opencl-devel Mailing List for Botan
Brought to you by:
randombit
Menu
▾
▴
opencl-devel — General discussion list for OpenCL
You can subscribe to this list here.
| 2002 |
Jan
|
Feb
|
Mar
|
Apr
(10) |
May
(4) |
Jun
(15) |
Jul
(7) |
Aug
(14) |
Sep
(3) |
Oct
(1) |
Nov
(2) |
Dec
(1) |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 2003 |
Jan
|
Feb
(5) |
Mar
(2) |
Apr
(4) |
May
(13) |
Jun
(12) |
Jul
(8) |
Aug
(1) |
Sep
|
Oct
(2) |
Nov
|
Dec
|
| 2004 |
Jan
|
Feb
(3) |
Mar
|
Apr
(20) |
May
(9) |
Jun
(8) |
Jul
(2) |
Aug
|
Sep
(4) |
Oct
(1) |
Nov
(1) |
Dec
|
|
From: Jack L. <ll...@ra...> - 2004-11-03 17:32:45
|
Just a quick FYI - I've realized that due to some changes I made in 1.4.2's build system, if you `make install` you get a useless build, because (among other things) config.h doesn't get installed, which breaks everything. Workaround is rm -f <whatever_prefix>/botan/*.h followed by cp build/include/botan/* <whatever_prefix>/botan/ I'll try to get 1.4.3 out quickly to resolve this. -Jack |
|
From: Patrick J S. <swe...@ho...> - 2004-10-19 21:00:09
|
Hi this is my first post so hopefully it works... =20
I'm developing a simple application which encrypts, then decrypts a =
buffer using DES/CBC and TripleDES/CBC.
Using the encryptor Pipe, it appears to be encrypting correctly, though =
at the moment I have no way of knowing because I get the following error =
when the decryptor Pipe is used:
Exception caught: Botan: Decoding error: PKCS7
I get the same error even if I leave the PKCS7 off of the =
"TripleDES/CBC" statement. Not sure what is going on here, any ideas? =
I'm operating FreeBSD 4.10 with Botan version 1.2.8.
Thanks for any help.
<-- Begin code -->
#include <iostream>
#include <iomanip>
#include <fstream>
#include <string>
#include <memory>
#include <botan/filters.h>
#include <botan/look_pk.h>
#include <botan/lookup.h>
#include <botan/dsa.h>
using namespace Botan;
int main() {
=20
byte buf[] =3D =
"abcdefghijklmnopqrstuvwxyz123456abcdefghijklmnopqrstuvwxyz123456";
try {
=20
LibraryInitializer init;
=20
/* this is the encryption/decryption key */
const char * ph =3D =
"aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa";
byte my_iv[] =3D "aaaaaaaaaaaaaaaa";
std::string passphrase(ph);
SymmetricKey key =3D OctetString(passphrase);
InitializationVector iv(my_iv, 8);
Pipe encryptor(get_cipher("TripleDES/CBC/PKCS7", iv, key, =
ENCRYPTION));
Pipe decryptor(get_cipher("TripleDES/CBC/PKCS7", iv, key, =
DECRYPTION));
std::cout << "Clear: " << buf << "\n";
=20
encryptor.start_msg();
encryptor.write(buf,64);
encryptor.end_msg();
encryptor.read(buf,64);
std::cout << "Encrypted: " << buf << "\n";
=20
decryptor.start_msg();
decryptor.write(buf,64);
decryptor.end_msg();
decryptor.read(buf,64);
std::cout << "Decrypted: " << buf << "\n";
=20
}
catch(std::exception& e) {
std::cout << "Exception caught: " << e.what() << std::endl;
}
return 1;
}
<-- end code --> |
|
From: JnMlMe <jn...@fr...> - 2004-09-08 15:41:52
|
HI,
I m working now on solaris 2.6 / gcc 2.95.3
I have an error during the compilation:
=20
=20
[xxx@solaris2.6]> make
g++ -m32 -mno-app-regs -Iinclude/ -O2 -finline-functions -mcpu=3Dv8
-Wa,-xarch=3Dv8plus -ansi -Wno-long-long -W -Wall -fPIC -c src/conf.cpp =
-o
build/lib/conf.o
/usr/ccs/bin/as: "/var/tmp/ccsqSuwc.s", line 6280: error: can't compute
value of an expression involving an external symbol
make: *** [build/lib/conf.o] Error 1
=20
Anybody have an idea ?
