Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
Feedback-ID: i56a14606:Fastmail
User-Agent: Cyrus-JMAP/3.7.0-alpha0-935-ge4ccd4c47b-fm-20220914.001-ge4ccd4c4
Mime-Version: 1.0
Message-Id: <c4ee195a-cf1f-44f6-b3b6-241d04e01fb9@www.fastmail.com>
In-Reply-To: <alpine.DEB.2.21.2209220223080.29493@angie.orcam.me.uk>
References: <alpine.DEB.2.21.2209220223080.29493@angie.orcam.me.uk>
Date:   Fri, 23 Sep 2022 12:31:39 +0200
From:   "Arnd Bergmann" <arnd@arndb.de>
To:     "Maciej W. Rozycki" <macro@orcam.me.uk>,
        "Paul Walmsley" <paul.walmsley@sifive.com>,
        "Palmer Dabbelt" <palmer@dabbelt.com>,
        "Albert Ou" <aou@eecs.berkeley.edu>
Cc:     linux-riscv@lists.infradead.org, linux-kernel@vger.kernel.org
Subject: Re: [PATCH] RISC-V: Make port I/O string accessors actually work
Content-Type: text/plain
Precedence: bulk

On Thu, Sep 22, 2022, at 11:56 PM, Maciej W. Rozycki wrote:
> Fix port I/O string accessors such as `insb', `outsb', etc. which use 
> the physical PCI port I/O address rather than the corresponding memory 
> mapping to get at the requested location, which in turn breaks at least 
> accesses made by our parport driver to a PCIe parallel port such as:
>
> PCI parallel port detected: 1415:c118, I/O at 0x1000(0x1008), IRQ 20
> parport0: PC-style at 0x1000 (0x1008), irq 20, using FIFO 
> [PCSPP,TRISTATE,COMPAT,EPP,ECP]

The patch looks correct to me, but I'm curious: Are you actually
using a parport device on your system, or just testing it to
make it work?

> +#define outsb(addr, buffer, count) __outsb(PCI_IOBASE + (addr), 
> buffer, count)
> +#define outsw(addr, buffer, count) __outsw(PCI_IOBASE + (addr), 
> buffer, count)
> +#define outsl(addr, buffer, count) __outsl(PCI_IOBASE + (addr), 
> buffer, count)
> 

I don't see anything actually risc-v specific in these definitions,
and it would be great to make the asm-generic version do the
right thing here. As far as I can tell, the only difference is
the barriers in the risc-v version, and we should really have the
same in the generic code anyway.

     Arnd