[PATCH] Call error_value_optimized_out from value_assign

[PATCH] Call error_value_optimized_out from value_assign

[Tom Tromey]

value_assign will throw an exception indicating that the value is
optimized out, but it seemed better to me for it to use the utility
function error_value_optimized_out.
---
 gdb/valops.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/gdb/valops.c b/gdb/valops.c
index 515e43ccdfa..fccadae6cf6 100644
--- a/gdb/valops.c
+++ b/gdb/valops.c
@@ -1225,8 +1225,7 @@ value_assign (struct value *toval, struct value *fromval)
 					   &unavail))
 	      {
 		if (optim)
-		  throw_error (OPTIMIZED_OUT_ERROR,
-			       _("value has been optimized out"));
+		  error_value_optimized_out ();
 		if (unavail)
 		  throw_error (NOT_AVAILABLE_ERROR,
 			       _("value is not available"));

base-commit: 8927b500532538cd86aae2d777c6452c51a210c6
-- 
2.53.0

Re: [PATCH] Call error_value_optimized_out from value_assign

[Tom de Vries]

On 4/21/26 9:41 PM, Tom Tromey wrote:
> value_assign will throw an exception indicating that the value is
> optimized out, but it seemed better to me for it to use the utility
> function error_value_optimized_out.

Hi Tom,

LGTM.

Approved-By: Tom de Vries <tdevries@suse.de>

I wondered about the NOT_AVAILABLE_ERROR, and I think we could do 
something similar, see unverified patch below.

I'll submit it as a separate patch, after you merge this one.  I'm not 
sure if it makes sense, but given that it touches the same code, you 
could also merge it into your patch.

Thanks,
- Tom

diff --git a/gdb/valops.c b/gdb/valops.c
index 515e43ccdfa..95888d3539a 100644
--- a/gdb/valops.c
+++ b/gdb/valops.c
@@ -1228,8 +1228,7 @@ value_assign (struct value *toval, struct value 
*fromval)
  		  throw_error (OPTIMIZED_OUT_ERROR,
  			       _("value has been optimized out"));
  		if (unavail)
-		  throw_error (NOT_AVAILABLE_ERROR,
-			       _("value is not available"));
+		  error_value_not_available ();
  	      }

  	    modify_field (type, buffer, value_as_long (fromval),
diff --git a/gdb/value.c b/gdb/value.c
index 2d90780f57a..332b3c4bbc7 100644
--- a/gdb/value.c
+++ b/gdb/value.c
@@ -1127,6 +1127,12 @@ error_value_optimized_out (void)
    throw_error (OPTIMIZED_OUT_ERROR, _("value has been optimized out"));
  }

+void
+error_value_not_available (void)
+{
+  throw_error (OPTIMIZED_OUT_ERROR, _("value is not available"));
+}
+
  void
  value::require_not_optimized_out () const
  {
@@ -1144,7 +1150,7 @@ void
  value::require_available () const
  {
    if (!m_unavailable.empty ())
-    throw_error (NOT_AVAILABLE_ERROR, _("value is not available"));
+    error_value_not_available ();
  }

  gdb::array_view<const gdb_byte>
diff --git a/gdb/value.h b/gdb/value.h
index 201db773a84..0267ee804e1 100644
--- a/gdb/value.h
+++ b/gdb/value.h
@@ -987,6 +987,10 @@ struct lval_funcs

  [[noreturn]] extern void error_value_optimized_out ();

+/* Throw an error complaining that the value is not available.  */
+
+[[noreturn]] extern void error_value_not_available ();
+
  /* Pointer to internal variable.  */
  #define VALUE_INTERNALVAR(val) (*((val)->deprecated_internalvar_hack ()))

Re: [PATCH] Call error_value_optimized_out from value_assign

[Tom Tromey]

>>>>> "Tom" == Tom de Vries <tdevries@suse.de> writes:

Tom> Approved-By: Tom de Vries <tdevries@suse.de>

Thanks.

Tom> I wondered about the NOT_AVAILABLE_ERROR, and I think we could do
Tom> something similar, see unverified patch below.

Makes sense.

Tom> I'll submit it as a separate patch, after you merge this one.  I'm not
Tom> sure if it makes sense, but given that it touches the same code, you
Tom> could also merge it into your patch.

Tom> +void
Tom> +error_value_not_available (void)
Tom> +{
Tom> +  throw_error (OPTIMIZED_OUT_ERROR, _("value is not available"));

Should use NOT_AVAILABLE_ERROR here.

Also (void) is a C-ism.

Tom

Re: [PATCH] Call error_value_optimized_out from value_assign

[Tom de Vries]

On 4/24/26 4:37 PM, Tom Tromey wrote:
>>>>>> "Tom" == Tom de Vries <tdevries@suse.de> writes:
> 
> Tom> Approved-By: Tom de Vries <tdevries@suse.de>
> 
> Thanks.
> 
> Tom> I wondered about the NOT_AVAILABLE_ERROR, and I think we could do
> Tom> something similar, see unverified patch below.
> 
> Makes sense.
> 
> Tom> I'll submit it as a separate patch, after you merge this one.  I'm not
> Tom> sure if it makes sense, but given that it touches the same code, you
> Tom> could also merge it into your patch.
> 
> Tom> +void
> Tom> +error_value_not_available (void)
> Tom> +{
> Tom> +  throw_error (OPTIMIZED_OUT_ERROR, _("value is not available"));
> 
> Should use NOT_AVAILABLE_ERROR here.
> 
> Also (void) is a C-ism.
> 

Thanks, I've pushed this ( 
https://sourceware.org/pipermail/gdb-patches/2026-April/226811.html ) 
with those two issues fixed.

Thanks,
- Tom

> Tom