Thks
=20
=20
=20
-----Message d'origine-----
De : ope...@li...
[mailto:ope...@li...] De la part de Jack =
Lloyd
Envoy=E9 : mardi 7 septembre 2004 03:28
=C0 : ope...@li...
Objet : Re: [Botan-devel] TripleDES /ECB Mode
=20
On Mon, Sep 06, 2004 at 03:46:52PM +0200, JnMlMe wrote:
> Hi,
>=20
> I need to decrypt a string who was encrypted with Triple DES (ECB =
Mode)
> under SUN Solaris.
>=20
> First, I would like to compile Botan under Windows 2K environment.
> I have a problem when i process the nmake command :
>=20
> =20
>=20
> D:\botan>perl configure.pl msvc-windows-i586 --module-set=3Dwin32
[...]
>=20
> CL /nologo -Iinclude /O2 /G5 /GX /GR /D_CONSOLE /c
src\adler32.cpp
> /Fob
> uild\lib\adler32.obj
> 'CL' n'est pas reconnu en tant que commande interne
> ou externe, un programme ex=E9cutable ou un fichier de commandes.
> NMAKE : fatal error U1077: 'CL' : return code '0x1'
> Stop.
=20
It looks like cl, the command line version of the compiler, isn't in =
your
PATH
variable. Have you tried using the command shell that comes with Visual
Studio?
Running that will automatically set up your environment with the right
variables for using the compiler and linker from the command line. =
Otherwise
you'll have to find where the CL binary is installed, and add that =
directory
to
PATH.
=20
> Second point, i have looked the encrypt and decrypt cpp examples =
files.
>=20
> These files used the CBC mode.
> Should i only changed the mode with ECB in line 117 ?
>=20
=20
No -- the examples are doing password hashing, message authentication, =
and
similiar things that you probably don't want to do. Well, you do want to =
do
them normally, for decent security, but in this case you won't be able =
to
decrypt the message trying to use them, so there is not much point.
=20
The following should start you off, but I don't know how the key is =
derived,
so
as-is it won't work (because currently the key is zero-length, which =
3DES
doesn't take). If you know the hex value of the key, just assign it to =
the
key variable (SymmetricKey key =3D "01234[...]").
=20
----- FILE ------
=20
#include <botan/botan.h>
using namespace Botan;
=20
int main(int argc, char* argv[])
{
try {
if(argc !=3D 2)
{
std::cout << "Usage: " << argv[0] << " input_file" << =
std::endl;
return 1;
}
=20
LibraryInitializer init;
=20
SymmetricKey key; // fill this in with whatever key you have
=20
Pipe pipe(get_cipher("TripleDES/ECB/NoPadding", key, DECRYPTION));
=20
DataSource_Stream input(argv[1]);
pipe.process_msg(input);
std::cout << pipe.read_all_as_string();
}
catch(std::exception& e)
{
std::cout << "Error: " << e.what() << std::endl;
return 1;
}
return 0;
}
=20
----- /FILE ------
=20
=20
-------------------------------------------------------
This SF.Net email is sponsored by BEA Weblogic Workshop
FREE Java Enterprise J2EE developer tools!
Get your free copy of BEA WebLogic Workshop 8.1 today.
http://ads.osdn.com/?ad_id=3D5047&alloc_id=3D10808&op=3Dclick
--=20
opencl-devel mailing list
https://lists.sourceforge.net/lists/listinfo/opencl-devel
____________
Virus checked by G DATA AntiVirusKit
Version: BD 14.0.534 from 04.09.2004
Virus news: www.antiviruslab.com
|
|
From: JnMlMe <jn...@fr...> - 2004-09-07 09:03:29
|
Hi
First, thanks for your very useful help.
I go forward in the process, but now, i have a problem during the
compilation project:
------ Rebuild All started: Project: DES, Configuration: Debug Win32 =
------
=20
Deleting intermediate files and output files for project 'DES',
configuration 'Debug|Win32'.
Compiling...
3des.cpp
Linking...
LINK : error LNK2020: unresolved token (0A00002E) _DebugHeapTag LINK : =
fatal
error LNK1120: 1 unresolved externals
=20
Build log was saved at " <file:///d:\Dev\Des\Des\Debug\BuildLog.htm>
file://d:\Dev\Des\Des\Debug\BuildLog.htm"
DES - 2 error(s), 0 warning(s)
=20
=20
---------------------- Done ----------------------
=20
Rebuild All: 0 succeeded, 1 failed, 0 skipped
=20
=20
I m using VS 2003.NET.
