Patchwork gdb/testsuite: Regenerate the testglue if it is not in path

login
register
mail settings
Submitter Shahab Vahedi
Date Feb. 11, 2020, 1:37 p.m.
Message ID <20200211133710.6120-1-shahab.vahedi@gmail.com>
Download mbox | patch
Permalink /patch/37964/
State New
Headers show

Comments

Shahab Vahedi - Feb. 11, 2020, 1:37 p.m.
From: Shahab Vahedi <shahab@synopsys.com>

For running the  DejaGnu  tests,  some  esoteric  configurations
may require a testglue.   This,  for  instance,  is  true  about
testing ARC  targets  which  uses  its  own  DejaGnu  board  and
a simulator which does not support returning the  pass  or  fail
through the exit code.  Therefore,  for  those  tests  that  use
"gdb_compile" a "gdb_tg.o" file is compiled and  linked  to  the
final executable.

There  are  tests  that  invoke  "gdb_compile"  from   different
directories.   Let's  take  a   look   at   an   example   test:
gdb.base/fullname.exp.  The purpose of this  test  is  to  build
the executable from different directories (absolute vs. relative
vs.  other) and then check if gdb can handle setting breakpoints
accordingly.

When  "gdb_compile"  generates  the  "gdb_tg.o",  it  does   not
do it again  for  the  same  test.   Although  this  might  seem
efficient, it can lead to  problems  when  changing  directories
before the next compile:

  gdb compile failed, arc-elf32-gcc: error: gdb_tg.o:
  No such file or directory

This patch checks if the wrapper file ("gdb_tg.o") is  still  in
reach and if it is not, it will stimulate the generation of  the
wrapper again.

It is worth mentioning that GCC's  DejaGnu  tests  handle  these
scenarios as well and they seem to be more efficient in doing so
by saving the library paths and manipulating them  if  necessary
[1].  However, for GDB tests, that  require  less  compilations,
I think the proposed solution should be fine compared to a  more
full fledged solution from GCC.  The glue file in  our  case  is
only 2 KiB.

Last but not least, I ran the x86_64 tests on an x86_64 host and
found no regression.

[1]
Avid  coders  may  look  for  "set_ld_library_path_env_vars"  in
gcc/testsuite/lib/target-libpath.exp.

gdb/testsuite/ChangeLog:

	* lib/gdb.exp (gdb_compile): Reset
	"gdb_wrapper_initialized" to 0 if "wrapper_file" does
	not exist.
---
 gdb/testsuite/lib/gdb.exp | 9 +++++++++
 1 file changed, 9 insertions(+)
Tom Tromey - Feb. 11, 2020, 3:39 p.m.
>>>>> "Shahab" == Shahab Vahedi <shahab.vahedi@gmail.com> writes:

Shahab> For running the  DejaGnu  tests,  some  esoteric  configurations
Shahab> may require a testglue.

Thank you for the patch.  I wasn't aware of this code at all :-)

Shahab> This patch checks if the wrapper file ("gdb_tg.o") is  still  in
Shahab> reach and if it is not, it will stimulate the generation of  the
Shahab> wrapper again.

I wonder if it would make sense to instead set gdb_wrapper_file to the
absolute path.

It could perhaps be put in the cache directory.  This might help avoid
recompiling it many times when running the tests in parallel.

What do you think of that?

Shahab> +    # if the wrapper is initialized but the wrapper file cannot be

GNU style is to start a comment with an upper-case letter, so "If".

Shahab> +    # found anymore, the wrapper file must be built again.
Shahab> +    if { $gdb_wrapper_initialized == 1 && \
Shahab> +	    [info exists gdb_wrapper_file] && \
Shahab> +	    ![file exists $gdb_wrapper_file] } {
Shahab> +	verbose "reinitializing the wrapper"
Shahab> +	set gdb_wrapper_initialized 0
Shahab> +    }
Shahab> +
Shahab>      if { $gdb_wrapper_initialized == 0 } { gdb_wrapper_init }

If you don't like the absolute path idea, maybe just having all this
logic be in the wrapper code would be better, and this could just call
gdb_wrapper_init unconditionally.

Tom
Shahab Vahedi - Feb. 11, 2020, 4:52 p.m.
On Tue, Feb 11, 2020 at 08:39:03AM -0700, Tom Tromey wrote:
> >>>>> "Shahab" == Shahab Vahedi <shahab.vahedi@gmail.com> writes:
> 
> Shahab> For running the  DejaGnu  tests,  some  esoteric  configurations
> Shahab> may require a testglue.
> 
> Thank you for the patch.  I wasn't aware of this code at all :-)
> 
> Shahab> This patch checks if the wrapper file ("gdb_tg.o") is  still  in
> Shahab> reach and if it is not, it will stimulate the generation of  the
> Shahab> wrapper again.
> 
> I wonder if it would make sense to instead set gdb_wrapper_file to the
> absolute path.

This is how "gdb_wrapper_file" is set in gdb/testsuite/lib/gdb.exp:

  proc gdb_wrapper_init { args } {
  ...
  global gdb_wrapper_file
  ...
      set result [build_wrapper "testglue.o"]
      if { $result != "" } {
          set gdb_wrapper_file [lindex $result 0]
          ...
  }

The "build_wrapper" is a procedure form DejaGnu  itself  (libgloss.exp),
and it does not provide the full path.  Maybe we  can  get  the  current
working directory from somewhere else and append it here?

> It could perhaps be put in the cache directory.  This might help avoid
> recompiling it many times when running the tests in parallel.
> 
> What do you think of that?

I absolutely like this idea and believe this is how it should  be  done,
but am not sure where to begin. Any pointers?

> 
> Shahab> +    # if the wrapper is initialized but the wrapper file cannot be
> 
> GNU style is to start a comment with an upper-case letter, so "If".

I will fix this in next patch (whatever approach we take).

> 
> Shahab> +    # found anymore, the wrapper file must be built again.
> Shahab> +    if { $gdb_wrapper_initialized == 1 && \
> Shahab> +	    [info exists gdb_wrapper_file] && \
> Shahab> +	    ![file exists $gdb_wrapper_file] } {
> Shahab> +	verbose "reinitializing the wrapper"
> Shahab> +	set gdb_wrapper_initialized 0
> Shahab> +    }
> Shahab> +
> Shahab>      if { $gdb_wrapper_initialized == 0 } { gdb_wrapper_init }
> 
> If you don't like the absolute path idea, maybe just having all this
> logic be in the wrapper code would be better, and this could just call
> gdb_wrapper_init unconditionally.

I agree with your suggestion and if we opt for  this  approach,  I  will
adapt it as such.

> 
> Tom

Cheers,
Shahab

Patch

diff --git a/gdb/testsuite/lib/gdb.exp b/gdb/testsuite/lib/gdb.exp
index d5e22957039..8e91c91e715 100644
--- a/gdb/testsuite/lib/gdb.exp
+++ b/gdb/testsuite/lib/gdb.exp
@@ -3812,6 +3812,15 @@  proc gdb_compile {source dest type options} {
     verbose "options are $options"
     verbose "source is $source $dest $type $options"
 
+    # if the wrapper is initialized but the wrapper file cannot be
+    # found anymore, the wrapper file must be built again.
+    if { $gdb_wrapper_initialized == 1 && \
+	    [info exists gdb_wrapper_file] && \
+	    ![file exists $gdb_wrapper_file] } {
+	verbose "reinitializing the wrapper"
+	set gdb_wrapper_initialized 0
+    }
+
     if { $gdb_wrapper_initialized == 0 } { gdb_wrapper_init }
 
     if {[target_info exists needs_status_wrapper] && \