* [PATCH v2] mbuf: fix bulk allocation when debug enabled
@ 2017-05-06 7:26 Gregory Etelson
2017-05-10 14:42 ` Olivier Matz
2017-05-10 15:13 ` [PATCH v3] " Olivier Matz
0 siblings, 2 replies; 5+ messages in thread
From: Gregory Etelson @ 2017-05-06 7:26 UTC (permalink / raw)
To: dev; +Cc: Olivier Matz
The debug assertions when allocating a raw mbuf are not correct since
commit 8f094a9ac5d7 ("mbuf: set mbuf fields while in pool"),
which triggers a panic when using this function in debug mode
Signed-off-by: Gregory Etelson <gregory@weka.io>
---
lib/librte_mbuf/rte_mbuf.h | 25 ++++++++++++-------------
1 file changed, 12 insertions(+), 13 deletions(-)
diff --git a/lib/librte_mbuf/rte_mbuf.h b/lib/librte_mbuf/rte_mbuf.h
index 9097f18..05b8300 100644
--- a/lib/librte_mbuf/rte_mbuf.h
+++ b/lib/librte_mbuf/rte_mbuf.h
@@ -788,6 +788,13 @@ rte_mbuf_refcnt_set(struct rte_mbuf *m, uint16_t new_value)
void
rte_mbuf_sanity_check(const struct rte_mbuf *m, int is_header);
+#define MBUF_RAW_ALLOC_CHECK(m_) do { \
+ RTE_ASSERT(rte_mbuf_refcnt_read(m_) == 1); \
+ RTE_ASSERT(m_->next == NULL); \
+ RTE_ASSERT(m_->nb_segs == 1); \
+ __rte_mbuf_sanity_check(m_, 0); \
+} while (0)
+
/**
* Allocate an unitialized mbuf from mempool *mp*.
*
@@ -815,11 +822,7 @@ static inline struct rte_mbuf *rte_mbuf_raw_alloc(struct rte_mempool *mp)
if (rte_mempool_get(mp, &mb) < 0)
return NULL;
m = (struct rte_mbuf *)mb;
- RTE_ASSERT(rte_mbuf_refcnt_read(m) == 1);
- RTE_ASSERT(m->next == NULL);
- RTE_ASSERT(m->nb_segs == 1);
- __rte_mbuf_sanity_check(m, 0);
-
+ MBUF_RAW_ALLOC_CHECK(m);
return m;
}
@@ -1152,26 +1155,22 @@ static inline int rte_pktmbuf_alloc_bulk(struct rte_mempool *pool,
switch (count % 4) {
case 0:
while (idx != count) {
- RTE_ASSERT(rte_mbuf_refcnt_read(mbufs[idx]) == 0);
- rte_mbuf_refcnt_set(mbufs[idx], 1);
+ MBUF_RAW_ALLOC_CHECK(mbufs[idx]);
rte_pktmbuf_reset(mbufs[idx]);
idx++;
/* fall-through */
case 3:
- RTE_ASSERT(rte_mbuf_refcnt_read(mbufs[idx]) == 0);
- rte_mbuf_refcnt_set(mbufs[idx], 1);
+ MBUF_RAW_ALLOC_CHECK(mbufs[idx]);
rte_pktmbuf_reset(mbufs[idx]);
idx++;
/* fall-through */
case 2:
- RTE_ASSERT(rte_mbuf_refcnt_read(mbufs[idx]) == 0);
- rte_mbuf_refcnt_set(mbufs[idx], 1);
+ MBUF_RAW_ALLOC_CHECK(mbufs[idx]);
rte_pktmbuf_reset(mbufs[idx]);
idx++;
/* fall-through */
case 1:
- RTE_ASSERT(rte_mbuf_refcnt_read(mbufs[idx]) == 0);
- rte_mbuf_refcnt_set(mbufs[idx], 1);
+ MBUF_RAW_ALLOC_CHECK(mbufs[idx]);
rte_pktmbuf_reset(mbufs[idx]);
idx++;
/* fall-through */
--
2.9.3
^ permalink raw reply related [flat|nested] 5+ messages in thread* Re: [PATCH v2] mbuf: fix bulk allocation when debug enabled
2017-05-06 7:26 [PATCH v2] mbuf: fix bulk allocation when debug enabled Gregory Etelson
@ 2017-05-10 14:42 ` Olivier Matz
2017-05-10 14:54 ` Thomas Monjalon
2017-05-10 15:13 ` [PATCH v3] " Olivier Matz
1 sibling, 1 reply; 5+ messages in thread
From: Olivier Matz @ 2017-05-10 14:42 UTC (permalink / raw)
To: Gregory Etelson; +Cc: dev, thomas
Hi,
On Sat, 06 May 2017 10:26:49 +0300, Gregory Etelson <gregory@weka.io> wrote:
> The debug assertions when allocating a raw mbuf are not correct since
> commit 8f094a9ac5d7 ("mbuf: set mbuf fields while in pool"),
> which triggers a panic when using this function in debug mode
>
Fixes: 8f094a9ac5d7 ("mbuf: set mbuf fields while in pool")
> Signed-off-by: Gregory Etelson <gregory@weka.io>
> ---
> lib/librte_mbuf/rte_mbuf.h | 25 ++++++++++++-------------
> 1 file changed, 12 insertions(+), 13 deletions(-)
>
> diff --git a/lib/librte_mbuf/rte_mbuf.h b/lib/librte_mbuf/rte_mbuf.h
> index 9097f18..05b8300 100644
> --- a/lib/librte_mbuf/rte_mbuf.h
> +++ b/lib/librte_mbuf/rte_mbuf.h
> @@ -788,6 +788,13 @@ rte_mbuf_refcnt_set(struct rte_mbuf *m, uint16_t new_value)
> void
> rte_mbuf_sanity_check(const struct rte_mbuf *m, int is_header);
>
> +#define MBUF_RAW_ALLOC_CHECK(m_) do { \
> + RTE_ASSERT(rte_mbuf_refcnt_read(m_) == 1); \
> + RTE_ASSERT(m_->next == NULL); \
> + RTE_ASSERT(m_->nb_segs == 1); \
> + __rte_mbuf_sanity_check(m_, 0); \
> +} while (0)
> +
Few nits:
- maybe 'm_' could be 'm'
- some parenthesis could be added for safety
Thomas, I know you're in hurry for 17.05, I think this fix should
go in. If you want, I can send a v3 with this small changes if you
need it now.
Regards,
Olivier
^ permalink raw reply [flat|nested] 5+ messages in thread* Re: [PATCH v2] mbuf: fix bulk allocation when debug enabled
2017-05-10 14:42 ` Olivier Matz
@ 2017-05-10 14:54 ` Thomas Monjalon
0 siblings, 0 replies; 5+ messages in thread
From: Thomas Monjalon @ 2017-05-10 14:54 UTC (permalink / raw)
To: Olivier Matz; +Cc: dev, Gregory Etelson
10/05/2017 16:42, Olivier Matz:
> Hi,
>
> On Sat, 06 May 2017 10:26:49 +0300, Gregory Etelson <gregory@weka.io> wrote:
> > The debug assertions when allocating a raw mbuf are not correct since
> > commit 8f094a9ac5d7 ("mbuf: set mbuf fields while in pool"),
> > which triggers a panic when using this function in debug mode
> >
>
> Fixes: 8f094a9ac5d7 ("mbuf: set mbuf fields while in pool")
>
> > Signed-off-by: Gregory Etelson <gregory@weka.io>
> > ---
> > lib/librte_mbuf/rte_mbuf.h | 25 ++++++++++++-------------
> > 1 file changed, 12 insertions(+), 13 deletions(-)
> >
> > diff --git a/lib/librte_mbuf/rte_mbuf.h b/lib/librte_mbuf/rte_mbuf.h
> > index 9097f18..05b8300 100644
> > --- a/lib/librte_mbuf/rte_mbuf.h
> > +++ b/lib/librte_mbuf/rte_mbuf.h
> > @@ -788,6 +788,13 @@ rte_mbuf_refcnt_set(struct rte_mbuf *m, uint16_t new_value)
> > void
> > rte_mbuf_sanity_check(const struct rte_mbuf *m, int is_header);
> >
> > +#define MBUF_RAW_ALLOC_CHECK(m_) do { \
> > + RTE_ASSERT(rte_mbuf_refcnt_read(m_) == 1); \
> > + RTE_ASSERT(m_->next == NULL); \
> > + RTE_ASSERT(m_->nb_segs == 1); \
> > + __rte_mbuf_sanity_check(m_, 0); \
> > +} while (0)
> > +
>
> Few nits:
> - maybe 'm_' could be 'm'
> - some parenthesis could be added for safety
>
> Thomas, I know you're in hurry for 17.05, I think this fix should
> go in. If you want, I can send a v3 with this small changes if you
> need it now.
Yes, please a quick v3 would be appreciated.
Thanks
^ permalink raw reply [flat|nested] 5+ messages in thread
* [PATCH v3] mbuf: fix bulk allocation when debug enabled
2017-05-06 7:26 [PATCH v2] mbuf: fix bulk allocation when debug enabled Gregory Etelson
2017-05-10 14:42 ` Olivier Matz
@ 2017-05-10 15:13 ` Olivier Matz
2017-05-10 16:17 ` Thomas Monjalon
1 sibling, 1 reply; 5+ messages in thread
From: Olivier Matz @ 2017-05-10 15:13 UTC (permalink / raw)
To: dev; +Cc: gregory, thomas
From: Gregory Etelson <gregory@weka.io>
The debug assertions when allocating a raw mbuf are not correct since
commit 8f094a9ac5d7 ("mbuf: set mbuf fields while in pool"),
which triggers a panic when using this function in debug mode
Change the expected number of segments to 1 instead of 0, and
factorize these sanity checks.