Here is the command line for the compiler:
/Od /I "D:\DEV\BOTAN\include" /AI "D:\DEV\DES\Debug" /D "WIN32" /D =
"_DEBUG"
/D "_MBCS" /FD /EHsc /MTd /GS /Fo"Debug/" /Fd"Debug/vc70.pdb" /W3 =
/nologo /c
/Zi /clr /TP
=20
Here, the command line for the linker:
/OUT:"D:\DEV\DES\Debug\DES.exe" /INCREMENTAL:NO /NOLOGO
/LIBPATH:"D:\DEV\BOTAN" /NODEFAULTLIB:"LIBC" /NODEFAULTLIB:"libcpmtd" =
/DEBUG
/ASSEMBLYDEBUG /PDB:"D:\DEV\DES\Debug/DES.pdb" /FIXED:No libbotan.lib
kernel32.lib user32.lib gdi32.lib winspool.lib comdlg32.lib advapi32.lib
shell32.lib ole32.lib oleaut32.lib uuid.lib odbc32.lib odbccp32.lib
=20
As you can see, I have ignore specific librairies : LIBC and libcpmtd.
=20
Have you an idea ?
Thks
JM
=20
=20
=20
-----Message d'origine-----
De : ope...@li...
[mailto:ope...@li...] De la part de Jack =
Lloyd
Envoy=E9 : mardi 7 septembre 2004 03:28
=C0 : ope...@li...
Objet : Re: [Botan-devel] TripleDES /ECB Mode
=20
On Mon, Sep 06, 2004 at 03:46:52PM +0200, JnMlMe wrote:
> Hi,
>=20
> I need to decrypt a string who was encrypted with Triple DES (ECB =
Mode)
> under SUN Solaris.
>=20
> First, I would like to compile Botan under Windows 2K environment.
> I have a problem when i process the nmake command :
>=20
> =20
>=20
> D:\botan>perl configure.pl msvc-windows-i586 --module-set=3Dwin32
[...]
>=20
> CL /nologo -Iinclude /O2 /G5 /GX /GR /D_CONSOLE /c
src\adler32.cpp
> /Fob
> uild\lib\adler32.obj
> 'CL' n'est pas reconnu en tant que commande interne
> ou externe, un programme ex=E9cutable ou un fichier de commandes.
> NMAKE : fatal error U1077: 'CL' : return code '0x1'
> Stop.
=20
It looks like cl, the command line version of the compiler, isn't in =
your
PATH
variable. Have you tried using the command shell that comes with Visual
Studio?
Running that will automatically set up your environment with the right
variables for using the compiler and linker from the command line. =
Otherwise
you'll have to find where the CL binary is installed, and add that =
directory
to
PATH.
=20
> Second point, i have looked the encrypt and decrypt cpp examples =
files.
>=20
> These files used the CBC mode.
> Should i only changed the mode with ECB in line 117 ?
>=20
=20
No -- the examples are doing password hashing, message authentication, =
and
similiar things that you probably don't want to do. Well, you do want to =
do
them normally, for decent security, but in this case you won't be able =
to
decrypt the message trying to use them, so there is not much point.
=20
The following should start you off, but I don't know how the key is =
derived,
so
as-is it won't work (because currently the key is zero-length, which =
3DES
doesn't take). If you know the hex value of the key, just assign it to =
the
key variable (SymmetricKey key =3D "01234[...]").
=20
----- FILE ------
=20
#include <botan/botan.h>
using namespace Botan;
=20
int main(int argc, char* argv[])
{
try {
if(argc !=3D 2)
{
std::cout << "Usage: " << argv[0] << " input_file" << =
std::endl;
return 1;
}
=20
LibraryInitializer init;
=20
SymmetricKey key; // fill this in with whatever key you have
=20
Pipe pipe(get_cipher("TripleDES/ECB/NoPadding", key, DECRYPTION));
=20
DataSource_Stream input(argv[1]);
pipe.process_msg(input);
std::cout << pipe.read_all_as_string();
}
catch(std::exception& e)
{
std::cout << "Error: " << e.what() << std::endl;
return 1;
}
return 0;
}
=20
----- /FILE ------
=20
=20
-------------------------------------------------------
This SF.Net email is sponsored by BEA Weblogic Workshop
FREE Java Enterprise J2EE developer tools!
