From: Alison Schofield <alison.schofield@intel.com>
To: nvdimm@lists.linux.dev, linux-cxl@vger.kernel.org
Cc: Dan Williams <dan.j.williams@intel.com>,
Alison Schofield <alison.schofield@intel.com>
Subject: [ndctl PATCH v2] test/cxl-dax-hmem.sh: validate dax_hmem vs CXL collisions
Date: Fri, 3 Apr 2026 19:51:21 -0700 [thread overview]
Message-ID: <20260404025123.2967169-1-alison.schofield@intel.com> (raw)
From: Dan Williams <dan.j.williams@intel.com>
Use the new "cxl_test.hmem_test" and "cxl_test.fail_autoassemble"
module options to implement a new cxl-dax-hmem.sh test. The test
checks dax_hmem takeover of Soft Reserve ranges that collide with
autoassembled CXL regions. It depends on the cxl_mock_mem driver
to launch multiple async probes before the dax_hmem driver attaches.
[as: do_skip on missing params, explicit param usage, robust unload,
check_dmesg, misc style]
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Tested-by: Alison Schofield <alison.schofield@intel.com>
---
Changes in v2:
- Add delay (cxl-list) between modprobe and dax lookup
In testing sometimes a daxctl right after modprobe fails to find
the devices.
test/cxl-dax-hmem.sh | 68 ++++++++++++++++++++++++++++++++++++++++++++
test/meson.build | 2 ++
2 files changed, 70 insertions(+)
create mode 100755 test/cxl-dax-hmem.sh
diff --git a/test/cxl-dax-hmem.sh b/test/cxl-dax-hmem.sh
new file mode 100755
index 000000000000..6f4ed5076870
--- /dev/null
+++ b/test/cxl-dax-hmem.sh
@@ -0,0 +1,68 @@
+#!/bin/bash
+# SPDX-License-Identifier: GPL-2.0
+# Copyright (C) 2026 Intel Corporation
+
+. $(dirname $0)/common
+
+rc=77
+
+set -ex
+
+trap 'err $LINENO' ERR
+
+check_prereq "jq"
+
+modinfo cxl_test | grep -q '^parm:.*hmem_test' || \
+ do_skip "cxl_test hmem_test module param not available"
+
+modinfo cxl_test | grep -q '^parm:.*fail_autoassemble' || \
+ do_skip "cxl_test fail_autoassemble module param not available"
+
+rc=1
+
+unload()
+{
+ modprobe -r dax_cxl 2>/dev/null || true
+ modprobe -r dax_hmem 2>/dev/null || true
+ modprobe -r cxl_mock_mem 2>/dev/null || true
+ modprobe -r cxl_test 2>/dev/null || true
+}
+
+find_dax_cxl()
+{
+ $DAXCTL list -R | jq -r \
+ '.[] | select(.path | contains("cxl_acpi.0")) | .path'
+}
+
+find_dax_hmem()
+{
+ $DAXCTL list -R | jq -r \
+ '.[] | select(.path | contains("hmem_platform.1")) | .path'
+}
+
+unload
+
+# Verify CXL autoassembly claims the Soft Reserve range before dax_hmem
+modprobe cxl_mock_mem
+modprobe cxl_test hmem_test=1
+$CXL list
+
+dax=$(find_dax_cxl)
+[[ -z "$dax" ]] && err $LINENO
+dax=$(find_dax_hmem)
+[[ -n "$dax" ]] && err $LINENO
+
+unload
+
+# Verify dax_hmem claims the Soft Reserve range when CXL autoassembly fails
+modprobe cxl_mock_mem
+modprobe cxl_test hmem_test=1 fail_autoassemble=1
+$CXL list
+
+dax=$(find_dax_cxl)
+[[ -n "$dax" ]] && err $LINENO
+dax=$(find_dax_hmem)
+[[ -z "$dax" ]] && err $LINENO
+
+unload
+check_dmesg "$LINENO"
diff --git a/test/meson.build b/test/meson.build
index 593edf552b36..4260a3fa4448 100644
--- a/test/meson.build
+++ b/test/meson.build
@@ -169,6 +169,7 @@ cxl_destroy_region = find_program('cxl-destroy-region.sh')
cxl_qos_class = find_program('cxl-qos-class.sh')
cxl_translate = find_program('cxl-translate.sh')
cxl_elc = find_program('cxl-elc.sh')
+cxl_dax_hmem = find_program('cxl-dax-hmem.sh')
tests = [
[ 'libndctl', libndctl, 'ndctl' ],
@@ -203,6 +204,7 @@ tests = [
[ 'cxl-qos-class.sh', cxl_qos_class, 'cxl' ],
[ 'cxl-translate.sh', cxl_translate, 'cxl' ],
[ 'cxl-elc.sh', cxl_elc, 'cxl' ],
+ [ 'cxl-dax-hmem.sh', cxl_dax_hmem, 'cxl' ],
]
if get_option('destructive').enabled()
--
2.37.3
next reply other threads:[~2026-04-04 2:51 UTC|newest]
Thread overview: 2+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-04-04 2:51 Alison Schofield [this message]
2026-04-06 15:22 ` [ndctl PATCH v2] test/cxl-dax-hmem.sh: validate dax_hmem vs CXL collisions Dave Jiang
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=20260404025123.2967169-1-alison.schofield@intel.com \
--to=alison.schofield@intel.com \
--cc=dan.j.williams@intel.com \
--cc=linux-cxl@vger.kernel.org \
--cc=nvdimm@lists.linux.dev \
/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