[Top] [All Lists]

Re: custom kernel on lemote-3a-itx (Loongson-3A) crashes in userspace

To: Mike Frysinger <>
Subject: Re: custom kernel on lemote-3a-itx (Loongson-3A) crashes in userspace
From: Huacai Chen <>
Date: Fri, 27 Feb 2015 21:37:44 +0800
Cc: Linux MIPS Mailing List <>
Dkim-signature: v=1; a=rsa-sha256; c=relaxed/relaxed;; s=20120113; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc:content-type; bh=8qvFtfeSRz36ZF2KelzU5ZhdBrAlrVtAWKuOZGBnQC0=; b=OLXbZcNAXwvLhZnqaWPNNjYwRVSkDc/GfGflEEAIa2StXHy8E/Uw3IdPV5JbHP181S v1ExlIzGH5UCnDHUzgEj6R8IiXluo60/x9PL0poce73qO0wOz6cxOUT//PIYD8zftrMZ RR2kGa4Wv/kT2v4tizrnfkR3qqMe5QJ8BDvqq6hzo7Xp7BIqL9Id8UPIl2ciCUl3brh6 iieeiIlaUnm8q2nAmi6wd2POzkLdCcGOW8AuSfBH5MiK19Fx8DabB53zgIbwY/v/WJG3 Ovb0p8JQvbeN+DJfbGHnnMpaCAQaPaujRmrOUj8bKQGiDXonbCHyiykPkTJBtESaKmY5 HspA==
In-reply-to: <20150226081425.GR6655@vapier>
List-archive: <>
List-help: <>
List-id: linux-mips <>
List-owner: <>
List-post: <>
List-software: Ecartis version 1.0.0
List-subscribe: <>
List-unsubscribe: <>
Original-recipient: rfc822;
References: <20150219194617.GT544@vapier> <> <20150226081425.GR6655@vapier>
We have also built good kernels with native GCC-4.6.4/4.7.2/4.8.3 (o32
binaries). So maybe this is an N32-related problems.


On Thu, Feb 26, 2015 at 4:14 PM, Mike Frysinger <> wrote:
> On 26 Feb 2015 15:47, Huacai Chen wrote:
>> Please try the toolchain here:
> thanks, but that's kind of dodgy and kind of defeats what i'm going for ;)
>  - looks like binary-only
>  - they're x86_64 binaries (i'm doing everything native here, so mips/n32)
>  - is pretty old (gcc-4.5 / gcc-4.6 / binutils-2.21 / binutils-2.22)
> that said, i installed Debian/mipsel(o32) into a chroot.  building with the 
> same
> sources & configs yielded a kernel that seems to be running OK -- i'm on
> mainline linux-3.19 now.  it too has an old toolchain: binutils-2.22 and
> gcc-4.6.4.  i guess there's a miscompilation / the kernel sources have 
> problems
> with newer gcc versions.
> installing older binutils in Gentoo was pretty easy, but trying to do gcc-4.6
> ran into an ICE while bootstrapping.  but at least it'll be easy to rule out
> weird assembler/linker problems.
> might try to put together a script to rebuild the objects one-by-one with the
> newer compiler to try and narrow down the problem in the source.
> -mike

<Prev in Thread] Current Thread [Next in Thread>