* [PATCH] app/dma-perf: fix physical address seg-fault
@ 2023-08-15 15:10 Vipin Varghese
2023-08-15 15:15 ` Varghese, Vipin
0 siblings, 1 reply; 6+ messages in thread
From: Vipin Varghese @ 2023-08-15 15:10 UTC (permalink / raw)
To: thomas, dev; +Cc: Ferruh.Yigit, cheng1.jiang, stable
do_cpu_mem_copy uses DPDK API rte_mbuf_data_iova to return
the start of the virtual address for both src and dst.
But in case of iova mode set as PA, this results in seg-fault.
This is because rte_memcpy VA address and not PA.
this fix checks the iova mode and invokes rte_memcpy with the right
arguments.
Bugzilla ID: 1269
Fixes: 623dc9364dc6 ("app/dma-perf: introduce DMA performance test")
Cc: cheng1.jiang@intel.com
Cc: stable@dpdk.org
Signed-off-by: Vipin Varghese <vipin.varghese@amd.com>
---
tested for both va and pa
CMD:
PA: dpdk-test-dma-perf --iova-mode=pa -- --config test.ini
VA: dpdk-test-dma-perf --iova-mode=va -- --config test.ini
DC: dpdk-test-dma-perf --iova-mode=dc -- --config test.ini
Log: fails for dc mode `EAL: invalid parameters for --iova-mode`
test.ini:
```
[case1]
type=CPU_MEM_COPY
mem_size=10
buf_size=64,8192,2,MUL
src_numa_node=0
dst_numa_node=0
cache_flush=0
test_seconds=2
lcore = 7
eal_args=--in-memory --no-pci
```
---
app/test-dma-perf/benchmark.c | 36 ++++++++++++++++++++++++++---------
1 file changed, 27 insertions(+), 9 deletions(-)
diff --git a/app/test-dma-perf/benchmark.c b/app/test-dma-perf/benchmark.c
index 0601e0d171..73200e1935 100644
--- a/app/test-dma-perf/benchmark.c
+++ b/app/test-dma-perf/benchmark.c
@@ -279,6 +279,10 @@ do_cpu_mem_copy(void *p)
struct rte_mbuf **srcs = para->srcs;
struct rte_mbuf **dsts = para->dsts;
uint32_t i;
+ bool isAddrPaMode = false;
+
+ if (rte_eal_iova_mode() == RTE_IOVA_PA)
+ isAddrPaMode = true;
worker_info->stop_flag = false;
worker_info->ready_flag = true;
@@ -286,16 +290,30 @@ do_cpu_mem_copy(void *p)
while (!worker_info->start_flag)
;
- while (1) {
- for (i = 0; i < nr_buf; i++) {
- /* copy buffer form src to dst */
- rte_memcpy((void *)(uintptr_t)rte_mbuf_data_iova(dsts[i]),
- (void *)(uintptr_t)rte_mbuf_data_iova(srcs[i]),
- (size_t)buf_size);
- worker_info->total_cpl++;
+ if (true == isAddrPaMode) {
+ while (1) {
+ for (i = 0; i < nr_buf; i++) {
+ /* copy buffer form src to dst */
+ rte_memcpy((void *)dsts[i],
+ (void *)srcs[i],
+ (size_t)buf_size);
+ worker_info->total_cpl++;
+ }
+ if (worker_info->stop_flag)
+ break;
+ }
+ } else {
+ while (1) {
+ for (i = 0; i < nr_buf; i++) {
+ /* copy buffer form src to dst */
+ rte_memcpy((void *)(uintptr_t)rte_mbuf_data_iova(dsts[i]),
+ (void *)(uintptr_t)rte_mbuf_data_iova(srcs[i]),
+ (size_t)buf_size);
+ worker_info->total_cpl++;
+ }
+ if (worker_info->stop_flag)
+ break;
}
- if (worker_info->stop_flag)
- break;
}
return 0;
--
2.34.1
^ permalink raw reply [flat|nested] 6+ messages in thread* RE: [PATCH] app/dma-perf: fix physical address seg-fault
2023-08-15 15:10 [PATCH] app/dma-perf: fix physical address seg-fault Vipin Varghese
@ 2023-08-15 15:15 ` Varghese, Vipin
2023-08-16 6:07 ` Anoob Joseph
0 siblings, 1 reply; 6+ messages in thread
From: Varghese, Vipin @ 2023-08-15 15:15 UTC (permalink / raw)
To: Varghese, Vipin, thomas, dev; +Cc: Yigit, Ferruh, cheng1.jiang, stable
[AMD Official Use Only - General]
Apologies, marking this as `NA`. After recheck of this logic without use of ` rte_mbuf_data_iova` will result in mbuf meta-data corruption.
