linux-kernel - Re: [PATCH] MMIO should have more priority then IO

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <20220628180919.GA1850423@bhelgaas>
Date:   Tue, 28 Jun 2022 13:09:19 -0500
From:   Bjorn Helgaas <helgaas@...nel.org>
To:     Ajay Kaher <akaher@...are.com>
Cc:     bhelgaas@...gle.com, tglx@...utronix.de, mingo@...hat.com,
        bp@...en8.de, dave.hansen@...ux.intel.com, x86@...nel.org,
        hpa@...or.com, linux-pci@...r.kernel.org,
        linux-kernel@...r.kernel.org, stable@...r.kernel.org,
        rostedt@...dmis.org, namit@...are.com, srivatsab@...are.com,
        srivatsa@...il.mit.edu, amakhalov@...are.com, anishs@...are.com,
        vsirnapalli@...are.com, er.ajay.kaher@...il.com,
        Matthew Wilcox <willy@...radead.org>
Subject: Re: [PATCH] MMIO should have more priority then IO

[+cc Matthew]

On Tue, Jun 28, 2022 at 09:59:21PM +0530, Ajay Kaher wrote:
> Port IO instructions (PIO) are less efficient than MMIO (memory
> mapped I/O). They require twice as many PCI accesses and PIO
> instructions are serializing. As a result, MMIO should be preferred
> when possible over PIO.
> 
> Bare metal test result
> 1 million reads using raw_pci_read() took:
> PIO: 0.433153 Sec.
> MMIO: 0.268792 Sec.
> 
> Virtual Machine test result
> 1 hundred thousand reads using raw_pci_read() took:
> PIO: 12.809 Sec.
> MMIO: took 8.517 Sec.
> 
> Signed-off-by: Ajay Kaher <akaher@...are.com>
> ---
>  arch/x86/pci/common.c          |  8 ++++----
>  1 files changed, 4 insertions(+), 4 deletions(-)
> 
> diff --git a/arch/x86/pci/common.c b/arch/x86/pci/common.c
> index 3507f456f..0b3383d9c 100644
> --- a/arch/x86/pci/common.c
> +++ b/arch/x86/pci/common.c
> @@ -40,20 +40,20 @@ const struct pci_raw_ops *__read_mostly raw_pci_ext_ops;
>  int raw_pci_read(unsigned int domain, unsigned int bus, unsigned int devfn,
>  						int reg, int len, u32 *val)
>  {
> +	if (raw_pci_ext_ops)
> +		return raw_pci_ext_ops->read(domain, bus, devfn, reg, len, val);
>  	if (domain == 0 && reg < 256 && raw_pci_ops)
>  		return raw_pci_ops->read(domain, bus, devfn, reg, len, val);
> -	if (raw_pci_ext_ops)
> -		return raw_pci_ext_ops->read(domain, bus, devfn, reg, len, val);
>  	return -EINVAL;

This organization of raw_pci_read() dates to b6ce068a1285 ("Change
pci_raw_ops to pci_raw_read/write"), by Matthew.  Cc'd him for
comment, since I think he considered the ordering at the time.

>  }
>  
>  int raw_pci_write(unsigned int domain, unsigned int bus, unsigned int devfn,
>  						int reg, int len, u32 val)
>  {
> +	if (raw_pci_ext_ops)
> +		return raw_pci_ext_ops->write(domain, bus, devfn, reg, len, val);
>  	if (domain == 0 && reg < 256 && raw_pci_ops)
>  		return raw_pci_ops->write(domain, bus, devfn, reg, len, val);
> -	if (raw_pci_ext_ops)
> -		return raw_pci_ext_ops->write(domain, bus, devfn, reg, len, val);
>  	return -EINVAL;
>  }
>  
> -- 
> 2.30.0
>