From: Junjie Cao <junjie.cao@linux.dev>
To: shrihari.s@samsung.com
Cc: jic23@kernel.org, linux-cxl@vger.kernel.org,
linux-pci@vger.kernel.org, qemu-devel@nongnu.org,
anisa.su887@gmail.com, mst@redhat.com,
marcel.apfelbaum@gmail.com, arun.george@samsung.com,
cpgs@samsung.com, dave@stgolabs.net, dongjoo.seo1@samsung.com,
gost.dev@samsung.com, s.neeraj@samsung.com,
vikash.k5@samsung.com, anisa.su@samsung.com,
junjie.cao@linux.dev
Subject: Re: [PATCH qemu v1 3/8] hw/cxl: Wire UIO capability into HDM decoder and DVSEC registers
Date: Wed, 4 Nov 2026 11:25:24 -0600 [thread overview]
Message-ID: <20261104172527.2172366-1-junjie.cao@linux.dev> (raw)
In-Reply-To: <1384010761.101793792283622.JavaMail.epsvc@epcpadp2new>
Hi Shrihari,
The switch/HB write mask opens UIG and UIW, but leaves ISP read-only.
CXL 3.2 section 8.2.4.20.7 makes ISP writable on a UIO-capable switch
or Host Bridge too.
With two UIO-enabled host bridges in a two-way CFMWS (4 KiB granularity),
the switch below the second HB advertises UIO, but writing decoder 0
Control with UIO=1, UIG=4, UIW=1 (two ways), ISP=1 reads back
0x00144000 rather than 0x01144000. Software cannot configure the second
position in the interleave set.
Could R_CXL_HDM_DECODER0_CTRL_ISP_MASK be included alongside UIG and UIW
in that switch/HB branch?
It may also be worth diagnosing a partially configured host bridge:
with x-uio=on and two RPs, one without SVC/flit, the HB reads 0x22382
but that RP reads 0x312. The HB UIO bit is defined to cover all its
root ports (8.2.4.20.1). A warning would make this inconsistent setup
easier to spot.
Many thanks,
Junjie
next prev parent reply other threads:[~2026-11-04 17:25 UTC|newest]
Thread overview: 12+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <CGME20261104113014epcas5p1dbf4b09e2849d28b46eb39db7990a54f@epcas5p1.samsung.com>
[not found] ` <20261104115242.2035807-1-shrihari.s@samsung.com>
2026-11-04 11:52 ` [PATCH qemu v1 1/8] hw/pci: Refactor flitmode from PCIESlot to PCIEPort Shrihari E S
2026-11-04 11:52 ` [PATCH qemu v1 2/8] hw/pci: Move 'x-256b-flit' property from cxl_root_port to pcie_root_port Shrihari E S
2026-11-04 11:52 ` [PATCH qemu v1 3/8] hw/cxl: Wire UIO capability into HDM decoder and DVSEC registers Shrihari E S
2026-11-04 17:25 ` Junjie Cao [this message]
2026-11-04 11:52 ` [PATCH qemu v1 4/8] hw/pci: Add PCIe Streamlined Virtual Channel (SVC) capability Shrihari E S
2026-11-04 17:25 ` Junjie Cao
2026-11-04 11:52 ` [PATCH qemu v1 5/8] hw/pci: Wire SVC initialization into port realize functions Shrihari E S
2026-11-04 17:25 ` Junjie Cao
2026-11-04 11:52 ` [PATCH qemu v1 6/8] hw/pci: Add PCIe Device3 capability support Shrihari E S
2026-11-04 11:52 ` [PATCH qemu v1 7/8] hw/cxl: Wire SVC and Dev3 capabilities to CXL Type 3 device Shrihari E S
2026-11-04 17:25 ` Junjie Cao
2026-11-04 11:52 ` [PATCH qemu v1 8/8] cxl: Add documentation for CXL UIO support Shrihari E S
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=20261104172527.2172366-1-junjie.cao@linux.dev \
--to=junjie.cao@linux.dev \
--cc=anisa.su887@gmail.com \
--cc=anisa.su@samsung.com \
--cc=arun.george@samsung.com \
--cc=cpgs@samsung.com \
--cc=dave@stgolabs.net \
--cc=dongjoo.seo1@samsung.com \
--cc=gost.dev@samsung.com \
--cc=jic23@kernel.org \
--cc=linux-cxl@vger.kernel.org \
--cc=linux-pci@vger.kernel.org \
--cc=marcel.apfelbaum@gmail.com \
--cc=mst@redhat.com \
--cc=qemu-devel@nongnu.org \
--cc=s.neeraj@samsung.com \
--cc=shrihari.s@samsung.com \
--cc=vikash.k5@samsung.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