Need to fix this in a different way.
> -----Original Message-----
> From: Vipin Varghese <vipin.varghese@amd.com>
> Sent: Tuesday, August 15, 2023 8:41 PM
> To: thomas@monjalon.net; dev@dpdk.org
> Cc: Yigit, Ferruh <Ferruh.Yigit@amd.com>; cheng1.jiang@intel.com;
> stable@dpdk.org
> Subject: [PATCH] app/dma-perf: fix physical address seg-fault
>
> Caution: This message originated from an External Source. Use proper caution
> when opening attachments, clicking links, or responding.
>
>
> do_cpu_mem_copy uses DPDK API rte_mbuf_data_iova to return the start of
> the virtual address for both src and dst.
> But in case of iova mode set as PA, this results in seg-fault.
> This is because rte_memcpy VA address and not PA.
>
> this fix checks the iova mode and invokes rte_memcpy with the right
> arguments.
>
> Bugzilla ID: 1269
> Fixes: 623dc9364dc6 ("app/dma-perf: introduce DMA performance test")
> Cc: cheng1.jiang@intel.com
>
> Cc: stable@dpdk.org
>
> Signed-off-by: Vipin Varghese <vipin.varghese@amd.com>
> ---
>
> tested for both va and pa
>
> CMD:
> PA: dpdk-test-dma-perf --iova-mode=pa -- --config test.ini
> VA: dpdk-test-dma-perf --iova-mode=va -- --config test.ini
> DC: dpdk-test-dma-perf --iova-mode=dc -- --config test.ini
>
> Log: fails for dc mode `EAL: invalid parameters for --iova-mode`
>
> test.ini:
> ```
> [case1]
> type=CPU_MEM_COPY
> mem_size=10
> buf_size=64,8192,2,MUL
> src_numa_node=0
> dst_numa_node=0
> cache_flush=0
> test_seconds=2
> lcore = 7
> eal_args=--in-memory --no-pci
> ```
> ---
> app/test-dma-perf/benchmark.c | 36 ++++++++++++++++++++++++++---------
> 1 file changed, 27 insertions(+), 9 deletions(-)
>
> diff --git a/app/test-dma-perf/benchmark.c b/app/test-dma-perf/benchmark.c
> index 0601e0d171..73200e1935 100644
> --- a/app/test-dma-perf/benchmark.c
> +++ b/app/test-dma-perf/benchmark.c
> @@ -279,6 +279,10 @@ do_cpu_mem_copy(void *p)
> struct rte_mbuf **srcs = para->srcs;
> struct rte_mbuf **dsts = para->dsts;
> uint32_t i;
> + bool isAddrPaMode = false;
> +
> + if (rte_eal_iova_mode() == RTE_IOVA_PA)
> + isAddrPaMode = true;
>
> worker_info->stop_flag = false;
> worker_info->ready_flag = true;
> @@ -286,16 +290,30 @@ do_cpu_mem_copy(void *p)
> while (!worker_info->start_flag)
> ;
>
> - while (1) {
> - for (i = 0; i < nr_buf; i++) {
> - /* copy buffer form src to dst */
> - rte_memcpy((void *)(uintptr_t)rte_mbuf_data_iova(dsts[i]),
> - (void *)(uintptr_t)rte_mbuf_data_iova(srcs[i]),
> - (size_t)buf_size);
> - worker_info->total_cpl++;
> + if (true == isAddrPaMode) {
> + while (1) {
> + for (i = 0; i < nr_buf; i++) {
