From: Zoran Zaric <Zoran.Zaric@amd.com>
To: gdb-patches@sourceware.org
Cc: Zoran Zaric <Zoran.Zaric@amd.com>
Subject: [PATCH 27/30] Add support for any location description in CFI
Date: Mon, 7 Dec 2020 19:00:28 +0000 [thread overview]
Message-ID: <20201207190031.13341-28-Zoran.Zaric@amd.com> (raw)
In-Reply-To: <20201207190031.13341-1-Zoran.Zaric@amd.com>
One of the main benefits of allowing location description to be on the
DWARF stack is that now CFI expression based register rules can be
defined using a location description operations. This allows a register
of one frame to be saved in any location, including any composite
location.
To fully support this feature, the execute_stack_op function in
dwarf2/frame.c needs to return a single struct value object instead of
just an address.
Function put_frame_register_bytes also needs to change to support any
location description.
This support is a one of the key features to truly support optimized
code.
gdb/ChangeLog:
* dwarf2/frame.c (execute_stack_op): Change to return a struct
value object.
(dwarf2_frame_cache): Change to call new execute_stack_op
definition.
(dwarf2_frame_prev_register): Change to call new execute_stack_op
definition.
* frame.c (put_frame_register_bytes): Add support for writing to
composite location description.
---
gdb/dwarf2/frame.c | 54 ++++++++++++++++++++++++++--------------------
gdb/frame.c | 36 +++++++++++++++++++++++++------
2 files changed, 61 insertions(+), 29 deletions(-)
diff --git a/gdb/dwarf2/frame.c b/gdb/dwarf2/frame.c
index 1469621133..e83031d1a5 100644
--- a/gdb/dwarf2/frame.c
+++ b/gdb/dwarf2/frame.c
@@ -235,16 +235,17 @@ register %s (#%d) at %s"),
}
}
-static CORE_ADDR
+static value *
execute_stack_op (const gdb_byte *exp, ULONGEST len, int addr_size,
struct frame_info *this_frame, CORE_ADDR initial,
- int initial_in_stack_memory, dwarf2_per_objfile *per_objfile)
+ int initial_in_stack_memory, dwarf2_per_objfile *per_objfile,
+ struct type* type = nullptr, bool as_lval = true)
{
scoped_value_mark free_values;
- struct type *type = address_type (per_objfile->objfile->arch (),
- addr_size);
+ struct type *init_type = address_type (per_objfile->objfile->arch (),
+ addr_size);
- struct value *init_value = value_at_lazy (type, initial);
+ struct value *init_value = value_at_lazy (init_type, initial);
std::vector<struct value *> init_values;
set_value_stack (init_value, initial_in_stack_memory);
@@ -254,10 +255,15 @@ execute_stack_op (const gdb_byte *exp, ULONGEST len, int addr_size,
= dwarf2_eval_exp (exp, len, true, per_objfile, nullptr,
this_frame, addr_size, &init_values, nullptr);
- if (VALUE_LVAL (result_val) == lval_memory)
- return value_address (result_val);
- else
- return value_as_address (result_val);
+ /* We need to clean up all the values that are not needed any more.
+ The problem with a value_ref_ptr class is that it disconnects the
+ RETVAL from the value garbage collection, so we need to make
+ a copy of that value on the stack to keep everything consistent.
