish the crypto book, and then I get back to the projects. No, I haven't been ignorning my e-mail, I'm just crazy busy. Sit tight! Nice TODO list brewing for LTC v1.15 and I also plan on releasing TFM 0.10 with AVR32 support.
• October 12th, 2006. Atmel has agreed to loan me an AVR32 development board and tools. I will port TomsFastMath to it and then do some ECC/RSA benchmarks on it. w00t w00t. Let's thank Atmel for being a good sport!
• October 10th, 2006. Toorcon talk mirrored at here. Video is now up. Slides to follow.
• October 4th, 2006. Secure Science has agreed to open source the UVLAN program I wrote last year. I gotta clean it up and [sadly] slap a GPL on it. Should be up in a week or so. Think this "encrypted layer 2 switching over UDP/IP" :-)
• October 4th, 2006. My toorcon talk turned into an MP3, w00t. Yes, I have a video but it's 155MB and I'm not posting here. I'll post the toorcon edition of the slides in a bit.
• September 28th, 2006. In San Diego. Fear! David Eder broke my PMAC .. :-( turns out there is an overrun bug in pmac_done which is now fixed [will be part of LTC 1.15]. Darn users, not supposed to find my flaws hehehehehe. Thanks for the help David!
• September 26th, 2006. E6300 stable at 2940MHz and IIRC 1.225V. Any higher and the FSB is too fast for the box to keep up. I've played movies, did bignum runs [at the same time] and played games over it. It's stable at 420x7 with this i965 Express chipset. Oh yeah, I'm heading out to San Diego this Thursday. I'll mostly be hanging out Horton Plaza like a junkie at the QuiKStop so if you're in the hood just give me a hollar. Oh yeah and my crypto book [writing with Simon] is coming closer to completion. I'm near finished the eight chapter and I have one left after this. Slow going but will pick up shortly. Peace out!
• September 23rd, 2006. Sweet jesus ... got my E6300 running at 2975MHz at 1.25V, that's right, a 1113MHz overclock at 0.1V UNDER the stock setting. w00t. If you have the means I highly recommend it. FREE SPEED
• September 20th, 2006. I've resigned from my current JOB to pursue other options. Update the C.V. w00t. No, I'm not looking for a new job or side contracts.
• September 20thish, 2006. My toorcon slides be up, try and find them. BTW you can overclock Core2 processors like mad. My 1.83GHz is stable at 2.6GHz (note: you need PC2-6400 memory for that to work). I probably won't see any of you at the con, but show up for my talk, or ELSE!
• September 12th, 2006. Core2 results here.
• September 12th, 2006. Got my core2 box. Install the gentoo on it. BTW ICH8 chipsets suck bad. Bad Intel, Bad.
• September 6th, 2006. I'm picking up an Intel Core 2 Duo box ($922 CDN) to do benchmarking work on. Don't say I don't do anything for y'all. Hehehe. Thanks to shoprbc for helping with the quote and building the box for me.
• September 1st, 2006. Andreas Lange and I are cooking up PKCS #1 v1.5 support (again) since apparently people still want it. v1.15 will support it (again). People, move to v2 already!!! hehehe.
• August 30th, 2006. Release of v1.14, go fetch it.
• August 29th, 2006. I've offered to donate EC-DSA and RSA-PSS code to the One Laptop Per Child project. Originally was to be based off LTC but they have severe space requirements and I will draft up custom Public Domain code for them. The code will be posted here shortly free for all uses.
• August 26th, 2006. Various improvements to LTC 1.14 (not yet released) like a bug fix to the DER flexi decoder, sped up GCM on SSE2 capable processors, cleaned up code here and there. I'll try to sneak out a LTC release before Toorcon.
  
• August 16th, 2006. WTF. A united flight today was divert to Boston. Fox news cleverly reported for A WHOLE HOUR that a female passenger had a screwdriver, matches, vaseline and a note "about al qaeda". Then they later report "that may not be correct," and that she was simply clausterphobic. I can't imagine how that could happen. I mean the airlines give you SO MUCH ROOM to exist on the plane in a humane fashion. Clearly someone who is clausterphobic must be a terrorist...
• August 14th, 2006. I fly later today to the states. Time to pack some Water with me, I mean, terrorism juice. Here's an open question to you all. Where do you draw the line in the sand? When will you say, no government, you can't do this "for our safety?" I feel sorry for the world. Sure this liquids on the plane is a "trivial" issue but it's a good example of how we all roll over whenever the government asks us to. Fight back. Pack terrorism juice! In short don't do this:
  
  
• August 8th, 2006. Back from Defcon. No arrest warrants. A few good talks, but overall when you have >5000 people in a given location you are bound to run into a few asses. It would also help if more than 1/1000th of the attendees were actually into hacking. Reading your hotmail account from the compo room not only gets you on the "wall of sheep" but calls into question your spending habits. Can't you read e-mail at home? My suggestion, if you plan on going to Defcon, bring a friend. Otherwise, you'll be hella bored.
• August 3rd, 2006. Hell froze over. Out of complete boredom I wrote a C++ wrapper for LTM. Check it out here.
• July 25th, 2006. Book sales are going ok, but not great. So now we plan to up the offer, if you buy the books at the currently offered price, we'll not only print it on paper but for a limited time, bind it with a cover! An extra 30 dollar value, yours free. ... :-) Kiddin, the books are perfect bound. I'm a bit peeved at the lack of reviews though. Good and/or bad would be appreciated.
