From: sashiko-bot@kernel.org
To: "Aaron Lewis" <aaronlewis@google.com>, kvm@vger.kernel.org
Subject: Re: [PATCH v4 5/5] vfio: selftests: Allow a size for vfio_dma_mapping_perf_test
Date: Sat, 14 Nov 2026 13:29:46 +0000 [thread overview]
Message-ID: <sashiko-outbox-214571@kernel.org> (raw)
In-Reply-To: <20261114132200.113697-6-aaronlewis@google.com>
Thank you for your contribution! Sashiko AI review found 1 potential
issue to consider:
- [Low] vfio: selftests: non-power-of-2 -b size aborts in
iova_allocator_alloc()
--
commit 7eb32375cc90a8bc43da0741f8174df52036eeb6
Author: Aaron Lewis <aaronlewis@google.com>
vfio: selftests: Allow a size for vfio_dma_mapping_perf_test
This commit adds a '-b' command-line option to vfio_dma_mapping_perf_test to
allow specifying the DMA region size to map and unmap.
> diff --git a/tools/testing/selftests/vfio/vfio_dma_mapping_perf_test.c b/tools/testing/selftests/vfio/vfio_dma_mapping_perf_test.c
> index 086a63a50df80..6e38cc134a49c 100644
> --- a/tools/testing/selftests/vfio/vfio_dma_mapping_perf_test.c
> +++ b/tools/testing/selftests/vfio/vfio_dma_mapping_perf_test.c
[ ... ]
> @@ -58,11 +136,16 @@ FIXTURE_TEARDOWN(vfio_dma_mapping_perf_test)
>
> TEST_F(vfio_dma_mapping_perf_test, dma_map_unmap)
> {
> - const u64 size = SZ_1G;
> + const u64 page_size = variant->page_size ?: getpagesize();
> + const u64 size = test_params.size;
> const int flags = variant->mmap_flags;
> struct dma_region region;
>
> - ksft_print_msg("mmap size = %lluG\n", (unsigned long long)(size / SZ_1G));
> + if (!IS_ALIGNED(size, page_size))
> + SKIP(return, "Size 0x%llx is not aligned to page size 0x%llx",
> + (unsigned long long)size, (unsigned long long)page_size);
> +
> + ksft_print_msg("mmap size = %s\n", test_params.size_str);
>
> TIME("mmap",
> region.vaddr = mmap(NULL, size, PROT_READ | PROT_WRITE, flags, -1, 0));
[Severity: Low]
Will passing a non-power-of-2 size such as "-b 3G" abort the test inside
iova_allocator_alloc()?
parse_size() accepts any positive multiplier and both dma_map_unmap and
dma_map_unmap_from_file only check IS_ALIGNED(size, page_size), but after
populating the mapping they pass size directly to iova_allocator_alloc(),
which requires a power of two:
tools/testing/selftests/vfio/lib/iova_allocator.c:iova_allocator_alloc() {
VFIO_ASSERT_GT(size, 0, "Invalid size arg, zero\n");
VFIO_ASSERT_EQ(size & (size - 1), 0, "Invalid size arg, non-power-of-2\n");
...
}
--
Sashiko AI review · https://sashiko.dev/#/patchset/20261114132200.113697-1-aaronlewis@google.com?part=5
prev parent reply other threads:[~2026-11-14 13:29 UTC|newest]
Thread overview: 7+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-11-14 13:21 [PATCH v4 0/5] Introduce vfio_dma_mapping_perf_test Aaron Lewis
2026-11-14 13:21 ` [PATCH v4 1/5] vfio: selftests: Assert the region was unmapped in iommu_unmap() Aaron Lewis
2026-11-14 13:21 ` [PATCH v4 2/5] vfio: selftests: Introduce vfio_dma_mapping_perf_test Aaron Lewis
2026-11-14 13:21 ` [PATCH v4 3/5] vfio: selftests: Add memfd test to vfio_dma_mapping_perf_test Aaron Lewis
2026-11-14 13:21 ` [PATCH v4 4/5] selftests: harness: Support custom command-line options Aaron Lewis
2026-11-14 13:22 ` [PATCH v4 5/5] vfio: selftests: Allow a size for vfio_dma_mapping_perf_test Aaron Lewis
2026-11-14 13:29 ` sashiko-bot [this message]
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=sashiko-outbox-214571@kernel.org \
--to=sashiko-bot@kernel.org \
--cc=aaronlewis@google.com \
--cc=kvm@vger.kernel.org \
--cc=sashiko-reviews@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