Get your free copy of BEA WebLogic Workshop 8.1 today.
http://ads.osdn.com/?ad_id=3D5047&alloc_id=3D10808&op=3Dclick
--=20
opencl-devel mailing list
https://lists.sourceforge.net/lists/listinfo/opencl-devel
|
|
From: Jack L. <ll...@ra...> - 2004-09-07 01:28:14
|
On Mon, Sep 06, 2004 at 03:46:52PM +0200, JnMlMe wrote:
> Hi,
>
> I need to decrypt a string who was encrypted with Triple DES (ECB Mode)
> under SUN Solaris.
>
> First, I would like to compile Botan under Windows 2K environment.
> I have a problem when i process the nmake command :
>
>
>
> D:\botan>perl configure.pl msvc-windows-i586 --module-set=win32
[...]
>
> CL /nologo -Iinclude /O2 /G5 /GX /GR /D_CONSOLE /c src\adler32.cpp
> /Fob
> uild\lib\adler32.obj
> 'CL' n'est pas reconnu en tant que commande interne
> ou externe, un programme exécutable ou un fichier de commandes.
> NMAKE : fatal error U1077: 'CL' : return code '0x1'
> Stop.
It looks like cl, the command line version of the compiler, isn't in your PATH
variable. Have you tried using the command shell that comes with Visual Studio?
Running that will automatically set up your environment with the right
variables for using the compiler and linker from the command line. Otherwise
you'll have to find where the CL binary is installed, and add that directory to
PATH.
> Second point, i have looked the encrypt and decrypt cpp examples files.
>
> These files used the CBC mode.
> Should i only changed the mode with ECB in line 117 ?
>
No -- the examples are doing password hashing, message authentication, and
similiar things that you probably don't want to do. Well, you do want to do
them normally, for decent security, but in this case you won't be able to
decrypt the message trying to use them, so there is not much point.
The following should start you off, but I don't know how the key is derived, so
as-is it won't work (because currently the key is zero-length, which 3DES
doesn't take). If you know the hex value of the key, just assign it to the
key variable (SymmetricKey key = "01234[...]").
----- FILE ------
#include <botan/botan.h>
using namespace Botan;
int main(int argc, char* argv[])
{
try {
if(argc != 2)
{
std::cout << "Usage: " << argv[0] << " input_file" << std::endl;
return 1;
}
LibraryInitializer init;
SymmetricKey key; // fill this in with whatever key you have
Pipe pipe(get_cipher("TripleDES/ECB/NoPadding", key, DECRYPTION));
DataSource_Stream input(argv[1]);
pipe.process_msg(input);
std::cout << pipe.read_all_as_string();
}
catch(std::exception& e)
{
std::cout << "Error: " << e.what() << std::endl;
return 1;
}
return 0;
}
----- /FILE ------
|
|
From: JnMlMe <jn...@fr...> - 2004-09-06 13:46:14
|
Hi,
I need to decrypt a string who was encrypted with Triple DES (ECB Mode)
under SUN Solaris.
First, I would like to compile Botan under Windows 2K environment.
I have a problem when i process the nmake command :
=20
D:\botan>perl configure.pl msvc-windows-i586 --module-set=3Dwin32
Disabling use of symlink()/link() due to Win32 FS limitations
AUTOCONFIG: Enabling module ml_win32
AUTOCONFIG: Enabling module tm_win32
=20
D:\botan>nmake
=20
Microsoft (R) Program Maintenance Utility Version 7.10.3077
Copyright (C) Microsoft Corporation. All rights reserved.
=20
CL /nologo -Iinclude /O2 /G5 /GX /GR /D_CONSOLE /c =
src\adler32.cpp
/Fob
uild\lib\adler32.obj
'CL' n'est pas reconnu en tant que commande interne
ou externe, un programme ex=E9cutable ou un fichier de commandes.
NMAKE : fatal error U1077: 'CL' : return code '0x1'
Stop.
=20
Second point, i have looked the encrypt and decrypt cpp examples files.
These files used the CBC mode.
Should i only changed the mode with ECB in line 117 ?
=20
=20
Thks for your response.
