d/freebsd-dev
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Fix single-stepping when the kernel was entered through the EPC syscall

Browse Source 
path. When the taken branch leaves the kernel and enters the process,
we still need to execute the instruction at that address. Don't raise
SIGTRAP when we branch into the process, but enable single-stepping
instead.
This commit is contained in:
Marcel Moolenaar 2010-02-06 20:46:14 +00:00
parent d180e2b3d4
commit 58ce165dfd
Notes: svn2git 2020-12-20 02:59:44 +00:00 
svn path=/head/; revision=203572
1 changed files with 5 additions and 4 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

9
sys/ia64/ia64/trap.c
View File

@ -803,7 +803,7 @@ trap(int vector, struct trapframe *tf)
* out of the gateway page we'll get back into the kernel
* and then we enable single stepping.
* Since this a rather round-about way of enabling single
* stepping, don't make things complicated even more by
* stepping, don't make things even more complicated by
* calling userret() and do_ast(). We do that later...
*/
tf->tf_special.psr &= ~IA64_PSR_LP;
@ -814,13 +814,14 @@ trap(int vector, struct trapframe *tf)
/*
* Don't assume there aren't any branches other than the
* branch that takes us out of the gateway page. Check the
* iip and raise SIGTRAP only when it's an user address.
* iip and enable single stepping only when it's an user
* address.
*/
if (tf->tf_special.iip >= VM_MAX_ADDRESS)
return;
tf->tf_special.psr &= ~IA64_PSR_TB;
sig = SIGTRAP;
break;
tf->tf_special.psr |= IA64_PSR_SS;
return;
case IA64_VEC_IA32_EXCEPTION:
case IA64_VEC_IA32_INTERCEPT: