This patch looks correct to me, so I applied. Don, speak up if you object... --david >>>>> On Thu, 26 Jul 2001 19:34:47 +0200, "Jose Luu" <jluu@mainsoft.com> said: Jose> This is the solution to the problem I described in june, Jose> see test case there: Jose> https://external-lists.valinux.com/archives//linux-ia64/2001-June/001792.htm Jose> l Jose> There was an oversight on the size of the long, which caused Jose> an improper restoration of the real time signals blocking mask Jose> of the 32 bit frame, 2 of these signals are used by the Jose> pthreads library for wakeup or cancellation. Jose> This patch can be applied to all 2.4.x series kernels Jose> including 2.4.7, since the file has not changed. Jose> Jose Jose> --- 2.4.5/arch/ia64/ia32/ia32_signal.c Tue Oct 10 02:54:53 Jose> 2000 +++ new-2.4.5/arch/ia64/ia32/ia32_signal.c Thu Jul 26 Jose> 12:49:42 2001 @@ -279,7 +279,7 @@ err |= Jose> setup_sigcontext_ia32(&frame->sc, &frame->fpstate, regs, set-> sig[0]); Jose> if (_IA32_NSIG_WORDS > 1) { - err |= Jose> __copy_to_user(frame->extramask, &set->sig[1], + err |= Jose> __copy_to_user(frame->extramask,(((char *) &set->sig) + 4) , Jose> sizeof(frame->extramask)); } Jose> _______________________________________________ Linux-IA64 Jose> mailing list Linux-IA64@linuxia64.org Jose> http://lists.linuxia64.org/lists/listinfo/linux-ia64Received on Mon Jul 30 18:28:41 2001
This archive was generated by hypermail 2.1.8 : 2005-08-02 09:20:04 EST