• July 24th, 2006. AMD and ATI plan to unite. Doh. Hopefully this doesn't spell the end of the Nforce chipsets. Oddly enough, the ATI HQ is in Markham ... where I work alone for IBM as a vendor (from AMD). Now I'll be closer to an AMD office. Woohoo.
• July 23rd, 2006. No updates to LTC or TFM in a while. No I still love the projects. Just haven't learned the lesson from my first book. These things take time!!! Still hoping for a mid Sept release date to the printers. Now only if my employer would give me more time to work on the book ... hmmm ... "strategic corporate synergisms" here we come!
• July 11th, 2006. My employer wants me at Linux World in August. So go I shall. I'll be there with the rest of the team from my dept. Say Hi to us. Oh and I put up my new book project (actually started earlier this year). It was supposed to come out on Sept 1st but due to the LTM book this will likely slide by a few weeks. Rest assured I'm working hard on it (along with my new co-author Simon Johnson) and my horde of technical reviewers (including Greg Rose, Jean-Luc Cooke and a few other seaoned pros).
• July 10th, 2006. Against better financial judgement I'm going to Defcon and Toorcon this year. I'll be presenting at Toorcon, no not about LTC but about ECC (patents, algorithms, and other neato things). See y'all there and BUY MY BOOK!
• July 5th, 2006. Got my 15 copies of the LTM book ... they are choice. I already have 10 of them spoken for. If you want a copy for your business or school please e-mail me.
• July 3rd, 2006. My BigNum book is now available for sale. Pick up your copy!
• June 30th, 2006. RSA got bought out. PWNED. Yet LibTomCrypt remains. Good work to all who have contributed to LT projects!
• June 17th, 2006. Release of 1.13. Sports updates to make it compile better on UNIX like OSes, F8 chaining mode, faster ECC and a few other minor fixes. Note I haven't documented F8 mode yet (I had people asking for the code). v1.14 will sport updated documentation and a few portability issues (see the TODO file).
• June 16th, 2006. All moved into my new location. Yipee. I've still got a few things on my TODO list for v1.13 before I release. I will also be adding F8 chaining mode support for you SRTP/3GPP folk.
• June 13th, 2006. So I was all set to do my driving test today [yeah I know, no license == lame]. I was doing one last practice with the driving instructor when at the end of the hour lesson WHAM! we got rear-ended while waiting to merge with traffic. The auto-karma-gods are out against me!!! hehehe. Also this week I'm moving to my new place in Toronto ... by "my place" I mean some dudes basement. So things will be halted for a week or so as I sort my "stuff" out. I've got to clean up the ECC code and make it all presentable for you youngins before I can make the next release anyways. Sit tight, relax and keep pre-ordering my books. I'm tired of living in a VAN down by the RIVER!
• June 10th, 2006. Some graphs of the new ECC on my AMD 885 box. Comparing all multipliers, just the fixed point multipliers and the fixed point memory requirements (bytes).
• June 9th, 2006. I sped up the ECC again. On an 885 with an 8-bit window I get ECC-192/224/256 in 374K/461K/572K cycles. With a 12-bit window I get 297K/369K/452K ... yes that's right, more than 8000 ECC-192 point muls per second in software! Keep in mind the fastest I got DUAL-threaded RSA-1024 was ~4000/sec. This is a single thread implementation, in theory dual-threads would get close to a 2x boost. Have to define MECC_FP to use it...
• June 8th, 2006. Mike Marin submitted a list of patches to make LTC a bit more friendly towards more traditional UNIX compilers. LTC 1.13 should be a hoot for you AIX, SGI and HP-UX folk.
• May 31st, 2006. LTM book sent to the printers tonight. Will take about a month to get back. mmm papery.
• May 29th, 2006. Got LTC 1.12 up with a series of ASN.1 fixes, source cleanups and the new ECC point multiplier. Enjoy.
• May 28th, 2006. Getting back from SeattleCon (mmm drunken stupor) and have to work on the book on Monday. In theory once the book is out I can get 1.12 out in a day or two. Sit tight. .... mmm got a flight in an hour. West sideeed!
• May 25th, 2006. ECC update... on my Opteron I got ECC-192 down to 391K cycles using 1MB of memory. That's almost twice as fast as dual-threaded RSA-1024 (and with 10-bits more security). Go ECC :-)
• May 24th, 2006. Got the fixed point ECC multiplier in place. Take a look at the results. It's fast, tunable and totally optional.
• May 23rd, 2006. Started working on that ECC FP multiplier. Got the baseline support in (e.g. manipulating the cache) but not enough to actually run point muls. It's tweakable too. Hoping to have prelim benchmarks by this weekend. Then lots of review before 1.12 release. I've got another book project underway as well. So my time is split fairly brutally.
• May 22nd, 2006. Submitted the LTM book (with many many many many changes) to Syngress. Barring any new typos or errors it should go to the printers shortly.
• May 18th, 2006. Still no LTC release... LTM book (see link above) is going to the publishers May 22nd and then printers in early June. There are enough changes in terms of presentation and grammar, that picking up a paper copy of the text (en lieu of a PDF copy from the archive) is not a bad idea. Not only will it be easier to read, but you will also be getting me money. And as we all know, I started the LibTom projects to get blind stinking rich. :-). Seriously though, I'm fixing the grammar and moving figures,etc around, also, the index actually has entries in it now and I'm finishing a few of the "discuss later" sections. None of these fixes will be in the public domain archive. The only way to experience the joygasm that is the LTM text is to pre-order a copy.
