[Buildroot] 回复: openssh totaly can't use on riscv/arm64 with latest buildroot

[jiao shuaipeng]

Dear All,

Any one has ever used the openssh8.3 built by latest buildroot on arm64/riscv platform?

Dose the openssh need support/dependency from the linux kernel side ?

Thanks a lot in advance.

Joseph
________________________________
发件人: jiao shuaipeng <jiaoshuaipeng at outlook.com>
发送时间: 2020年8月7日 0:28
收件人: Thomas Petazzoni <thomas.petazzoni at bootlin.com>
抄送: buildroot at busybox.net <buildroot at busybox.net>; shuaipeng.jiao at montage-tech.com <shuaipeng.jiao at montage-tech.com>; Jiaoshuaipeng at yeah.net <Jiaoshuaipeng at yeah.net>
主题: 回复: [Buildroot] openssh totaly can't use on riscv/arm64 with latest buildroot

Dear Thomas,

Thanks for your reply.

I have ever tried to run "sshd --help" after linux is booting up inside the qemu. I see the similar phenomenon, the program is stuck.

So it should not "the lack of entropy in the random number generator ".

Thanks

________________________________
发件人: Thomas Petazzoni <thomas.petazzoni at bootlin.com>
发送时间: 2020年8月6日 23:12
收件人: jiao shuaipeng <jiaoshuaipeng at outlook.com>
抄送: buildroot at busybox.net <buildroot at busybox.net>; shuaipeng.jiao at montage-tech.com <shuaipeng.jiao at montage-tech.com>; jiaoshuaipeng at yeah.net <jiaoshuaipeng at yeah.net>
主题: Re: [Buildroot] openssh totaly can't use on riscv/arm64 with latest buildroot

On Thu, 6 Aug 2020 14:31:33 +0000
jiao shuaipeng <jiaoshuaipeng at outlook.com> wrote:

> Any one ever use openssh8.3 on RISCV/ARM64 with latest buildroot commit ?
>
> We found the the binary of openssh can generated successfully with the buildroot.
> But when we run the application sshd or "ssh-keygen -a", it will be stuck.
> I have used the GDB to find root cause why the program is stuch, i see the stach frame is corrupted.
>
> we are using the config "qemu_aarch64_virt_defconfig" and "qemu_riscv32_virt_defconfig", and enable the openssh with "make menucofig". qemu version is 5.0.50.  linux kernel version is 5.1.12.

Could this simply due to the lack of entropy in the random number
generator ?

Thomas
--
Thomas Petazzoni, CTO, Bootlin
Embedded Linux and Kernel engineering
https://bootlin.com
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.busybox.net/pipermail/buildroot/attachments/20200807/ed77da56/attachment-0002.html>