lkml.org 
[lkml]   [2011]   [Jul]   [1]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
SubjectRe: [PATCH v8 05/12] mmc: mmc_test: add test for non-blocking transfers
From
On 28 June 2011 10:11, Per Forlin <per.forlin@linaro.org> wrote:
> Add four tests for read and write performance per
> different transfer size, 4k to 4M.
>  * Read using blocking mmc request
>  * Read using non-blocking mmc request
>  * Write using blocking mmc request
>  * Write using non-blocking mmc request
>
> The host dirver must support pre_req() and post_req()
> in order to run the non-blocking test cases.
>
> Signed-off-by: Per Forlin <per.forlin@linaro.org>
> ---
>  drivers/mmc/card/mmc_test.c |  311 +++++++++++++++++++++++++++++++++++++++++-
>  1 files changed, 303 insertions(+), 8 deletions(-)
>
> diff --git a/drivers/mmc/card/mmc_test.c b/drivers/mmc/card/mmc_test.c
> +static int mmc_test_nonblock_transfer(struct mmc_test_card *test,
> +                                     struct scatterlist *sg, unsigned sg_len,
> +                                     unsigned dev_addr, unsigned blocks,
> +                                     unsigned blksz, int write, int count)
> +{
> +       struct mmc_request mrq1;
> +       struct mmc_command cmd1;
> +       struct mmc_command stop1;
> +       struct mmc_data data1;
> +
> +       struct mmc_request mrq2;
> +       struct mmc_command cmd2;
> +       struct mmc_command stop2;
> +       struct mmc_data data2;
> +
> +       struct mmc_test_async_req test_areq[2];
> +       struct mmc_async_req *done_areq;
> +       struct mmc_async_req *cur_areq = &test_areq[0].areq;
> +       struct mmc_async_req *other_areq = &test_areq[1].areq;
> +       int i;
> +       int ret;
> +
> +       test_areq[0].test = test;
> +       test_areq[1].test = test;
> +
> +       if (!test->card->host->ops->pre_req ||
> +               !test->card->host->ops->post_req)
> +               return -RESULT_UNSUP_HOST;
Remove this error check. It is fine to run this test without these
hooks, but there will be no performance gain compared to blocking
request.

...

> +/*
> + * Multiple blocking write 4k to 4 MB chunks
> + */
> +static int mmc_test_profile_mult_write_blocking_perf(struct mmc_test_card *test)
> +{
> +       unsigned int bs[] = {1 << 12, 1 << 13, 1 << 14, 1 << 15, 1 << 16,
> +                            1 << 17, 1 << 18, 1 << 19, 1 << 20, 1 << 22};
> +       struct mmc_test_multiple_rw test_data = {
> +               .bs = bs,
> +               .size = 128*1024*1024,
I got this comment from Linus W.
Use TEST_AREA_MAX_SIZE instead of hard coded 128*1024*1024. update all
relevant functions.

/Per
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/

\
 
 \ /
  Last update: 2011-07-01 15:31    [from the cache]
©2003-2011 Jasper Spaans