[FFmpeg-cvslog] ppc: fix 32-bit PIC build

[Mans Rullgard]

ffmpeg | branch: master | Mans Rullgard <mans at mansr.com> | Sun Sep 25 12:53:44 2011 +0100| [6e4a35ced96cdf31a9d3bd82fd147554750af839] | committer: Mans Rullgard

ppc: fix 32-bit PIC build

On 32-bit ppc, the GOT pointer must be loaded manually.
This adds a "get_got" assembler macro to compute the
GOT address.  The "movrel" macro is updated to take an
additional parameter containing the GOT address since
no register is reserved for this purpose on ppc32.
These changes have no effect on ppc64 builds.

Signed-off-by: Mans Rullgard <mans at mansr.com>

> http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commit;h=6e4a35ced96cdf31a9d3bd82fd147554750af839
---

 libavcodec/ppc/asm.S           |   19 ++++++++++++++++---
 libavcodec/ppc/fft_altivec_s.S |    7 ++++---
 2 files changed, 20 insertions(+), 6 deletions(-)

diff --git a/libavcodec/ppc/asm.S b/libavcodec/ppc/asm.S
index 5cbbf97..4d4285b 100644
--- a/libavcodec/ppc/asm.S
+++ b/libavcodec/ppc/asm.S
@@ -44,10 +44,13 @@ X(\name):
 L(\name):
 .endm
 
-.macro movrel rd, sym
+.macro movrel rd, sym, gp
     ld      \rd, \sym at got(r2)
 .endm
 
+.macro get_got rd
+.endm
+
 #else /* ARCH_PPC64 */
 
 #define PTR  .int
@@ -65,15 +68,25 @@ X(\name):
 \name:
 .endm
 
-.macro movrel rd, sym
+.macro movrel rd, sym, gp
 #if CONFIG_PIC
-    lwz     \rd, \sym at got(r2)
+    lwz     \rd, \sym at got(\gp)
 #else
     lis     \rd, \sym at ha
     la      \rd, \sym at l(\rd)
 #endif
 .endm
 
+.macro get_got rd
+#if CONFIG_PIC
+    bcl     20, 31, .Lgot\@
+.Lgot\@:
+    mflr    \rd
+    addis   \rd, \rd, _GLOBAL_OFFSET_TABLE_ - .Lgot\@@ha
+    addi    \rd, \rd, _GLOBAL_OFFSET_TABLE_ - .Lgot\@@l
+#endif
+.endm
+
 #endif /* ARCH_PPC64 */
 
 #if HAVE_IBM_ASM
diff --git a/libavcodec/ppc/fft_altivec_s.S b/libavcodec/ppc/fft_altivec_s.S
index ab33900..958d7df 100644
--- a/libavcodec/ppc/fft_altivec_s.S
+++ b/libavcodec/ppc/fft_altivec_s.S
@@ -353,6 +353,7 @@ extfunc ff_fft_calc\interleave\()_altivec
     mflr    r0
     stp     r0, 2*PS(r1)
     stpu    r1, -(160+16*PS)(r1)
+    get_got r11
     addi    r6, r1, 16*PS
     stvm    r6, v20, v21, v22, v23, v24, v25, v26, v27, v28, v29
     mfvrsave r0
@@ -360,14 +361,14 @@ extfunc ff_fft_calc\interleave\()_altivec
     li      r6, 0xfffffffc
     mtvrsave r6
 
-    movrel  r6, fft_data
+    movrel  r6, fft_data, r11
     lvm     r6, v14, v15, v16, v17, v18, v19, v20, v21
     lvm     r6, v22, v23, v24, v25, v26, v27, v28, v29
 
     li      r9, 16
-    movrel  r12, X(ff_cos_tabs)
+    movrel  r12, X(ff_cos_tabs), r11
 
-    movrel  r6, fft_dispatch_tab\interleave\()_altivec
+    movrel  r6, fft_dispatch_tab\interleave\()_altivec, r11
     lwz     r3, 0(r3)
     subi    r3, r3, 2
     slwi    r3, r3, 2+ARCH_PPC64