On 01/06/15 03:56, Ouyang, Changchun wrote:
>> -Original Message-
>> From: Vlad Zolotarov [mailto:vladz at cloudius-systems.com]
>> Sent: Monday, January 5, 2015 6:10 PM
>> To: Ouyang, Changchun;dev at dpdk.org
>> Subject: Re: [dpdk-dev] [PATCH v4 4/6] ether: Check VMDq RSS mode
>>
>>
>> O
Add defconfig and rte.vars.mk files for Tile architecture.
Signed-off-by: Zhigang Lu
Signed-off-by: Cyril Chemparathy
---
config/defconfig_tile-tilegx-linuxapp-gcc | 78 +++
mk/arch/tile/rte.vars.mk | 59 +++
mk/machine/tilegx/rte
On architectures that do not rely on RTE_COMPILE_TIME_CPUFLAGS, the
compile_time_flags[] array can end up being zero sized. This results in a
compiler complaint in the subsequent loop. Pulling out the array size
computation silences this complaint.
Signed-off-by: Zhigang Lu
Signed-off-by: Cyril
Test all defined CPU flags for supported architectures so that we
do not have to include conditional compilation for each architecture
in app test case.
Signed-off-by: Zhigang Lu
Signed-off-by: Cyril Chemparathy
---
app/test/test_cpuflags.c | 78
This patch adds empty functions for CPU flags operations to support DPDK,
since tile processor doesn't have CPU flag hardware registers.
Signed-off-by: Zhigang Lu
Signed-off-by: Cyril Chemparathy
---
.../common/include/arch/tile/rte_cpuflags.h| 78 ++
1 file changed,
Signed-off-by: Zhigang Lu
Signed-off-by: Cyril Chemparathy
---
.../common/include/arch/tile/rte_common_vect.h | 49 ++
1 file changed, 49 insertions(+)
create mode 100644 lib/librte_eal/common/include/arch/tile/rte_common_vect.h
diff --git a/lib/librte_eal/common/includ
This patch splits vector operations from DPDK and push them
to architecture specific arch directories, so that other processor
architecture can implement its own vector functions to support DPDK.
Signed-off-by: Zhigang Lu
Signed-off-by: Cyril Chemparathy
---
lib/librte_eal/common/Makefile
This patch adds CPU TSC read operations for TileGx.
Signed-off-by: Zhigang Lu
Signed-off-by: Cyril Chemparathy
---
.../common/include/arch/tile/rte_cycles.h | 64 ++
1 file changed, 64 insertions(+)
create mode 100644 lib/librte_eal/common/include/arch/tile/rte_cyc
Signed-off-by: Zhigang Lu
Signed-off-by: Cyril Chemparathy
---
.../common/include/arch/tile/rte_memcpy.h | 97 ++
1 file changed, 97 insertions(+)
create mode 100644 lib/librte_eal/common/include/arch/tile/rte_memcpy.h
diff --git a/lib/librte_eal/common/include/arc
Signed-off-by: Zhigang Lu
Signed-off-by: Cyril Chemparathy
---
.../common/include/arch/tile/rte_prefetch.h| 62 ++
1 file changed, 62 insertions(+)
create mode 100644 lib/librte_eal/common/include/arch/tile/rte_prefetch.h
diff --git a/lib/librte_eal/common/include/a
TileGx uses generic spinlock operations.
Signed-off-by: Zhigang Lu
Signed-off-by: Cyril Chemparathy
---
.../common/include/arch/tile/rte_spinlock.h| 47 ++
1 file changed, 47 insertions(+)
create mode 100644 lib/librte_eal/common/include/arch/tile/rte_spinlock.h
di
This patch adds architecture specific byte swap and endianness
operations for TileGx.
Signed-off-by: Zhigang Lu
Signed-off-by: Cyril Chemparathy
---
.../common/include/arch/tile/rte_byteorder.h | 70 ++
1 file changed, 70 insertions(+)
create mode 100644 lib/librte_ea
This patch adds architecture specific memory barrier operations for
TileGx.
