From: Bruce Richardson <bruce.richardson@intel.com>
To: Tejasree Kondoj <ktejasree@marvell.com>
Cc: Akhil Goyal <gakhil@marvell.com>,
Chengwen Feng <fengchengwen@huawei.com>,
Kevin Laatz <kevin.laatz@intel.com>,
Vidya Sagar Velumuri <vvelumuri@marvell.com>,
Anoob Joseph <anoobj@marvell.com>, <dev@dpdk.org>
Subject: Re: [PATCH 2/2] test/dma: add functions to verify zero and one fill
Date: Mon, 22 Jun 2026 10:27:28 +0100 [thread overview]
Message-ID: <ajkAAEyYhFu5xPP2@bricha3-mobl1.ger.corp.intel.com> (raw)
In-Reply-To: <20260622030623.86094-3-ktejasree@marvell.com>
On Mon, Jun 22, 2026 at 08:36:23AM +0530, Tejasree Kondoj wrote:
> Add test cases to verify zero fill and one fill
>
> Signed-off-by: Vidya Sagar Velumuri <vvelumuri@marvell.com>
> ---
> app/test/test.h | 4 +++
> app/test/test_dmadev.c | 57 ++++++++++++++++++++++++------------------
> 2 files changed, 37 insertions(+), 24 deletions(-)
>
Not a bad idea. One piece of feedback inline below. Otherwise:
Acked-by: Bruce Richardson <bruce.richardson@intel.com>
> diff --git a/app/test/test.h b/app/test/test.h
> index 1f12fc5397..5fc2d7d048 100644
> --- a/app/test/test.h
> +++ b/app/test/test.h
> @@ -28,6 +28,10 @@
>
> #include <rte_test.h>
>
> +#ifndef ARRAY_SIZE
> +#define ARRAY_SIZE(x) (sizeof(x) / sizeof((x)[0]))
> +#endif
> +
> #define TEST_ASSERT RTE_TEST_ASSERT
>
> #define TEST_ASSERT_EQUAL RTE_TEST_ASSERT_EQUAL
> diff --git a/app/test/test_dmadev.c b/app/test/test_dmadev.c
> index b30f2214e5..d4299e501d 100644
> --- a/app/test/test_dmadev.c
> +++ b/app/test/test_dmadev.c
> @@ -931,42 +931,51 @@ test_completion_handling(int16_t dev_id, uint16_t vchan)
> static int
> test_enqueue_fill(int16_t dev_id, uint16_t vchan)
> {
> + uint64_t pattern[3] = {0x0, 0xfedcba9876543210, 0xffffffffffffffff};
> const unsigned int lengths[] = {8, 64, 1024, 50, 100, 89};
> + unsigned int i, j, k;
> struct rte_mbuf *dst;
> char *dst_data;
> - uint64_t pattern = 0xfedcba9876543210;
> - unsigned int i, j;
>
> dst = rte_pktmbuf_alloc(pool);
> if (dst == NULL)
> ERR_RETURN("Failed to allocate mbuf\n");
> dst_data = rte_pktmbuf_mtod(dst, char *);
>
> - for (i = 0; i < RTE_DIM(lengths); i++) {
> - /* reset dst_data */
> - memset(dst_data, 0, rte_pktmbuf_data_len(dst));
> + for (k = 0; k < ARRAY_SIZE(pattern); k++) {
> + for (i = 0; i < RTE_DIM(lengths); i++) {
> + /* reset dst_data */
> + memset(dst_data, 0, rte_pktmbuf_data_len(dst));
> +
I don't think we should reset between each run, because of the zero-fill
operation, which we cannot detect if it runs properly. I'd suggest
therefore:
* zero the buffer outside the main loop
* Run the non-zero patterns first, checking each one, without reset
* Do the zero-pattern at the end, so we can see that zeroing works.
> + /* perform the fill operation */
> + int id = rte_dma_fill(dev_id, vchan, pattern[k],
> + rte_pktmbuf_iova(dst), lengths[i], RTE_DMA_OP_FLAG_SUBMIT);
> + if (id < 0) {
> + if (id == -ENOTSUP) {
> + rte_pktmbuf_free(dst);
> + return 0;
> + }
> + ERR_RETURN("Error with rte_dma_fill\n");
> + }
> + await_hw(dev_id, vchan);
>
> - /* perform the fill operation */
> - int id = rte_dma_fill(dev_id, vchan, pattern,
> - rte_pktmbuf_iova(dst), lengths[i], RTE_DMA_OP_FLAG_SUBMIT);
> - if (id < 0)
> - ERR_RETURN("Error with rte_dma_fill\n");
> - await_hw(dev_id, vchan);
> + if (rte_dma_completed(dev_id, vchan, 1, NULL, NULL) != 1)
> + ERR_RETURN("Error: fill operation failed (length: %u)\n",
> + lengths[i]);
> + /* check the data from the fill operation is correct */
> + for (j = 0; j < lengths[i]; j++) {
> + char pat_byte = ((char *)&pattern[k])[j % 8];
>
> - if (rte_dma_completed(dev_id, vchan, 1, NULL, NULL) != 1)
> - ERR_RETURN("Error: fill operation failed (length: %u)\n", lengths[i]);
> - /* check the data from the fill operation is correct */
> - for (j = 0; j < lengths[i]; j++) {
> - char pat_byte = ((char *)&pattern)[j % 8];
> - if (dst_data[j] != pat_byte)
> - ERR_RETURN("Error with fill operation (lengths = %u): got (%x), not (%x)\n",
> - lengths[i], dst_data[j], pat_byte);
> + if (dst_data[j] != pat_byte)
> + ERR_RETURN("Error with fill operation (lengths = %u): got (%x), not (%x)\n",
> + lengths[i], dst_data[j], pat_byte);
> + }
> + /* check that the data after the fill operation was not written to */
> + for (; j < rte_pktmbuf_data_len(dst); j++)
> + if (dst_data[j] != 0)
> + ERR_RETURN("Error, fill operation wrote too far (lengths = %u): got (%x), not (%x)\n",
> + lengths[i], dst_data[j], 0);
> }
> - /* check that the data after the fill operation was not written to */
> - for (; j < rte_pktmbuf_data_len(dst); j++)
> - if (dst_data[j] != 0)
> - ERR_RETURN("Error, fill operation wrote too far (lengths = %u): got (%x), not (%x)\n",
> - lengths[i], dst_data[j], 0);
> }
>
> rte_pktmbuf_free(dst);
> --
> 2.34.1
>
prev parent reply other threads:[~2026-06-22 9:27 UTC|newest]
Thread overview: 4+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-06-22 3:06 [PATCH 0/2] test/dma: improve dmadev tests Tejasree Kondoj
2026-06-22 3:06 ` [PATCH 1/2] test/dma: update the sg test to verify wrap around case Tejasree Kondoj
2026-06-22 3:06 ` [PATCH 2/2] test/dma: add functions to verify zero and one fill Tejasree Kondoj
2026-06-22 9:27 ` Bruce Richardson [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=ajkAAEyYhFu5xPP2@bricha3-mobl1.ger.corp.intel.com \
--to=bruce.richardson@intel.com \
--cc=anoobj@marvell.com \
--cc=dev@dpdk.org \
--cc=fengchengwen@huawei.com \
--cc=gakhil@marvell.com \
--cc=kevin.laatz@intel.com \
--cc=ktejasree@marvell.com \
--cc=vvelumuri@marvell.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