Commit Graph

1039 Commits

Author SHA1 Message Date
zzz
cb72bb0427 * Shutdown:
- Implement and call shutdown for BandwidthRefiller
    - Implement and register shutdown hook for Timestamper
    - Implement and register shutdown hook for Jetty console server
    - Fix UPnP-SSDPNotifySocket thread not stopping
    - Fix all but one UDP PacketHandler threads not stopping
2011-06-16 19:04:23 +00:00
zzz
518fdd8c03 AsyncFortunaRandomSource: Refactor refiller for concurrent and shutdown
Add shutdown hook for FortunaRandomSource
2011-06-16 12:42:25 +00:00
zzz
c655d23815 * Updates:
- Add the router version to the zip file comment in the updater
    - Add a class to extract the zip file comment
    - Require the sud version header to match the zip file comment
      to prevent spoofing of the version number, since the version
      number in the header is not covered by the sud signature.
2011-06-15 13:30:24 +00:00
zzz
3c88c854bf more cleanup at shutdown 2011-06-14 19:27:46 +00:00
zzz
21b76f1ca1 concurrentify 2011-06-14 19:15:04 +00:00
zzz
d50cf13085 Register shutdown hooks for caches, in the first step of cleaning up
resources on shutdown, which is important in Android as the JVM
isn't going away. More to do.
2011-06-14 19:08:15 +00:00
zzz
bb8b428598 * jbigi / NBI / wrapper / installer:
jbigi and wrapper files for arm.
    Compiled on trimslice with gcc version 4.4.5 (Ubuntu/Linaro 4.4.4-14ubuntu5) 
    Log postinstall errors to postinstall.log.
    java version "1.6.0_18"
    OpenJDK Runtime Environment (IcedTea6 1.8.7) (6b18-1.8.7-0ubuntu2.1)
    OpenJDK Zero VM (build 14.0-b16, mixed mode)
    GMP 4.3.2 (half the size of 5.0.2, and no speed difference) LGPLv3
    Wrapper 3.5.9 GPLv2
    All binaries stripped.

    As on the Android emulator, the stock BigInteger.modPow()
    has some serious bug on arm.
    Without the libjbigi checked in here, the JVM crashes almost immediately.
2011-06-14 18:11:43 +00:00
zzz
730965812b final 2011-06-14 18:08:40 +00:00
kytv
ffffd2deb5 Add jcpuid binaries for Intel macs 2011-06-12 16:33:34 +00:00
zzz
550bb1aaae fix the other gzip direction on android too 2011-06-11 17:05:24 +00:00
zzz
4fedb3fb5a javadoc thx kytv 2011-06-10 23:30:49 +00:00
zzz
dba92b0c42 actually use the 64 bit cpu detection 2011-06-10 23:24:54 +00:00
zzz
f7751de1c3 propagate from branch 'i2p.cpuid' (head be49d2042be9311c16ec5cc4b5accd1366b93556)
to branch 'i2p.i2p' (head 0d92feaa69315781a8e98190c6d11b6b91355f8b)
2011-06-10 21:59:42 +00:00
zzz
0b4e430c70 * NativeBigI: Add support for atom, core2, corei, nano, pentiumm, geode
* JBigI:
      - Add new libjbigi-linux-xxx.so files built by sponge, now that we have
        CPUID.java support for them (thanks hottuna).
        See http://zzz.i2p/topics/306 for discussion and test results.

        64-bit processors (atom, core2, corei, nano):
        Built with GMP 5.0.2. License is LGPLv3.
        Built by sponge with GCC 4.4.4, downloaded from
        http://sponge.i2p/files/jbigi/gmp-5.0.2/
        For 64-bit processors, both performance testing and
        the GMP changelog http://gmplib.org/gmp5.0.html led us to use 5.0.2
        for both the 32- and 64-bit versions, even though the files are twice as big.
        5.0.x contains specific optimizations for atom and nano.
        All 64-bit libs have _64 appended.

