This is the mail archive of the binutils@sourceware.org mailing list for the binutils project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

[PATCH 3/3] binutils/testsuite: Tighten the error message check with `strip-13'


Avoid false positives and actually verify both that an `unsupported 
relocation type 0x8f' message is produced and that no other message is, 
except for the final `bad value', in the `strip-13' test.  This ensures 
that it is a relocation processing error and not a different issue that 
has caused `strip' to terminate unsuccessfully, and that the number 
representing the unsupported relocation has not been clobbered.

	binutils/
	* testsuite/binutils-all/strip-13.d: Also expect `unsupported 
	relocation type 0x8f' error message.
---
Hi,

 This causes a `sparc-linux' target regression, because an `unsupported 
relocation type 0x8f' error message is printed twice for some reason, 
which is likely a backend oddity similar to one I recently fixed with the 
o32 MIPS backend with commit 75def2abc3da ("PR binutils/22875: MIPS: 
Remove duplicate unsupported relocation processing").  I'm leaving it up 
to the SPARC maintainers to sort out.

 No other issues appeared across my usual targets.  OK to apply?

  Maciej
---
 binutils/testsuite/binutils-all/strip-13.d |    3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

binutils-strip-13-test-reloc-unsupp.diff
Index: binutils/binutils/testsuite/binutils-all/strip-13.d
===================================================================
--- binutils.orig/binutils/testsuite/binutils-all/strip-13.d	2018-04-04 20:33:25.000000000 +0100
+++ binutils/binutils/testsuite/binutils-all/strip-13.d	2018-04-04 20:34:57.526821144 +0100
@@ -1,6 +1,7 @@
 #PROG: strip
 #strip: -g
-#error: .* bad value
+#error: \A[^\n]*: unsupported relocation type 0x8f\n
+#error:   [^\n]*: bad value\Z
 #not-target: h8300-* ip2k-* m6811-* m68hc11-* rx-*
 # The H8300-*, IP2K and 68HC11 targets use 16-bit addressing, so `.dc.a'
 #  does not work for manual relocation data construction.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]