Re: alignment and thread problem - solaris

2000-01-21 Thread Richard Levitte - VMS Whacker
From: Randy Wright <[EMAIL PROTECTED]> rw26> I know that malloc is not reentrant. However, if somewhere in the openssl rw26> libraries malloc is also called, perhaps that is when I get this crash. Yes, OpenSSL uses malloc by default. I think it is generally assumed that any OS that supports th

Re: alignment and thread problem - solaris

2000-01-21 Thread Randy Wright
On Thu, 20 Jan 2000, Richard Levitte - VMS Whacker wrote: > rw26> I'm using openssl-0.9.4 on solirs 2.6 on a Sun E3500 (sparc) > [...] > > Noted. > > rw26> The other problem is with my multithreading. I'm opening a thread > rw26> for each new accepted socket. When I do this I I call SSL_new, s

Re: alignment and thread problem - solaris

2000-01-21 Thread Goetz Babin-Ebell
At 15:55 19.01.00 -0500, you wrote: Hello, (sorry for my last mail, mixed cancel with send...) >I'm using openssl-0.9.4 on solirs 2.6 on a Sun E3500 (sparc) >with the egcs gcc compiler v. 2.91.60. The application is multithreaded >using pthreads. > >Anyway, when I run my app, I call n = SSL_read

Re: alignment and thread problem - solaris

2000-01-19 Thread Richard Levitte - VMS Whacker
rw26> I'm using openssl-0.9.4 on solirs 2.6 on a Sun E3500 (sparc) [...] Noted. rw26> Anyway, when I run my app, I call n = SSL_read( ssl, (char *)&ch, 1 ); rw26> to retrieve a byte into an int named ch. I find that my correct rw26> byte is returned in the most significant byte of the int. No s

Re: alignment and thread problem - solaris

2000-01-19 Thread Goetz Babin-Ebell
At 15:55 19.01.00 -0500, you wrote: Hello, >I'm using openssl-0.9.4 on solirs 2.6 on a Sun E3500 (sparc) >with the egcs gcc compiler v. 2.91.60. The application is multithreaded >using pthreads. > >Anyway, when I run my app, I call n = SSL_read( ssl, (char *)&ch, 1 ); >to retrieve a byte into an