> + /* copy buffer form src to dst */
> + rte_memcpy((void *)dsts[i],
> + (void *)srcs[i],
> + (size_t)buf_size);
> + worker_info->total_cpl++;
> + }
> + if (worker_info->stop_flag)
> + break;
> + }
> + } else {
> + while (1) {
> + for (i = 0; i < nr_buf; i++) {
> + /* copy buffer form src to dst */
> + rte_memcpy((void
> *)(uintptr_t)rte_mbuf_data_iova(dsts[i]),
> + (void *)(uintptr_t)rte_mbuf_data_iova(srcs[i]),
> + (size_t)buf_size);
> + worker_info->total_cpl++;
> + }
> + if (worker_info->stop_flag)
> + break;
> }
> - if (worker_info->stop_flag)
> - break;
> }
>
> return 0;
> --
> 2.34.1
^ permalink raw reply [flat|nested] 6+ messages in thread* RE: [PATCH] app/dma-perf: fix physical address seg-fault
2023-08-15 15:15 ` Varghese, Vipin
@ 2023-08-16 6:07 ` Anoob Joseph
2023-08-16 7:24 ` Varghese, Vipin
0 siblings, 1 reply; 6+ messages in thread
From: Anoob Joseph @ 2023-08-16 6:07 UTC (permalink / raw)
To: Varghese, Vipin
Cc: Yigit, Ferruh, cheng1.jiang, stable, thomas, dev,
Jerin Jacob Kollanukkaran
Hi Vipin,
Would using 'rte_pktmbuf_mtod' in 'do_cpu_mem_copy' function resolve the issue?
Thanks,
Anoob
> -----Original Message-----
> From: Varghese, Vipin <Vipin.Varghese@amd.com>
> Sent: Tuesday, August 15, 2023 8:45 PM
> To: Varghese, Vipin <Vipin.Varghese@amd.com>; thomas@monjalon.net;
> dev@dpdk.org
> Cc: Yigit, Ferruh <Ferruh.Yigit@amd.com>; cheng1.jiang@intel.com;
> stable@dpdk.org
> Subject: [EXT] RE: [PATCH] app/dma-perf: fix physical address seg-fault
>
> External Email
>
> ----------------------------------------------------------------------
> [AMD Official Use Only - General]
>
> Apologies, marking this as `NA`. After recheck of this logic without use of `
> rte_mbuf_data_iova` will result in mbuf meta-data corruption.
>
> Need to fix this in a different way.
>
> > -----Original Message-----
> > From: Vipin Varghese <vipin.varghese@amd.com>
> > Sent: Tuesday, August 15, 2023 8:41 PM
> > To: thomas@monjalon.net; dev@dpdk.org
> > Cc: Yigit, Ferruh <Ferruh.Yigit@amd.com>; cheng1.jiang@intel.com;
> > stable@dpdk.org
> > Subject: [PATCH] app/dma-perf: fix physical address seg-fault
> >
> > Caution: This message originated from an External Source. Use proper
> > caution when opening attachments, clicking links, or responding.
> >
> >
> > do_cpu_mem_copy uses DPDK API rte_mbuf_data_iova to return the start
> > of the virtual address for both src and dst.
> > But in case of iova mode set as PA, this results in seg-fault.
> > This is because rte_memcpy VA address and not PA.
> >
> > this fix checks the iova mode and invokes rte_memcpy with the right
> > arguments.
