All of lore.kernel.org
 help / color / mirror / Atom feed
From: Glenn Washburn <development@efficientek.com>
To: grub-devel@gnu.org, Daniel Kiper <dkiper@net-space.pl>
Cc: Robbie Harwood <rharwood@redhat.com>,
	Peter Jones <pjones@redhat.com>,
	Glenn Washburn <development@efficientek.com>
Subject: [PATCH v6 11/14] gdb: Fix issue with breakpoints defined before the GRUB image is loaded
Date: Tue, 10 Jan 2023 15:50:38 -0600	[thread overview]
Message-ID: <20230110215041.1247699-12-development@efficientek.com> (raw)
In-Reply-To: <20230110215041.1247699-1-development@efficientek.com>

On some platforms, notably x86, software breakpoints set in GDB before
the GRUB image is loaded will be cleared when the image is loaded. This
is because the breakpoints work by overwriting the memory of the break-
point location with a special instruction which when hit will cause the
debugger to stop execution. Just before execution is resumed by the
debugger, the original instruction bytes are put back. When a breakpoint
is set before the GRUB image is loaded, the special debugger instruction
will be written to memory and when the GRUB image is loaded by the
firmware, which has no knowledge of the debugger, the debugger instruction
is overwritten. To the GDB user, GDB will show the breakpoint as set, but
it will never be hit. Furthermore, GDB now becomes confused, such that
even deleting and re-setting the breakpoint after the GRUB image is loaded
will not allow for a working breakpoint.

To work around this, in run_on_start, first a watchpoint is set on _start,
which will be triggered when the firmware starts loading the GRUB image.
When the _start watchpoint is hit, the current breakpoints are saved to a
file and then deleted by GDB before they can be overwritten by the firmware
and confuse GDB. Then a temporary software breakpoint is set on _start,
which will get triggered when the firmware hands off to GRUB to execute. In
that breakpoint load the previously saved and deleted breakpoints now that
there is no worry of them getting overwritten by the firmware.

Note that watchpoints are generally types of hardware breakpoints on x86, so
its deleted as soon as it gets triggered so that a minimal set of hardware
breakpoints are used, allowing more for the user.

Signed-off-by: Glenn Washburn <development@efficientek.com>
---
 grub-core/gdb_grub.in | 33 +++++++++++++++++++++++++++------
 1 file changed, 27 insertions(+), 6 deletions(-)

diff --git a/grub-core/gdb_grub.in b/grub-core/gdb_grub.in
index 281dfb5927..8e89bbf368 100644
--- a/grub-core/gdb_grub.in
+++ b/grub-core/gdb_grub.in
@@ -59,14 +59,35 @@ end
 define run_on_start
 	# TODO: Add check to see if _start symbol is defined, if not, then
 	# the symbols have not yet been loaded and this command will not work.
-	thbreak _start
+	watch *_start
+	set $break_efi_start_bpnum = $bpnum
 	commands
 		silent
-
-		runtime_load_module
-
-		if $is_user_command("onstart")
-			onstart
+		delete $break_efi_start_bpnum
+
+		# Save the breakpoints here before the GRUB image is loaded
+		# into memory, then delete them. Later they will be reloaded
+		# once the GRUB image has been loaded. This avoids the issue
+		# where the loading of the GRUB image overwrites the software
+		# breakpoints, thus confusing GDB and effectively clearing
+		# those breakpoints.
+		save breakpoints .early-breakpoints.gdb
+		delete breakpoints
+
+		tbreak _start
+		commands
+			silent
+
+			# Reload the breakpoints now that the GRUB image has
+			# finished being loaded into memory.
+			source .early-breakpoints.gdb
+
+			runtime_load_module
+
+			if $is_user_command("onstart")
+				onstart
+			end
+			continue
 		end
 		continue
 	end
-- 
2.34.1



  parent reply	other threads:[~2023-01-10 21:51 UTC|newest]

Thread overview: 18+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-01-10 21:50 [PATCH v6 00/14] GDB script fixes and improvements Glenn Washburn
2023-01-10 21:50 ` [PATCH v6 01/14] gdb: Fix redirection issue in dump_module_sections Glenn Washburn
2023-01-10 21:50 ` [PATCH v6 02/14] gdb: Prevent wrapping when writing to .segments.tmp Glenn Washburn
2023-01-10 21:50 ` [PATCH v6 03/14] gdb: If no modules have been loaded, do not try to load module symbols Glenn Washburn
2023-01-10 21:50 ` [PATCH v6 04/14] gdb: Move runtime module loading into runtime_load_module Glenn Washburn
2023-01-10 21:50 ` [PATCH v6 05/14] gdb: Conditionally run GDB script logic for dynamically or statically positioned GRUB Glenn Washburn
2023-01-10 21:50 ` [PATCH v6 06/14] gdb: Only connect to remote target once when first sourced Glenn Washburn
2023-01-10 21:50 ` [PATCH v6 07/14] gdb: Replace module symbol loading implementation with Python one Glenn Washburn
2023-01-10 21:50 ` [PATCH v6 08/14] gdb: Add functions to make loading from dynamically positioned targets easier Glenn Washburn
2023-01-10 21:50 ` [PATCH v6 09/14] gdb: Add more support for debugging on EFI platforms Glenn Washburn
2023-02-20 20:06   ` Robbie Harwood
2023-02-21  3:07     ` Glenn Washburn
2023-03-02 18:44       ` Daniel Kiper
2023-01-10 21:50 ` [PATCH v6 10/14] gdb: Allow running user-defined commands at GRUB start Glenn Washburn
2023-01-10 21:50 ` Glenn Washburn [this message]
2023-01-10 21:50 ` [PATCH v6 12/14] gdb: Add extra early initialization symbols for i386-pc Glenn Washburn
2023-01-10 21:50 ` [PATCH v6 13/14] gdb: Modify gdb prompt when running gdb_grub script Glenn Washburn
2023-01-10 21:50 ` [PATCH v6 14/14] docs: Add debugging chapter to development documentation Glenn Washburn

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=20230110215041.1247699-12-development@efficientek.com \
    --to=development@efficientek.com \
    --cc=dkiper@net-space.pl \
    --cc=grub-devel@gnu.org \
    --cc=pjones@redhat.com \
    --cc=rharwood@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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.