=20
JM
=20
____________
Virus checked by G DATA AntiVirusKit
Version: BD 14.0.534 from 04.09.2004
Virus news: www.antiviruslab.com
|
|
From: Sean R. <sra...@bl...> - 2004-07-02 21:23:28
|
enjoy :-) On Fri, 2004-07-02 at 21:52, Jack Lloyd wrote: > I'm going to be on vacation all next week, and probably won't be checking > email. What, you say? Why would you need to email me about Botan?!? After all, > the code is bug-free, has every feature anyone could want, and the > documentation is flawless! :) > > Anyway, if something comes up, don't expect a response before the 12th. > > -Jack > > > ------------------------------------------------------- > This SF.Net email sponsored by Black Hat Briefings & Training. > Attend Black Hat Briefings & Training, Las Vegas July 24-29 - > digital self defense, top technical experts, no vendor pitches, > unmatched networking opportunities. Visit www.blackhat.com -- Dr. Sean Radford, MBBS, MSc sra...@bl... http://bladesys.demon.co.uk/ |
|
From: Jack L. <ll...@ra...> - 2004-07-02 20:52:29
|
I'm going to be on vacation all next week, and probably won't be checking email. What, you say? Why would you need to email me about Botan?!? After all, the code is bug-free, has every feature anyone could want, and the documentation is flawless! :) Anyway, if something comes up, don't expect a response before the 12th. -Jack |
|
From: <vaclav.ovsik@i.cz> - 2004-06-29 12:03:49
|
Hi On Wed, Jun 23, 2004 at 11:05:18PM -0400, Ramin Javanbakht wrote: > Are there any open source libraries written in C that provides the same > functionalities as Botan (support for different symmetric and asymmetric > algorithms)? What about www.openssl.org ? -- Zito |
|
From: Ken P. <pe...@ca...> - 2004-06-28 16:49:11
|
Ramin, I have seen some good open source routines in C for various components (such as AES, SHA, etc.) but not a complete library. Botan (which I highly recommend) and Crypto++ (which I do not) are the only completely open source cryptographic libraries that I am aware of but these are both in C++. If you would like a pointer to the AES routine it is: http://fp.gladman.plus.com/ I have not come across any open source routines for the asymmetric algorithms in C. But, of course, that alone is no proof that they do not exist. Sorry I don't have better news for you. Ken -----Original Message----- From: ope...@li... [mailto:ope...@li...] On Behalf Of Ramin Javanbakht Sent: Wednesday, June 23, 2004 8:05 PM To: ope...@li... Subject: [Botan-devel] Crypto library in C Hello, Are there any open source libraries written in C that provides the same functionalities as Botan (support for different symmetric and asymmetric algorithms)? Thanks and Regards, Ramin |
|
From: Ramin J. <ra...@ki...> - 2004-06-24 03:05:38
|
Hello, Are there any open source libraries written in C that provides the same functionalities as Botan (support for different symmetric and asymmetric algorithms)? Thanks and Regards, Ramin |
|
From: Jack L. <ll...@ra...> - 2004-06-12 10:32:08
|
This is just a quick note for people on the dev list who aren't on announce: I've release Botan 1.3.14, it's the first release candidate for Botan 1.4.0. Please test Botan on all systems you care about, and make sure your applications are happy with the new verison. Let me know about any problems (success reports are good too). Known good/known bad systems are listed in the email sent to announce. -Jack |
|
From: Jack L. <ll...@ra...> - 2004-06-02 13:57:11
|
Nice! Thanks. -Jack On Wed, Jun 02, 2004 at 10:40:12AM +0100, Sean Radford wrote: > Hi Jack, > > Would be more than happy for you to mention one of the companies that I > am a director of as a user of Botan. > > Company: Aegeus Technology Ltd (http://www.aegeus-technology.com) > Location: UK > Market: Identity & Access Management > Use of Botan: Research implementation of Simple Public Key > Infrastructure (SPKI) with aim to be released as an Open Source project. > > > Kind regards, > > Sean > > > On Tue, 2004-06-01 at 19:05, Jack Lloyd wrote: > > Hi all, > > > > I'm creating a list of projects/companies using Botan, which is currently a > > little sparse (http://botan.randombit.net/users.html). If you're using Botan > > for a project (research, internal, commercial, F/OSS, whatever) and willing to > > have your project/company/whatever listed with a brief description of what it > > uses Botan for, I would appreciate it (a quick blurb+link would be great). I > > know from emails I've received privately that there are quite a few users who > > aren't on the current list, but I don't want to put something on the list that > > would get anyone (including myself!) in hot water with, say, a commercial > > software company that didn't want to admit to using open source code. > > > > Thanks, > > Jack > > > > > > ------------------------------------------------------- > > This SF.Net email is sponsored by: Oracle 10g > > Get certified on the hottest thing ever to hit the market... Oracle 10g. > > Take an Oracle 10g class now, and we'll give you the exam FREE. > > http://ads.osdn.com/?ad_id=3149&alloc_id=8166&op=click > -- > Dr. Sean Radford, MBBS, MSc > sra...@bl... > http://bladesys.demon.co.uk/ > > > > ------------------------------------------------------- > This SF.Net email is sponsored by the new InstallShield X. > >From Windows to Linux, servers to mobile, InstallShield X is the one > installation-authoring solution that does it all. Learn more and > evaluate today! http://www.installshield.com/Dev2Dev/0504 > -- > opencl-devel mailing list > https://lists.sourceforge.net/lists/listinfo/opencl-devel |
|
From: <Han...@br...> - 2004-06-02 12:32:29
|
Hi Jack,
Botan is used within Portalsphere, a CORBA based application server=
.