Signed-off-by: Zhigang Lu
Signed-off-by: Cyril Chemparathy
---
.../common/include/arch/tile/rte_atomic.h | 62 ++
1 file changed, 62 insertions(+)
create mode 100644 lib/librte_eal/common/i
On 01/06/15 04:01, Ouyang, Changchun wrote:
>
>> -Original Message-
>> From: Vlad Zolotarov [mailto:vladz at cloudius-systems.com]
>> Sent: Monday, January 5, 2015 6:12 PM
>> To: Ouyang, Changchun; dev at dpdk.org
>> Subject: Re: [dpdk-dev] [PATCH v4 6/6] testpmd: Set Rx VMDq RSS mode
>>
>
On 01/06/15 03:54, Ouyang, Changchun wrote:
>
>> -Original Message-
>> From: Vlad Zolotarov [mailto:vladz at cloudius-systems.com]
>> Sent: Monday, January 5, 2015 6:07 PM
>> To: Ouyang, Changchun; dev at dpdk.org
>> Subject: Re: [dpdk-dev] [PATCH v4 3/6] ixgbe: Get VF queue number
>>
>>
>
On 01/06/15 03:11, Ouyang, Changchun wrote:
>
>> -Original Message-
>> From: Vlad Zolotarov [mailto:vladz at cloudius-systems.com]
>> Sent: Monday, January 5, 2015 9:02 PM
>> To: Richardson, Bruce; Ouyang, Changchun
>> Cc: dev at dpdk.org
>> Subject: Re: [dpdk-dev] [PATCH v3 0/6] Enable VF
On 01/06/15 03:11, Ouyang, Changchun wrote:
>
>> -Original Message-
>> From: Vlad Zolotarov [mailto:vladz at cloudius-systems.com]
>> Sent: Monday, January 5, 2015 9:02 PM
>> To: Richardson, Bruce; Ouyang, Changchun
>> Cc: dev at dpdk.org
>> Subject: Re: [dpdk-dev] [PATCH v3 0/6] Enable VF
Changes in v4
Move common functions in eal_pci.c to librte_eal/common/
eal_common_pci.c file.
Following functions are moved to eal_common_pci.c file.
void *pci_map_resource(void *requested_addr, const int vfio_fd,
const char *devname, off_t offset, size_t size);
int pci_addr_comparison(stru
Changes in v4
Make rte_eal_hugepage_init and rte_eal_hugepage_attach as
wrapper functions for BSD.
Changes in v3
Changed subject to be more explicit on file name inclusion.
Changes in v2
Use common function names rte_eal_hugepage_init and
rte_eal_hugepage_attach for BSD and Linux. Update comments
Changes in v4
Removed extern declaration of eal_tsc_resolution_hz,
instead provided _set_ API.
Make set_tsc_freq_from_clock as wrapper function for BSD.
Changes in v3
Changed subject to be more explicit on file name inclusion.
Changes in v2
Use common function name set_tsc_freq_from_sysctl for BS
Changes in v4
Implement cpu_detected() for BSD.
Have common RTE_LOG for Linux and BSD in rte_eal_cpu_init().
Remove RTE_EXEC_ENV_BSDAPP in common file.
Changes in v3
Changed subject to be more explicit on file name inclusion.
Changes in v2
None
Changes in v1
Move common function in eal_lcore.c t
Changes in v4
Remove eal_externs.h file, instead use _get_ and _set_ APIS
to access those variables.
Split eal_common.c into eal_common_system.c and
and eal_common_runtime.c
rte_eal prefix functions are moved to _runtime_ and
eal prefix functions are moved to _system_ files respectively
Changes i
Changes in v4
None
Changes in v3
Changed subject to be more explicit on file name inclusion.
Changes in v2
None
Changes in v1
eal_thread.c has minor differences between Linux and BSD, move
entire file into common directory.
Use RTE_EXEC_ENV_BSDAPP to differentiate on minor differences.
Rename ea
Changes in v4
None
Changes in v3
Changed subject to be more explicit on file name inclusion.
Changes in v2
Remove rte_dump_registers() function since it is not implemented.
Fix comment for _rte_exit()
Changes in v1
eal_debug.c has no difference between Linux and BSD, move
entire file into common
Fix v3 PATCH review comments from Thomas.
eal_common.c is further split into eal_common_system.c and
eal_common_runtime.c files with appropriate functions.
eal_externs.h file is removed and _get_ and _set_ APIs are
used to access variables.
Functions with same signature in Linux and BSD but differe
Fix v3 PATCH review comments from Thomas.
eal_common.c is further split into eal_common_system.c and
eal_common_runtime.c files with appropriate functions.
eal_externs.h file is removed and _get_ and _set_ APIs are
used to access variables.
Functions with same signature in Linux and BSD but differe
2014-12-27 15:13, Stephen Hemminger:
> This is a patch to address the conflict between
> and the definitions in . It has two side effects
> worth discussion:
> 1. It forces inclusion of net/ethernet.h
> 2. It has definition to deal with the differing structure elements
> in the two versio
2014-12-30 04:45, Sujith Sankar:
> On 29/12/14 1:45 pm, "Wu, Jingjing" wrote:
> >I found that in perfect fdir is also supported in enic driver.
