From: Pedro Alves <pedro@palves.net>
To: gdb-patches@sourceware.org
Subject: [PATCH 15/23] Make breakpoint_address_bits look at the location kind
Date: Mon, 16 May 2022 19:40:22 +0100 [thread overview]
Message-ID: <20220516184030.665489-16-pedro@palves.net> (raw)
In-Reply-To: <20220516184030.665489-1-pedro@palves.net>
Software watchpoints allocate a special dummy location using
software_watchpoint_add_no_memory_location, and then
breakpoint_address_bits checks whether the location is that special
location to decide whether the location has a meaninful address to
print.
Introduce a new bp_loc_software_watchpoint location kind, and make
breakpoint_address_bits use bl_address_is_meaningful instead, which
returns false for bp_loc_other, which is in accordance with we
document for bp_location::address:
/* (... snip ...) Valid for all types except
bp_loc_other. */
CORE_ADDR address = 0;
Rename software_watchpoint_add_no_memory_location to
add_dummy_location, and simplify it. This will be used by catchpoints
too in a following patch.
Note that neither "info breakpoints" nor "maint info breakpoints"
actually prints the addresses of watchpoints, but I think it would be
useful to do so in "maint info breakpoints". This approach let's us
implement that in the future.
Change-Id: I50e398f66ef618c31ffa662da755eaba6295aed7
---
gdb/breakpoint.c | 59 ++++++++++++++++++------------------------------
gdb/breakpoint.h | 1 +
2 files changed, 23 insertions(+), 37 deletions(-)
diff --git a/gdb/breakpoint.c b/gdb/breakpoint.c
index 6e24137ec4b..6dfb89f51bc 100644
--- a/gdb/breakpoint.c
+++ b/gdb/breakpoint.c
@@ -164,6 +164,8 @@ static std::vector<symtab_and_line> bkpt_probe_decode_location
struct event_location *location,
struct program_space *search_pspace);
+static bool bl_address_is_meaningful (bp_location *loc);
+
/* update_global_location_list's modes of operation wrt to whether to
insert locations now. */
enum ugll_insert_mode
@@ -1830,34 +1832,18 @@ extract_bitfield_from_watchpoint_value (struct watchpoint *w, struct value *val)
return bit_val;
}
-/* Allocate a dummy location and add it to B, which must be a software
- watchpoint. This is required because even if a software watchpoint
- is not watching any memory, bpstat_stop_status requires a location
- to be able to report stops. */
+/* Allocate a dummy location and add it to B. This is required
+ because bpstat_stop_status requires a location to be able to report
+ stops. */
static void
-software_watchpoint_add_no_memory_location (struct breakpoint *b,
- struct program_space *pspace)
+add_dummy_location (struct breakpoint *b,
+ struct program_space *pspace)
{
- gdb_assert (b->type == bp_watchpoint && b->loc == NULL);
+ gdb_assert (b->loc == NULL);
- b->loc = b->allocate_location ();
+ b->loc = new bp_location (b, bp_loc_other);
b->loc->pspace = pspace;
- b->loc->address = -1;
- b->loc->length = -1;
-}
-
-/* Returns true if B is a software watchpoint that is not watching any
- memory (e.g., "watch $pc"). */
-
-static bool
-is_no_memory_software_watchpoint (struct breakpoint *b)
-{
- return (b->type == bp_watchpoint
- && b->loc != NULL
- && b->loc->next == NULL
- && b->loc->address == -1
- && b->loc->length == -1);
}
/* Assuming that B is a watchpoint:
@@ -2197,7 +2183,7 @@ update_watchpoint (struct watchpoint *b, int reparse)
else
b->type = bp_watchpoint;
- loc_type = (b->type == bp_watchpoint? bp_loc_other
+ loc_type = (b->type == bp_watchpoint? bp_loc_software_watchpoint
: bp_loc_hardware_watchpoint);
for (bp_location *bl : b->locations ())
bl->loc_type = loc_type;
@@ -2208,7 +2194,7 @@ update_watchpoint (struct watchpoint *b, int reparse)
bpstat_stop_status requires a location to be able to report
stops, so make sure there's at least a dummy one. */
if (b->type == bp_watchpoint && b->loc == NULL)
- software_watchpoint_add_no_memory_location (b, frame_pspace);
+ add_dummy_location (b, frame_pspace);
}
else if (!within_current_scope)
{
@@ -3840,11 +3826,12 @@ detach_breakpoints (ptid_t ptid)
/* This function must physically remove breakpoints locations
from the specified ptid, without modifying the breakpoint
- package's state. Locations of type bp_loc_other are only
- maintained at GDB side. So, there is no need to remove
- these bp_loc_other locations. Moreover, removing these
+ package's state. Locations of type bp_loc_other and
+ bp_loc_software_watchpoint are only maintained at GDB side,
+ so there is no need to remove them. Moreover, removing these
would modify the breakpoint package's state. */
- if (bl->loc_type == bp_loc_other)
+ if (bl->loc_type == bp_loc_other
+ || bl->loc_type == bp_loc_software_watchpoint)
continue;
if (bl->inserted)
@@ -5798,7 +5785,8 @@ bpstat_what (bpstat *bs_head)
{
/* Some catchpoints are implemented with breakpoints.
