Mirror of the gdb-patches mailing list
 help / color / mirror / Atom feed
From: Pedro Alves <pedro@palves.net>
To: gdb-patches@sourceware.org
Subject: [PATCH 3/7] gdb/testsuite: Let gdb_simple_compile take empty default flags
Date: Tue, 11 Aug 2026 02:45:54 +0100	[thread overview]
Message-ID: <20260811014558.3606251-4-pedro@palves.net> (raw)
In-Reply-To: <20260811014558.3606251-1-pedro@palves.net>

I noticed this while renaming nodebug to symtab.

gdb_simple_compile and its wrapper gdb_can_simple_compile take a
default_compile_flags argument that, when left unspecified, defaults
to a debug build ("debug nowarnings quiet").  The check for
"unspecified argument" is:

    if { $default_compile_flags == "" } {

which makes the default indistinguishable from a caller explicitly
passing "", so there was no way to ask for a build with no default
flags at all -- passing an empty list still gets you the default
flags.

Fix this by using a distinct "DEFAULT" sentinel for the unspecified
case instead.

Note no caller passes an explicit empty default_compile_flags today.

Change-Id: I30235fa1d28eb93dfdf08248af26669a7336d708
---
 gdb/testsuite/lib/gdb.exp | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/gdb/testsuite/lib/gdb.exp b/gdb/testsuite/lib/gdb.exp
index d34ac30f2f0..b268152bdb7 100644
--- a/gdb/testsuite/lib/gdb.exp
+++ b/gdb/testsuite/lib/gdb.exp
@@ -6293,7 +6293,7 @@ gdb_caching_proc universal_compile_options_assembly {} {
 # Return 1 if code can be compiled
 # Leave the file name of the resulting object in the upvar object.
 
-proc gdb_simple_compile {name code {type object} {compile_flags {}} {object obj} {default_compile_flags {}}} {
+proc gdb_simple_compile {name code {type object} {compile_flags {}} {object obj} {default_compile_flags DEFAULT}} {
     upvar $object obj
 
     switch -regexp -- $type {
@@ -6331,7 +6331,7 @@ proc gdb_simple_compile {name code {type object} {compile_flags {}} {object obj}
     }
     set src [standard_temp_file $name.$ext]
     set obj [standard_temp_file $name.$postfix]
-    if { $default_compile_flags == "" } {
+    if { $default_compile_flags eq "DEFAULT" } {
 	set compile_flags [concat $compile_flags {debug nowarnings quiet}]
     } else {
 	set compile_flags [concat $compile_flags $default_compile_flags]
@@ -6357,7 +6357,7 @@ proc gdb_simple_compile {name code {type object} {compile_flags {}} {object obj}
 # Return 1 if code can be compiled
 # Delete all created files and objects.
 
-proc gdb_can_simple_compile {name code {type object} {compile_flags ""} {default_compile_flags ""}} {
+proc gdb_can_simple_compile {name code {type object} {compile_flags ""} {default_compile_flags DEFAULT}} {
     set ret [gdb_simple_compile $name $code $type $compile_flags temp_obj \
 		 $default_compile_flags]
     file delete $temp_obj
-- 
2.54.0


  parent reply	other threads:[~2026-08-11  1:47 UTC|newest]

Thread overview: 9+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-08-11  1:45 [PATCH 0/7] gdb_compile nodebug => symtab Pedro Alves
2026-08-11  1:45 ` [PATCH 1/7] gdb/testsuite: Don't build a program for cvexpr.exp's last tests Pedro Alves
2026-08-11  1:45 ` [PATCH 2/7] gdb/testsuite: Remove gdb_can_simple_compile_nodebug Pedro Alves
2026-08-11  1:45 ` Pedro Alves [this message]
2026-08-11  1:45 ` [PATCH 4/7] gdb/testsuite: Don't pass "nodebug" to object compilations Pedro Alves
2026-08-11  1:45 ` [PATCH 5/7] gdb/testsuite: Rename nodebug => symtab Pedro Alves
2026-08-11  1:45 ` [PATCH 6/7] gdb/testsuite: Pass "symtab" to testcases that rely on minsyms Pedro Alves
2026-08-12 21:41   ` Luis
2026-08-11  1:45 ` [PATCH 7/7] gdb/testsuite: Handle "symtab" option on windows-msvc Pedro Alves

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=20260811014558.3606251-4-pedro@palves.net \
    --to=pedro@palves.net \
    --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