All of the core Botan library components are exposed via a Tcl scriptin=
g
interface allowing developers to create and use Botan objects as part o=
f an
applications business logic.
Thanks,
Hany=
|
|
From: Sean R. <sra...@bl...> - 2004-06-02 10:16:28
|
Hi Jack, Would be more than happy for you to mention one of the companies that I am a director of as a user of Botan. Company: Aegeus Technology Ltd (http://www.aegeus-technology.com) Location: UK Market: Identity & Access Management Use of Botan: Research implementation of Simple Public Key Infrastructure (SPKI) with aim to be released as an Open Source project. Kind regards, Sean On Tue, 2004-06-01 at 19:05, Jack Lloyd wrote: > Hi all, > > I'm creating a list of projects/companies using Botan, which is currently a > little sparse (http://botan.randombit.net/users.html). If you're using Botan > for a project (research, internal, commercial, F/OSS, whatever) and willing to > have your project/company/whatever listed with a brief description of what it > uses Botan for, I would appreciate it (a quick blurb+link would be great). I > know from emails I've received privately that there are quite a few users who > aren't on the current list, but I don't want to put something on the list that > would get anyone (including myself!) in hot water with, say, a commercial > software company that didn't want to admit to using open source code. > > Thanks, > Jack > > > ------------------------------------------------------- > This SF.Net email is sponsored by: Oracle 10g > Get certified on the hottest thing ever to hit the market... Oracle 10g. > Take an Oracle 10g class now, and we'll give you the exam FREE. > http://ads.osdn.com/?ad_id=3149&alloc_id=8166&op=click -- Dr. Sean Radford, MBBS, MSc sra...@bl... http://bladesys.demon.co.uk/ |
|
From: Jack L. <ll...@ra...> - 2004-06-01 18:05:50
|
Hi all, I'm creating a list of projects/companies using Botan, which is currently a little sparse (http://botan.randombit.net/users.html). If you're using Botan for a project (research, internal, commercial, F/OSS, whatever) and willing to have your project/company/whatever listed with a brief description of what it uses Botan for, I would appreciate it (a quick blurb+link would be great). I know from emails I've received privately that there are quite a few users who aren't on the current list, but I don't want to put something on the list that would get anyone (including myself!) in hot water with, say, a commercial software company that didn't want to admit to using open source code. Thanks, Jack |
|
From: Jack L. <ll...@ra...> - 2004-05-24 17:58:43
|
Nope, that should do what you want. I'll see if I can figure out a reasonably clean way to disable es_file without having to edit the library. Shouldn't be too hard. At the very least I can have a controlling macro in config.h; no run-time control but at least it is a little more self-contained. I'd still like to figure out why it's causing crashes, but for now that will have to wait, I guess. -Jack On Mon, May 24, 2004 at 09:48:01AM -0700, Anthony Carathimas wrote: > I can get my own botan test program to work by editing the slow_poll funtion > in src/es_file by have j initialize to 2 and including the es_capi module. > This I believe skips checking /dev/urandom and /dev/random and simple uses > the capi module. Can you think of any problems with using this modification? > > Thanks > > Anthony Carathimas > > > > > > ------------------------------------------------------- > This SF.Net email is sponsored by: Oracle 10g > Get certified on the hottest thing ever to hit the market... Oracle 10g. > Take an Oracle 10g class now, and we'll give you the exam FREE. > http://ads.osdn.com/?ad_id=3149&alloc_id=8166&op=click > -- > opencl-devel mailing list > https://lists.sourceforge.net/lists/listinfo/opencl-devel |
|
From: Anthony C. <ag...@sa...> - 2004-05-24 16:50:48