For those, we need to step over the breakpoint. */
- if (bs->bp_location_at->loc_type != bp_loc_other)
+ if (bs->bp_location_at->loc_type == bp_loc_software_breakpoint
+ || bs->bp_location_at->loc_type == bp_loc_hardware_breakpoint)
this_action = BPSTAT_WHAT_SINGLE;
}
break;
@@ -6557,16 +6545,12 @@ breakpoint_address_bits (struct breakpoint *b)
{
int print_address_bits = 0;
- /* Software watchpoints that aren't watching memory don't have an
- address to print. */
- if (is_no_memory_software_watchpoint (b))
- return 0;
-
for (bp_location *loc : b->locations ())
{
- int addr_bit;
+ if (!bl_address_is_meaningful (loc))
+ continue;
- addr_bit = gdbarch_addr_bit (loc->gdbarch);
+ int addr_bit = gdbarch_addr_bit (loc->gdbarch);
if (addr_bit > print_address_bits)
print_address_bits = addr_bit;
}
@@ -7161,6 +7145,7 @@ bp_location_from_bp_type (bptype type)
case bp_access_watchpoint:
return bp_loc_hardware_watchpoint;
case bp_watchpoint:
+ return bp_loc_software_watchpoint;
case bp_catchpoint:
case bp_tracepoint:
case bp_fast_tracepoint:
diff --git a/gdb/breakpoint.h b/gdb/breakpoint.h
index a4ead8b4d4e..7375e976dc6 100644
--- a/gdb/breakpoint.h
+++ b/gdb/breakpoint.h
@@ -314,6 +314,7 @@ enum bp_loc_type
{
bp_loc_software_breakpoint,
bp_loc_hardware_breakpoint,
+ bp_loc_software_watchpoint,
bp_loc_hardware_watchpoint,
bp_loc_other /* Miscellaneous... */
};
--
2.36.0
next prev parent reply other threads:[~2022-05-16 18:41 UTC|newest]
Thread overview: 31+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-05-16 18:40 [PATCH 00/23] More breakpoints cleanups Pedro Alves
2022-05-16 18:40 ` [PATCH 01/23] add_location_to_breakpoint -> breakpoint::add_location Pedro Alves
2022-05-16 18:40 ` [PATCH 02/23] Make structs breakpoint/base_breakpoint/catchpoint be abstract Pedro Alves
2022-05-16 18:40 ` [PATCH 03/23] ranged_breakpoint: don't use init_raw_breakpoint Pedro Alves
2022-05-16 18:40 ` [PATCH 04/23] ranged_breakpoint, use install_breakpoint Pedro Alves
2022-05-16 18:40 ` [PATCH 05/23] ranged_breakpoint - move initialization to ctor Pedro Alves
2022-05-16 18:40 ` [PATCH 06/23] Make a few functions work with base_breakpoint instead of breakpoint Pedro Alves
2022-05-16 18:40 ` [PATCH 07/23] More breakpoint_ops parameter elimination Pedro Alves
2022-05-16 18:40 ` [PATCH 08/23] Remove "internal" parameter from a couple functions Pedro Alves
2022-05-16 18:40 ` [PATCH 09/23] init_breakpoint_sal -> base_breakpoint::base_breakpoint Pedro Alves
2022-05-20 13:40 ` Lancelot SIX
2022-05-20 19:20 ` Pedro Alves
2022-05-16 18:40 ` [PATCH 10/23] Make ada_catchpoint_location's owner ctor parameter be ada_catchpoint Pedro Alves
2022-05-16 18:40 ` [PATCH 11/23] Convert init_ada_exception_catchpoint to a ctor Pedro Alves
2022-05-16 18:40 ` [PATCH 12/23] Refactor set_internal_breakpoint / internal_breakpoint ctor Pedro Alves
2022-05-16 18:40 ` [PATCH 13/23] Refactor momentary breakpoints, eliminate set_raw_breakpoint{, _without_location} Pedro Alves
2022-05-16 18:40 ` [PATCH 14/23] Make exception_catchpoint inherit base_breakpoint instead of catchpoint Pedro Alves
2022-05-16 18:40 ` Pedro Alves [this message]
2022-05-16 18:40 ` [PATCH 16/23] Make catchpoint inherit breakpoint, eliminate init_raw_breakpoint Pedro Alves
2022-05-16 18:40 ` [PATCH 17/23] Move common bits of catchpoint/exception_catchpoint to breakpoint's ctor Pedro Alves
2022-05-16 18:40 ` [PATCH 18/23] Move add_location(sal) to base_breakpoint Pedro Alves
2022-05-16 18:40 ` [PATCH 19/23] Add/tweak intro comments of struct breakpoint and several subclasses Pedro Alves
2022-05-16 18:40 ` [PATCH 20/23] Momentary breakpoints should have no breakpoint number Pedro Alves
2022-05-20 16:00 ` Tom Tromey
2022-05-20 19:25 ` Pedro Alves
2022-05-16 18:40 ` [PATCH 21/23] Make sure momentary breakpoints are always thread-specific Pedro Alves
2022-05-16 18:40 ` [PATCH 22/23] Test "set multiple-symbols on" creating multiple breakpoints Pedro Alves
2022-05-16 18:40 ` [PATCH 23/23] Rename base_breakpoint -> code_breakpoint Pedro Alves
2022-05-20 16:05 ` Tom Tromey
2022-05-20 16:06 ` [PATCH 00/23] More breakpoints cleanups Tom Tromey
2022-05-20 19:43 ` Pedro Alves
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=20220516184030.665489-16-pedro@palves.net \
--to=pedro@palves.net \
--cc=gdb-patches@sourceware.org \
/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;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).