Fixes: 8f094a9ac5d7 ("mbuf: set mbuf fields while in pool")
Signed-off-by: Gregory Etelson <gregory@weka.io>
Signed-off-by: Olivier Matz <olivier.matz@6wind.com>
---
lib/librte_mbuf/rte_mbuf.h | 25 ++++++++++++-------------
1 file changed, 12 insertions(+), 13 deletions(-)
diff --git a/lib/librte_mbuf/rte_mbuf.h b/lib/librte_mbuf/rte_mbuf.h
index 9097f1859..1cb03109c 100644
--- a/lib/librte_mbuf/rte_mbuf.h
+++ b/lib/librte_mbuf/rte_mbuf.h
@@ -788,6 +788,13 @@ rte_mbuf_refcnt_set(struct rte_mbuf *m, uint16_t new_value)
void
rte_mbuf_sanity_check(const struct rte_mbuf *m, int is_header);
+#define MBUF_RAW_ALLOC_CHECK(m) do { \
+ RTE_ASSERT(rte_mbuf_refcnt_read(m) == 1); \
+ RTE_ASSERT((m)->next == NULL); \
+ RTE_ASSERT((m)->nb_segs == 1); \
+ __rte_mbuf_sanity_check(m, 0); \
+} while (0)
+
/**
* Allocate an unitialized mbuf from mempool *mp*.
*
@@ -815,11 +822,7 @@ static inline struct rte_mbuf *rte_mbuf_raw_alloc(struct rte_mempool *mp)
if (rte_mempool_get(mp, &mb) < 0)
return NULL;
m = (struct rte_mbuf *)mb;
- RTE_ASSERT(rte_mbuf_refcnt_read(m) == 1);
- RTE_ASSERT(m->next == NULL);
- RTE_ASSERT(m->nb_segs == 1);
- __rte_mbuf_sanity_check(m, 0);
-
+ MBUF_RAW_ALLOC_CHECK(m);
return m;
}
@@ -1152,26 +1155,22 @@ static inline int rte_pktmbuf_alloc_bulk(struct rte_mempool *pool,
switch (count % 4) {
case 0:
while (idx != count) {
- RTE_ASSERT(rte_mbuf_refcnt_read(mbufs[idx]) == 0);
- rte_mbuf_refcnt_set(mbufs[idx], 1);
+ MBUF_RAW_ALLOC_CHECK(mbufs[idx]);
rte_pktmbuf_reset(mbufs[idx]);
idx++;
/* fall-through */
case 3:
- RTE_ASSERT(rte_mbuf_refcnt_read(mbufs[idx]) == 0);
- rte_mbuf_refcnt_set(mbufs[idx], 1);
+ MBUF_RAW_ALLOC_CHECK(mbufs[idx]);
rte_pktmbuf_reset(mbufs[idx]);
idx++;
/* fall-through */
case 2:
- RTE_ASSERT(rte_mbuf_refcnt_read(mbufs[idx]) == 0);
- rte_mbuf_refcnt_set(mbufs[idx], 1);
+ MBUF_RAW_ALLOC_CHECK(mbufs[idx]);
rte_pktmbuf_reset(mbufs[idx]);
idx++;
/* fall-through */
case 1:
- RTE_ASSERT(rte_mbuf_refcnt_read(mbufs[idx]) == 0);
- rte_mbuf_refcnt_set(mbufs[idx], 1);
+ MBUF_RAW_ALLOC_CHECK(mbufs[idx]);
rte_pktmbuf_reset(mbufs[idx]);
idx++;
/* fall-through */
--
2.11.0
^ permalink raw reply related [flat|nested] 5+ messages in thread* Re: [PATCH v3] mbuf: fix bulk allocation when debug enabled
2017-05-10 15:13 ` [PATCH v3] " Olivier Matz
@ 2017-05-10 16:17 ` Thomas Monjalon
0 siblings, 0 replies; 5+ messages in thread
From: Thomas Monjalon @ 2017-05-10 16:17 UTC (permalink / raw)
To: gregory; +Cc: dev, Olivier Matz
10/05/2017 17:13, Olivier Matz:
> From: Gregory Etelson <gregory@weka.io>
>
> The debug assertions when allocating a raw mbuf are not correct since
> commit 8f094a9ac5d7 ("mbuf: set mbuf fields while in pool"),
> which triggers a panic when using this function in debug mode
>
> Change the expected number of segments to 1 instead of 0, and
> factorize these sanity checks.
>
> Fixes: 8f094a9ac5d7 ("mbuf: set mbuf fields while in pool")
>
> Signed-off-by: Gregory Etelson <gregory@weka.io>
> Signed-off-by: Olivier Matz <olivier.matz@6wind.com>
Applied, thanks
^ permalink raw reply [flat|nested] 5+ messages in thread
end of thread, other threads:[~2017-05-10 16:17 UTC | newest]
Thread overview: 5+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2017-05-06 7:26 [PATCH v2] mbuf: fix bulk allocation when debug enabled Gregory Etelson
2017-05-10 14:42 ` Olivier Matz
2017-05-10 14:54 ` Thomas Monjalon
2017-05-10 15:13 ` [PATCH v3] " Olivier Matz
2017-05-10 16:17 ` Thomas Monjalon
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.