• May 9th, 2006. Still working on the LTM (and another) book projects with Syngress. I've fixed all reported bugs for LTC but I'm not ready to release it (needs more testing). I do plan on adding an extra added value bonus gourmet goody. Fixed point ECC point mul accelerator. Something I was going to sell but I see the value in it. Imagine doing a 256-bit point mul with nothing more than 32 ADD and DBLs (instead of 256 DBL and 64 ADDs). The plugin will be a neato hack that you just "register and forget", it will learn what bases you re-use and either pre-compute and use the fixed point or use the baseline sliding window method. You'll see.
• Apr 25th, 2006. Syngress and I are working on publishing the LTM book. Score. More details to follow. (Yes, this means that I can't work on the LTM book publicly anymore...)
• Apr 15th, 2006. While doing some "writing" I found bugs in the ASN.1 code in LTC. So far I've found two minor bugs and have a few other things to add. I'll likely get an update release out in a week or so to address this. The ASN.1 errors shouldn't crop up in anything people are doing really, hopefully.
• Apr 6th, 2006. Uploaded the ZIP files for LTC, TFM and LTM which I seemingly missed when I did the mass scp command. Whoops. Bah, y'all should be using tar anyways. :-)
• Apr 6th, 2006. Talking with Brian Gladman it appears that my LRW implementation MAY NOT be what IEEE will settle on. Users are strongly encouraged to keep that in mind. The final implementation likely will require no API changes but the code will obviously not be binary compatible.
  
  
• Apr 4th, 2006. Release of LTC 1.11, TFM 0.09 and LTM 0.39. Enjoy. ... And it's my b-day on friday. w00t w00t
• Apr 2nd, 2006. More work on LTC. I found out that Intel CC v9 doesn't like LTC_FAST_TYPE (in pelican.c). Anyone would be welcomed to fix that. Since I work for the "green" I won't spend too much time on it (specially since the source is correct). I'm checking into supporting PGI. Actually any compiler vendor who wants to send me an x86 32/64 compiler with a non-commercial license would be more than welcomed to e-mail me. It's also my b-day on Friday (the 7th) so if all goes to plan I won't be in any shape whatsoever to do a release :-) and the week after I'm back in Sunnyvale. I'll see what I can do before the 7th. Peace!
• Mar 22nd, 2006. LTC v1.11 and TFM 0.09 are being put through final audits, reviews and test builds. I'm travelling next week so most likely releases will be the weak after. Some goodies. I've added GMP (GNU MP) support to LTC through my third math plugin. I've cleaned up the LRW code a bit, optimized the ECC plugins with TFM, corrected various typos in documentation and comments and fixed a bug in the DER flexi decoder. In TFM I've made partial mults (e.g. 640-bit x 640-bit) slighty faster. Still not 100% optimal but faster than before. The trick is to avoid odd sizes :-).
• Mar 16th, 2006. You may have noticed that libtomcrypt.org has gone a bit "awry". It seems someone decided to register the name and point it to a web.archive.org page. Well, no matter, Lance James of Phishing Exposed and SecureScience fame has bought me the .com and .net versions of the name. The site is now fully up again and will be moving to 100mbit hosting shortly. LT development continues and LTC 1.11 already sports a slew of bug fixes and a few optimizations. Enjoy and lets all thank Lance for putting this up admist his rather busy personal and business life.
  
  
• Feb 11th, 2006. It turns out the fix I think John was telling me was wrong for RSA it missed another code path. I fixed up the RSA import function by just making the variable usage simpler. I also missed some NULL values for the cipher descriptors since I added LRW values. NOTE I've just re-released v1.10 since they're trivial changes and frankly it's the weekend. If you're at work using LTC or something I'm sorry, but you also should be out and about with your mates (I'm in the UK now, notice the lingo). So just grab the latest and check the changelog if you're not sure. I added a comment to this affect in it. Cheers!
  
  
• Feb 10th, 2006. Release of v1.10 corrects a severe error in the RSA import which leaks small amounts of memory (one mp_int of value zero) and I sped up the LRW/GCM code greatly. Highly recommended you upgrade. Thanks to John Kuhns for the bug report about RSA.
• Jan 28th, 2006. Victory is mine! I managed to speed up gcm_gf_mult() which in turns speeds up LRW and GCM state creation. Took me 5 hours to track down a simple off by one bug in the damn multiplier. All works now. wee.
• Jan 27th, 2006. patch for LRW that fixes the fact that it doesn't use tables when enabled... e.g. slow. The buggy code produces the right results but is slow. This patch fixes it.
• Jan 26th, 2006. LTC 1.09 is out. Features LRW mode for you disk encryption folk and some cleanups to the build/headers.
• Jan 24th, 2006 Public Service Announcement. My XPC Shuttle case is *C*R*A*P*. I bought one and it's unstable in 64-bit linux and WinXP Pro. It mis-auto-detects memory and is otherwise the worse motherboard I've ever seen. Oh no, I'm not bitter. I'm just 484$ poorer. I'll try to refund it tommorow. Oh and LRW is coming.
• Jan 19th, 2006 As of a few days ago I work for the company that goes by the name AMD. I still can work on my projects here but obviously at a lower pace. Oh and "my LT* projects are my doing and not that of my employer and do not represent the intentions or whatever of AMD, in other words, what I say w.r.t. my LT* projects are my thoughts and send complaints my way." I'm hoping to get LRW in next week for a release on Saturday or Friday [depending on how the week goes]. Stay tuned.