|
I can get my own botan test program to work by editing the slow_poll funtion in src/es_file by have j initialize to 2 and including the es_capi module. This I believe skips checking /dev/urandom and /dev/random and simple uses the capi module. Can you think of any problems with using this modification? Thanks Anthony Carathimas |
|
From: Jack L. <ll...@ra...> - 2004-05-23 14:22:46
|
On Sat, May 22, 2004 at 06:11:15PM -0600, Carathimas, Anthony G wrote: > Actually your test program doesn't work when I compile with -mno-cygwin > param., it's results are: > > g++ -mno-cygwin test.cpp -o test > ./test > 0 0 > > So I guess I need a way to compile botan with out using /dev/urandom. No, this is what would be expected - without Cygwin, no /dev/urandom, so the read fails, so it then Botan will continue down the list trying other entropy sources. Based on your report of the crash in Botan when reading /dev/urandom through an ifstream, I would have expected that test.cpp would result in a crash as well. The fact that it doesn't is rather worrisome. -J |
|
From: Carathimas, A. G <ag...@sa...> - 2004-05-23 00:11:35
|
Actually your test program doesn't work when I compile with -mno-cygwin param., it's results are: g++ -mno-cygwin test.cpp -o test ./test 0 0 So I guess I need a way to compile botan with out using /dev/urandom. |
|
From: Jack L. <ll...@ra...> - 2004-05-22 19:08:22
|
On Sat, May 22, 2004 at 12:32:56PM -0600, Carathimas, Anthony G wrote: > I believe I am using the newest version of cygwin which has: > gcc version 3.3.1 > > your sample program returned: > > 1024 1024 Hmmm, that's rather disturbing then. Why would it be failing inside the library? ... > > but Botan using /dev/urandom represents another problem for me because I > want my program to not require cygwin being installed. It doesn't. If it's not running under Cygwin, it will attempt to open /dev/urandom, which will fail, and then attempt to get entropy from the next entropy source on the list (probably the CryptoAPI one). As long as it can find some entropy somewhere, it's happy. Jack |
|
From: Carathimas, A. G <ag...@sa...> - 2004-05-22 18:33:12
|
I believe I am using the newest version of cygwin which has: gcc version 3.3.1 your sample program returned: 1024 1024 but Botan using /dev/urandom represents another problem for me because I want my program to not require cygwin being installed. How would I do this? Would using gcc-window-i586 work, I will have to tell my developers to rename the result library 'libbotan.lib' to 'libbotan.a' so that cygwin ld will find it. I am not able to try this until monday, so I'll give you my results then. |
|
From: Jack L. <ll...@ra...> - 2004-05-20 23:39:52
|
On Thu, May 20, 2004 at 04:23:57PM -0700, Anthony Carathimas wrote:
> I don't want my program to have the dependence on cygwin so I tried compiling
> using these configurations:
>
> ./configure.pl gcc-cygwin-i586 --noauto --modules=tm_hard
> ./configure.pl gcc-cygwin-i586 --noauto --modules=es_capi,es_win32,tm_hard
> ./configure.pl gcc-cygwin-i586 --noauto --modules=es_capi,tm_hard
>
>
> Botan compiles but my program seg faults. Here is log of running my test
> program in gdb:
>
>
> (gdb) run
> Starting program: /home/Administrator/test.exe
>
>
> ************* Begin testing *************
>
> CryptoLib object count (should be 0): 0
>
> Program received signal SIGSEGV, Segmentation fault.
> 0x005670a2 in std::locale::operator=(std::locale const&) ()
> (gdb) where
> #0 0x005670a2 in std::locale::operator=(std::locale const&) ()
> #1 0x00563c8d in std::ios_base::_M_init() ()
> #2 0x005e0ea9 in std::basic_ios<char, std::char_traits<char>
> >::init(std::basic
> _streambuf<char, std::char_traits<char> >*) ()
> #3 0x004a4fff in Botan::File_EntropySource::slow_poll(unsigned char*,
> unsigned)
> ()
> (gdb) quit
> The program is running. Exit anyway? (y or n) y
>
>
> Any suggestions?
