From: Vivek Sharma <vivek.sharma@caviumnetworks.com>
To: "cristian.dumitrescu@intel.com" <cristian.dumitrescu@intel.com>
Cc: "Sharma, Vivek" <Vivek.Sharma@cavium.com>,
"stable@dpdk.org" <stable@dpdk.org>,
"dev@dpdk.org" <dev@dpdk.org>
Subject: Re: [dpdk-stable] [PATCH v1] eal: use correct data type for slab operations
Date: Mon, 24 Sep 2018 15:50:28 +0000 [thread overview]
Message-ID: <bacff11a-b786-772f-dbb8-d7f92c516e6d@caviumnetworks.com> (raw)
In-Reply-To: <1537803523-23347-1-git-send-email-vivek.sharma@caviumnetworks.com>
+ dev.
+cristian(Bitmap maintainer)
Please review.
Thanks!
On Monday 24 September 2018 09:08 PM, Vivek Sharma wrote:
> Currently, slab operations use unsigned long data type for 64-bit slab
> related operations. On target 'i686-native-linuxapp-gcc', unsigned long
> is 32-bit and thus, slab operations breaks on this target. Changing slab
> operations to use unsigned long long for correct functioning on all targets.
>
> Fixes: de3cfa2c9823 ("sched: initial import")
> Fixes: 693f715da45c ("remove extra parentheses in return statement")
> CC: stable@dpdk.org
>
> Signed-off-by: Vivek Sharma <vivek.sharma@caviumnetworks.com>
> ---
> lib/librte_eal/common/include/rte_bitmap.h | 14 +++++++-------
> test/test/test_bitmap.c | 18 ++++++++++++++++++
> 2 files changed, 25 insertions(+), 7 deletions(-)
>
> diff --git a/lib/librte_eal/common/include/rte_bitmap.h b/lib/librte_eal/common/include/rte_bitmap.h
> index d9facc6..7a36ce7 100644
> --- a/lib/librte_eal/common/include/rte_bitmap.h
> +++ b/lib/librte_eal/common/include/rte_bitmap.h
> @@ -88,7 +88,7 @@ __rte_bitmap_index1_inc(struct rte_bitmap *bmp)
> static inline uint64_t
> __rte_bitmap_mask1_get(struct rte_bitmap *bmp)
> {
> - return (~1lu) << bmp->offset1;
> + return (~1llu) << bmp->offset1;
> }
>
> static inline void
> @@ -317,7 +317,7 @@ rte_bitmap_get(struct rte_bitmap *bmp, uint32_t pos)
> index2 = pos >> RTE_BITMAP_SLAB_BIT_SIZE_LOG2;
> offset2 = pos & RTE_BITMAP_SLAB_BIT_MASK;
> slab2 = bmp->array2 + index2;
> - return (*slab2) & (1lu << offset2);
> + return (*slab2) & (1llu << offset2);
> }
>
> /**
> @@ -342,8 +342,8 @@ rte_bitmap_set(struct rte_bitmap *bmp, uint32_t pos)
> slab2 = bmp->array2 + index2;
> slab1 = bmp->array1 + index1;
>
> - *slab2 |= 1lu << offset2;
> - *slab1 |= 1lu << offset1;
> + *slab2 |= 1llu << offset2;
> + *slab1 |= 1llu << offset1;
> }
>
> /**
> @@ -370,7 +370,7 @@ rte_bitmap_set_slab(struct rte_bitmap *bmp, uint32_t pos, uint64_t slab)
> slab1 = bmp->array1 + index1;
>
> *slab2 |= slab;
> - *slab1 |= 1lu << offset1;
> + *slab1 |= 1llu << offset1;
> }
>
> static inline uint64_t
> @@ -408,7 +408,7 @@ rte_bitmap_clear(struct rte_bitmap *bmp, uint32_t pos)
> slab2 = bmp->array2 + index2;
>
> /* Return if array2 slab is not all-zeros */
> - *slab2 &= ~(1lu << offset2);
> + *slab2 &= ~(1llu << offset2);
> if (*slab2){
> return;
> }
> @@ -424,7 +424,7 @@ rte_bitmap_clear(struct rte_bitmap *bmp, uint32_t pos)
> index1 = pos >> (RTE_BITMAP_SLAB_BIT_SIZE_LOG2 + RTE_BITMAP_CL_BIT_SIZE_LOG2);
> offset1 = (pos >> RTE_BITMAP_CL_BIT_SIZE_LOG2) & RTE_BITMAP_SLAB_BIT_MASK;
> slab1 = bmp->array1 + index1;
> - *slab1 &= ~(1lu << offset1);
> + *slab1 &= ~(1llu << offset1);
>
> return;
> }
> diff --git a/test/test/test_bitmap.c b/test/test/test_bitmap.c
> index c3169e9..95c5184 100644
> --- a/test/test/test_bitmap.c
> +++ b/test/test/test_bitmap.c
> @@ -101,6 +101,7 @@ test_bitmap_slab_set_get(struct rte_bitmap *bmp)
> static int
> test_bitmap_set_get_clear(struct rte_bitmap *bmp)
> {
> + uint64_t val;
> int i;
>
> rte_bitmap_reset(bmp);
> @@ -124,6 +125,23 @@ test_bitmap_set_get_clear(struct rte_bitmap *bmp)
> }
> }
>
> + rte_bitmap_reset(bmp);
> +
> + /* Alternate slab set test */
> + for (i = 0; i < MAX_BITS; i++) {
> + if (i % RTE_BITMAP_SLAB_BIT_SIZE)
> + rte_bitmap_set(bmp, i);
> + }
> +
> + for (i = 0; i < MAX_BITS; i++) {
> + val = rte_bitmap_get(bmp, i);
> + if (((i % RTE_BITMAP_SLAB_BIT_SIZE) && !val) ||
> + (!(i % RTE_BITMAP_SLAB_BIT_SIZE) && val)) {
> + printf("Failed to get set bit.\n");
> + return TEST_FAILED;
> + }
> + }
> +
> return TEST_SUCCESS;
> }
>
>
next prev parent reply other threads:[~2018-09-24 15:50 UTC|newest]
Thread overview: 6+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-09-24 15:38 Vivek Sharma
2018-09-24 15:50 ` Vivek Sharma [this message]
2018-09-24 17:12 ` Dumitrescu, Cristian
2018-10-08 15:57 ` Dumitrescu, Cristian
2018-09-25 9:53 ` [dpdk-stable] [PATCH v2] eal: use correct data type for bitmap " Vivek Sharma
2018-09-25 12:38 ` [dpdk-stable] [PATCH v1] eal: use correct data type for " Luca Boccassi
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=bacff11a-b786-772f-dbb8-d7f92c516e6d@caviumnetworks.com \
--to=vivek.sharma@caviumnetworks.com \
--cc=Vivek.Sharma@cavium.com \
--cc=cristian.dumitrescu@intel.com \
--cc=dev@dpdk.org \
--cc=stable@dpdk.org \
/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;
as well as URLs for NNTP newsgroup(s).