From: Jonathan Cameron via <qemu-devel@nongnu.org>
To: "Michael S. Tsirkin" <mst@redhat.com>
Cc: <qemu-devel@nongnu.org>, Fan Ni <nifan.cxl@gmail.com>,
<linuxarm@huawei.com>, <linux-cxl@vger.kernel.org>
Subject: Re: [PATCH qemu] hw/pci-bridge/pxb-cxl: Drop RAS capability from host bridge.
Date: Thu, 15 Feb 2024 17:42:12 +0000 [thread overview]
Message-ID: <20240215174212.000057c9@Huawei.com> (raw)
In-Reply-To: <20240215111117-mutt-send-email-mst@kernel.org>
On Thu, 15 Feb 2024 11:11:47 -0500
"Michael S. Tsirkin" <mst@redhat.com> wrote:
> On Thu, Feb 15, 2024 at 03:52:06PM +0000, Jonathan Cameron wrote:
> > This CXL component isn't allowed to have a RAS capability.
> > Whilst this should be harmless as software is not expected to look
> > here, good to clean it up.
> >
> > Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
>
> Isn't this device migrateable? If yes you need compat
> handling.
Not yet. Migrating these is broken in lots of ways :(
Given they are functional emulation only I've never cared that much.
We'll need to fix this as part of adding support for virtualization
use cases which start to make sense when dynamic capacity
support lands in kernel + QEMU (probably later this year)
Jonathan
>
>
> > ---
> > I've been carrying this on my tree for quite a while.
> > This wasn't in previous fixes set because it's low priority and
> > rebasing it across cleanup series that followed those fixes was
> > too fiddly to bother.
> >
> > include/hw/cxl/cxl_component.h | 1 +
> > hw/cxl/cxl-component-utils.c | 21 +++++++++++++++++----
> > hw/pci-bridge/pci_expander_bridge.c | 2 +-
> > 3 files changed, 19 insertions(+), 5 deletions(-)
> >
> > diff --git a/include/hw/cxl/cxl_component.h b/include/hw/cxl/cxl_component.h
> > index 0e5d35c263..5012fab6f7 100644
> > --- a/include/hw/cxl/cxl_component.h
> > +++ b/include/hw/cxl/cxl_component.h
> > @@ -25,6 +25,7 @@ enum reg_type {
> > CXL2_TYPE3_DEVICE,
> > CXL2_LOGICAL_DEVICE,
> > CXL2_ROOT_PORT,
> > + CXL2_RC,
> > CXL2_UPSTREAM_PORT,
> > CXL2_DOWNSTREAM_PORT,
> > CXL3_SWITCH_MAILBOX_CCI,
> > diff --git a/hw/cxl/cxl-component-utils.c b/hw/cxl/cxl-component-utils.c
> > index 84ab503325..cd116c0401 100644
> > --- a/hw/cxl/cxl-component-utils.c
> > +++ b/hw/cxl/cxl-component-utils.c
> > @@ -297,6 +297,7 @@ void cxl_component_register_init_common(uint32_t *reg_state,
> > caps = 3;
> > break;
> > case CXL2_ROOT_PORT:
> > + case CXL2_RC:
> > /* + Extended Security, + Snoop */
> > caps = 5;
> > break;
> > @@ -326,8 +327,19 @@ void cxl_component_register_init_common(uint32_t *reg_state,
> > CXL_##reg##_REGISTERS_OFFSET); \
> > } while (0)
> >
> > + switch (type) {
> > + case CXL2_DEVICE:
> > + case CXL2_TYPE3_DEVICE:
> > + case CXL2_LOGICAL_DEVICE:
> > + case CXL2_ROOT_PORT:
> > + case CXL2_UPSTREAM_PORT:
> > + case CXL2_DOWNSTREAM_PORT:
> > init_cap_reg(RAS, 2, CXL_RAS_CAPABILITY_VERSION);
> > - ras_init_common(reg_state, write_msk);
> > + ras_init_common(reg_state, write_msk);
> > + break;
> > + default:
> > + break;
> > + }
> >
> > init_cap_reg(LINK, 4, CXL_LINK_CAPABILITY_VERSION);
> >
> > @@ -335,9 +347,10 @@ void cxl_component_register_init_common(uint32_t *reg_state,
> > return;
> > }
> >
> > - init_cap_reg(HDM, 5, CXL_HDM_CAPABILITY_VERSION);
> > - hdm_init_common(reg_state, write_msk, type);
> > -
> > + if (type != CXL2_ROOT_PORT) {
> > + init_cap_reg(HDM, 5, CXL_HDM_CAPABILITY_VERSION);
> > + hdm_init_common(reg_state, write_msk, type);
> > + }
> > if (caps < 5) {
> > return;
> > }
> > diff --git a/hw/pci-bridge/pci_expander_bridge.c b/hw/pci-bridge/pci_expander_bridge.c
> > index 535889f7c2..0411ad31ea 100644
> > --- a/hw/pci-bridge/pci_expander_bridge.c
> > +++ b/hw/pci-bridge/pci_expander_bridge.c
> > @@ -290,7 +290,7 @@ static void pxb_cxl_dev_reset(DeviceState *dev)
> > uint32_t *write_msk = cxl_cstate->crb.cache_mem_regs_write_mask;
> > int dsp_count = 0;
> >
> > - cxl_component_register_init_common(reg_state, write_msk, CXL2_ROOT_PORT);
> > + cxl_component_register_init_common(reg_state, write_msk, CXL2_RC);
> > /*
> > * The CXL specification allows for host bridges with no HDM decoders
> > * if they only have a single root port.
> > --
> > 2.39.2
>
next prev parent reply other threads:[~2024-02-15 17:42 UTC|newest]
Thread overview: 4+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-02-15 15:52 [PATCH qemu] hw/pci-bridge/pxb-cxl: Drop RAS capability from host bridge Jonathan Cameron via
2024-02-15 16:11 ` Michael S. Tsirkin
2024-02-15 17:42 ` Jonathan Cameron via [this message]
2024-03-12 15:30 ` Michael S. Tsirkin
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=20240215174212.000057c9@Huawei.com \
--to=qemu-devel@nongnu.org \
--cc=Jonathan.Cameron@Huawei.com \
--cc=linux-cxl@vger.kernel.org \
--cc=linuxarm@huawei.com \
--cc=mst@redhat.com \
--cc=nifan.cxl@gmail.com \
/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).