This is the mail archive of the
mailing list for the Archer project.
Re: [PATCH 0/4] 'catch syscall' feature
- From: Phil Muldoon <pmuldoon at redhat dot com>
- To: Sérgio Durigan Júnior<sergiodj at linux dot vnet dot ibm dot com>
- Cc: Project Archer <archer at sourceware dot org>
- Date: Mon, 13 Oct 2008 13:01:34 +0100
- Subject: Re: [PATCH 0/4] 'catch syscall' feature
- References: <1222799256.30389.31.camel@miki>
Sérgio Durigan Júnior wrote:
For now the feature is only implemented for PPC32 and PPC64, but in a
future not so distant I intend to send patches for x86 and x86_64 too.
I looked at this patch, it looks great. Beyond the comments that others
have added I have only a few small nitpick comments. As I cannot run
this patch on my architecture, I'll have to guess a bit. What happens if
a user attempts to use it without specific architecture support? What
error message is written? The code that sets the breakpoint is:
+ Let's check if there's a syscall name. */
+ if (arg != NULL)
+ syscall_number = gdbarch_syscall_number_from_name (current_gdbarch,
+ (const char *) arg);
+ if (syscall_number == UNKNOWN_SYSCALL)
+ error (_("Invalid syscall name '%s'."), arg);
+ /* Now let's create the catchpoint */
+ create_syscall_event_catchpoint (tempflag, syscall_number);
But if I were to run this say on x86. would it print "Invalid syscall
name" for every (valid, but not implemented in GDB yet) syscall? If so,
I think gate the feature with "not supported on your architecture" or
some other message?
The other thing is, is this test included in the patch only gated to run
on Linux architectures that have syscall support added into GDB?
I'm really looking forward to seeing this implemented n x86 and x864 ;)