        32-bit processors (pentiumm, geode):
        Built with GMP 4.3.2. License is LGPLv3.
        Built by sponge with GCC 4.4.4, downloaded from
        http://sponge.i2p/files/jbigi/gmp-4.3.2/
        The 4.3.2 files are half the size of those built with GMP 5.0.2, and there was
        little or no performance difference between the two versions for 32 bit processors.
2011-06-09 16:04:06 +00:00
dev
6cf1591688 Large but simple refactor of CPUInfo interfaces and CPUInfoImpl classes.
Added interface and class for VIA cpus. 
Added support for VIA c5 cpus and some Via nano cpus (a few might be missing).
2011-06-09 15:00:30 +00:00
zzz
3d18255eb1 * LogManager: Use DataHelper methods for loading and storing config 2011-06-08 20:16:48 +00:00
zzz
7e84451ac3 Android:
- Import the main_rules file and remove the copied rules
  - Set the version code and name in the apk badging
  - Build target is 8 to allow SD install, MIN-SDK is still 3
  - Fix NPE on new install
  - Remove some debug code from NBigI
2011-06-05 22:38:35 +00:00
zzz
48f4c7286d - Remove libjbigi-windows-athlon64.dll which was a duplicate of libjbigi-windows-athlon.dll;
NBI now tries athlon as a fallback for all 64-bit processors.
2011-06-05 13:31:26 +00:00
zzz
d1e42233a8 * DSAEngine: Add code for alternate implementation using Java libs;
disabled by default. Add test code to verify identical results
                 and compare speed.
2011-06-05 12:40:59 +00:00
zzz
d308d7da97 remove unnecessary zeroing code, javadoc 2011-06-05 11:19:32 +00:00
zzz
4d34078678 * Data: Remove duplicate signature verification code
in RouterInfo and LeaseSet
2011-06-05 11:18:35 +00:00
zzz
4e16ef35a2 * KeyGenerator: Restore old return type to not break ABI (thx kytv) 2011-06-04 12:04:18 +00:00
zzz
885d549e84 * NBigI: Recognize Android 2011-06-04 12:03:35 +00:00
zzz
219d7fd8c3 fix ElGamalTest compile; fix jbigi.jar path to speed up the test 2011-06-02 20:29:15 +00:00
zzz
fe15db51d8 - Logging tweaks inspired by Android
- Disable reusing Deflater on Android
2011-06-02 18:29:06 +00:00
zzz
761ad38bcc * HMAC:
- Javadocs and cleanups
      - Use SimpleByteCache
      - Comments and speculation
2011-06-02 13:37:35 +00:00
zzz
2a5ed938bb update generateKeyPair() return type to make it easier 2011-06-02 13:34:26 +00:00
zzz
c5c4e3c7ce * ElGamalAESEngine: Fixups required after SessionKey enforcement
* SessionKey: Enforce data size and prevent reuse like the other SimpleDataStructures
2011-06-02 13:33:03 +00:00
zzz
94af6550fa HMAC256 is used by Syndie 2011-06-02 13:30:33 +00:00
zzz
43332bb6d0 * Crypto:
- Use java.security.MessageDigest instead of bundled GNU SHA-256 code
        if available, which it should always be.
        5 to 20% faster on Oracle JVM; 40 to 60% on Harmony;
        5 to 15% on JamVM; 20x (!) on GIJ.
      - Use java.security.MessageDigest instead of bundled Bitzi SHA-1 code
        if available on non-Oracle JVMs, which it should always be.
        Not faster on Oracle JVM; 30 to 60% faster on Harmony;
        15 to 20% on JamVM; 10-15x (!) on GIJ.
2011-06-01 11:44:10 +00:00
zzz
3677cadcca * jbigi: Add support for solaris 2011-05-31 20:21:51 +00:00
zzz
d4bf2523a6 * I2CP:
- Append I2CP Version information to the Get/Set Date Messages,
        so that both the router and client are aware of the other side's version,
        and future protocol changes will be easier to implement.
        Previously, router version information was not available to the client,
        so when router and client were in different JVMs,
        old clients would work with new routers
        but new clients would not work with old routers.
        After this change, we can design future changes so that new clients
        will work with old routers.
        This is an enhancement to the old protocol version byte sent by the client,
        which we have never changed and probably never will.
      - Prevent a client from setting the router's clock
      - Javadocs
2011-05-30 16:31:09 +00:00
zzz
393d813d05 Don't re-append the query string to the redirect URL,
which caused eepget to repeatedly retry the new redirect
of an i2paddresshelper query by the HTTP proxy.

This was clearly non-compliant (the Location: header requires
an absolute URI including any query string).

This was part of the original redirect handling added by
jrandom to eepget back when it was in syndie;
appending that changelog just in case this hackery was
required by Freenet USKs, which I doubt.

Also, reset the redirect counter when retrying.

==============

	Revision: 62a8cac8565958d111d0a47a443c964918d2e3ef
	Parent:   277fe109ea4416f7d8a6dba6e09cd4f605a8ce1a
	Author:   jrandom@i2p.net
	Date:     10/22/2006 15:06:30
	Branch:   i2p.syndie

	Changelog: 
	    ...
	    - add support for HTTP 301 redirects to EepGet (so freenet USKs work as
	      intended)
2011-05-27 21:41:49 +00:00
zzz
2c3be29016 * JBigI:
- Refactor to handle 64 bit libs and a list of fallback libs
      - Logging tweaks
      - Only check for one file name in library path
2011-05-27 15:04:40 +00:00
zzz
150cb30339 minor NBI tweaks and comments 2011-05-25 22:13:18 +00:00
zzz
724d8de9ae call the 64 bit cpuid, even if we dont use it yet 2011-05-25 21:35:21 +00:00
zzz
ddcd8cbb10 prep for new 32 bit cpus 2011-05-25 21:23:40 +00:00
zzz
f5f4f14b7e Drop libjbigi-linux-k63.so and libjbigi-freebsd-k63.so that were identical
to k62 versions; add mapping in NativeBigInteger.
2011-05-25 21:14:16 +00:00
zzz
6616ccd3b4 cleanup in prep for more changes 2011-05-25 21:05:20 +00:00
zzz
27b48034c5 b32 cleanups 2011-05-25 13:34:32 +00:00
kytv
4fd8da8041 Fix errors that tests/scripts/checkxml.sh found 2011-05-24 22:33:28 +00:00
zzz
7ab95d0144 javadoc 2011-05-23 19:16:33 +00:00
zzz
96cf598691 javadoc 2011-05-22 18:34:41 +00:00
zzz
e42b78e177 propagate from branch 'i2p.i2p.zzz.naming' (head 9626e0df682c8d5f706d2c814158ba451f3ebeb5)
to branch 'i2p.i2p' (head a3969e6c9c4fd5bfd69cd716ce0df191ad2af634)
2011-05-20 14:12:07 +00:00
zzz
4b19801cdb try to prevent out-of-order logging at shutdown 2011-05-20 14:09:36 +00:00
zzz
e2c98ac134 propagate from branch 'i2p.i2p.zzz.jrobin159' (head 08df645290d8e156d88930ddaf71b3f3d8af419d)
to branch 'i2p.i2p' (head 69c9e6bd0c6894c335f7d9892aa28ac7d6389907)
2011-05-20 13:52:35 +00:00
zzz
ff520b74dc propagate from branch 'i2p.i2p.libgmp' (head 41a57954f7e78e57f8eb5e30e8ebde7fd75b15dc)
to branch 'i2p.i2p' (head d174e772ebd4f3e3eac5c9b5abd7f736c2527eb2)
2011-05-20 13:13:26 +00:00
zzz
20452c9387 * Certificate: Fix creation bug; improve router cert generation 2011-05-20 12:46:50 +00:00
zzz
7377086aae 0.8.6 2011-05-16 19:58:19 +00:00
zzz
5ed22d5c16 cleanups 2011-04-28 18:04:31 +00:00