xen-devel.lists.xenproject.org archive mirror
 help / color / mirror / Atom feed
From: Paul Durrant <xadimgnik@gmail.com>
To: "'Oleksandr'" <olekstysh@gmail.com>, <xen-devel@lists.xenproject.org>
Cc: "'Oleksandr Tyshchenko'" <oleksandr_tyshchenko@epam.com>,
	"'Jan Beulich'" <jbeulich@suse.com>,
	"'Andrew Cooper'" <andrew.cooper3@citrix.com>,
	"'Roger Pau Monné'" <roger.pau@citrix.com>,
	"'Julien Grall'" <julien@xen.org>,
	"'Stefano Stabellini'" <sstabellini@kernel.org>,
	"'Wei Liu'" <wl@xen.org>, "'Julien Grall'" <julien.grall@arm.com>
Subject: RE: [PATCH V2 01/23] x86/ioreq: Prepare IOREQ feature for making it common
Date: Wed, 4 Nov 2020 09:59:35 -0000	[thread overview]
Message-ID: <007501d6b291$34160a30$9c421e90$@xen.org> (raw)
In-Reply-To: <3dd55087-0c07-c9f3-e80a-8b136c226475@gmail.com>

> -----Original Message-----
> From: Oleksandr <olekstysh@gmail.com>
> Sent: 04 November 2020 09:06
> To: paul@xen.org; xen-devel@lists.xenproject.org
> Cc: 'Oleksandr Tyshchenko' <oleksandr_tyshchenko@epam.com>; 'Jan Beulich' <jbeulich@suse.com>; 'Andrew
> Cooper' <andrew.cooper3@citrix.com>; 'Roger Pau Monné' <roger.pau@citrix.com>; 'Julien Grall'
> <julien@xen.org>; 'Stefano Stabellini' <sstabellini@kernel.org>; 'Wei Liu' <wl@xen.org>; 'Julien
> Grall' <julien.grall@arm.com>
> Subject: Re: [PATCH V2 01/23] x86/ioreq: Prepare IOREQ feature for making it common
> 
> 
> On 20.10.20 10:13, Paul Durrant wrote:
> 
> Hi Paul.
> 
> Sorry for the late response.
> 
> >> +
> >> +/* Called when target domain is paused */
> >> +static inline void arch_hvm_destroy_ioreq_server(struct hvm_ioreq_server *s)
> >> +{
> >> +    p2m_set_ioreq_server(s->target, 0, s);
> >> +}
> >> +
> >> +/*
> >> + * Map or unmap an ioreq server to specific memory type. For now, only
> >> + * HVMMEM_ioreq_server is supported, and in the future new types can be
> >> + * introduced, e.g. HVMMEM_ioreq_serverX mapped to ioreq server X. And
> >> + * currently, only write operations are to be forwarded to an ioreq server.
> >> + * Support for the emulation of read operations can be added when an ioreq
> >> + * server has such requirement in the future.
> >> + */
> >> +static inline int hvm_map_mem_type_to_ioreq_server(struct domain *d,
> >> +                                                   ioservid_t id,
> >> +                                                   uint32_t type,
> >> +                                                   uint32_t flags)
> >> +{
> >> +    struct hvm_ioreq_server *s;
> >> +    int rc;
> >> +
> >> +    if ( type != HVMMEM_ioreq_server )
> >> +        return -EINVAL;
> >> +
> >> +    if ( flags & ~XEN_DMOP_IOREQ_MEM_ACCESS_WRITE )
> >> +        return -EINVAL;
> >> +
> >> +    spin_lock_recursive(&d->arch.hvm.ioreq_server.lock);
> >> +
> >> +    s = get_ioreq_server(d, id);
> >> +
> >> +    rc = -ENOENT;
> >> +    if ( !s )
> >> +        goto out;
> >> +
> >> +    rc = -EPERM;
> >> +    if ( s->emulator != current->domain )
> >> +        goto out;
> >> +
> >> +    rc = p2m_set_ioreq_server(d, flags, s);
> >> +
> >> + out:
> >> +    spin_unlock_recursive(&d->arch.hvm.ioreq_server.lock);
> >> +
> >> +    if ( rc == 0 && flags == 0 )
> >> +    {
> >> +        struct p2m_domain *p2m = p2m_get_hostp2m(d);
> >> +
> >> +        if ( read_atomic(&p2m->ioreq.entry_count) )
> >> +            p2m_change_entry_type_global(d, p2m_ioreq_server, p2m_ram_rw);
> >> +    }
> >> +
> >> +    return rc;
> >> +}
> >> +
> > The above doesn't really feel right to me. It's really an entry point into the ioreq server code and
> as such I think it ought to be left in the common code. I suggest replacing the p2m_set_ioreq_server()
> function with an arch specific function (also taking the type) which you can then implement here.
> 
> Agree that it ought to be left in the common code.
> 
> However, I am afraid I didn't entirely get you suggestion how this
> function could be split. On Arm struct p2m_domain doesn't contain IOREQ
> fields (p2m->ioreq.XXX), nor p2m_change_entry_type_global() is used, so
> they should be abstracted together with p2m_set_ioreq_server().
> 
> So the whole "if ( rc == 0 && flags == 0 )" check should be folded into
> arch_p2m_set_ioreq_server implementation on x86? This in turn raises a
> question can we put a spin_unlock after.
> 

Hi Oleksandr,

I think the code as it stands is really a bit of a layering violation. I don't really see a problem with retaining the ioreq server lock around the call to p2m_change_entry_type_global() so I'd just fold that into p2m_set_ioreq_server().

