[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <20191008153924.GA20318@infradead.org>
Date: Tue, 8 Oct 2019 08:39:24 -0700
From: Christoph Hellwig <hch@...radead.org>
To: Anup Patel <anup@...infault.org>
Cc: Christoph Hellwig <hch@...radead.org>,
Albert Ou <aou@...s.berkeley.edu>,
Alan Kao <alankao@...estech.com>,
Greg Kroah-Hartman <gregkh@...uxfoundation.org>,
Palmer Dabbelt <palmer@...ive.com>,
"linux-kernel@...r.kernel.org List" <linux-kernel@...r.kernel.org>,
Mike Rapoport <rppt@...ux.ibm.com>,
Atish Patra <atish.patra@....com>, Gary Guo <gary@...yguo.net>,
Paul Walmsley <paul.walmsley@...ive.com>,
linux-riscv <linux-riscv@...ts.infradead.org>,
Thomas Gleixner <tglx@...utronix.de>,
Allison Randal <allison@...utok.net>
Subject: Re: [PATCH v2 3/3] RISC-V: Move SBI related macros under uapi.
On Thu, Oct 03, 2019 at 11:00:05AM +0530, Anup Patel wrote:
> These defines are indeed part of KVM userspace API because we will
> be forwarding SBI calls not handled by KVM RISC-V kernel module to
> KVM userspace (QEMU/KVMTOOL). The forwarded SBI call details
> are passed to userspace via "struct kvm_run" of KVM_RUN ioctl.
At best your are passing through a hardware interface. We don't expose
e.g. the nvme headers to userspace either. We keep the headers clean
enough that userspace can copy them (and a few projects do), but they
really are not a kernel interface in any classic way.
Powered by blists - more mailing lists