• Jan 14th, 2005 err... 2006. Ok new years, new plans. :-) I'm adding LRW to the list of modes (like cbc/ctr) supported by LTC and of course it'll be pluggable. I also found a possibly annoying overflow in LTM/TFM which in the case of LTM shouldn't ever cause a problem. I'll be releasing a new tripplet release the week of January 23rd. (and yes, I've added the "easy" button already).
• Dec 26th, 2005. Been over four years now :-) [Dec 21st, 2001 was the first release]. Not dying :-). I'm looking into adding NLS as another Enc+Auth mode. I'm also adding an "Easy Button" to the LTC build so you can trivially turn off the unlikely to be used features (e.g. most enc+auth modes, ciphers, hashes) with something as simple as "CFLAGS=-DLTC_EASY make install" for platforms like my poor IBM PPC 405 which take FOREVER to build... :-) This is my way of compromising [e.g. wholesale removal of most algos]. No ETA on LTC 1.09 as I've had no bug reports from 1.08 [that's not always a good thing...] and I really haven't worked with the code in a good three weeks or so.
• Nov 25th, 2005. The cool folk of RoboDesign hooked me up with new LibTom logos. Feel free to use them on your own website to link back to ltc.org or just advertise who provides your crypto :-)
• Nov 24th, 2005. Release of LTC 1.08 sports fixes to the ASN.1 SEQUENCE decoder, addition of SET and SET OF and fixes to the PK build status.
• Nov 22nd, 2005. LTC 1.08 will be released this Thursday, it sports several key fixes to the ASN.1 code as well as SET and SET OF. I'm doing a release instead of patches because there are new files. STRONGLY SUGGESTED that you fetch it when available.
• Nov 20th, 2005. I'm getting around to SET and SET OF [SET encoder done, the decoders are done but not tested and I still have SET OF to code]. Now look at this it has to be the worst and ugliest script ever but the output sure is nice :-).
• Nov 19th, 2005. Working on SET and SET OF support for the ASN.1 routines. Fortunately they share quite a bit of code, the only new routines are for SET decoding and SETOF encoding.
• Nov 18th, 2005. Release of LTM 0.37, TFM 0.07 and LTC 1.07. Enjoy.
• Nov 9th, 2005. Improvements to TFM (yes, over the last public release) and LTC (both not yet released) and my semi-cool PPC kit have produced this neato benchmark list. So yeah TFM 0.07 and LTC 1.07 will sport proper support for the 32-bit PPC cores.
• Oct 31st, 2005. Inspiration from a commercial source has led me to re-factor the ECC again. Now the built in point add/dbl/mul code will NOT be directly called from LTC functions. This means if you have an ECC plugin you won't waste your precious code space with my ECC stuff ;-). I also plan on doing an optimized ECC plugin for some curves with TFM (hint: reduction algos :-)). Stay tuned. [and yes, the ASN.1 Flexidecoder is still on the table.]
• Oct 31st, 2005. LTC 1.07 and TFM 0.06 are coming along. Lots of fixes to LTC to make it more smooth and easy to work with (e.g. maintain). Also I've added my newly patented PointChek(tm) technology to ecc_import(). In this routine I now use my patented "does the bivariate equation hold true" technology to ensure imported keys are valid. Licensing will be available shortly.
• Oct 26th, 2005. Site has been down for a few days. Dan has moved the site to a new location (dunno, ask him yourself). Everything seems to be back up. My ECC work isn't going well. I have refactored the ECC code but my 4NAF and other tricks have NOT paid off. I'm still going to try fixed-point ECC but right now I'm gonna work on the ASN.1 stuff first.
• Oct 17th, 2005. (later in the day). New challenge. The current P-256 point multiplier with TFM clocks in at 2.75M cycles on my AMDX2 box. I want to beat 1M cycles before I release (using less than 32KB of tables). I think it's doable as fixed point alone can get me a 2.6x improvement without changing anything (using less than 16KB of memory). Stay tuned. Expect numbers within the next week or two.
• Oct 17th, 2005. Change of plans again, the updates I made for VC7 support weren't that huge so LTC 1.07 will include the new (faster) ECC and ASN.1 code. Release date is not fixed but I'm aiming at the middle of November.
• Oct 12th, 2005. New development (including the ECC code) is gonna stop for this release. I have a gig to provide updated MSVC makefiles and a demo (the demo will be private). So LTC 1.07 development will stop, I'll update the build files for MSVC and test them. LTC 1.08 will sport the new ASN.1 and ECC code I promised for LTC 1.07.
• Oct 10th, 2005. Work progresses on LTC 1.07. Among other things I'm re-organizing the ECC source code so it's more stable to work with. I'm also starting in on a new point multiplier (using wNAF and jacobian+affine point addition) should speed things up for low latency applications. The time invariant algo will be the same speed though.
• Oct 8th, 2005. I've uploaded a beauty of a function der_decode_sequence() (and the encoder is pretty too). Oh and progress has been made on LTC 1.07. It will be more friendly towards external accelerators and feature a few new ASN.1 tricks.
• Sept 26th, 2005. Got Toorcon code up here. It is neat.
• Sept 20th, 2005. Got back from Toorcon. Phew, that was fun. In between "events" I got some coding in. I've updated odds and ends as well as got a headstart on the "flexidecoder". I'm still lacking a SET type (which means updating all the SEQUENCE code too) but the flexidecoder will be handy.