> >
> > Bugzilla ID: 1269
> > Fixes: 623dc9364dc6 ("app/dma-perf: introduce DMA performance test")
> > Cc: cheng1.jiang@intel.com
> >
> > Cc: stable@dpdk.org
> >
> > Signed-off-by: Vipin Varghese <vipin.varghese@amd.com>
> > ---
> >
> > tested for both va and pa
> >
> > CMD:
> > PA: dpdk-test-dma-perf --iova-mode=pa -- --config test.ini
> > VA: dpdk-test-dma-perf --iova-mode=va -- --config test.ini
> > DC: dpdk-test-dma-perf --iova-mode=dc -- --config test.ini
> >
> > Log: fails for dc mode `EAL: invalid parameters for --iova-mode`
> >
> > test.ini:
> > ```
> > [case1]
> > type=CPU_MEM_COPY
> > mem_size=10
> > buf_size=64,8192,2,MUL
> > src_numa_node=0
> > dst_numa_node=0
> > cache_flush=0
> > test_seconds=2
> > lcore = 7
> > eal_args=--in-memory --no-pci
> > ```
> > ---
> > app/test-dma-perf/benchmark.c | 36
> > ++++++++++++++++++++++++++---------
> > 1 file changed, 27 insertions(+), 9 deletions(-)
> >
> > diff --git a/app/test-dma-perf/benchmark.c
> > b/app/test-dma-perf/benchmark.c index 0601e0d171..73200e1935 100644
> > --- a/app/test-dma-perf/benchmark.c
> > +++ b/app/test-dma-perf/benchmark.c
> > @@ -279,6 +279,10 @@ do_cpu_mem_copy(void *p)
> > struct rte_mbuf **srcs = para->srcs;
> > struct rte_mbuf **dsts = para->dsts;
> > uint32_t i;
> > + bool isAddrPaMode = false;
> > +
> > + if (rte_eal_iova_mode() == RTE_IOVA_PA)
> > + isAddrPaMode = true;
> >
> > worker_info->stop_flag = false;
> > worker_info->ready_flag = true; @@ -286,16 +290,30 @@
> > do_cpu_mem_copy(void *p)
> > while (!worker_info->start_flag)
> > ;
> >
> > - while (1) {
> > - for (i = 0; i < nr_buf; i++) {
> > - /* copy buffer form src to dst */
> > - rte_memcpy((void *)(uintptr_t)rte_mbuf_data_iova(dsts[i]),
> > - (void *)(uintptr_t)rte_mbuf_data_iova(srcs[i]),
> > - (size_t)buf_size);
> > - worker_info->total_cpl++;
> > + if (true == isAddrPaMode) {
> > + while (1) {
> > + for (i = 0; i < nr_buf; i++) {
> > + /* copy buffer form src to dst */
> > + rte_memcpy((void *)dsts[i],
> > + (void *)srcs[i],
> > + (size_t)buf_size);
> > + worker_info->total_cpl++;
> > + }
> > + if (worker_info->stop_flag)
> > + break;
> > + }
> > + } else {
> > + while (1) {
> > + for (i = 0; i < nr_buf; i++) {
> > + /* copy buffer form src to dst */
> > + rte_memcpy((void
> > *)(uintptr_t)rte_mbuf_data_iova(dsts[i]),
> > + (void *)(uintptr_t)rte_mbuf_data_iova(srcs[i]),
> > + (size_t)buf_size);
> > + worker_info->total_cpl++;
> > + }
> > + if (worker_info->stop_flag)
> > + break;
> > }
> > - if (worker_info->stop_flag)
> > - break;
> > }
> >
> > return 0;
> > --
> > 2.34.1
^ permalink raw reply [flat|nested] 6+ messages in thread* RE: [PATCH] app/dma-perf: fix physical address seg-fault
2023-08-16 6:07 ` Anoob Joseph
@ 2023-08-16 7:24 ` Varghese, Vipin
0 siblings, 0 replies; 6+ messages in thread
From: Varghese, Vipin @ 2023-08-16 7:24 UTC (permalink / raw)
To: Anoob Joseph
Cc: Yigit, Ferruh, cheng1.jiang, stable, thomas, dev,
Jerin Jacob Kollanukkaran
[AMD Official Use Only - General]
Hi Anoob,
> -----Original Message-----
> From: Anoob Joseph <anoobj@marvell.com>
> Sent: Wednesday, August 16, 2023 11:38 AM
> To: Varghese, Vipin <Vipin.Varghese@amd.com>
> Cc: Yigit, Ferruh <Ferruh.Yigit@amd.com>; cheng1.jiang@intel.com;
> stable@dpdk.org; thomas@monjalon.net; dev@dpdk.org; Jerin Jacob
> Kollanukkaran <jerinj@marvell.com>
> Subject: RE: [PATCH] app/dma-perf: fix physical address seg-fault
>
> [AMD Official Use Only - General]
>
> Caution: This message originated from an External Source. Use proper caution
> when opening attachments, clicking links, or responding.
