New ARI warning Wed Aug 29 01:57:16 UTC 2018

New ARI warning Wed Aug 29 01:57:16 UTC 2018

[GDB Administrator]

120a121,122
> gdb/csky-tdep.c:1612: gettext: trailing new line: A message should not have a trailing new line
gdb/csky-tdep.c:1612:    warning (_("Invalid breakpoint address 0x%x is an odd number.\n"),
> gdb/csky-tdep.c:2234: deprecate: read_pc: Replace READ_PC() with frame_pc_unwind; at present the inferior function call code still uses this
gdb/csky-tdep.c:2234:  set_gdbarch_read_pc (gdbarch, csky_read_pc);
gdb/csky-tdep.c:2234:  set_gdbarch_read_pc (gdbarch, csky_read_pc);

Re: New ARI warning Wed Aug 29 01:57:16 UTC 2018

[Simon Marchi]

On 2018-08-28 09:57 PM, GDB Administrator wrote:
> 120a121,122
>> gdb/csky-tdep.c:1612: gettext: trailing new line: A message should not have a trailing new line
> gdb/csky-tdep.c:1612:    warning (_("Invalid breakpoint address 0x%x is an odd number.\n"),

I pushed this obvious patch to fix this warning:

From 2362e7f76adb395df19c867cd27c75f77a0ade6f Mon Sep 17 00:00:00 2001
From: Simon Marchi <simon.marchi@ericsson.com>
Date: Wed, 29 Aug 2018 10:26:11 -0400
Subject: [PATCH] Remove newline at end of warning message

... to fix this ARI warning:

  gdb/csky-tdep.c:1612: gettext: trailing new line: A message should not have a trailing new line
  gdb/csky-tdep.c:1612:    warning (_("Invalid breakpoint address 0x%x is an odd number.\n"),

gdb/ChangeLog:

	* csky-tdep.c (csky_memory_insert_breakpoint): Remove newline at
	end of warning message.
---
 gdb/ChangeLog   | 5 +++++
 gdb/csky-tdep.c | 2 +-
 2 files changed, 6 insertions(+), 1 deletion(-)

diff --git a/gdb/ChangeLog b/gdb/ChangeLog
index 373e1b1..32054c6 100644
--- a/gdb/ChangeLog
+++ b/gdb/ChangeLog
@@ -1,3 +1,8 @@
+2018-08-29  Simon Marchi  <simon.marchi@ericsson.com>
+
+	* csky-tdep.c (csky_memory_insert_breakpoint): Remove newline at
+	end of warning message.
+
 2018-08-29  Alan Hayward  <alan.hayward@arm.com>

 	PR gdb/22943:
diff --git a/gdb/csky-tdep.c b/gdb/csky-tdep.c
index 997ca93..972ac59 100644
--- a/gdb/csky-tdep.c
+++ b/gdb/csky-tdep.c
@@ -1609,7 +1609,7 @@ csky_memory_insert_breakpoint (struct gdbarch *gdbarch,

   /* Sanity-check bp_address.  */
   if (bp_tgt->reqstd_address % 2)
-    warning (_("Invalid breakpoint address 0x%x is an odd number.\n"),
+    warning (_("Invalid breakpoint address 0x%x is an odd number."),
 	     (unsigned int) bp_tgt->reqstd_address);
   scoped_restore restore_memory
     = make_scoped_restore_show_memory_breakpoints (1);
-- 
2.7.4

Re: New ARI warning Wed Aug 29 01:57:16 UTC 2018

[Simon Marchi]

On 2018-08-28 09:57 PM, GDB Administrator wrote:
>> gdb/csky-tdep.c:2234: deprecate: read_pc: Replace READ_PC() with frame_pc_unwind; at present the inferior function call code still uses this
> gdb/csky-tdep.c:2234:  set_gdbarch_read_pc (gdbarch, csky_read_pc);
> gdb/csky-tdep.c:2234:  set_gdbarch_read_pc (gdbarch, csky_read_pc);

I am not sure about this one.  At first, I thought it was just a name clash,
the test looking for the "read_pc" function and wrongfully finding this line.
However, the check does purposefully look for "set_gdbarch_read_pc":

 997 BEGIN { doc["read_pc"] = "\
 998 Replace READ_PC() with frame_pc_unwind; \
 999 at present the inferior function call code still uses this"
1000     category["read_pc"] = ari_deprecate
1001 }
1002 /(^|[^_[:alnum:]])read_pc[[:space:]]*\(/ || \
1003 /(^|[^_[:alnum:]])set_gdbarch_read_pc[[:space:]]*\(/ || \
1004 /(^|[^_[:alnum:]])TARGET_READ_PC[[:space:]]*\(/ {
1005     fail("read_pc")
1006 }

Is this gdbarch method really deprecated?

Simon

Re: New ARI warning Wed Aug 29 01:57:16 UTC 2018

[Tom Tromey]

>>>>> "Simon" == Simon Marchi <simon.marchi@ericsson.com> writes:

Simon> I am not sure about this one.  At first, I thought it was just a name clash,
Simon> the test looking for the "read_pc" function and wrongfully finding this line.
Simon> However, the check does purposefully look for "set_gdbarch_read_pc":
[...]
Simon> Is this gdbarch method really deprecated?

I don't know but I see other callers in-tree, which seems a bit peculiar.
Another unfinished deprecation I guess.  Maybe we should be filing bugs
for these.

Tom