diff mbox series

[FFmpeg-devel] lavc/aacpsdsp: fix clobber on RISC-V LP64D/ILP32D

Message ID 20221006184612.51719-1-remi@remlab.net
State Accepted
Commit 105921251ab35b870887e0c7348016f4dff3ec5b
Headers show
Series [FFmpeg-devel] lavc/aacpsdsp: fix clobber on RISC-V LP64D/ILP32D | expand

Checks

Context Check Description
andriy/make_x86 success Make finished
andriy/make_fate_x86 success Make fate finished

Commit Message

Rémi Denis-Courmont Oct. 6, 2022, 6:46 p.m. UTC
Although the DSP function only uses single precision from RISC-V F, the
caller may leave double precision values in the spilled registers if the
calling convention supports double precision hardware floats. Then, we
need to save and restore FS registers as double precision.

Conversely, we do not need to save anything at all if an integer calling
convention is in use. However we can assume that single precision floats
are supported, since the Zve32f extension implies the F extension.
So for the sake of simplicity, we always save at least single precision
values.

In theory, we should even save quadruple precision values if the LP64Q
ABI is in use. I have yet to see a compiler that supports it though.
---
 libavcodec/riscv/aacpsdsp_rvv.S | 10 ++++++----
 1 file changed, 6 insertions(+), 4 deletions(-)
diff mbox series

Patch

diff --git a/libavcodec/riscv/aacpsdsp_rvv.S b/libavcodec/riscv/aacpsdsp_rvv.S
index 1d6e73fd2d..80bd19f6ad 100644
--- a/libavcodec/riscv/aacpsdsp_rvv.S
+++ b/libavcodec/riscv/aacpsdsp_rvv.S
@@ -55,9 +55,10 @@  endfunc
 
 func ff_ps_hybrid_analysis_rvv, zve32f
         /* We need 26 FP registers, for 20 scratch ones. Spill fs0-fs5. */
-        addi    sp, sp, -32
+        addi    sp, sp, -48
         .irp n, 0, 1, 2, 3, 4, 5
-        fsw     fs\n, (4 * \n)(sp)
+HWD     fsd     fs\n, (8 * \n)(sp)
+NOHWD   fsw     fs\n, (4 * \n)(sp)
         .endr
 
         .macro input, j, fd0, fd1, fd2, fd3
@@ -142,9 +143,10 @@  func ff_ps_hybrid_analysis_rvv, zve32f
         bnez       a4, 1b
 
         .irp n, 5, 4, 3, 2, 1, 0
-        flw     fs\n, (4 * \n)(sp)
+HWD     fld     fs\n, (8 * \n)(sp)
+NOHWD   flw     fs\n, (4 * \n)(sp)
         .endr
-        addi    sp, sp, 32
+        addi    sp, sp, 48
         ret
         .purgem input
         .purgem filter