From: "Andrew Burgess" <aburgess@broadcom.com>
To: gdb-patches@sourceware.org
Cc: "Pedro Alves" <palves@redhat.com>
Subject: Re: [PATCH] Extra error message from update_watchpoint
Date: Tue, 29 Oct 2013 16:44:00 -0000 [thread overview]
Message-ID: <526FE5E9.3000909@broadcom.com> (raw)
In-Reply-To: <52616D4B.3020209@redhat.com>
On 18/10/2013 6:18 PM, Pedro Alves wrote:
> On 10/18/2013 10:20 AM, Andrew Burgess wrote:
>> @@ -1946,8 +1946,13 @@ update_watchpoint (struct watchpoint *b, int reparse)
>> }
>> }
>> else if (!b->base.ops->works_in_software_mode (&b->base))
>> - error (_("Expression cannot be implemented with "
>> - "read/access watchpoint."));
>> + {
>> + if (!can_use_hw_watchpoints)
>> + error (_("Software read/access watchpoints not supported."));
>
> Hmm, is this really "not supported", or rather "explicitly disabled"?
> I think I'd prefer:
>
> error (_("Hardware watchpoints support disabled. "
> "See set/show can-use-hw-watchpoints."));
>
> which also hints towards the toggle. WDYT?
I used that error message as it matches the one I already added in:
https://sourceware.org/ml/gdb-patches/2013-10/msg00563.html
but I don't really mind.
Ideally you might want both bits of the error message, you've
turned off H/W watchpoints so you need to be told that read/access
watchpoints are not supported in S/W, but I agree, we should also
let the user know that H/W watchpoints have been turned off and
could be turned back on again. That seems like too much for one
error message though... I'm happy to go with your suggestion.
Updated patch changes both error messages to match, OK to apply?
Thanks,
Andrew
gdb/ChangeLog
2013-10-29 Andrew Burgess <aburgess@broadcom.com>
* breakpoint.c (update_watchpoint): Update error message and add
an additional error message.
gdb/testsuite/ChangeLog
2013-10-29 Andrew Burgess <aburgess@broadcom.com>
* gdb.base/watchpoint.exp (test_no_hw_watchpoints): Add additional
tests and update expected error message.
(test_watch_register_location): New tests.
(do_tests): Call test_watch_register_location.
* gdb.base/watchpoints.exp: Update expected error message.
diff --git a/gdb/breakpoint.c b/gdb/breakpoint.c
index 608463d..68b348d 100644
--- a/gdb/breakpoint.c
+++ b/gdb/breakpoint.c
@@ -1805,7 +1805,8 @@ update_watchpoint (struct watchpoint *b, int reparse)
if (b->base.ops->works_in_software_mode (&b->base))
b->base.type = bp_watchpoint;
else
- error (_("Software read/access watchpoints not supported."));
+ error (_("Hardware watchpoint support disabled. "
+ "See set/show can-use-hw-watchpoints."));
}
}
else if (within_current_scope && b->exp)
@@ -1946,8 +1947,14 @@ update_watchpoint (struct watchpoint *b, int reparse)
}
}
else if (!b->base.ops->works_in_software_mode (&b->base))
- error (_("Expression cannot be implemented with "
- "read/access watchpoint."));
+ {
+ if (!can_use_hw_watchpoints)
+ error (_("Hardware watchpoint support disabled. "
+ "See set/show can-use-hw-watchpoints."));
+ else
+ error (_("Expression cannot be implemented with "
+ "read/access watchpoint."));
+ }
else
b->base.type = bp_watchpoint;
diff --git a/gdb/testsuite/gdb.base/watchpoint.exp b/gdb/testsuite/gdb.base/watchpoint.exp
index e0d4f81..0ff5802 100644
--- a/gdb/testsuite/gdb.base/watchpoint.exp
+++ b/gdb/testsuite/gdb.base/watchpoint.exp
@@ -825,8 +825,12 @@ proc test_no_hw_watchpoints {} {
# refrains from using them.
#
gdb_test "rwatch ival3" \
- "Expression cannot be implemented with read/access watchpoint..*" \
+ "Hardware watchpoint support disabled\. See set/show can-use-hw-watchpoints\." \
"rwatch disallowed when can-set-hw-watchpoints cleared"
+ gdb_test "awatch ival3" \
+ "Hardware watchpoint support disabled\. See set/show can-use-hw-watchpoints\." \
+ "awatch disallowed when can-set-hw-watchpoints cleared"
+
# Re-enable hardware watchpoints if necessary.
if ![target_info exists gdb,no_hardware_watchpoints] {
@@ -879,6 +883,22 @@ proc test_watchpoint_in_big_blob {} {
gdb_test_no_output "delete \$bpnum" "delete watch buf"
}
+proc test_watch_register_location {} {
+ global no_hw
+
+ if {!$no_hw && ![target_info exists gdb,no_hardware_watchpoints]} {
+ # Non-memory read/access watchpoints are not supported, they would
+ # require software read/access watchpoint support (which is not
+ # currently available).
+ gdb_test "rwatch \$pc" \
+ "Expression cannot be implemented with read/access watchpoint..*" \
+ "rwatch disallowed for register based expression"
+ gdb_test "awatch \$pc" \
+ "Expression cannot be implemented with read/access watchpoint..*" \
+ "awatch disallowed for register based expression"
+ }
+}
+
# Start with a fresh gdb.
set prev_timeout $timeout
@@ -940,6 +960,8 @@ proc do_tests {} {
test_wide_location_1
test_wide_location_2
+
+ test_watch_register_location
}
# On targets that can do hardware watchpoints, run the tests twice:
diff --git a/gdb/testsuite/gdb.base/watchpoints.exp b/gdb/testsuite/gdb.base/watchpoints.exp
index b70e86c..e429f91 100644
--- a/gdb/testsuite/gdb.base/watchpoints.exp
+++ b/gdb/testsuite/gdb.base/watchpoints.exp
@@ -41,10 +41,10 @@ with_test_prefix "before inferior start" {
# and read watchpoints require hardware watchpoint support, with this
# turned off these can't be created.
gdb_test "awatch ival1" \
- "Software read/access watchpoints not supported." \
+ "Hardware watchpoint support disabled\. See set/show can-use-hw-watchpoints\." \
"create access watchpoint"
gdb_test "rwatch ival1" \
- "Software read/access watchpoints not supported." \
+ "Hardware watchpoint support disabled\. See set/show can-use-hw-watchpoints\." \
"create read watchpoint"
}
next prev parent reply other threads:[~2013-10-29 16:44 UTC|newest]
Thread overview: 15+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-10-18 9:20 Andrew Burgess
2013-10-18 17:18 ` Pedro Alves
2013-10-29 16:44 ` Andrew Burgess [this message]
2013-10-29 17:09 ` Pedro Alves
2013-10-29 17:19 ` Eli Zaretskii
2013-10-29 17:38 ` Andrew Burgess
2013-10-29 17:52 ` Pedro Alves
2013-10-29 19:09 ` Eli Zaretskii
2013-10-30 11:49 ` Andrew Burgess
2013-10-30 15:41 ` Eli Zaretskii
2013-10-30 17:13 ` Andrew Burgess
2013-10-30 18:41 ` Eli Zaretskii
2013-10-31 13:05 ` Andrew Burgess
2013-10-30 18:48 ` Pedro Alves
2013-10-30 19:19 ` Eli Zaretskii
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=526FE5E9.3000909@broadcom.com \
--to=aburgess@broadcom.com \
--cc=gdb-patches@sourceware.org \
--cc=palves@redhat.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox