[2/5] Change linetables to be objfile-independent

Message ID 20230308-submit-constify-linetable-v1-2-ca4057478141@tromey.com
State New
Headers
Series Make line tables independent of objfile |

Commit Message

Tom Tromey March 8, 2023, 3:42 p.m. UTC
  This changes linetables to not add the text offset to the addresses
they contain.  I did this in a few steps, necessarily combined
together in one patch: I renamed the 'pc' member to 'm_pc', added the
appropriate accessors, and then recompiled.  Then I fixed all the
errors.  Where possible I generally chose to use the raw_pc accessor,
as it is less expensive.

Note that this patch discounts the possibility that the text section
offset might cause wraparound in the addresses in the line table.
However, this was already discounted -- in particular,
objfile_relocate1 did not re-sort the table in this scenario.  (There
was a bug open about this, but as far as I can tell this has never
happened, it's not even clear what inspired that bug.)
---
 gdb/buildsym.c            |  4 +--
 gdb/coffread.c            |  1 -
 gdb/dbxread.c             |  9 ++++---
 gdb/disasm.c              | 27 +++++++++++++-------
 gdb/dwarf2/read.c         |  5 ++--
 gdb/jit.c                 |  2 +-
 gdb/mdebugread.c          |  9 +++----
 gdb/mi/mi-symbol-cmds.c   |  6 +++--
 gdb/objfiles.c            | 16 ------------
 gdb/python/py-linetable.c |  3 ++-
 gdb/record-btrace.c       |  6 ++++-
 gdb/symmisc.c             |  4 +--
 gdb/symtab.c              | 64 +++++++++++++++++++++++++++++++----------------
 gdb/symtab.h              | 19 +++++++++++---
 gdb/xcoffread.c           | 11 ++++----
 15 files changed, 110 insertions(+), 76 deletions(-)
  

Comments

Simon Marchi March 11, 2023, 3:48 a.m. UTC | #1
> diff --git a/gdb/objfiles.c b/gdb/objfiles.c
> index 8e7be8e2e25..2c8f5d14444 100644
> --- a/gdb/objfiles.c
> +++ b/gdb/objfiles.c
> @@ -616,22 +616,6 @@ objfile_relocate1 (struct objfile *objfile,
>  
>    /* OK, get all the symtabs.  */
>    {
> -    for (compunit_symtab *cust : objfile->compunits ())
> -      {
> -	for (symtab *s : cust->filetabs ())
> -	  {
> -	    struct linetable *l;
> -
> -	    /* First the line table.  */
> -	    l = s->linetable ();
> -	    if (l)
> -	      {
> -		for (int i = 0; i < l->nitems; ++i)
> -		  l->item[i].pc += delta[SECT_OFF_TEXT (objfile)];
> -	      }
> -	  }
> -      }
> -
It's nice to see the objfile_relocate1 function shrinking, it's like a
progress meter of your project :).

> diff --git a/gdb/symmisc.c b/gdb/symmisc.c
> index 6ff06ccaa93..d1458784e5c 100644
> --- a/gdb/symmisc.c
> +++ b/gdb/symmisc.c
> @@ -263,7 +263,7 @@ dump_symtab_1 (struct symtab *symtab, struct ui_file *outfile)
>        for (int i = 0; i < len; i++)
>  	{
>  	  gdb_printf (outfile, " line %d at ", l->item[i].line);
> -	  gdb_puts (paddress (gdbarch, l->item[i].pc), outfile);
> +	  gdb_puts (paddress (gdbarch, l->item[i].raw_pc ()), outfile);

Here, do we want to see the raw of the relocated PC?  Maybe both, if
different?  The one you want depends if you are trying to match this
information against addresses from the ELF file (unrelocated) or against
addresses from the inferior (relocated).

> diff --git a/gdb/symtab.h b/gdb/symtab.h
> index 69f0eaa0f88..84d12617f31 100644
> --- a/gdb/symtab.h
> +++ b/gdb/symtab.h
> @@ -1547,18 +1547,29 @@ struct rust_vtable_symbol : public symbol
>  
>  struct linetable_entry
>  {
> +  /* Set the (unrelocated) PC for this entry.  */
> +  void set_pc (CORE_ADDR pc)
> +  { m_pc = pc; }

For symmetry with the getter, should this one be named "set_raw_pc"?

Simon
  
Tom Tromey March 11, 2023, 3:36 p.m. UTC | #2
Simon> It's nice to see the objfile_relocate1 function shrinking, it's
Simon> like a progress meter of your project :).

Yeah... though this one is a refresh of some ancient patches.
The other remaining parts are super hard, I've made several failed
attempts already :(

>> gdb_printf (outfile, " line %d at ", l->item[i].line);
>> -	  gdb_puts (paddress (gdbarch, l->item[i].pc), outfile);
>> +	  gdb_puts (paddress (gdbarch, l->item[i].raw_pc ()), outfile);

Simon> Here, do we want to see the raw of the relocated PC?  Maybe both, if
Simon> different?  The one you want depends if you are trying to match this
Simon> information against addresses from the ELF file (unrelocated) or against
Simon> addresses from the inferior (relocated).

Now that you mention it, I see that minsym printing relocates the
address, so I'll change this one to match.  That way it will also match
blocks.

>> +  /* Set the (unrelocated) PC for this entry.  */
>> +  void set_pc (CORE_ADDR pc)
>> +  { m_pc = pc; }

Simon> For symmetry with the getter, should this one be named "set_raw_pc"?

I'll change it.

Tom
  
Simon Marchi March 11, 2023, 7:13 p.m. UTC | #3
On 3/11/23 10:36, Tom Tromey wrote:
> Simon> It's nice to see the objfile_relocate1 function shrinking, it's
> Simon> like a progress meter of your project :).
> 
> Yeah... though this one is a refresh of some ancient patches.
> The other remaining parts are super hard, I've made several failed
> attempts already :(
> 
>>> gdb_printf (outfile, " line %d at ", l->item[i].line);
>>> -	  gdb_puts (paddress (gdbarch, l->item[i].pc), outfile);
>>> +	  gdb_puts (paddress (gdbarch, l->item[i].raw_pc ()), outfile);
> 
> Simon> Here, do we want to see the raw of the relocated PC?  Maybe both, if
> Simon> different?  The one you want depends if you are trying to match this
> Simon> information against addresses from the ELF file (unrelocated) or against
> Simon> addresses from the inferior (relocated).
> 
> Now that you mention it, I see that minsym printing relocates the
> address, so I'll change this one to match.  That way it will also match
> blocks.
> 
>>> +  /* Set the (unrelocated) PC for this entry.  */
>>> +  void set_pc (CORE_ADDR pc)
>>> +  { m_pc = pc; }
> 
> Simon> For symmetry with the getter, should this one be named "set_raw_pc"?
> 
> I'll change it.

Hi Tom,

I see these regressions:

FAIL: gdb.base/async.exp: stepi&
FAIL: gdb.base/async.exp: nexti&
FAIL: gdb.base/async.exp: finish&
FAIL: gdb.base/consecutive.exp: stopped at bp, 2nd instr (missing hex prefix)
FAIL: gdb.base/disasm-optim.exp: Disassemble main with source (pattern 2)
FAIL: gdb.dwarf2/dw2-disasm-over-non-stmt.exp: opt=s: disassemble /s main
FAIL: gdb.dwarf2/dw2-disasm-over-non-stmt.exp: opt=s: check_disassembly_results 16 line_label_1 (no disassembly for 16)
FAIL: gdb.dwarf2/dw2-disasm-over-non-stmt.exp: opt=s: check_disassembly_results 16 line_label_2 (no disassembly for 16)
FAIL: gdb.dwarf2/dw2-disasm-over-non-stmt.exp: opt=s: check_disassembly_results 17 line_label_3 (no disassembly for 17)
FAIL: gdb.dwarf2/dw2-disasm-over-non-stmt.exp: opt=s: check_disassembly_results 17 line_label_4 (no disassembly for 17)
FAIL: gdb.dwarf2/dw2-disasm-over-non-stmt.exp: opt=s: check_disassembly_results 17 line_label_5 (no disassembly for 17)
FAIL: gdb.dwarf2/dw2-disasm-over-non-stmt.exp: opt=s: check_disassembly_results 18 line_label_6 (no disassembly for 18)
FAIL: gdb.reverse/consecutive-precsave.exp: stopped at bp, 2nd instr (missing hex prefix)
FAIL: gdb.reverse/consecutive-reverse.exp: stopped at bp, 2nd instr (missing hex prefix)

Simon
  
Tom Tromey March 12, 2023, 3:24 a.m. UTC | #4
Simon> FAIL: gdb.base/async.exp: stepi&
Simon> FAIL: gdb.base/async.exp: nexti&
Simon> FAIL: gdb.base/async.exp: finish&
Simon> FAIL: gdb.base/consecutive.exp: stopped at bp, 2nd instr (missing hex prefix)
Simon> FAIL: gdb.base/disasm-optim.exp: Disassemble main with source (pattern 2)
Simon> FAIL: gdb.dwarf2/dw2-disasm-over-non-stmt.exp: opt=s: disassemble /s main
Simon> FAIL: gdb.dwarf2/dw2-disasm-over-non-stmt.exp: opt=s: check_disassembly_results 16 line_label_1 (no disassembly for 16)
Simon> FAIL: gdb.dwarf2/dw2-disasm-over-non-stmt.exp: opt=s: check_disassembly_results 16 line_label_2 (no disassembly for 16)
Simon> FAIL: gdb.dwarf2/dw2-disasm-over-non-stmt.exp: opt=s: check_disassembly_results 17 line_label_3 (no disassembly for 17)
Simon> FAIL: gdb.dwarf2/dw2-disasm-over-non-stmt.exp: opt=s: check_disassembly_results 17 line_label_4 (no disassembly for 17)
Simon> FAIL: gdb.dwarf2/dw2-disasm-over-non-stmt.exp: opt=s: check_disassembly_results 17 line_label_5 (no disassembly for 17)
Simon> FAIL: gdb.dwarf2/dw2-disasm-over-non-stmt.exp: opt=s: check_disassembly_results 18 line_label_6 (no disassembly for 18)
Simon> FAIL: gdb.reverse/consecutive-precsave.exp: stopped at bp, 2nd instr (missing hex prefix)
Simon> FAIL: gdb.reverse/consecutive-reverse.exp: stopped at bp, 2nd instr (missing hex prefix)

I tried these with git master and they all work fine for me.  I wonder
what the difference could be.  Could you run one of them perhaps & send
me the gdb.log?  Maybe that would give a clue.

Tom
  
Simon Marchi March 12, 2023, 1:40 p.m. UTC | #5
On 3/11/23 22:24, Tom Tromey wrote:
> Simon> FAIL: gdb.base/async.exp: stepi&
> Simon> FAIL: gdb.base/async.exp: nexti&
> Simon> FAIL: gdb.base/async.exp: finish&
> Simon> FAIL: gdb.base/consecutive.exp: stopped at bp, 2nd instr (missing hex prefix)
> Simon> FAIL: gdb.base/disasm-optim.exp: Disassemble main with source (pattern 2)
> Simon> FAIL: gdb.dwarf2/dw2-disasm-over-non-stmt.exp: opt=s: disassemble /s main
> Simon> FAIL: gdb.dwarf2/dw2-disasm-over-non-stmt.exp: opt=s: check_disassembly_results 16 line_label_1 (no disassembly for 16)
> Simon> FAIL: gdb.dwarf2/dw2-disasm-over-non-stmt.exp: opt=s: check_disassembly_results 16 line_label_2 (no disassembly for 16)
> Simon> FAIL: gdb.dwarf2/dw2-disasm-over-non-stmt.exp: opt=s: check_disassembly_results 17 line_label_3 (no disassembly for 17)
> Simon> FAIL: gdb.dwarf2/dw2-disasm-over-non-stmt.exp: opt=s: check_disassembly_results 17 line_label_4 (no disassembly for 17)
> Simon> FAIL: gdb.dwarf2/dw2-disasm-over-non-stmt.exp: opt=s: check_disassembly_results 17 line_label_5 (no disassembly for 17)
> Simon> FAIL: gdb.dwarf2/dw2-disasm-over-non-stmt.exp: opt=s: check_disassembly_results 18 line_label_6 (no disassembly for 18)
> Simon> FAIL: gdb.reverse/consecutive-precsave.exp: stopped at bp, 2nd instr (missing hex prefix)
> Simon> FAIL: gdb.reverse/consecutive-reverse.exp: stopped at bp, 2nd instr (missing hex prefix)
> 
> I tried these with git master and they all work fine for me.  I wonder
> what the difference could be.  Could you run one of them perhaps & send
> me the gdb.log?  Maybe that would give a clue.

This for instance:


 68 disassemble /s main^M
 69 Dump of assembler code for function main:^M
 70 gdb_expect_list pattern: /disasm-optim\.c:^M
 71 24/
 72 Cannot access memory at address 0x1130^M
 73 (gdb) FAIL: gdb.base/disasm-optim.exp: Disassemble main with source (pattern 2)

Looks like we're trying to read from the inferior at an unrelocated
address.  From some previous bug, I recall that you were using a system
with no-PIE as a default, so that could be why you don't see it.

Simon
  
Tom Tromey March 16, 2023, 2:02 p.m. UTC | #6
>>>>> "Simon" == Simon Marchi via Gdb-patches <gdb-patches@sourceware.org> writes:

Simon> Looks like we're trying to read from the inferior at an unrelocated
Simon> address.  From some previous bug, I recall that you were using a system
Simon> with no-PIE as a default, so that could be why you don't see it.

I tried building this executable with -fPIE and it still seems to work
for me :(

Tom
  
Simon Marchi March 16, 2023, 2:45 p.m. UTC | #7
On 3/16/23 10:02, Tom Tromey wrote:
>>>>>> "Simon" == Simon Marchi via Gdb-patches <gdb-patches@sourceware.org> writes:
> 
> Simon> Looks like we're trying to read from the inferior at an unrelocated
> Simon> address.  From some previous bug, I recall that you were using a system
> Simon> with no-PIE as a default, so that could be why you don't see it.
> 
> I tried building this executable with -fPIE and it still seems to work
> for me :(

Ok, I'll look into it then.

I can reproduce it like hand like this:

    $ gcc -fno-stack-protector -c \
        -o /home/smarchi/build/binutils-gdb/gdb/testsuite/outputs/gdb.base/disasm-optim/disasm-optim0.o \
        /home/smarchi/src/binutils-gdb/gdb/testsuite/gdb.base/disasm-optim.S \
        -fPIE
    $ gcc /home/smarchi/build/binutils-gdb/gdb/testsuite/outputs/gdb.base/disasm-optim/disasm-optim0.o \
        -pie -o disasm-optim

    $ gdb ./gdb -nx -q --data-directory=data-directory disasm-optim
    Reading symbols from disasm-optim...
    (gdb) start

    This GDB supports auto-downloading debuginfo from the following URLs:
      <https://debuginfod.archlinux.org>
    Enable debuginfod for this session? (y or [n]) n
    Debuginfod has been disabled.
    To make this setting permanent, add 'set debuginfod enabled off' to .gdbinit.
    Temporary breakpoint 1 at 0x1120: file disasm-optim.c, line 25.
    Starting program: /home/smarchi/build/binutils-gdb/gdb/disasm-optim
    [Thread debugging using libthread_db enabled]
    Using host libthread_db library "/usr/lib/../lib/libthread_db.so.1".

    Temporary breakpoint 1, main () at disasm-optim.c:25
    25      disasm-optim.c: No such file or directory.
    (gdb) disassemble main
    Dump of assembler code for function main:
    => 0x0000555555555120 <+0>:     mov    0x2ef2(%rip),%eax        # 0x555555558018 <y>
       0x0000555555555126 <+6>:     test   %eax,%eax
       0x0000555555555128 <+8>:     js     0x555555555140 <main+32>
       0x000055555555512a <+10>:    lea    0xa(%rax),%edx
       0x000055555555512d <+13>:    test   %eax,%eax
       0x000055555555512f <+15>:    mov    $0x1,%eax
       0x0000555555555134 <+20>:    cmovne %edx,%eax
       0x0000555555555137 <+23>:    mov    %eax,0x2ed7(%rip)        # 0x555555558014 <x>
       0x000055555555513d <+29>:    xor    %eax,%eax
       0x000055555555513f <+31>:    ret
       0x0000555555555140 <+32>:    add    %eax,%eax
       0x0000555555555142 <+34>:    jmp    0x555555555137 <main+23>
    End of assembler dump.
    (gdb) disassemble /s main
    Dump of assembler code for function main:
    Cannot access memory at address 0x1120

If I link with -no-pie, then it works.

Simon
  
Tom Tromey March 16, 2023, 11:43 p.m. UTC | #8
Simon> I can reproduce it like hand like this:
[...]

I think I have a fix for it.  I'll send it tomorrow.

Tom
  
Simon Marchi March 17, 2023, 2:45 a.m. UTC | #9
On 3/16/23 19:43, Tom Tromey wrote:
> Simon> I can reproduce it like hand like this:
> [...]
> 
> I think I have a fix for it.  I'll send it tomorrow.
> 
> Tom

Ok, I worked on that today and also have some patches.  One fixes
do_mixed_source_and_assembly, and the other makes "maint info
line-table" show relocated addresses again, it fixes a test that relied
on that (I guess we could also change the test).

If you want to take a peek:

https://review.lttng.org/c/binutils-gdb/+/9665

Simon
  
Tom Tromey March 17, 2023, 1:09 p.m. UTC | #10
Simon> https://review.lttng.org/c/binutils-gdb/+/9665

Lots of patches there!

#1 Renaming is fine but the "_raw" name was chosen to match psymtabs.
psymtabs are essentially obsolete now (depending on the state of the
stabs removal thread) but as long as they are around the naming should
match.

#2 Making a new header seems great to me.

#3 Ditto

#4 For this one I think my fix is better, it introduces a new type and
prevents this sort of bug.

#5 Looks good

Tom
  
Tom Tromey March 17, 2023, 10:21 p.m. UTC | #11
Tom> #1 Renaming is fine but the "_raw" name was chosen to match psymtabs.

Well, I have this partly wrong, sorry about that.

psymbols themselves use the other name:

  CORE_ADDR unrelocated_address () const

but psymtabs use:

  CORE_ADDR raw_text_low () const
  CORE_ADDR raw_text_high () const

and so do minsyms:

  CORE_ADDR value_raw_address () const

I think one set or the other should be renamed.  Which one do you
prefer?  I could go either way.  Anyway let me know and I'll rename
stuff.

Tom
  
Simon Marchi March 20, 2023, 2:57 p.m. UTC | #12
On 3/17/23 18:21, Tom Tromey wrote:
> Tom> #1 Renaming is fine but the "_raw" name was chosen to match psymtabs.
> 
> Well, I have this partly wrong, sorry about that.
> 
> psymbols themselves use the other name:
> 
>   CORE_ADDR unrelocated_address () const
> 
> but psymtabs use:
> 
>   CORE_ADDR raw_text_low () const
>   CORE_ADDR raw_text_high () const
> 
> and so do minsyms:
> 
>   CORE_ADDR value_raw_address () const
> 
> I think one set or the other should be renamed.  Which one do you
> prefer?  I could go either way.  Anyway let me know and I'll rename
> stuff.

I prefer "unrelocated", since it's a bit more specific and precise than
"raw". "raw" suggests that it's a value before some kind of processing,
but it doesn't tell which processing that it.  With "unrelocated", you
know it's before relocation.

Simon
  

Patch

diff --git a/gdb/buildsym.c b/gdb/buildsym.c
index 6594da68131..33c51beed57 100644
--- a/gdb/buildsym.c
+++ b/gdb/buildsym.c
@@ -655,7 +655,7 @@  buildsym_compunit::record_line (struct subfile *subfile, int line,
 	  linetable_entry *last = &subfile->line_vector_entries.back ();
 	  last_line = last->line;
 
-	  if (last->pc != pc)
+	  if (last->raw_pc () != pc)
 	    break;
 
 	  subfile->line_vector_entries.pop_back ();
@@ -670,7 +670,7 @@  buildsym_compunit::record_line (struct subfile *subfile, int line,
   linetable_entry &e = subfile->line_vector_entries.back ();
   e.line = line;
   e.is_stmt = (flags & LEF_IS_STMT) != 0;
-  e.pc = pc;
+  e.set_pc (pc);
   e.prologue_end = (flags & LEF_PROLOGUE_END) != 0;
 }
 
diff --git a/gdb/coffread.c b/gdb/coffread.c
index 45542bc2432..81df2579619 100644
--- a/gdb/coffread.c
+++ b/gdb/coffread.c
@@ -1463,7 +1463,6 @@  enter_linenos (file_ptr file_offset, int first_line,
       if (L_LNNO32 (&lptr) && L_LNNO32 (&lptr) <= last_line)
 	{
 	  CORE_ADDR addr = lptr.l_addr.l_paddr;
-	  addr += objfile->text_section_offset ();
 	  record_line (get_current_subfile (),
 		       first_line + L_LNNO32 (&lptr),
 		       gdbarch_addr_bits_remove (gdbarch, addr));
diff --git a/gdb/dbxread.c b/gdb/dbxread.c
index 02c7e3118ad..1e88121f11c 100644
--- a/gdb/dbxread.c
+++ b/gdb/dbxread.c
@@ -2454,7 +2454,8 @@  process_one_symbol (int type, int desc, CORE_ADDR valu, const char *name,
 	      CORE_ADDR addr = last_function_start + valu;
 
 	      record_line (get_current_subfile (), 0,
-			   gdbarch_addr_bits_remove (gdbarch, addr));
+			   gdbarch_addr_bits_remove (gdbarch, addr)
+			   - objfile->text_section_offset ());
 	    }
 
 	  within_function = 0;
@@ -2662,12 +2663,14 @@  process_one_symbol (int type, int desc, CORE_ADDR valu, const char *name,
 			   last_function_start : valu;
 
 	  record_line (get_current_subfile (), desc,
-		       gdbarch_addr_bits_remove (gdbarch, addr));
+		       gdbarch_addr_bits_remove (gdbarch, addr)
+		       - objfile->text_section_offset ());
 	  sline_found_in_function = 1;
 	}
       else
 	record_line (get_current_subfile (), desc,
-		     gdbarch_addr_bits_remove (gdbarch, valu));
+		     gdbarch_addr_bits_remove (gdbarch, valu)
+		     - objfile->text_section_offset ());
       break;
 
     case N_BCOMM:
diff --git a/gdb/disasm.c b/gdb/disasm.c
index 49053dcfbad..2acde04ffe9 100644
--- a/gdb/disasm.c
+++ b/gdb/disasm.c
@@ -32,6 +32,7 @@ 
 #include "gdbsupport/gdb_optional.h"
 #include "valprint.h"
 #include "cli/cli-style.h"
+#include "objfiles.h"
 
 /* Disassemble functions.
    FIXME: We should get rid of all the duplicate code in gdb that does
@@ -592,17 +593,21 @@  do_mixed_source_and_assembly_deprecated
   mle = (struct deprecated_dis_line_entry *)
     alloca (nlines * sizeof (struct deprecated_dis_line_entry));
 
+  struct objfile *objfile = symtab->compunit ()->objfile ();
+  low -= objfile->text_section_offset ();
+  high -= objfile->text_section_offset ();
+
   /* Copy linetable entries for this function into our data
      structure, creating end_pc's and setting out_of_order as
      appropriate.  */
 
   /* First, skip all the preceding functions.  */
 
-  for (i = 0; i < nlines - 1 && le[i].pc < low; i++);
+  for (i = 0; i < nlines - 1 && le[i].raw_pc () < low; i++);
 
   /* Now, copy all entries before the end of this function.  */
 
-  for (; i < nlines - 1 && le[i].pc < high; i++)
+  for (; i < nlines - 1 && le[i].raw_pc () < high; i++)
     {
       if (le[i] == le[i + 1])
 	continue;		/* Ignore duplicates.  */
@@ -614,19 +619,19 @@  do_mixed_source_and_assembly_deprecated
       mle[newlines].line = le[i].line;
       if (le[i].line > le[i + 1].line)
 	out_of_order = 1;
-      mle[newlines].start_pc = le[i].pc;
-      mle[newlines].end_pc = le[i + 1].pc;
+      mle[newlines].start_pc = le[i].pc (objfile);
+      mle[newlines].end_pc = le[i + 1].pc (objfile);
       newlines++;
     }
 
   /* If we're on the last line, and it's part of the function,
      then we need to get the end pc in a special way.  */
 
-  if (i == nlines - 1 && le[i].pc < high)
+  if (i == nlines - 1 && le[i].raw_pc () < high)
     {
       mle[newlines].line = le[i].line;
-      mle[newlines].start_pc = le[i].pc;
-      sal = find_pc_line (le[i].pc, 0);
+      mle[newlines].start_pc = le[i].pc (objfile);
+      sal = find_pc_line (le[i].pc (objfile), 0);
       mle[newlines].end_pc = sal.end;
       newlines++;
     }
@@ -733,6 +738,10 @@  do_mixed_source_and_assembly (struct gdbarch *gdbarch,
 
   htab_up dis_line_table (allocate_dis_line_table ());
 
+  struct objfile *objfile = main_symtab->compunit ()->objfile ();
+  low -= objfile->text_section_offset ();
+  high -= objfile->text_section_offset ();
+
   pc = low;
 
   /* The prologue may be empty, but there may still be a line number entry
@@ -746,10 +755,10 @@  do_mixed_source_and_assembly (struct gdbarch *gdbarch,
   first_le = NULL;
 
   /* Skip all the preceding functions.  */
-  for (i = 0; i < nlines && le[i].pc < low; i++)
+  for (i = 0; i < nlines && le[i].raw_pc () < low; i++)
     continue;
 
-  if (i < nlines && le[i].pc < high)
+  if (i < nlines && le[i].raw_pc () < high)
     first_le = &le[i];
 
   /* Add lines for every pc value.  */
diff --git a/gdb/dwarf2/read.c b/gdb/dwarf2/read.c
index a9758c305b3..e1e2b4a805d 100644
--- a/gdb/dwarf2/read.c
+++ b/gdb/dwarf2/read.c
@@ -17980,10 +17980,9 @@  class lnp_state_machine
   }
 
   /* Handle DW_LNE_set_address.  */
-  void handle_set_address (CORE_ADDR baseaddr, CORE_ADDR address)
+  void handle_set_address (CORE_ADDR address)
   {
     m_op_index = 0;
-    address += baseaddr;
     m_address = gdbarch_adjust_dwarf2_line (m_gdbarch, address, false);
   }
 
@@ -18458,7 +18457,7 @@  dwarf_decode_lines_1 (struct line_header *lh, struct dwarf2_cu *cu,
 
 		    state_machine.check_line_address (cu, line_ptr,
 						      lowpc - baseaddr, address);
-		    state_machine.handle_set_address (baseaddr, address);
+		    state_machine.handle_set_address (address);
 		  }
 		  break;
 		case DW_LNE_define_file:
diff --git a/gdb/jit.c b/gdb/jit.c
index 7f4b9f0fd37..f9f2c92553c 100644
--- a/gdb/jit.c
+++ b/gdb/jit.c
@@ -491,7 +491,7 @@  jit_symtab_line_mapping_add_impl (struct gdb_symbol_callbacks *cb,
   stab->linetable->nitems = nlines;
   for (i = 0; i < nlines; i++)
     {
-      stab->linetable->item[i].pc = (CORE_ADDR) map[i].pc;
+      stab->linetable->item[i].set_pc ((CORE_ADDR) map[i].pc);
       stab->linetable->item[i].line = map[i].line;
       stab->linetable->item[i].is_stmt = 1;
     }
diff --git a/gdb/mdebugread.c b/gdb/mdebugread.c
index 793795c903a..85dceb54fe7 100644
--- a/gdb/mdebugread.c
+++ b/gdb/mdebugread.c
@@ -2152,7 +2152,7 @@  parse_external (EXTR *es, int bigend, const section_offsets &section_offsets,
 
 static void
 parse_lines (FDR *fh, PDR *pr, struct linetable *lt, int maxlines,
-	     CORE_ADDR textlow, CORE_ADDR lowest_pdr_addr)
+	     CORE_ADDR lowest_pdr_addr)
 {
   unsigned char *base;
   int j, k;
@@ -2183,7 +2183,7 @@  parse_lines (FDR *fh, PDR *pr, struct linetable *lt, int maxlines,
 	halt = base + fh->cbLine;
       base += pr->cbLineOffset;
 
-      adr = textlow + pr->adr - lowest_pdr_addr;
+      adr = pr->adr - lowest_pdr_addr;
 
       l = adr >> 2;		/* in words */
       for (lineno = pr->lnLow; base < halt;)
@@ -4011,7 +4011,6 @@  mdebug_expand_psymtab (legacy_psymtab *pst, struct objfile *objfile)
 	      else
 		{
 		  /* Handle encoded stab line number.  */
-		  valu += section_offsets[SECT_OFF_TEXT (objfile)];
 		  record_line (get_current_subfile (), sh.index,
 			       gdbarch_addr_bits_remove (gdbarch, valu));
 		}
@@ -4163,7 +4162,7 @@  mdebug_expand_psymtab (legacy_psymtab *pst, struct objfile *objfile)
 		}
 
 	      parse_lines (fh, pr_block.data (), lines, maxlines,
-			   pst->text_low (objfile), lowest_pdr_addr);
+			   lowest_pdr_addr);
 	      if (lines->nitems < fh->cline)
 		lines = shrink_linetable (lines);
 
@@ -4540,7 +4539,7 @@  add_line (struct linetable *lt, int lineno, CORE_ADDR adr, int last)
     return lineno;
 
   lt->item[lt->nitems].line = lineno;
-  lt->item[lt->nitems++].pc = adr << 2;
+  lt->item[lt->nitems++].set_pc (adr << 2);
   return lineno;
 }
 
diff --git a/gdb/mi/mi-symbol-cmds.c b/gdb/mi/mi-symbol-cmds.c
index 16947dad51b..08e6e15e14d 100644
--- a/gdb/mi/mi-symbol-cmds.c
+++ b/gdb/mi/mi-symbol-cmds.c
@@ -50,14 +50,16 @@  mi_cmd_symbol_list_lines (const char *command, char **argv, int argc)
      already sorted by increasing values in the symbol table, so no
      need to perform any other sorting.  */
 
-  gdbarch = s->compunit ()->objfile ()->arch ();
+  struct objfile *objfile = s->compunit ()->objfile ();
+  gdbarch = objfile->arch ();
 
   ui_out_emit_list list_emitter (uiout, "lines");
   if (s->linetable () != NULL && s->linetable ()->nitems > 0)
     for (i = 0; i < s->linetable ()->nitems; i++)
       {
 	ui_out_emit_tuple tuple_emitter (uiout, NULL);
-	uiout->field_core_addr ("pc", gdbarch, s->linetable ()->item[i].pc);
+	uiout->field_core_addr ("pc", gdbarch,
+				s->linetable ()->item[i].pc (objfile));
 	uiout->field_signed ("line", s->linetable ()->item[i].line);
       }
 }
diff --git a/gdb/objfiles.c b/gdb/objfiles.c
index 8e7be8e2e25..2c8f5d14444 100644
--- a/gdb/objfiles.c
+++ b/gdb/objfiles.c
@@ -616,22 +616,6 @@  objfile_relocate1 (struct objfile *objfile,
 
   /* OK, get all the symtabs.  */
   {
-    for (compunit_symtab *cust : objfile->compunits ())
-      {
-	for (symtab *s : cust->filetabs ())
-	  {
-	    struct linetable *l;
-
-	    /* First the line table.  */
-	    l = s->linetable ();
-	    if (l)
-	      {
-		for (int i = 0; i < l->nitems; ++i)
-		  l->item[i].pc += delta[SECT_OFF_TEXT (objfile)];
-	      }
-	  }
-      }
-
     for (compunit_symtab *cust : objfile->compunits ())
       {
 	struct blockvector *bv = cust->blockvector ();
diff --git a/gdb/python/py-linetable.c b/gdb/python/py-linetable.c
index 7e5a6536fde..e42bcc2c792 100644
--- a/gdb/python/py-linetable.c
+++ b/gdb/python/py-linetable.c
@@ -422,7 +422,8 @@  ltpy_iternext (PyObject *self)
       item = &(symtab->linetable ()->item[iter_obj->current_index]);
     }
 
-  obj = build_linetable_entry (item->line, item->pc);
+  struct objfile *objfile = symtab->compunit ()->objfile ();
+  obj = build_linetable_entry (item->line, item->pc (objfile));
   iter_obj->current_index++;
 
   return obj;
diff --git a/gdb/record-btrace.c b/gdb/record-btrace.c
index 61de8491bb9..55fe25e84ae 100644
--- a/gdb/record-btrace.c
+++ b/gdb/record-btrace.c
@@ -44,6 +44,7 @@ 
 #include "cli/cli-style.h"
 #include "async-event.h"
 #include <forward_list>
+#include "objfiles.h"
 
 static const target_info record_btrace_target_info = {
   "record-btrace",
@@ -722,6 +723,9 @@  btrace_find_line_range (CORE_ADDR pc)
   if (nlines <= 0)
     return btrace_mk_line_range (symtab, 0, 0);
 
+  struct objfile *objfile = symtab->compunit ()->objfile ();
+  pc -= objfile->text_section_offset ();
+
   range = btrace_mk_line_range (symtab, 0, 0);
   for (i = 0; i < nlines - 1; i++)
     {
@@ -733,7 +737,7 @@  btrace_find_line_range (CORE_ADDR pc)
 	 possibly adding more line numbers to the range.  At the time this
 	 change was made I was unsure how to test this so chose to go with
 	 maintaining the existing experience.  */
-      if ((lines[i].pc == pc) && (lines[i].line != 0)
+      if ((lines[i].raw_pc () == pc) && (lines[i].line != 0)
 	  && (lines[i].is_stmt == 1))
 	range = btrace_line_range_add (range, lines[i].line);
     }
diff --git a/gdb/symmisc.c b/gdb/symmisc.c
index 6ff06ccaa93..d1458784e5c 100644
--- a/gdb/symmisc.c
+++ b/gdb/symmisc.c
@@ -263,7 +263,7 @@  dump_symtab_1 (struct symtab *symtab, struct ui_file *outfile)
       for (int i = 0; i < len; i++)
 	{
 	  gdb_printf (outfile, " line %d at ", l->item[i].line);
-	  gdb_puts (paddress (gdbarch, l->item[i].pc), outfile);
+	  gdb_puts (paddress (gdbarch, l->item[i].raw_pc ()), outfile);
 	  if (l->item[i].is_stmt)
 	    gdb_printf (outfile, "\t(stmt)");
 	  gdb_printf (outfile, "\n");
@@ -996,7 +996,7 @@  maintenance_print_one_line_table (struct symtab *symtab, void *data)
 	  else
 	    uiout->field_string ("line", _("END"));
 	  uiout->field_core_addr ("address", objfile->arch (),
-				  item->pc);
+				  item->raw_pc ());
 	  uiout->field_string ("is-stmt", item->is_stmt ? "Y" : "");
 	  uiout->field_string ("prologue-end", item->prologue_end ? "Y" : "");
 	  uiout->text ("\n");
diff --git a/gdb/symtab.c b/gdb/symtab.c
index f26dd2cb187..df7d3b2a398 100644
--- a/gdb/symtab.c
+++ b/gdb/symtab.c
@@ -329,6 +329,14 @@  search_domain_name (enum search_domain e)
 
 /* See symtab.h.  */
 
+CORE_ADDR
+linetable_entry::pc (const struct objfile *objfile) const
+{
+  return m_pc + objfile->text_section_offset ();
+}
+
+/* See symtab.h.  */
+
 call_site *
 compunit_symtab::find_call_site (CORE_ADDR pc) const
 {
@@ -3126,6 +3134,7 @@  find_pc_sect_line (CORE_ADDR pc, struct obj_section *section, int notcurrent)
     }
 
   bv = cust->blockvector ();
+  struct objfile *objfile = cust->objfile ();
 
   /* Look at all the symtabs that share this blockvector.
      They all have the same apriori range, that we found was right;
@@ -3152,18 +3161,21 @@  find_pc_sect_line (CORE_ADDR pc, struct obj_section *section, int notcurrent)
 
       /* Is this file's first line closer than the first lines of other files?
 	 If so, record this file, and its first line, as best alternate.  */
-      if (item->pc > pc && (!alt || item->pc < alt->pc))
+      if (item->pc (objfile) > pc
+	  && (!alt || item->raw_pc () < alt->raw_pc ()))
 	alt = item;
 
       auto pc_compare = [](const CORE_ADDR & comp_pc,
 			   const struct linetable_entry & lhs)->bool
       {
-	return comp_pc < lhs.pc;
+	return comp_pc < lhs.raw_pc ();
       };
 
       struct linetable_entry *first = item;
       struct linetable_entry *last = item + len;
-      item = std::upper_bound (first, last, pc, pc_compare);
+      item = std::upper_bound (first, last,
+			       pc - objfile->text_section_offset (),
+			       pc_compare);
       if (item != first)
 	prev = item - 1;		/* Found a matching item.  */
 
@@ -3177,7 +3189,7 @@  find_pc_sect_line (CORE_ADDR pc, struct obj_section *section, int notcurrent)
 	 save prev if it represents the end of a function (i.e. line number
 	 0) instead of a real line.  */
 
-      if (prev && prev->line && (!best || prev->pc > best->pc))
+      if (prev && prev->line && (!best || prev->raw_pc () > best->raw_pc ()))
 	{
 	  best = prev;
 	  best_symtab = iter_s;
@@ -3192,7 +3204,7 @@  find_pc_sect_line (CORE_ADDR pc, struct obj_section *section, int notcurrent)
 	  if (!best->is_stmt)
 	    {
 	      struct linetable_entry *tmp = best;
-	      while (tmp > first && (tmp - 1)->pc == tmp->pc
+	      while (tmp > first && (tmp - 1)->raw_pc () == tmp->raw_pc ()
 		     && (tmp - 1)->line != 0 && !tmp->is_stmt)
 		--tmp;
 	      if (tmp->is_stmt)
@@ -3200,16 +3212,16 @@  find_pc_sect_line (CORE_ADDR pc, struct obj_section *section, int notcurrent)
 	    }
 
 	  /* Discard BEST_END if it's before the PC of the current BEST.  */
-	  if (best_end <= best->pc)
+	  if (best_end <= best->pc (objfile))
 	    best_end = 0;
 	}
 
       /* If another line (denoted by ITEM) is in the linetable and its
 	 PC is after BEST's PC, but before the current BEST_END, then
 	 use ITEM's PC as the new best_end.  */
-      if (best && item < last && item->pc > best->pc
-	  && (best_end == 0 || best_end > item->pc))
-	best_end = item->pc;
+      if (best && item < last && item->raw_pc () > best->raw_pc ()
+	  && (best_end == 0 || best_end > item->pc (objfile)))
+	best_end = item->pc (objfile);
     }
 
   if (!best_symtab)
@@ -3232,11 +3244,11 @@  find_pc_sect_line (CORE_ADDR pc, struct obj_section *section, int notcurrent)
       val.is_stmt = best->is_stmt;
       val.symtab = best_symtab;
       val.line = best->line;
-      val.pc = best->pc;
-      if (best_end && (!alt || best_end < alt->pc))
+      val.pc = best->pc (objfile);
+      if (best_end && (!alt || best_end < alt->pc (objfile)))
 	val.end = best_end;
       else if (alt)
-	val.end = alt->pc;
+	val.end = alt->pc (objfile);
       else
 	val.end = bv->global_block ()->end ();
     }
@@ -3387,6 +3399,7 @@  find_pcs_for_symtab_line (struct symtab *symtab, int line,
 {
   int start = 0;
   std::vector<CORE_ADDR> result;
+  struct objfile *objfile = symtab->compunit ()->objfile ();
 
   /* First, collect all the PCs that are at this line.  */
   while (1)
@@ -3410,7 +3423,7 @@  find_pcs_for_symtab_line (struct symtab *symtab, int line,
 	  break;
 	}
 
-      result.push_back (symtab->linetable ()->item[idx].pc);
+      result.push_back (symtab->linetable ()->item[idx].pc (objfile));
       start = idx + 1;
     }
 
@@ -3436,7 +3449,7 @@  find_line_pc (struct symtab *symtab, int line, CORE_ADDR *pc)
   if (symtab != NULL)
     {
       l = symtab->linetable ();
-      *pc = l->item[ind].pc;
+      *pc = l->item[ind].pc (symtab->compunit ()->objfile ());
       return true;
     }
   else
@@ -3640,6 +3653,8 @@  skip_prologue_using_lineinfo (CORE_ADDR func_addr, struct symtab *symtab)
   if (!find_pc_partial_function (func_addr, NULL, &func_start, &func_end))
     return func_addr;
 
+  struct objfile *objfile = symtab->compunit ()->objfile ();
+
   /* Linetable entries are ordered by PC values, see the commentary in
      symtab.h where `struct linetable' is defined.  Thus, the first
      entry whose PC is in the range [FUNC_START..FUNC_END[ is the
@@ -3647,12 +3662,13 @@  skip_prologue_using_lineinfo (CORE_ADDR func_addr, struct symtab *symtab)
   for (i = 0; i < l->nitems; i++)
     {
       struct linetable_entry *item = &(l->item[i]);
+      CORE_ADDR item_pc = item->pc (objfile);
 
