[<prev] [next>] [thread-next>] [day] [month] [year] [list]
Message-ID: <20250804170916.3212221-1-christian.bruel@foss.st.com>
Date: Mon, 4 Aug 2025 19:09:13 +0200
From: Christian Bruel <christian.bruel@...s.st.com>
To: <mani@...nel.org>, <kwilczynski@...nel.org>, <kishon@...nel.org>,
<arnd@...db.de>, <gregkh@...uxfoundation.org>, <shuah@...nel.org>
CC: <linux-pci@...r.kernel.org>, <linux-kernel@...r.kernel.org>,
<linux-kselftest@...r.kernel.org>,
Christian Bruel
<christian.bruel@...s.st.com>
Subject: [PATCH 0/3] Skip IRQ tests if irq legitimately is out of range
'pci_endpoint_test' fails for architectures allowing less than 32 MSI
registers and that doesnt support MSI-X, avoid reporting false errors
because of out-of-range irqs.
e.g for an EP configured with 8 msi_interrupts and no msix we can have
./pci_endpoint_test -t MSI_TEST
# PASSED: 1 / 1 tests passed.
# 1 skipped test(s) detected. Consider enabling relevant config options to improve coverage.
# Totals: pass:0 fail:0 xfail:0 xpass:0 skip:1 error:0
instead of
# FAILED: 0 / 1 tests passed
# Totals: pass:0 fail:1 xfail:0 xpass:0 skip:0 error:0
An alternative could have been to implement VARIANTs so that the harness
runs only the supported tests, but that seems quite heavy considering the
huge number of possible interrupts.
Another alternative could also have been to use a new ioctl to get the
allocated number of irqs from the driver, but that doesn't seem to be
more efficient than just using -EINVAL when the
irq is out of range.
thank you for your feedback
Christian Bruel (3):
misc: pci_endpoint_test: Skip IRQ tests if irq is out of range
misc: pci_endpoint_test: Cleanup extra 0 initialization
selftests: pci_endpoint: Skip IRQ test if irq is out of range.
drivers/misc/pci_endpoint_test.c | 14 ++++++--------
.../selftests/pci_endpoint/pci_endpoint_test.c | 4 ++++
2 files changed, 10 insertions(+), 8 deletions(-)
--
2.34.1
Powered by blists - more mailing lists