This ensures that it overrides glibc correctly for rv32 on dunfell

Signed-off-by: Khem Raj <raj.khem@gmail.com>

Signed-off-by: Alistair Francis <alistair.francis@wdc.com>

Kernel 5.8 already on stable state in linux-yocto
To support both kernel 5.8 and 5.4 for riscv32
renamed bbappend file

Signed-off-by: sungwon.pino <sungwon.pino@gmail.com>

Signed-off-by: Alistair Francis <alistair.francis@wdc.com>

No longer support Linux Standard Base (LSB) since version 3.0
core-image-lsb-sdk recipes are removed
So remove the related content

https://www.yoctoproject.org/docs/3.0/ref-manual/ref-manual.html#migration-3.0-lsb-support-removed



Signed-off-by: sungwon.pino <sungwon.pino@gmail.com>

Signed-off-by: Khem Raj <raj.khem@gmail.com>

Update the glibc RV32 to a more recent patch version. Remove some
backported patches that don't apply as we already have them merged.

Signed-off-by: Alistair Francis <alistair.francis@wdc.com>

Signed-off-by: sungwon.pino <sungwon.pino@gmail.com>

The list of people was generated with:

git log | grep Signed | sort | uniq

Also update the LICENSE file to point to the new file.

Signed-off-by: Alistair Francis <alistair.francis@wdc.com>

Signed-off-by: Alistair Francis <alistair.francis@wdc.com>

Patch has been moved to oe-core

Signed-off-by: Khem Raj <raj.khem@gmail.com>

OE-Core now has 1.14 since dunfell release which supports riscv64 out of
box

Signed-off-by: Khem Raj <raj.khem@gmail.com>

Fixes build failure on rv64

Signed-off-by: Khem Raj <raj.khem@gmail.com>

Signed-off-by: Alistair Francis <alistair.francis@wdc.com>

Update the RISC-V fork of OpenOCD to the SHA.

Signed-off-by: Alistair Francis <alistair.francis@wdc.com>

Signed-off-by: Alistair Francis <alistair.francis@wdc.com>

Fixes
NOTE: Multiple providers are available for virtual/bootloader (u-boot, u-boot-juno, u-boot-orangepi-i96, u-boot-poplar, u-boot-socfpga)
Consider defining a PREFERRED_PROVIDER entry to match virtual/bootloader

Signed-off-by: Khem Raj <raj.khem@gmail.com>

Signed-off-by: Khem Raj <raj.khem@gmail.com>

Define __NR_futex in terms for __NR_futex_time64 for rv32

Signed-off-by: Khem Raj <raj.khem@gmail.com>

Signed-off-by: Khem Raj <raj.khem@gmail.com>

Signed-off-by: Khem Raj <raj.khem@gmail.com>

libunwind is not yet ported to rv32

Signed-off-by: Khem Raj <raj.khem@gmail.com>

Not buildable yet

Signed-off-by: Khem Raj <raj.khem@gmail.com>

They still need porting to rv32

Signed-off-by: Khem Raj <raj.khem@gmail.com>

It is defined to 0.10+XXX in main recipe
Fixes
NOTE: preferred version riscv of openocd not available (for item openocd)
NOTE: versions of openocd available: 0.10+gitrAUTOINC+ded6799025 0.10+gitrAUTOINC+f93ede5401

Signed-off-by: Khem Raj <raj.khem@gmail.com>

This brings in needed riscv64 support

Signed-off-by: Khem Raj <raj.khem@gmail.com>

Fixes
../include/libunwind.h:29:3: error: #error "Unsupported arch"

Signed-off-by: Khem Raj <raj.khem@gmail.com>

The real issue is in binutils/ld inserting R_RISCV_NONE which is then
caught by textrel checker, there are binutils bugs to address this
[1] [2]

[1] https://sourceware.org/bugzilla/show_bug.cgi?id=24676
[2] https://sourceware.org/bugzilla/show_bug.cgi?id=24673



Signed-off-by: Khem Raj <raj.khem@gmail.com>

Signed-off-by: Khem Raj <raj.khem@gmail.com>

Signed-off-by: Khem Raj <raj.khem@gmail.com>

Signed-off-by: Khem Raj <raj.khem@gmail.com>

Build QA complains
ERROR: eudev-3.2.9-r0 do_package_qa: QA Issue: eudev: ELF binary /usr/bin/udevadm has relocations in .text
eudev: ELF binary /usr/libexec/udevadm has relocations in .text [textrel]

Signed-off-by: Khem Raj <raj.khem@gmail.com>

The rv32 patch is needed across all openssl 1.1.1 versions

Signed-off-by: Khem Raj <raj.khem@gmail.com>

Signed-off-by: Khem Raj <raj.khem@gmail.com>

Signed-off-by: Khem Raj <raj.khem@gmail.com>

it still is needed with mesa 20

Signed-off-by: Khem Raj <raj.khem@gmail.com>

We still use 1.13 for risv64, so cheat a little so we can re-use recipes
from oe-core to pretend to be 1.14

Signed-off-by: Khem Raj <raj.khem@gmail.com>

currently meta-riscv does not parse unless meta-oe is present.
It does not appear that anything explictly depends on openocd
in the layer.

Signed-off-by: Jeremy Puhlman <jpuhlman@mvista.com>

Signed-off-by: Khem Raj <raj.khem@gmail.com>

Signed-off-by: Alistair Francis <alistair.francis@wdc.com>