  Paul



  reply	other threads:[~2020-11-04  9:59 UTC|newest]

Thread overview: 109+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-10-15 16:44 [PATCH V2 00/23] IOREQ feature (+ virtio-mmio) on Arm Oleksandr Tyshchenko
2020-10-15 16:44 ` [PATCH V2 01/23] x86/ioreq: Prepare IOREQ feature for making it common Oleksandr Tyshchenko
2020-10-20  7:13   ` Paul Durrant
2020-11-04  9:06     ` Oleksandr
2020-11-04  9:59       ` Paul Durrant [this message]
2020-11-12 10:58   ` Jan Beulich
2020-11-13 11:09     ` Oleksandr
2020-11-13 11:20       ` Jan Beulich
2020-11-13 12:45         ` Oleksandr
2020-11-13 14:31           ` Jan Beulich
2020-10-15 16:44 ` [PATCH V2 02/23] xen/ioreq: Make x86's IOREQ feature common Oleksandr Tyshchenko
2020-10-20  7:57   ` Paul Durrant
2020-11-10 16:45     ` Oleksandr
2020-11-17 14:47       ` Oleksandr
2020-11-17 15:29         ` Paul Durrant
2020-11-17 16:29           ` Oleksandr
2020-11-17 19:43             ` Paul Durrant
2020-11-12 11:11   ` Jan Beulich
2020-11-13 13:11     ` Oleksandr
2020-10-15 16:44 ` [PATCH V2 03/23] xen/ioreq: Make x86's hvm_ioreq_needs_completion() common Oleksandr Tyshchenko
2020-10-20  7:59   ` Paul Durrant
2020-10-15 16:44 ` [PATCH V2 04/23] xen/ioreq: Provide alias for the handle_mmio() Oleksandr Tyshchenko
2020-10-20  9:14   ` Paul Durrant
2020-10-20 10:05     ` Jan Beulich
2020-10-20 10:38       ` Paul Durrant
2020-11-10 19:44         ` Oleksandr
2020-11-11  7:27           ` Jan Beulich
2020-11-11  8:09             ` Oleksandr
2020-11-11  8:16               ` Jan Beulich
2020-10-15 16:44 ` [PATCH V2 05/23] xen/ioreq: Make x86's hvm_mmio_first(last)_byte() common Oleksandr Tyshchenko
2020-10-20  9:15   ` Paul Durrant
2020-10-15 16:44 ` [PATCH V2 06/23] xen/ioreq: Make x86's hvm_ioreq_(page/vcpu/server) structs common Oleksandr Tyshchenko
2020-10-15 16:44 ` [PATCH V2 07/23] xen/ioreq: Move x86's ioreq_gfn(server) to struct domain Oleksandr Tyshchenko
2020-11-12 11:21   ` Jan Beulich
2020-11-13 14:05     ` Oleksandr
2020-11-18 12:09   ` Oleksandr
2020-11-18 12:29     ` Paul Durrant
2020-10-15 16:44 ` [PATCH V2 08/23] xen/ioreq: Introduce ioreq_params to abstract accesses to arch.hvm.params Oleksandr Tyshchenko
2020-10-20 10:41   ` Paul Durrant
2020-11-10 20:00     ` Oleksandr
2020-10-15 16:44 ` [PATCH V2 09/23] xen/dm: Make x86's DM feature common Oleksandr Tyshchenko
2020-11-12 11:32   ` Jan Beulich
2020-11-13 14:28     ` Oleksandr
2020-10-15 16:44 ` [PATCH V2 10/23] xen/mm: Make x86's XENMEM_resource_ioreq_server handling common Oleksandr Tyshchenko
2020-11-12 11:40   ` Jan Beulich
2020-11-13 15:00     ` Oleksandr
2020-10-15 16:44 ` [PATCH V2 11/23] xen/ioreq: Move x86's io_completion/io_req fields to struct vcpu Oleksandr Tyshchenko
2020-10-20 10:55   ` Paul Durrant
2020-11-10 20:59     ` Oleksandr
2020-11-11  8:04   ` Jan Beulich
2020-11-11  8:14     ` Oleksandr
2020-10-15 16:44 ` [PATCH V2 12/23] xen/ioreq: Remove "hvm" prefixes from involved function names Oleksandr Tyshchenko
2020-11-12 11:45   ` Jan Beulich
2020-11-12 12:14     ` Paul Durrant
2020-11-13 15:53     ` Oleksandr
2020-11-23 14:39       ` Oleksandr
2020-11-23 14:56         ` Jan Beulich
2020-11-23 15:47           ` Oleksandr
2020-11-23 15:54             ` Paul Durrant
2020-11-23 16:36               ` Oleksandr