       /* Don't use line numbers of zero, they mark special entries in
 	 the table.  See the commentary on symtab.h before the
 	 definition of struct linetable.  */
-      if (item->line > 0 && func_start <= item->pc && item->pc < func_end)
-	return item->pc;
+      if (item->line > 0 && func_start <= item_pc && item_pc < func_end)
+	return item_pc;
     }
 
   return func_addr;
@@ -3679,18 +3695,22 @@  skip_prologue_using_linetable (CORE_ADDR func_addr)
     {
       struct linetable *linetable = prologue_sal.symtab->linetable ();
 
+      struct objfile *objfile = prologue_sal.symtab->compunit ()->objfile ();
+      start_pc -= objfile->text_section_offset ();
+      end_pc -= objfile->text_section_offset ();
+
       auto it = std::lower_bound
 	(linetable->item, linetable->item + linetable->nitems, start_pc,
 	 [] (const linetable_entry &lte, CORE_ADDR pc) -> bool
 	 {
-	   return lte.pc < pc;
+	   return lte.raw_pc () < pc;
 	 });
 
       for (;
-	   it < linetable->item + linetable->nitems && it->pc <= end_pc;
+	   it < linetable->item + linetable->nitems && it->raw_pc () <= end_pc;
 	   it++)
 	if (it->prologue_end)
-	  return {it->pc};
+	  return {it->pc (objfile)};
     }
 
   return {};
@@ -3924,18 +3944,20 @@  skip_prologue_using_sal (struct gdbarch *gdbarch, CORE_ADDR func_addr)
 	 do this.  */
       if (prologue_sal.symtab->language () != language_asm)
 	{
+	  struct objfile *objfile
+	    = prologue_sal.symtab->compunit ()->objfile ();
 	  struct linetable *linetable = prologue_sal.symtab->linetable ();
 	  int idx = 0;
 
 	  /* Skip any earlier lines, and any end-of-sequence marker
 	     from a previous function.  */
-	  while (linetable->item[idx].pc != prologue_sal.pc
+	  while (linetable->item[idx].pc (objfile) != prologue_sal.pc
 		 || linetable->item[idx].line == 0)
 	    idx++;
 
 	  if (idx+1 < linetable->nitems
 	      && linetable->item[idx+1].line != 0
-	      && linetable->item[idx+1].pc == start_pc)
+	      && linetable->item[idx+1].pc (objfile) == start_pc)
 	    return start_pc;
 	}
 
diff --git a/gdb/symtab.h b/gdb/symtab.h
index 69f0eaa0f88..84d12617f31 100644
--- a/gdb/symtab.h
+++ b/gdb/symtab.h
@@ -1547,18 +1547,29 @@  struct rust_vtable_symbol : public symbol
 
 struct linetable_entry
 {
+  /* Set the (unrelocated) PC for this entry.  */
+  void set_pc (CORE_ADDR pc)
+  { m_pc = pc; }
+
+  /* Return the unrelocated PC for this entry.  */
+  CORE_ADDR raw_pc () const
+  { return m_pc; }
+
+  /* Return the relocated PC for this entry.  */
+  CORE_ADDR pc (const struct objfile *objfile) const;
+
   bool operator< (const linetable_entry &other) const
   {
-    if (pc == other.pc
+    if (m_pc == other.m_pc
 	&& (line != 0) != (other.line != 0))
       return line == 0;
-    return pc < other.pc;
+    return m_pc < other.m_pc;
   }
 
   /* Two entries are equal if they have the same line and PC.  The
      other members are ignored.  */
   bool operator== (const linetable_entry &other) const
-  { return line == other.line && pc == other.pc; }
+  { return line == other.line && m_pc == other.m_pc; }
 
   /* The line number for this entry.  */
   int line;
@@ -1571,7 +1582,7 @@  struct linetable_entry
   bool prologue_end : 1;
 
   /* The address for this entry.  */
-  CORE_ADDR pc;
+  CORE_ADDR m_pc;
 };
 
 /* The order of entries in the linetable is significant.  They should
diff --git a/gdb/xcoffread.c b/gdb/xcoffread.c
index 93054bdeb19..45578311705 100644
--- a/gdb/xcoffread.c
+++ b/gdb/xcoffread.c
@@ -432,7 +432,7 @@  arrange_linetable (std::vector<linetable_entry> &old_linetable)
 	  linetable_entry &e = fentries.back ();
 	  e.line = ii;
 	  e.is_stmt = 1;
-	  e.pc = old_linetable[ii].pc;
+	  e.set_pc (old_linetable[ii].raw_pc ());
 	}
     }
 
@@ -457,7 +457,7 @@  arrange_linetable (std::vector<linetable_entry> &old_linetable)
 	 extra line to cover the function prologue.  */
       int jj = entry.line;
       if (jj + 1 < old_linetable.size ()
-	  && old_linetable[jj].pc != old_linetable[jj + 1].pc)
+	  && old_linetable[jj].raw_pc () != old_linetable[jj + 1].raw_pc ())
 	{
 	  new_linetable.push_back (old_linetable[jj]);
 	  new_linetable.back ().line = old_linetable[jj + 1].line;
@@ -790,15 +790,16 @@  enter_line_range (struct subfile *subfile, unsigned beginoffset,
       if (addr < startaddr || (endaddr && addr >= endaddr))
 	return;
 
+      CORE_ADDR record_addr = (gdbarch_addr_bits_remove (gdbarch, addr)
+			       - objfile->text_section_offset ());
       if (int_lnno.l_lnno == 0)
 	{
 	  *firstLine = read_symbol_lineno (int_lnno.l_addr.l_symndx);
-	  record_line (subfile, 0, gdbarch_addr_bits_remove (gdbarch, addr));
+	  record_line (subfile, 0, record_addr);
 	  --(*firstLine);
 	}
       else
-	record_line (subfile, *firstLine + int_lnno.l_lnno,
-		     gdbarch_addr_bits_remove (gdbarch, addr));
+	record_line (subfile, *firstLine + int_lnno.l_lnno, record_addr);
       curoffset += linesz;
     }
 }