>
>
> Hi Vipin,
>
> Would using 'rte_pktmbuf_mtod' in 'do_cpu_mem_copy' function resolve the
> issue?
https://patchwork.dpdk.org/project/dpdk/patch/20230816071810.1796-1-vipin.varghese@amd.com/
thanks, that is what we tried and tested.
>
> Thanks,
> Anoob
>
> > -----Original Message-----
> > From: Varghese, Vipin <Vipin.Varghese@amd.com>
> > Sent: Tuesday, August 15, 2023 8:45 PM
> > To: Varghese, Vipin <Vipin.Varghese@amd.com>; thomas@monjalon.net;
> > dev@dpdk.org
> > Cc: Yigit, Ferruh <Ferruh.Yigit@amd.com>; cheng1.jiang@intel.com;
> > stable@dpdk.org
> > Subject: [EXT] RE: [PATCH] app/dma-perf: fix physical address
> > seg-fault
> >
> > External Email
> >
> > ----------------------------------------------------------------------
> > [AMD Official Use Only - General]
> >
> > Apologies, marking this as `NA`. After recheck of this logic without
> > use of ` rte_mbuf_data_iova` will result in mbuf meta-data corruption.
> >
> > Need to fix this in a different way.
> >
> > > -----Original Message-----
> > > From: Vipin Varghese <vipin.varghese@amd.com>
> > > Sent: Tuesday, August 15, 2023 8:41 PM
> > > To: thomas@monjalon.net; dev@dpdk.org
> > > Cc: Yigit, Ferruh <Ferruh.Yigit@amd.com>; cheng1.jiang@intel.com;
> > > stable@dpdk.org
> > > Subject: [PATCH] app/dma-perf: fix physical address seg-fault
> > >
> > > Caution: This message originated from an External Source. Use proper
> > > caution when opening attachments, clicking links, or responding.
> > >
> > >
> > > do_cpu_mem_copy uses DPDK API rte_mbuf_data_iova to return the start
> > > of the virtual address for both src and dst.
> > > But in case of iova mode set as PA, this results in seg-fault.
> > > This is because rte_memcpy VA address and not PA.
> > >
> > > this fix checks the iova mode and invokes rte_memcpy with the right
> > > arguments.