2020-10-15 16:44 ` [PATCH V2 13/23] xen/ioreq: Use guest_cmpxchg64() instead of cmpxchg() Oleksandr Tyshchenko
2020-10-15 16:44 ` [PATCH V2 14/23] arm/ioreq: Introduce arch specific bits for IOREQ/DM features Oleksandr Tyshchenko
2020-11-12 11:48   ` Jan Beulich
2020-11-13 15:48     ` Oleksandr
2020-10-15 16:44 ` [PATCH V2 15/23] xen/arm: Stick around in leave_hypervisor_to_guest until I/O has completed Oleksandr Tyshchenko
2020-10-15 16:44 ` [PATCH V2 16/23] xen/mm: Handle properly reference in set_foreign_p2m_entry() on Arm Oleksandr Tyshchenko
2020-11-12 12:18   ` Jan Beulich
2020-11-13 17:00     ` Oleksandr
2020-10-15 16:44 ` [PATCH V2 17/23] xen/ioreq: Introduce domain_has_ioreq_server() Oleksandr Tyshchenko
2020-10-20 10:51   ` Paul Durrant
2020-11-10 20:53     ` Oleksandr
2020-11-11  8:08       ` Jan Beulich
2020-11-11  8:41         ` Oleksandr
2020-11-11 13:27           ` Jan Beulich
2020-11-11 16:28             ` Paul Durrant
2020-11-11 17:33               ` Oleksandr
2020-11-11 17:31             ` Oleksandr
2020-10-15 16:44 ` [PATCH V2 18/23] xen/dm: Introduce xendevicemodel_set_irq_level DM op Oleksandr Tyshchenko
2020-10-15 16:44 ` [PATCH V2 19/23] xen/arm: io: Abstract sign-extension Oleksandr Tyshchenko
2020-10-15 16:44 ` [PATCH V2 20/23] xen/ioreq: Make x86's send_invalidate_req() common Oleksandr Tyshchenko
2020-11-12 11:55   ` Jan Beulich
2020-11-13 16:03     ` Oleksandr
2020-10-15 16:44 ` [PATCH V2 21/23] xen/arm: Add mapcache invalidation handling Oleksandr Tyshchenko
2020-10-16  6:29   ` Jan Beulich
2020-10-16  8:41     ` Julien Grall
2020-10-16  8:56       ` Jan Beulich
2020-11-11  0:03       ` Oleksandr
2020-11-11 19:27         ` Julien Grall
2020-11-11 19:42           ` Oleksandr
2020-10-15 16:44 ` [PATCH V2 22/23] libxl: Introduce basic virtio-mmio support on Arm Oleksandr Tyshchenko
2020-10-15 16:44 ` [PATCH V2 23/23] [RFC] libxl: Add support for virtio-disk configuration Oleksandr Tyshchenko
2020-11-09  6:45   ` Wei Chen
2020-11-11  0:53     ` Oleksandr
2020-11-11  4:28       ` Wei Chen
2020-11-13 17:38         ` Oleksandr
2020-10-29  7:41 ` [PATCH V2 00/23] IOREQ feature (+ virtio-mmio) on Arm Masami Hiramatsu
2020-10-29 18:48   ` Oleksandr Tyshchenko
2020-10-29 19:53     ` Stefano Stabellini
2020-10-29 21:13       ` Oleksandr Tyshchenko
2020-10-29 21:34         ` Stefano Stabellini
2020-10-30 11:34         ` Masami Hiramatsu
2020-10-31 21:10           ` Oleksandr Tyshchenko
2020-11-02  7:23             ` Wei Chen
2020-11-02 18:05               ` Oleksandr
2020-11-03 14:30             ` Masami Hiramatsu
2020-11-03 21:09               ` Oleksandr
2020-10-29 20:03     ` Alex Bennée
2020-10-29 20:10       ` Stefano Stabellini
2020-10-29 22:06       ` Oleksandr Tyshchenko

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to='007501d6b291$34160a30$9c421e90$@xen.org' \
    --to=xadimgnik@gmail.com \
    --cc=andrew.cooper3@citrix.com \
    --cc=jbeulich@suse.com \
    --cc=julien.grall@arm.com \
    --cc=julien@xen.org \
    --cc=oleksandr_tyshchenko@epam.com \
    --cc=olekstysh@gmail.com \
    --cc=paul@xen.org \
    --cc=roger.pau@citrix.com \
    --cc=sstabellini@kernel.org \
    --cc=wl@xen.org \
    --cc=xen-devel@lists.xenproject.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).