• Sept 6th, 2005. If you have had trouble downloading LibTomMath that is because the paths were relative and you were going to "www.libtomcrypt.com/math/download.html" which won't work. I've fixed the page to use absolute paths now. You can go to here to fetch it.
• August 29th, 2005. Slides for Toorcon have been finalized. I've also done a bit of work on LTC [nothing new to release yet]. In particular I put error checking on the cipher accelerators. Also TFM 0.06 will have a totally cleaned up testing/timing harness then what is currently available.
• August 24th, 2005. Got my Nocona (dual-core P4 EMT64) box. It sucks :-) ops per second and cycles per operation (RSA decrypt, 2.8Ghz Nocona). Close ups raw and normalized to 1.8Ghz.
• August 10th, 2005. The stipend has been taken. I look forward to seeing the talk (I won't mention who it is) at Toorcon.
• August 7th, 2005. I've updated the TODO list and fixed the makefiles w.r.t. "testprof" ((makefile.diff makefile.shared.diff)
• August 5th, 2005. HEY AMD some free (cycles/op) press (ops/sec) for you. (RSA private operations, DC == dual core with TFM).
• August 5th, 2005. In case you haven't noticed LTC is pluggable. The point of that is to promote others to develop stuff so I don't have to [e.g. HW accel]. Now I've got my mind set on future projects. Currently a TLS implementation is the likely candidate (scope: server+client, limited ciphersuites (mostly just RSA_AES_CBC stuff), time frame for first release: ~March 2006). Anyone got better ideas for a future project?
• August 4th, 2005. Nobody has bothered to really submit conformance info so far... So for kicks I tried a little test against OpenSSL. It seems my OAEP code can decrypt what OpenSSL produces. Since it can also decode what I produce likely my encoder is compliant as well. No word on PSS yet... hehehe
• August 3rd, 2005. Posted a FAQ about how to build and use new releases. Please read it.
• August 1st, 2005. Release of LTC 1.06, TFM 0.05 and LTM 0.36 ... yipee!
• July 31st, 2005. More eyecandy. This is comparing doing RSA private key operations with TFM, LTM and finally independently with OpenSSL. cycle counts and Operations per second (on 2.2Ghz AMD64, 1.8Ghz AMD32 and 3.2Ghz P4 Prescott).
• July 31st, 2005. LTC, TFM and LTM are on the last phase of the development cycle (that being design, develop, test, document). In particular I'm happy about how well LTC is working with the math descriptors even though it has made the build process slightly more complicated (all documented). Releases will be on time tommorow.
• July 29th, 2005. I've pushed the ASN.1 work (see the TODO) to 1.07 since there are already tons of changes in place. I'm also releasing a new TFM and LTM shortly that have updated build scripts (re: more configurable!). So August 1st will see the release of LTC 1.06, TFM 0.05 and LTM 0.36 all at the same time. It's a treo-magnifique! ;-)
• July 26th, 2005. LTC 1.06 will be delayed unfortunately. We'll play this by year.
• July 24th, 2005. All three PK schemes work with the descriptors and I've added TFM to the list of descriptors. Some prelim performance data for RSA-1024 private key operations yielded me AMD64="LTC+TFM=1738/sec, OpenSSL=1387/sec", P4="LTC+TFM=392/sec, OpenSSL=136/sec". So it's just a "tad" faster. UPDATE I've uploaded a chart of TFM vs. LTM vs. SSL for RSA private key operations various RSA key sizes. Note that OpenSSL only "out of the box" can test 1024 and 2048 bit RSA keys... But you can clearly see who is on top. As suspected I have some tuning to do to make TFM a bit more scalable but it never fails to outperform LTM which makes me happy.
• July 23rd, 2005. I've ported ECC to the new framework. Now all three PK schemes can use the math plugins. Code still need cleaning up and I have lots of testing/documentation todo still. Just FYI, there have been 3023 lines removed, 2866 lines added since the release of 1.05.
• July 23rd, 2005. Ok enough teasing. Go fetch your own copy of TFM 0.04 already. Enjoy.
• July 22nd, 2005. I've uploaded a PNG dump of a graph comparing "fp_exptmod()" calls to TFM 0.04 on five different build/platform combinations. Note the AMD64 ownage. ;-)
• July 21st, 2005. TFM 0.04 is getting ready to ship. I have some prelim speeds for exptmod here. UPDATE I've uploaded PPC32 timings to the same text file. Wee.
• July 20th, 2005. With help from PeerSec (the MatrixSSL folk) I have debugged and tested my PPC32 code for TomsFastMath. It works fine. Hint: don't overwrite link registers... how was I supposed to know "r1" was reserved hehehehe
• July 17th, 2005. Good news, I've managed to get the RSA code working through LTM descriptor, now it's onto DSA then ECC then testing then documentation .... ugggh. Progress good ;-) ... UPDATE With a small burst of energy I've ported the DSA code as well. The ECC code will be a project for another day... :-)
• July 16th, 2005. I've finished the first pass of the LTM descriptor for LTC. I haven't tried compiling it though but it's fairly simple. First thing tommorow I'm going to start making some globals for the math descriptors then start with the ASN code. After that it's RSA, DSA then ECC. Tons of documentation ahead but at least one battle is done. I may ship LTC 1.06 with only a descriptor for LTM depending on time. A TFM one shouldn't be that hard after all is said and done. UPDATE. I have ported the majority of the RSA code to the new math api. Actually I'm cheating and making LTM macros that map to the new math descriptors. The math descriptors are a bit messy so I'll spend time before release cleaning it up. Essentially a "math descriptor" has pointers to about 50 functions which is a bit overwelming at first but they're all very simple.