> > >
> > > Bugzilla ID: 1269
> > > Fixes: 623dc9364dc6 ("app/dma-perf: introduce DMA performance test")
> > > Cc: cheng1.jiang@intel.com
> > >
> > > Cc: stable@dpdk.org
> > >
> > > Signed-off-by: Vipin Varghese <vipin.varghese@amd.com>
> > > ---
> > >
> > > tested for both va and pa
> > >
> > > CMD:
> > > PA: dpdk-test-dma-perf --iova-mode=pa -- --config test.ini
> > > VA: dpdk-test-dma-perf --iova-mode=va -- --config test.ini
> > > DC: dpdk-test-dma-perf --iova-mode=dc -- --config test.ini
> > >
> > > Log: fails for dc mode `EAL: invalid parameters for --iova-mode`
> > >
> > > test.ini:
> > > ```
> > > [case1]
> > > type=CPU_MEM_COPY
> > > mem_size=10
> > > buf_size=64,8192,2,MUL
> > > src_numa_node=0
> > > dst_numa_node=0
> > > cache_flush=0
> > > test_seconds=2
> > > lcore = 7
> > > eal_args=--in-memory --no-pci
> > > ```
> > > ---
> > > app/test-dma-perf/benchmark.c | 36
> > > ++++++++++++++++++++++++++---------
> > > 1 file changed, 27 insertions(+), 9 deletions(-)
> > >
> > > diff --git a/app/test-dma-perf/benchmark.c
> > > b/app/test-dma-perf/benchmark.c index 0601e0d171..73200e1935 100644
> > > --- a/app/test-dma-perf/benchmark.c
> > > +++ b/app/test-dma-perf/benchmark.c
> > > @@ -279,6 +279,10 @@ do_cpu_mem_copy(void *p)
> > > struct rte_mbuf **srcs = para->srcs;
> > > struct rte_mbuf **dsts = para->dsts;
> > > uint32_t i;
> > > + bool isAddrPaMode = false;
> > > +
> > > + if (rte_eal_iova_mode() == RTE_IOVA_PA)
> > > + isAddrPaMode = true;
> > >
> > > worker_info->stop_flag = false;
> > > worker_info->ready_flag = true; @@ -286,16 +290,30 @@
> > > do_cpu_mem_copy(void *p)
> > > while (!worker_info->start_flag)
> > > ;
> > >
> > > - while (1) {
> > > - for (i = 0; i < nr_buf; i++) {
> > > - /* copy buffer form src to dst */
> > > - rte_memcpy((void *)(uintptr_t)rte_mbuf_data_iova(dsts[i]),
> > > - (void *)(uintptr_t)rte_mbuf_data_iova(srcs[i]),
> > > - (size_t)buf_size);
> > > - worker_info->total_cpl++;
> > > + if (true == isAddrPaMode) {
> > > + while (1) {
> > > + for (i = 0; i < nr_buf; i++) {
> > > + /* copy buffer form src to dst */
> > > + rte_memcpy((void *)dsts[i],
> > > + (void *)srcs[i],
> > > + (size_t)buf_size);
> > > + worker_info->total_cpl++;
> > > + }
> > > + if (worker_info->stop_flag)
> > > + break;
> > > + }
> > > + } else {
> > > + while (1) {
> > > + for (i = 0; i < nr_buf; i++) {
> > > + /* copy buffer form src to dst */
> > > + rte_memcpy((void
> > > *)(uintptr_t)rte_mbuf_data_iova(dsts[i]),
> > > + (void *)(uintptr_t)rte_mbuf_data_iova(srcs[i]),
> > > + (size_t)buf_size);
> > > + worker_info->total_cpl++;
> > > + }
> > > + if (worker_info->stop_flag)
> > > + break;
> > > }
> > > - if (worker_info->stop_flag)
> > > - break;
> > > }
> > >
> > > return 0;
> > > --
> > > 2.34.1
^ permalink raw reply [flat|nested] 6+ messages in thread
* [PATCH] app/dma-perf: fix physical address seg-fault
@ 2023-08-16 7:18 Vipin Varghese
0 siblings, 0 replies; 6+ messages in thread
From: Vipin Varghese @ 2023-08-16 7:18 UTC (permalink / raw)
To: thomas, dev; +Cc: Ferruh.Yigit, cheng1.jiang, stable
do_cpu_mem_copy uses DPDK API rte_mbuf_data_iova to return
the start of the virtual address for both src and dst.
But in case of iova mode set as PA, this results in seg-fault.
This is because rte_memcpy VA address and not PA.
this fix checks the iova mode and invokes rte_memcpy with the right
arguments.