Yow. Give me the versions of gcc, libstdc++, and the Cygwin libs you're using,
I'll see if I can reproduce. You might want to check and see if you're running
the latest Cygwin junk, that's all I've tested on. I've got to say, this looks
an awful lot like a Cygwin or libstdc++ bug. Does the following progrm work on
your system?
--------------- CUT HERE ------------------
#include <fstream>
#include <cstdio>
typedef unsigned int u32bit;
typedef unsigned char byte;
int main()
{
byte output[1024];
u32bit length = sizeof(output);
std::ifstream random_source("/dev/urandom");
random_source.read((char*)output, length);
u32bit read1 = random_source.gcount();
u32bit read2 = random_source.gcount();
std::printf("%d %d\n", read1, read2);
}
--------------- CUT HERE ------------------
I am suspicious about this is either a plain and simple libstdc++ bug (Botan
doesn't touch anything even remotely near the locale code), or a weird
interaction between iostreams and Cygwin's handling of /dev/ psuedo-devices (on
Cygwin there are no block/character devices, the names are interpolated somehow
inside the Cygwin libs). OTOH, if the above works, it's possible I've borked
something up somehow.
>
> It would also be nice if I could use
>
I think your mail got truncated here...
Jack
|
|
From: Anthony C. <ag...@sa...> - 2004-05-20 23:27:11
|
I don't want my program to have the dependence on cygwin so I tried compiling using these configurations: ./configure.pl gcc-cygwin-i586 --noauto --modules=tm_hard ./configure.pl gcc-cygwin-i586 --noauto --modules=es_capi,es_win32,tm_hard ./configure.pl gcc-cygwin-i586 --noauto --modules=es_capi,tm_hard Botan compiles but my program seg faults. Here is log of running my test program in gdb: (gdb) run Starting program: /home/Administrator/test.exe ************* Begin testing ************* CryptoLib object count (should be 0): 0 Program received signal SIGSEGV, Segmentation fault. 0x005670a2 in std::locale::operator=(std::locale const&) () (gdb) where #0 0x005670a2 in std::locale::operator=(std::locale const&) () #1 0x00563c8d in std::ios_base::_M_init() () #2 0x005e0ea9 in std::basic_ios<char, std::char_traits<char> >::init(std::basic _streambuf<char, std::char_traits<char> >*) () #3 0x004a4fff in Botan::File_EntropySource::slow_poll(unsigned char*, unsigned) () (gdb) quit The program is running. Exit anyway? (y or n) y Any suggestions? It would also be nice if I could use |
|
From: Jack L. <ll...@ra...> - 2004-05-04 20:33:20
|
1.3.12 is very broken. It's an easy fix, just delete line 33 from pkcs10.h.= My release QA has not been very good of late. I'll try to do a fixed 1.3.13 by the end of the week. -Jack ----- Forwarded message from Jack Lloyd <ll...@ra...> ----- From: Jack Lloyd <ll...@ra...> To: V?clav Ovs?k <vaclav.ovsik@i.cz> Date: Tue, 4 May 2004 16:29:45 -0400 Subject: Re: [Botan-devel] Perl Botan binding... On Tue, May 04, 2004 at 03:33:01PM +0200, V?clav Ovs?k wrote: [...] >=20 > Compiling 1.3.12 with gcc 3.0.4 ends for this time like with 2.95.4: > g++-3.0 -Iinclude/ -O3 -march=3Di686 -ansi -Wno-long-long -fpermissive -W= -Wall -fPIC -c src/pkcs10.cpp -o build/lib/pkcs10.o > In file included from src/pkcs10.cpp:6: > include/botan/pkcs10.h: In member function `std::vector<Botan::OID,=20 > std::allocator<Botan::OID> > Botan::PKCS10_Request::get_policies() con= st': > include/botan/pkcs10.h:33: `policies' undeclared (first use this function) > include/botan/pkcs10.h:33: (Each undeclared identifier is reported only o= nce=20 > for each function it appears in.) > make: *** [build/lib/pkcs10.o] Error 1 >=20 > Something is bad? <kicks self hard>. 1.3.12 is broken, I obviously copied and pasted a little= too much. I'll put out a fixed 1.3.13 soon. For now, just delete line 33 in pkcs10.h and I think it will work. That function isn't supposed to be there= at all. -Jack ----- End forwarded message ----- |