• July 16th, 2005. I've got a headstart on working on LTC (pending access to a PPC box to test TFM 0.04). The todo file has been updated with what I've done so far.
• July 13th, 2005. I am wrapping up TFM 0.04 this week. Thanks to some code from the NSS folk at Mozilla (which I then further optimized) the code is very fast. If you want an idea check out this page and halve the cycle count for the SSE2 core. Once I get the PPC testing finished I'll update the manual and make a release.
• July 13th, 2005. Updated. Tom will be at ToorCon this year. I will be presenting on the subject of "Pluggable LibTomCrypt". So if you want to learn how to exploit LTC for your hardware or asm implementations of ciphers, hashes and/or bignum math, please see my talk. This is contingent on me getting the TFM and LTC releases out the door. So you can tell what I'll be doing shortly.
• July 12th, 2005. Shortened the TODO list for LTC and also started putting more thought into the pluggable math. I am planning to submit a ToorCon talk on "Pluggable LibTomCrypt" so I really need TFM 0.04 and LTC 1.06 out the door quickly. Also read this to learn my thoughts on pluggable math and what impacts it has on LTC users [as early as LTC 1.06!!!]
• July 11th, 2005. Back from LSM in France. Phew, what a trip! I've added to the LTC TO-DO list (see above) and also have TomsFastMath 0.04 in the works (faster, better, stronger). Also Dave "of Toorcon fame" has agreed to a limited printing of the LibTomMath book so I'm spending the next couple weeks going through that. The goal is to have some printed copies for Toor this year. Of course keep submitting bug reports/patches/sugestions/etc.
• June 27th, 2005. I pushed a few things onto 1.06 and have released 1.05 today. It fixes a nasty bug in the OAEP decoder and adds two new ASN.1 types (UTCTIME and CHOICE). As for the ECC there are no external functions yet, right now I'm not too worried because if the point multiplier didn't work EC-DSA would likely fail through the 1000s of random signatures I've performed. I'm still adding ecc tv_gen support just now it's slated for 1.06. I've updated the TODO list as well.
• June 26th, 2005. Sorry I'm behind again, no LTC 1.05 yet. Give me a few days. Meanwhile check out story.
• June 23rd, 2005. Not really any progress. Nelson from the NSS [mozilla] crew submitted bug fixes and an update to TomsFastMath ... I'll be releasing TFM 0.04 shortly after I release LTC 1.05 on the weekend. Recall the eventual goal is to make LTC pluggable for math. I've done a quick port of the ECC to TFM [this isn't in the CVS] and it clocks in at over 2x faster. On my AMD64 at 2.2Ghz with ECC-224 GF(p) I hit 1200 point multiplies per second.
• June 20th, 2005. I've tested the CHOICE code, added more ASN.1 documentation (and fixed numerous other documentation errors). I've updated the TODO list. I'm on track for the June 26th release of 1.05.
• June 19th, 2005. I've fixed the OAEP error and added CHOICE/UTCTIME to the list of support ASN.1 types. I've also re-arranged my TODO list so you can see where this is going.
• June 17th, 2005. Heads up, I found a buffer overflow in the OAEP decoder. Will be fixed for LTC 1.05. :-)
• June 15th, 2005. Formal (in order of precedence) TODO list for LTC 1.05. I aim to have the code updates finished by June 26th and welcome volunteers to run the test scripts on their boxes at that time.
• June 15th, 2005. Release of LTC 1.04 which fixes minor issues in the DSA and ECC public key code [read the changelog].
• June 8th, 2005. A day early, fully tested with GCC 3.4.4 and ICC 8.0 [on an AMD64 and Prescott P4] ... LibTomCrypt 1.03 has been released. As to why this took so long, well just check out this sort of function [img] or [img] or [img]. You can make the full 1370 page reference manual (bz2 pdf)with "make doxy" and see the size of the library for yourself [while admiring how small the compiled code is... ;-)].
• June 7th, 2005. After 2600 new lines, 1200 removed lines and a score of new ASN.1 functionality LTC 1.03 is essentially ready. Currently I have about 6 items of documentation to address and a few minor coding issues. Expect v1.03 for this June 9th.
• May 20th, 2005. This website was down for a few days. Fortunately Dan was all over it and got it back up as soon as he could. I'm back from the west coast and plan to get cracking on LTC shortly. There is still development to be done then I have to add more testing routines, document new API and get beta testers...
• May 12th, 2005. Work goes SLOOOOWLY on 1.03. I'll be missing the May 14th release date. Whaddaya gonna do. C'est la vie. There is no planned release date. This goes for all LibTom projects.
• May 7th, 2005. You might notice no updates today. I fell behind. Lots of good changes since 1.02 though (like totally new ASN.1 API). I still have coding todo let alone testing/documentation. Realistically we're talking May 14th at the earliest. Peace out.
• May 2nd, 2005. The signature scheme I used for ECC turns out to be really insecure. Of course since people audit my code they noticed this right? Nope. LTC now uses ECDSA for ECC signatures. At least it's a standard.
