Skip to contentSkip to conversation

New: opencv3 SIGILL on Cortex-A5 with VFPv4-D16

Buildroot bugzilla at busybox.net 1 participant 4 messages Jul 1, 2019 – Feb 5, 2020

lore
Share & tools
lore
#0 lore raw

https://bugs.busybox.net/show_bug.cgi?id=11996

Bug ID: 11996
Summary: opencv3 SIGILL on Cortex-A5 with VFPv4-D16
Product: buildroot
Version: 2019.02.3
Hardware: All
OS: Linux
Status: NEW
Severity: normal
Priority: P5
Component: Other
Assignee: unassigned at buildroot.uclibc.org
Reporter: jcowgill+busybox at jcowgill.uk
CC: buildroot at uclibc.org
Target Milestone: ---

I recently upgraded from 2018.02 to 2019.02 and discovered that opencv3 has broken on my board (it worked with 2018.02). The board has a Cortex-A5 with VFPv4-D16 enabled but without NEON and VFPv4 (so it only has 16 double floating point registers).

These are the target options I have set:
BR2_arm=y
BR2_BINFMT_ELF=y
BR2_cortex_a5=y
BR2_ARM_ENABLE_NEON=n
BR2_ARM_ENABLE_VFP=y
BR2_ARM_EABIHF=y
BR2_ARM_FPU_VFPV4D16=y
BR2_ARM_INSTRUCTIONS_THUMB2=y

I can see that my application gets a SIGILL because it tries to load a value into the d20 register which doesn't exist in VFPv4-D16.

0xb661d3d2 in cv::interpolateLanczos4 (coeffs=0xbef4602c, x=0.03125)
at
/home/jcowgill/workspace/bsp/buildroot/output/build/opencv3-3.4.3/modules/imgproc/src/imgwarp.cpp:176 176 /home/jcowgill/workspace/bsp/buildroot/output/build/opencv3-3.4.3/modules/imgproc/src/imgwarp.cpp:
No such file or directory.
(gdb) disassemble
...
0xb661d3cc <+796>: ldr r3, [pc, #240] ; (0xb661d4c0
<cv::initInterTab2D(int, bool)+1040>)

   0xb661d3ce <+798>:   vmov    s14, r8
=> 0xb661d3d2 <+802>:   vldr    d20, [r10]
   0xb661d3d6 <+806>:   vldr    d19, [r9]
   0xb661d3da <+810>:   add     r3, pc
   0xb661d3dc <+812>:   mov     r2, r7

========

I think this code in opencv3.mk is wrong (or maybe the ARM fpu config options are wrong):

ifeq ($(BR2_ARCH_IS_64):$(BR2_ARM_CPU_HAS_VFPV3),:y) OPENCV3_CONF_OPTS += -DENABLE_VFPV3=ON
else
OPENCV3_CONF_OPTS += -DENABLE_VFPV3=OFF endif

Apparently I have BR2_ARM_CPU_HAS_VFPV3 set (even though I only have the -D16 version), but when you pass -DENABLE_VFPV3=ON to OpenCV, it passes -mfpu=vfpv3 to the compiler and this is what's causing the SIGILL.

Signature · 2 lines
You are receiving this mail because:
You are on the CC list for the bug.
#1 lore raw

[Buildroot] [Bug 11996] opencv3 SIGILL on Cortex-A5 with VFPv4-D16

https://bugs.busybox.net/show_bug.cgi?id=11996

--- Comment #1 from Arnout Vandecappelle <[email protected]> ---

Review

Quoted 12 lines

I think this code in opencv3.mk is wrong (or maybe the ARM fpu config options are wrong):

ifeq ($(BR2_ARCH_IS_64):$(BR2_ARM_CPU_HAS_VFPV3),:y) OPENCV3_CONF_OPTS += -DENABLE_VFPV3=ON
else
OPENCV3_CONF_OPTS += -DENABLE_VFPV3=OFF endif

Apparently I have BR2_ARM_CPU_HAS_VFPV3 set (even though I only have the

(quoted diff) +0 -2
-D16 version), but when you pass -DENABLE_VFPV3=ON to OpenCV, it passes 
-mfpu=vfpv3 to the compiler and this is what's causing the SIGILL.

If the only thing that ENABLE_VFPV3 does is to set the -mfpu=... option, then we should just always set it to OFF, because we already pass that option.

Same for NEON, by the way.

Signature · 2 lines
You are receiving this mail because:
You are on the CC list for the bug.
#2 lore raw

[Buildroot] [Bug 11996] opencv3 SIGILL on Cortex-A5 with VFPv4-D16

In reply to bugzilla at busybox.net

https://bugs.busybox.net/show_bug.cgi?id=11996

--- Comment #2 from Thomas Petazzoni <[email protected]> --- So, I did a bit more research on this, but couldn't come to a useful conclusion. The handling of CPU optimizations in opencv3 is complicated. ENABLE_VFPV3 and ENABLE_NEON seem to be obsolete options, but the replacement CPU_BASELINE is not very clear.

With a VFPv3-D16 case, I get:

-- CPU/HW features:
-- Baseline:

--       requested:                 DETECT
--       disabled:                  VFPV3 NEON

which is quite expected.

When NEON is enabled as the FPU, I get:

-- CPU/HW features:

--     Baseline:                    NEON
--       requested:                 DETECT
--       disabled:                  VFPV3 NEON

So NEON seems to be detected, but it's also listed in the "disabled" features... which doesn't make a lot of sense.

Then, if I use VFPv3 as the FPU, I get:

-- CPU/HW features:
-- Baseline:

--       requested:                 DETECT
--       disabled:                  VFPV3 NEON

I.e, it doesn't detect that I have VFPv3.

(Of course the tests above are after removing ENABLE_VFPV3/ENABLE_NEON).

Signature · 2 lines
You are receiving this mail because:
You are on the CC list for the bug.
#3 lore raw

[Buildroot] [Bug 11996] opencv3 SIGILL on Cortex-A5 with VFPv4-D16

In reply to bugzilla at busybox.net

https://bugs.busybox.net/show_bug.cgi?id=11996

Thomas Petazzoni <[email protected]> changed:

What |Removed |Added
----------------------------------------------------------------------------

             Status|NEW                         |RESOLVED
         Resolution|---                         |FIXED

--- Comment #3 from Thomas Petazzoni <[email protected]> --- Fixed by
https://git.buildroot.org/buildroot/commit/?id=4d0f3dd8709f9f31a81eaa8a62cb35ef6ddddeaf

Signature · 2 lines
You are receiving this mail because:
You are on the CC list for the bug.