+ The value_ref_ptr will clean up after itself at the end of this block. */
+ value_ref_ptr value_holder = value_ref_ptr::new_reference (result_val);
+ free_values.free_to_mark ();
+
+ return value_copy(result_val);
}
\f
@@ -988,10 +994,14 @@ dwarf2_frame_cache (struct frame_info *this_frame, void **this_cache)
break;
case CFA_EXP:
- cache->cfa =
- execute_stack_op (fs.regs.cfa_exp, fs.regs.cfa_exp_len,
- cache->addr_size, this_frame, 0, 0,
- cache->per_objfile);
+ {
+ struct value *value
+ = execute_stack_op (fs.regs.cfa_exp, fs.regs.cfa_exp_len,
+ cache->addr_size, this_frame, 0, 0,
+ cache->per_objfile);
+ cache->cfa = value_address (value);
+ }
+
break;
default:
@@ -1189,24 +1199,22 @@ dwarf2_frame_prev_register (struct frame_info *this_frame, void **this_cache,
return frame_unwind_got_register (this_frame, regnum, realnum);
case DWARF2_FRAME_REG_SAVED_EXP:
- addr = execute_stack_op (cache->reg[regnum].loc.exp.start,
+ return execute_stack_op (cache->reg[regnum].loc.exp.start,
cache->reg[regnum].loc.exp.len,
- cache->addr_size,
- this_frame, cache->cfa, 1,
- cache->per_objfile);
- return frame_unwind_got_memory (this_frame, regnum, addr);
+ cache->addr_size, this_frame,
+ cache->cfa, 1, cache->per_objfile,
+ register_type (gdbarch, regnum));
case DWARF2_FRAME_REG_SAVED_VAL_OFFSET:
addr = cache->cfa + cache->reg[regnum].loc.offset;
return frame_unwind_got_constant (this_frame, regnum, addr);
case DWARF2_FRAME_REG_SAVED_VAL_EXP:
- addr = execute_stack_op (cache->reg[regnum].loc.exp.start,
+ return execute_stack_op (cache->reg[regnum].loc.exp.start,
cache->reg[regnum].loc.exp.len,
- cache->addr_size,
- this_frame, cache->cfa, 1,
- cache->per_objfile);
- return frame_unwind_got_constant (this_frame, regnum, addr);
+ cache->addr_size, this_frame,
+ cache->cfa, 1, cache->per_objfile,
+ register_type (gdbarch, regnum), false);
case DWARF2_FRAME_REG_UNSPECIFIED:
/* GCC, in its infinite wisdom decided to not provide unwind
diff --git a/gdb/frame.c b/gdb/frame.c
index 4618da6c81..0068bf8455 100644
--- a/gdb/frame.c
+++ b/gdb/frame.c
@@ -1567,25 +1567,49 @@ put_frame_register_bytes (struct frame_info *frame, int regnum,
{
int curr_len = register_size (gdbarch, regnum) - offset;
+ struct value *value = frame_unwind_register_value (frame->next,
+ regnum);
+
if (curr_len > len)
curr_len = len;
- if (curr_len == register_size (gdbarch, regnum))
+ /* Compute value is a special new case. The problem is that
+ the computed callback mechanism only supports a struct
+ value arguments, so we need to make one. */
+ if (value != NULL && VALUE_LVAL (value) == lval_computed)
+ {
+ struct value *from_value;
+ const struct lval_funcs *funcs = value_computed_funcs (value);
+ struct type * reg_type = register_type (gdbarch, regnum);
+
+ if (funcs->write == NULL)
+ error (_("Attempt to assign to an unmodifiable value."));
+
+ from_value = allocate_value (reg_type);
+ memcpy (value_contents_raw (from_value), myaddr,
+ TYPE_LENGTH (reg_type));
+
+ set_value_offset (value, offset);
+
+ funcs->write (value, from_value);
+ release_value (from_value);
+ }
+ else if (curr_len == register_size (gdbarch, regnum))
{
put_frame_register (frame, regnum, myaddr);
}
else
{
- struct value *value = frame_unwind_register_value (frame->next,
- regnum);
gdb_assert (value != NULL);
- memcpy ((char *) value_contents_writeable (value) + offset, myaddr,
- curr_len);
+ memcpy ((char *) value_contents_writeable (value) + offset,
+ myaddr, curr_len);
put_frame_register (frame, regnum, value_contents_raw (value));
- release_value (value);
}
+ if (value != NULL)
+ release_value (value);
+
myaddr += curr_len;
len -= curr_len;
offset = 0;
--
2.17.1
next prev parent reply other threads:[~2020-12-07 19:01 UTC|newest]
Thread overview: 75+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-12-07 19:00 [PATCH 00/30] Allow location description on the DWARF stack Zoran Zaric
2020-12-07 19:00 ` [PATCH 01/30] Replace the symbol needs evaluator with a parser Zoran Zaric
2021-01-21 21:16 ` Tom Tromey