Bugzilla ID: 1269
Fixes: 623dc9364dc6 ("app/dma-perf: introduce DMA performance test")
Cc: cheng1.jiang@intel.com
Cc: stable@dpdk.org
Signed-off-by: Vipin Varghese <vipin.varghese@amd.com>
---
tested for both va and pa
CMD:
PA: dpdk-test-dma-perf --iova-mode=pa -- --config test.ini
VA: dpdk-test-dma-perf --iova-mode=va -- --config test.ini
DC: dpdk-test-dma-perf --iova-mode=dc -- --config test.ini
Log: fails for dc mode `EAL: invalid parameters for --iova-mode`
test.ini:
```
[case1]
type=CPU_MEM_COPY
mem_size=10
buf_size=64,8192,2,MUL
src_numa_node=0
dst_numa_node=0
cache_flush=0
test_seconds=2
lcore = 7
eal_args=--in-memory --no-pci
```
---
app/test-dma-perf/benchmark.c | 39 +++++++++++++++++++++++++++--------
1 file changed, 30 insertions(+), 9 deletions(-)
diff --git a/app/test-dma-perf/benchmark.c b/app/test-dma-perf/benchmark.c
index 0601e0d171..5573acc9f9 100644
--- a/app/test-dma-perf/benchmark.c
+++ b/app/test-dma-perf/benchmark.c
@@ -279,6 +279,10 @@ do_cpu_mem_copy(void *p)
struct rte_mbuf **srcs = para->srcs;
struct rte_mbuf **dsts = para->dsts;
uint32_t i;
+ bool isAddrPaMode = false;
+
+ if (rte_eal_iova_mode() == RTE_IOVA_PA)
+ isAddrPaMode = true;
worker_info->stop_flag = false;
worker_info->ready_flag = true;
@@ -286,16 +290,33 @@ do_cpu_mem_copy(void *p)
while (!worker_info->start_flag)
;
- while (1) {
- for (i = 0; i < nr_buf; i++) {
- /* copy buffer form src to dst */
- rte_memcpy((void *)(uintptr_t)rte_mbuf_data_iova(dsts[i]),
- (void *)(uintptr_t)rte_mbuf_data_iova(srcs[i]),
- (size_t)buf_size);
- worker_info->total_cpl++;
+ if (true == isAddrPaMode) {
+ while (1) {
+ for (i = 0; i < nr_buf; i++) {
+ void *src = rte_pktmbuf_mtod(dsts[i], void *);
+ void *dst = rte_pktmbuf_mtod(srcs[i], void *);
+
+ /* copy buffer form src to dst */
+ rte_memcpy(dst,
+ src,
+ (size_t)buf_size);
+ worker_info->total_cpl++;
+ }
+ if (worker_info->stop_flag)
+ break;
+ }
+ } else {
+ while (1) {
+ for (i = 0; i < nr_buf; i++) {
+ /* copy buffer form src to dst */
+ rte_memcpy((void *)(uintptr_t)rte_mbuf_data_iova(dsts[i]),
+ (void *)(uintptr_t)rte_mbuf_data_iova(srcs[i]),
+ (size_t)buf_size);
+ worker_info->total_cpl++;
+ }
+ if (worker_info->stop_flag)
+ break;
}
- if (worker_info->stop_flag)
- break;
}
return 0;
--
2.34.1
^ permalink raw reply [flat|nested] 6+ messages in thread* [PATCH] app/dma-perf: fix physical address seg-fault
@ 2023-08-15 15:04 Vipin Varghese
0 siblings, 0 replies; 6+ messages in thread
From: Vipin Varghese @ 2023-08-15 15:04 UTC (permalink / raw)
To: thomas, dev; +Cc: Ferruh.Yigit, cheng1.jiang, stable
do_cpu_mem_copy uses DPDK API rte_mbuf_data_iova to return
the start of the virtual address for both src and dst.
But in case of iova mode set as PA, this results in seg-fault.