> >
> >During the R1.8 development, we defined a new dev_ops call filter_ctrl,
> >which can be used to control kinds of filters, flow director is included
2015-01-02 09:07, Lyn M:
> In my original post, I thought my choice of hex portmask -p 0x3 was causing
> this issue -- now I know that was not the case. But, I am still curious
> about how the hex portmask is determined. Since I only have two ports
> bound to igb_uio, my hex postmask will always
Hi Neil,
Any further comments on that one?
Konstantin
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Ananyev, Konstantin
> Sent: Thursday, December 18, 2014 3:02 PM
> To: Neil Horman
> Cc: dev at dpdk.org
> Subject: Re: [dpdk-dev] [PATCH 10/17] librte_acl: add AVX2 as new
> rte_acl_cl
On Tue, Jan 06, 2015 at 10:02:25AM +0100, Thomas Monjalon wrote:
> 2015-01-02 09:07, Lyn M:
> > In my original post, I thought my choice of hex portmask -p 0x3 was causing
> > this issue -- now I know that was not the case. But, I am still curious
> > about how the hex portmask is determined. Sin
Inline
On Mon, Jan 5, 2015 at 12:38 PM, Thomas Monjalon
wrote:
> 2015-01-05 10:56, Ravi Kerur:
> > On Mon, Jan 5, 2015 at 7:59 AM, Thomas Monjalon <
> thomas.monjalon at 6wind.com>
> > wrote:
> > > 2014-12-25 10:33, Ravi Kerur:
> > > > Move common functions in eal.c to librte_eal/common directo
I have added wrapper function for BSD.
On Mon, Jan 5, 2015 at 1:17 PM, Thomas Monjalon
wrote:
> 2014-12-30 11:37, Ravi Kerur:
> > Use common function names rte_eal_hugepage_init and
> > rte_eal_hugepage_attach for BSD and Linux.
>
> I'd like to be sure that using hugepage wording instead of cont
On Tue, Jan 06, 2015 at 09:57:40AM +, Ananyev, Konstantin wrote:
>
> Hi Neil,
> Any further comments on that one?
> Konstantin
>
No, I'm good. You're comment regarding compiler support makes sense (though its
really unfortunate that we have to do that). Still need to address the ifdefery
ar
> -Original Message-
> From: Vlad Zolotarov [mailto:vladz at cloudius-systems.com]
> Sent: Monday, January 5, 2015 6:12 PM
> To: Ouyang, Changchun; dev at dpdk.org
> Subject: Re: [dpdk-dev] [PATCH v4 6/6] testpmd: Set Rx VMDq RSS mode
>
>
> On 01/05/15 04:38, Ouyang, Changchun wrote:
>
> -Original Message-
> From: Vlad Zolotarov [mailto:vladz at cloudius-systems.com]
> Sent: Monday, January 5, 2015 6:10 PM
> To: Ouyang, Changchun; dev at dpdk.org
> Subject: Re: [dpdk-dev] [PATCH v4 4/6] ether: Check VMDq RSS mode
>
>
> On 01/05/15 03:00, Ouyang, Changchun wrote:
> >
>
> -Original Message-
> From: Vlad Zolotarov [mailto:vladz at cloudius-systems.com]
> Sent: Monday, January 5, 2015 6:07 PM
> To: Ouyang, Changchun; dev at dpdk.org
> Subject: Re: [dpdk-dev] [PATCH v4 3/6] ixgbe: Get VF queue number
>
>
> On 01/05/15 04:59, Ouyang, Changchun wrote:
> >
>
> -Original Message-
> From: Vlad Zolotarov [mailto:vladz at cloudius-systems.com]
> Sent: Monday, January 5, 2015 9:02 PM
> To: Richardson, Bruce; Ouyang, Changchun
> Cc: dev at dpdk.org
> Subject: Re: [dpdk-dev] [PATCH v3 0/6] Enable VF RSS for Niantic
>
>
> On 01/05/15 12:38, Bruce R
> -Original Message-
> From: Richardson, Bruce
> Sent: Monday, January 5, 2015 6:38 PM
> To: Ouyang, Changchun
> Cc: Vlad Zolotarov; dev at dpdk.org
> Subject: Re: [dpdk-dev] [PATCH v3 0/6] Enable VF RSS for Niantic
>
> On Thu, Dec 25, 2014 at 01:46:54AM +, Ouyang, Changchun wrote:
>
> -Original Message-
> From: Richardson, Bruce
> Sent: Monday, January 5, 2015 6:29 PM
> To: Ouyang, Changchun
> Cc: Vlad Zolotarov; dev at dpdk.org
> Subject: Re: [dpdk-dev] [PATCH v3 5/6] ixgbe: Config VF RSS
>
> On Fri, Dec 26, 2014 at 01:52:25AM +, Ouyang, Changchun wrote:
> >
>
40 matches
Mail list logo