2021-01-21 21:48 ` Zoran Zaric
2021-02-23 14:15 ` Zoran Zaric
2020-12-07 19:00 ` [PATCH 02/30] Move frame context info to dwarf_expr_context Zoran Zaric
2021-01-21 21:23 ` Tom Tromey
2020-12-07 19:00 ` [PATCH 03/30] Remove get_frame_cfa from dwarf_expr_context Zoran Zaric
2021-01-21 21:23 ` Tom Tromey
2020-12-07 19:00 ` [PATCH 04/30] Move compilation unit info to dwarf_expr_context Zoran Zaric
2021-01-21 21:28 ` Tom Tromey
2021-02-23 14:21 ` Zoran Zaric
2020-12-07 19:00 ` [PATCH 05/30] Move dwarf_call " Zoran Zaric
2021-01-21 21:30 ` Tom Tromey
2020-12-07 19:00 ` [PATCH 06/30] Move get_object_address " Zoran Zaric
2021-01-21 21:31 ` Tom Tromey
2021-02-23 14:33 ` Zoran Zaric
2020-12-07 19:00 ` [PATCH 07/30] Move read_mem " Zoran Zaric
2021-01-21 21:34 ` Tom Tromey
2020-12-07 19:00 ` [PATCH 08/30] Move push_dwarf_reg_entry_value to expr.c Zoran Zaric
2021-01-21 21:35 ` Tom Tromey
2020-12-07 19:00 ` [PATCH 09/30] Inline get_reg_value method of dwarf_expr_context Zoran Zaric
2021-01-21 21:36 ` Tom Tromey
2020-12-07 19:00 ` [PATCH 10/30] Remove empty frame and full evaluators Zoran Zaric
2021-01-21 21:37 ` Tom Tromey
2020-12-07 19:00 ` [PATCH 11/30] Merge evaluate_for_locexpr_baton evaluator Zoran Zaric
2021-02-08 21:21 ` Tom Tromey
2020-12-07 19:00 ` [PATCH 12/30] Move piece_closure and its support to expr.c Zoran Zaric
2021-02-08 21:32 ` Tom Tromey
2021-02-09 14:53 ` Zoran Zaric
2020-12-07 19:00 ` [PATCH 13/30] Make DWARF evaluator return a single struct value Zoran Zaric
2021-02-08 21:35 ` Tom Tromey
2021-02-09 14:55 ` Zoran Zaric
2021-02-09 17:13 ` Tom Tromey
2020-12-07 19:00 ` [PATCH 14/30] Simplify dwarf_expr_context class interface Zoran Zaric
2021-02-08 21:38 ` Tom Tromey
2021-02-09 14:56 ` Zoran Zaric
2021-02-23 14:38 ` Zoran Zaric
2020-12-07 19:00 ` [PATCH 15/30] Add as_lval argument to expression evaluator Zoran Zaric
2021-02-08 21:41 ` Tom Tromey
2021-02-09 15:25 ` Zoran Zaric
2021-02-09 20:33 ` Tom Tromey
2020-12-07 19:00 ` [PATCH 16/30] Add new register access interface to expr.c Zoran Zaric
2021-02-09 19:37 ` Tom Tromey
2020-12-07 19:00 ` [PATCH 17/30] Add new memory " Zoran Zaric
2021-02-09 19:45 ` Tom Tromey
2021-02-23 15:35 ` Zoran Zaric
2020-12-07 19:00 ` [PATCH 18/30] Add new classes that model DWARF stack element Zoran Zaric
2021-02-08 21:54 ` Tom Tromey
2021-02-09 17:34 ` Zoran Zaric
2021-02-09 20:36 ` Tom Tromey
2021-02-09 21:07 ` Tom Tromey
2021-02-09 21:26 ` Zoran Zaric
2021-02-23 14:57 ` Zoran Zaric
2020-12-07 19:00 ` [PATCH 19/30] Add new location description access interface Zoran Zaric
2021-02-08 21:46 ` Tom Tromey
2021-02-09 16:00 ` Zoran Zaric
2021-02-09 17:30 ` Zoran Zaric
2021-02-23 14:49 ` Zoran Zaric
2020-12-07 19:00 ` [PATCH 20/30] Add dwarf_entry factory class to expr.c Zoran Zaric
2021-02-09 19:54 ` Tom Tromey
2020-12-07 19:00 ` [PATCH 21/30] Change DWARF stack to use new dwarf_entry classes Zoran Zaric
2021-02-09 20:11 ` Tom Tromey
2020-12-07 19:00 ` [PATCH 22/30] Remove dwarf_expr_context from expr.h interface Zoran Zaric
2020-12-07 19:00 ` [PATCH 23/30] Rename and update the piece_closure structure Zoran Zaric
2020-12-07 19:00 ` [PATCH 24/30] Move read_addr_from_reg function to frame.c Zoran Zaric
2020-12-07 19:00 ` [PATCH 25/30] Add frame info check to DW_OP_reg operations Zoran Zaric
2020-12-07 19:00 ` [PATCH 26/30] Remove DWARF expression composition check Zoran Zaric
2020-12-07 19:00 ` Zoran Zaric [this message]
2020-12-07 19:00 ` [PATCH 28/30] Add DWARF operations for byte and bit offset Zoran Zaric
2020-12-07 19:00 ` [PATCH 29/30] Add support for DW_OP_LLVM_undefined operation Zoran Zaric
2020-12-07 19:00 ` [PATCH 30/30] Add support for nested composite locations Zoran Zaric
2020-12-08 14:48 ` [PATCH 00/30] Allow location description on the DWARF stack Metzger, Markus T
2020-12-08 16:17 ` Simon Marchi
2020-12-09 0:30 ` Tye, Tony
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=20201207190031.13341-28-Zoran.Zaric@amd.com \
--to=zoran.zaric@amd.com \
--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).