netdev - Re: [PATCH v3] vringh: IOMEM support

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]
Date:   Fri, 28 Apr 2023 11:00:04 +0900
From:   Shunsuke Mie <mie@...l.co.jp>
To:     Jason Wang <jasowang@...hat.com>
Cc:     "Michael S. Tsirkin" <mst@...hat.com>,
        Rusty Russell <rusty@...tcorp.com.au>, kvm@...r.kernel.org,
        virtualization@...ts.linux-foundation.org, netdev@...r.kernel.org,
        linux-kernel@...r.kernel.org
Subject: Re: [PATCH v3] vringh: IOMEM support

2023年4月26日(水) 15:10 Jason Wang <jasowang@...hat.com>:
>
> On Tue, Apr 25, 2023 at 6:23 PM Shunsuke Mie <mie@...l.co.jp> wrote:
> >
> > Introduce a new memory accessor for vringh. It is able to use vringh to
> > virtio rings located on io-memory region.
>
> Is there a user for this? It would be better if you can describe the
> use cases for this. Maybe you can post the user or at least a link to
> the git as a reference.
This is for the following patch.
https://lore.kernel.org/virtualization/20230427175244.GA261197@bhelgaas/T/#m30a258509caca2413a21f9a9ce0f5fd31d3bd006
I'd like to add a description of how this feature will be used in the
next version.
> >
> > Signed-off-by: Shunsuke Mie <mie@...l.co.jp>
> > ---
> >
> > Changes from v2: https://lore.kernel.org/virtualization/20230202090934.549556-1-mie@igel.co.jp/
> > - Focus on an adding io memory APIs
> > Remove vringh API unification and some fixes.
> > - Rebase on next-20230414
> >
> >  drivers/vhost/Kconfig  |   6 ++
> >  drivers/vhost/vringh.c | 129 +++++++++++++++++++++++++++++++++++++++++
> >  include/linux/vringh.h |  33 +++++++++++
> >  3 files changed, 168 insertions(+)
> >
> > diff --git a/drivers/vhost/Kconfig b/drivers/vhost/Kconfig
> > index b455d9ab6f3d..4b0dbb4a8ab3 100644
> > --- a/drivers/vhost/Kconfig
> > +++ b/drivers/vhost/Kconfig
> > @@ -6,6 +6,12 @@ config VHOST_IOTLB
> >           This option is selected by any driver which needs to support
> >           an IOMMU in software.
> >
> > +config VHOST_RING_IOMEM
> > +       tristate
> > +       select VHOST_IOMEM
> > +       help
> > +         This option enables vringh APIs to supports io memory space.
>
> There's no specific Kconfig for all the existing accessors. Any reason
> I/O memory is special or do you care about the size of the module?
I followed the IOTLB that is used for vhost and vringh. However, this code has
little effect on the code size and dependencies if included, so I
would like to remove
the Kconfig.
> > +
> >  config VHOST_RING
> >         tristate
> >         select VHOST_IOTLB
> > diff --git a/drivers/vhost/vringh.c b/drivers/vhost/vringh.c
> > index 955d938eb663..ce5a88eecc05 100644
> > --- a/drivers/vhost/vringh.c
> > +++ b/drivers/vhost/vringh.c
> > @@ -1604,4 +1604,133 @@ EXPORT_SYMBOL(vringh_need_notify_iotlb);
> >
> >  #endif
> >
>
> [...]
>
> >
> > base-commit: d3f2cd24819158bb70701c3549e586f9df9cee67
> > prerequisite-patch-id: 760abbe8c981c52ccc421b8139e8999ab71619aa
> > prerequisite-patch-id: 99d8679ab4569545c8af401e84142c66312e953e
> > prerequisite-patch-id: aca81516aba75b58c8422d37c2dc7db2f61ffe92
> > prerequisite-patch-id: 3d76136200c4e55ba2c41681325f242859dd6dbd
> > prerequisite-patch-id: 47a994feb68d95412d81b0fd1fa27bc7ba05ae18
> > prerequisite-patch-id: a2f7fc3f35358f70b6dad4c919ce293b10295c4f
> > prerequisite-patch-id: 70e2ee32b945be96a0388f0ff564651ac9335220
> > prerequisite-patch-id: 2023690f9c47017b56d7f036332a5ca3ece6bde8
> > prerequisite-patch-id: 211e113fec6c450d13fbdb437ecfad67dec0a157
> > prerequisite-patch-id: f2bcd3168933886e4cd4c39e47446d1bd7cb2691
> > prerequisite-patch-id: 37b131560808733a0b8878e85a3d2a46d6ab02ca
> > prerequisite-patch-id: 79b0219a715cb5ace227d55666d62fdb2dcc6ffe
> > prerequisite-patch-id: 30f1740cd48a19aa1c3c93e625c740cae2845478
> > prerequisite-patch-id: 31989e4a521f2fc6f68c4ccdb6960035e87666a7
> > prerequisite-patch-id: 3948bb3e0c045e2ffff06a714d17bab16c94775d
> > prerequisite-patch-id: cf28e0115b9111bcb77aa9c710d98b2be93c7e89
> > prerequisite-patch-id: ebf2349c0ae1296663854eee2da0b43fe8972f9b
> > prerequisite-patch-id: fc570921d885a2a6000800b4022321e63f1650a5
> > prerequisite-patch-id: 1fd5219fef17c2bf2d76000207b25aae58c368f3
> > prerequisite-patch-id: 34e5f078202762fe69df471e97b51b1341cbdfa9
> > prerequisite-patch-id: 7fa5151b9e0488b48c2b9d1219152cfb047d6586
> > prerequisite-patch-id: 33cca272767af04ae9abe7af2f6cbb9972cc0b77
> > prerequisite-patch-id: bb1a6befc899dd97bcd946c2d76ce73675a1fa45
> > prerequisite-patch-id: 10be04dd92fa451d13676e91d9094b63cd7fbcf8
> > prerequisite-patch-id: 87b86eb4ce9501bba9c04ec81094ac9202392431
> > prerequisite-patch-id: a5ced28762bf6bd6419dae0e4413d02ccafd72c2
> > prerequisite-patch-id: 2db4c9603e00d69bb0184dabcc319e7f74f30305
> > prerequisite-patch-id: 41933f9d53e5e9e02efd6157b68ee7d92b10cfa2
> > prerequisite-patch-id: df3295b4cdde3a45eaf4c40047179698a4224d05
> > prerequisite-patch-id: 9e2fca9ab0ba2b935daa96f1745ff4c909792231
> > prerequisite-patch-id: 8948378099ba4d61e10a87e617d69ed2fc4104ae
> > prerequisite-patch-id: 5e7466f3f0d74880d1a574a1bd91b12091dcf3f5
> > prerequisite-patch-id: 902899e1cd53b7fcc7971f630aed103830fc3e3d
> > prerequisite-patch-id: 42126b180500f9ff123db78748972c6ece18ac57
> > prerequisite-patch-id: 5236a03ef574074f3c1009a52612051862b31eff
> > prerequisite-patch-id: adae1aa80df65bd02a9e3f4db490cf801c1c6119
> > prerequisite-patch-id: 22806fcabb973ee5f04ee6212db6161aab5bcbfc
> > prerequisite-patch-id: 6eb14cfdc2cf31e90556f6afe7361427a332e8dc
>
> These seem meaningless?
I'm sorry, that didn't make sense. I'll remove it.
> Thanks
>
> > --
> > 2.25.1
> >
>
Best regards,
Shunsuke