• April 26th, 2005. I have all but got the finanicing in my hands. I'm going to LSM'05. Oh yeah, if you use GCC 3.4 try using "-fno-regmove" when you build. It speeds up the ciphers/hashes on my P4 box (AES-128 by 120 cycles/block). YMMV it doesn't work so well on the AMD64.
• April 21st, 2005. I may be financing the trip to LSM'05 myself. Don't count me out just yet. :-)
• April 19th, 2005. Minor update of LibTomCrypt. Should build on non-x86 GCC platforms. If you run a non x86 (32 or 64-bit) box please get back to me whether v1.02 builds properly (usually a "make test ; ./test" is sufficient). Thanks
• April 17th, 2005. Here are the (unfinished) slides for the talk I was proposing to do at LibreSoftwareMeeting (Dijon, France) this year. I pulled out since I couldn't finance the trip. The talk is about the "LibTom Projects".
• April 17th, 2005. Et voila. v1.01 has been released. Enjoy.
• April 10th, 2005. I'm sitting on v1.01 at the moment. It has loads of goodies and fixes in it.
• March 20th, 2005. LibTom projects are now stored in a local CVS (at iahu.ca). If you'd like to get developer access please email me a request. Additionally if any other OSS developers want access to an AMD64 box (not for LibTom projects) just ask. I'm willing to give shell accounts to responsible individuals.
• March 12th, 2005. Release of LTM 0.35. Fixes numerous bugs. Get it.
• March 1st, 2005. I've added a "PayPal" donate button. If you like LibTom stuff and want to donate now you can via PayPal. No "suggested" donation and everything is appreciated. Thanks!
• February 12th, 2005. Incremental release of LTM v0.34. Fixed a few bugs and added new reduction code.
• January 18th, 2005. I just noticed, the LT projects (all 7 of them) hit over 100,000 lines of C and TeX code recently. Cool.
• January 11th, 2005. I've uploaded some patches for 1.00 that fix two buffer overflows in LTC. I'd suggest ya head to the download page to get them. Peace.
• December 31st, 2004. Three years and ten days since I started LibTomCrypt. I present, v1.00 for your use and am signing off. The project can always be improved and I encourage other developers who are trying to develop their talents to take up the project and make it even better than it already is. Salut my many friends and users.
  
  
  
  
• December 23rd, 2004. I uploaded RC1. Enjoy. You'd better. **shakes fist**
• December 22nd, 2004. I got the word from my college. I'm now officially a graduate of ``computer science''. W00h00
• December 20th, 2004. RC1 code freeze. Fixed several cygwin/msvc related build problems (I don't develop in windows...). RC1 is on track to be released on the 23rd. Also can someone explain to me how "Best Ph4rM1es known to man...V1c0d1n V14gr4 and much more!" could ever entice ANYONE to buy from that person? I'll live the rest of my days puzzled.
• December 19th, 2004. I put up LibTomZip since it does no good just sitting on my hard disk. Enjoy.
• November 23rd, 2004. Got a replacement cell phone, same # as listed on my C.V.
• November 21st, 2004. Posted a brief notes on the future release schedules of LTC, LTM and TFM. If you are a LTC user please take note of the rc1/rc2 schedule. I need your help to ensure that the release schedule goes smoothly!
• November 15th, 2004. Updated my list of notable downloaders (of 0.98 and 0.99).
• November 10th, 2004. Got back from France safe and sound [and 1800$ poorer... sadly that's another story (related open letter)...]. Also keen notage is Entrust, RSA Security and Sun as recent LTC 0.99 downloaders ;-)
• November 3rd, 2004. Bush likely to win, I'm fleeing the continent for France. Be back November 10th. ;-)
• November 2nd, 2004. Added new entry to my C.V. ;-) Yet another LibTom consulting gig. Thanks to Blunk Microsystems for using LibTom projects!
• November 1st, 2004. I've given up on using iahu.ca for mail. My brother is "fixing" it too often. Please send all future email to tomstdenis@gmail.com.
• October 31st, 2004. Released patchset 001 for v0.99 which fixes a small bug in the "small variant" of mp_div(). Recommended. You can get it from the download page.
• October 29th, 2004. I goofed in the LTC release. Please if you grabbed a copy before 11PM [-5GMT] fetch another. See this for more info.
• October 29th, 2004. Release of LTC 0.99 and LTM 0.32. Many cool updates, fixes and build configurations now. I suggest y'all grab a copy.
• October 17th, 2004. I've got a new cell number. Please don't call the old one ;-) the new number is on my C.V. page.
• October 8th, 2004. A partial list of "notable downloaders" of LibTomCrypt 0.98 is available here.
• October 3rd, 2004. Dan recovered quickly (was at Toorcon on Saturday but didn't present). My slides are available in OpenOffice and PDF formats ;-).
• September 23rd, 2004. Dan "Effugas" [the dude who hosts this place] is missing Toorcon due to illness. Let's all take a minute to wish Dan a speedy recovery.
• September 18th, 2004. Release of TFM v0.02. Fixes a bunch of bugs and builds cleaner. What more could you ask for? ;-)
• September 14th, 2004. Made a TomsFastMath section to the website, released the second patchset for LTC 0.98 (recommended!).
• August 25th, 2004. Released TomsFastMath as public domain. Enjoy.
• August 22nd, 2004. Announced the release date of TomsFastMath. Even posted numbers (speed) up there. Check it out.
• August 13th, 2004. PLEASE read the following text if you're a LTC or LTM user.