This is because rte_memcpy VA address and not PA.
this fix checks the mode and invoked rte_memcpy accrodingly.
Bugzilla ID: 1269
Fixes: 623dc9364dc6 ("app/dma-perf: introduce DMA performance test")
Cc: cheng1.jiang@intel.com
Cc: stable@dpdk.org
Signed-off-by: Vipin Varghese <vipin.varghese@amd.com>
---
tested for both va and pa
CMD:
PA: dpdk-test-dma-perf --iova-mode=pa -- --config test.ini
VA: dpdk-test-dma-perf --iova-mode=va -- --config test.ini
DC: dpdk-test-dma-perf --iova-mode=dc -- --config test.ini
Log: fails for dc mode `EAL: invalid parameters for --iova-mode`
test.ini:
```
[case1]
type=CPU_MEM_COPY
mem_size=10
buf_size=64,8192,2,MUL
src_numa_node=0
dst_numa_node=0
cache_flush=0
test_seconds=2
lcore = 7
eal_args=--in-memory --no-pci
```
---
app/test-dma-perf/benchmark.c | 36 ++++++++++++++++++++++++++---------
1 file changed, 27 insertions(+), 9 deletions(-)
diff --git a/app/test-dma-perf/benchmark.c b/app/test-dma-perf/benchmark.c
index 0601e0d171..73200e1935 100644
--- a/app/test-dma-perf/benchmark.c
+++ b/app/test-dma-perf/benchmark.c
@@ -279,6 +279,10 @@ do_cpu_mem_copy(void *p)
struct rte_mbuf **srcs = para->srcs;
struct rte_mbuf **dsts = para->dsts;
uint32_t i;
+ bool isAddrPaMode = false;
+
+ if (rte_eal_iova_mode() == RTE_IOVA_PA)
+ isAddrPaMode = true;
worker_info->stop_flag = false;
worker_info->ready_flag = true;
@@ -286,16 +290,30 @@ do_cpu_mem_copy(void *p)
while (!worker_info->start_flag)
;
- while (1) {
- for (i = 0; i < nr_buf; i++) {
- /* copy buffer form src to dst */
- rte_memcpy((void *)(uintptr_t)rte_mbuf_data_iova(dsts[i]),
- (void *)(uintptr_t)rte_mbuf_data_iova(srcs[i]),
- (size_t)buf_size);
- worker_info->total_cpl++;
+ if (true == isAddrPaMode) {
+ while (1) {
+ for (i = 0; i < nr_buf; i++) {
+ /* copy buffer form src to dst */
+ rte_memcpy((void *)dsts[i],
+ (void *)srcs[i],
+ (size_t)buf_size);
+ worker_info->total_cpl++;
+ }
+ if (worker_info->stop_flag)
+ break;
+ }
+ } else {
+ while (1) {
+ for (i = 0; i < nr_buf; i++) {
+ /* copy buffer form src to dst */
+ rte_memcpy((void *)(uintptr_t)rte_mbuf_data_iova(dsts[i]),
+ (void *)(uintptr_t)rte_mbuf_data_iova(srcs[i]),
+ (size_t)buf_size);
+ worker_info->total_cpl++;
+ }
+ if (worker_info->stop_flag)
+ break;
}
- if (worker_info->stop_flag)
- break;
}
return 0;
--
2.34.1
^ permalink raw reply [flat|nested] 6+ messages in thread
end of thread, other threads:[~2023-08-16 7:24 UTC | newest]
Thread overview: 6+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2023-08-15 15:10 [PATCH] app/dma-perf: fix physical address seg-fault Vipin Varghese
2023-08-15 15:15 ` Varghese, Vipin
2023-08-16 6:07 ` Anoob Joseph
2023-08-16 7:24 ` Varghese, Vipin
-- strict thread matches above, loose matches on Subject: below --
2023-08-16 7:18 Vipin Varghese
2023-08-15 15:04 Vipin Varghese
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).