This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: [PATCH] Fix PPC{,64} debug info for __thread variables
- From: David Edelsohn <dje at watson dot ibm dot com>
- To: Jakub Jelinek <jakub at redhat dot com>
- Cc: Alan Modra <amodra at bigpond dot net dot au>, gcc-patches at gcc dot gnu dot org
- Date: Fri, 12 Mar 2004 10:24:58 -0500
- Subject: Re: [PATCH] Fix PPC{,64} debug info for __thread variables
- References: <20040312123048.GL3822@sunsite.ms.mff.cuni.cz>
+ case 8:
+ fputs ("\t.llong\t", file);
I thought the DOUBLE_INT_ASM_OP for powerpc-linux and
powerpc64-linux was ".quad". Also, this function probably should use
DOUBLE_INT_ASM_OP instead of duplicating the string. If it needs
".llong", I would like to know why.
+/* Emit a dtp-relative reference to a TLS variable. */
+
+#ifdef HAVE_AS_TLS
+#define ASM_OUTPUT_DWARF_DTPREL(FILE, SIZE, X) \
+ rs6000_output_dwarf_dtprel (FILE, SIZE, X)
+#endif
I guess defining this in rs6000.h instead of linux.h and
linux64.h, or at least sysv4.h, is forward-looking when other systems
support TLS?
* config/rs6000/rs6000-protos.h (rs6000_output_dwarf_dtprel): Add
prototype.
* config/rs6000/rs6000.c (rs6000_output_dwarf_dtprel): New.
* config/rs6000/rs6000.h (ASM_OUTPUT_DWARF_DTPREL): Define.
Okay with the DOUBLE_INT_ASM_OP change, assuming .quad change is correct.
Otherwise, there probably is something wrong.
Thanks, David