• August 13th, 2004. Released minor patchset for LTC 0.98. Go grab your copy today!
• August 6th, 2004. LTC 0.98.
• July 30th, 2004. My rant. Read it or else. [link]
• July 23rd, 2004. Released v0.97b which closes some security holes (in pkcs #5 and hmac) and adds some new functionality.
• July 14th, 2004. Added another user to my users list. They're also donating a copy of Practical Cryptography to me as a gift. Free stuff rocks. Oh and LibTomZip v0.01 is just right around the corner.
• June 28th, 2004. Updated the SSE2 patches. Much faster (faster than OpenSSL on my P4 for exptmod). Includes pre-made mpi.c for LTC.
• June 27th, 2004. Some good news for a change... ;-) SSE2 patches against LTM. See LTM download page for more info.
• June 23rd, 2004. Today's mishap in style. ;-)
• June 23rd, 2004. Released v0.97a which is just a pre-patched copy of v0.97 for those not wanting to apply the patches themselves [avoids the risk of messing it up]. Strongly recommended for LTC users.
• June 23rd, 2004. Added another patch to the set. Fixes a bug in mycrypt_macros.h for BIG_ENDIAN systems.
• June 23rd, 2004. NOTICE! I've flaged v0.97 due to an incorrect PKCS #1 v2.0 encoder/decoder. Please apply the patches and be careful! (hint: report any odd error codes you notice).
• June 21st, 2004. Released patches against LTC 0.97 which clean up a few errors in the header files.
• June 19th, 2004. Release of LTC 0.97. Fixes a bunch of bugs, improves the stack usage of the code and added a few changes to the manual.
• June 18th, 2004. Updated my C.V., users and features pages. I am preparing LTC 0.97 for release shortly.
• June 13th, 2004. I've patched a bug in pkcs_1_v15_es_decode(). Head over to the download page to get the patch.
• June 13th, 2004. I've updated the "users" list of LT projects a bit. Also please check out my CV. ;-)
• June 6th, 2004. Released patch that fixes buffer overflow in tim_exptmod() of LTC 0.96. Please apply it.
• May 30th, 2004. Release of v0.96. More PKCS #1 support, updated RSA code, optimized aes, sha1, sha256, md5 for size.
• May 22nd, 2004. Released new patches for PKCS #1 code (apply to fresh 0.95 source) that fixes bugs and cleans up the API
• May 15th, 2004. Released patch for the PKCS OAEP decoder so it can distinguish between failures and invalid packets. Non-critical but a good patch to have none the less.
• May 12th, 2004. Release of v0.95. PKCS support emerging and fixed up various other things. Refactored code base too ;-)
• May 5th, 2004. The LibTom foundation of all that is good is pleased to announce the initial release of the LibTomFloat multiple precision floating point library. Hop on over to float.libtomcrypt.com to grab a copy today!
• Feb 21st, 2004. Patch for AES and WHIRLPOOL released. It reduces the code size of the two when SMALL_CODE is defined.
• Feb 20th, 2004. LibTomCrypt v0.94 released. Added PMAC and WHIRLPOOL, fixed OCB and PK routines.
• Jan 30th, 2004. Added my new paper to the collective.
• Jan 25th, 2004. Added patch for OCB code from LTC 0.93 [see download page].
• Jan 25th, 2004. Simpler site. Release of LTC v0.93, LTM v0.29 and LTP v0.03
• Jan 11th, 2004. I forgot to announce LibTomPoly when I released it. LTP is a polynomial basis library using LTM.
• Dec 24th, 2003. LibTomCrypt v0.92 released. Merry er... Happy Holidays!
• Dec 22nd, 2003. LibTom 2003 Year in review available at newstands everywhere!
• Sept 25th, 2003. Release of v0.91. Fixes numerous bugs and updates.
• Sept 10th, 2003. If you were on the mailing list prior to today please re-register.
• Sept 7th, 2003. Release of v0.90. Added RIPEMD-128/160 hashes, new DH primes [faster], few bug fixes and officially released as public domain.
• August 25th, 2003. Started a new mailing list (again)
• July 16th, 2003. Release of v0.89. Several bug fixes, merged in LibTomMath v0.24.
• July 10th, 2003. Release of v0.88. Mostly optimizations and new LibTomMath [bug fixes from v0.22].
• July 4th, 2003. Release of a new project LibTomNet.
• June 25th, 2003. LibTomMath Book has been released [draft edition]. Check it out.
• June 19th, 2003. Release of v0.87. Mostly optimizations.
• June 15th, 2003. Release of v0.86. Bug fixes and updates.
• June 11th, 2003. Release of v0.85. Fully TDCAL now.
• June 8th, 2003. First draft of the ``official'' TDCAL license now online.
• June 1st, 2003. Release of LibTomCrypt v0.84 [slight bug fixes]
• April 6th, 2003. Added users page to the site here.
• March 29th, 2003. Release of v0.83 of LibTomCrypt. Fixes several severe exploitable bugs and sports an optimized ECC mulmod routine as promised.
• March 22nd, 2003. v0.15 of LibTomMath released. Check it out.
• March 18th, 2003. Starting to use Bit Keeper for code management. Though I'm the only developer Bit Keeper will let people see how the code evolves over time. Wayne Scott has spent considerably time getting previous releases setup in the repository for free. Check out the repository at bkbits.net.
• March 15th, 2003. Going public once again. New site and releases. LibTomCrypt 0.82 and LibTomMath 0.14

Hosted by: