[dpdk-dev] [PATCH v2 00/12] Introducing the TILE-Gx platform

On Mon, 22 Jun 2015 11:58:32 -0700 Cyril Chemparathy wrote: > This series adds support for the EZchip TILE-Gx family of SoCs. The > architecture port in itself is fairly straight forward due to its > reliance on generics for the most part. > > In addition to adding TILE-

[dpdk-dev] [PATCH v2 08/12] mempool: allow config override on element alignment

On Tue, 23 Jun 2015 00:31:06 + "Ananyev, Konstantin" wrote: > > +#define RTE_MEMPOOL_ALIGN_MASK (RTE_MEMPOOL_ALIGN - 1) > > I am probably a bit late with my comments, but why not make it a > runtime decision then? I know we can't add a new parameter to > mempool_xmem_create() without A

[dpdk-dev] [PATCH v4 0/9] Improve cast alignment for strict aligned machines

On Tue, 23 Jun 2015 17:36:40 +0200 Olivier MATZ wrote: > Series > Acked-by: Olivier Matz > Thanks Olivier.

[dpdk-dev] [PATCH] lib: fix RTE_MBUF_METADATA macros

On Mon, 22 Jun 2015 22:16:59 +0200 Thomas Monjalon wrote: > 2015-06-05 15:31, Dumitrescu, Cristian: > > > Fix RTE_MBUF_METADATA macros to allow for unaligned accesses to > > > meta-data fields. > > > Forcing aligned accesses is not really required, so this is > > > removing an unneeded constraint

[dpdk-dev] [PATCH] lib: fix RTE_MBUF_METADATA macros

On Mon, 22 Jun 2015 22:16:59 +0200 Thomas Monjalon wrote: > 2015-06-05 15:31, Dumitrescu, Cristian: > > > Fix RTE_MBUF_METADATA macros to allow for unaligned accesses to > > > meta-data fields. > > > Forcing aligned accesses is not really required, so this is > > > removing an unneeded constraint

[dpdk-dev] [PATCH v2 12/12] maintainers: claim responsibility for TILE-Gx platform

Change-Id: I6491108ff86c1249bf4ffa4d4624c01b4594805e Signed-off-by: Cyril Chemparathy --- MAINTAINERS | 4 1 file changed, 4 insertions(+) diff --git a/MAINTAINERS b/MAINTAINERS index 9362c19..bffc1e6 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -113,6 +113,10 @@ M: Bruce Richardson M

[dpdk-dev] [PATCH v2 11/12] tile: Add TILE-Gx mPIPE poll mode driver.

This commit adds a poll mode driver for the mPIPE hardware present on TILE-Gx SoCs. Change-Id: I1b9a9ef2c9f1c96810ec58b4d2ae77b870a6ec94 Signed-off-by: Cyril Chemparathy --- config/defconfig_tile-tilegx-linuxapp-gcc |1 + drivers/net/Makefile |1 + drivers/net/mpipe

[dpdk-dev] [PATCH v2 10/12] tile: initial TILE-Gx support.

This commit adds support for the TILE-Gx platform, as well as the TILE CPU architecture. This architecture port is fairly simple due to its reliance on generics for most arch stuff. Change-Id: I809fcf740e25ba5976a6b7736c1673515338cf80 Signed-off-by: Cyril Chemparathy --- config/defconfig_tile

[dpdk-dev] [PATCH v2 09/12] tile: add page sizes for TILE-Gx/Mx platforms

This patch adds a few new page sizes that are supported on the TILE-Gx and TILE-Mx platforms. Change-Id: I7da783703149c4f96b84d0017fa4f1191fc029c6 Signed-off-by: Cyril Chemparathy --- lib/librte_eal/common/eal_common_memzone.c | 8 + lib/librte_eal/common/include/rte_memory.h | 16

[dpdk-dev] [PATCH v2 08/12] mempool: allow config override on element alignment

: I9cd789d92b0bc9c8f44a633de59bb04d45d927a7 Signed-off-by: Cyril Chemparathy --- lib/librte_mempool/rte_mempool.c | 16 +--- lib/librte_mempool/rte_mempool.h | 6 ++ 2 files changed, 15 insertions(+), 7 deletions(-) diff --git a/lib/librte_mempool/rte_mempool.c b/lib/librte_mempool/rte_mempool.c index 002d3a8

[dpdk-dev] [PATCH v2 07/12] memzone: allow multiple pagesizes to be requested

on PPC), thereby allowing for a broader range of hugepages on architectures that support it. Change-Id: Ic3713f61da49629a570fe4de34a8aaf5e2e0a19b Signed-off-by: Cyril Chemparathy --- lib/librte_eal/common/eal_common_memzone.c | 58 ++ 1 file changed, 27 insertions

[dpdk-dev] [PATCH v2 06/12] memzone: refactor rte_memzone_reserve() variants

then called by all three variants of rte_memzone_reserve(). Change-Id: Id26c25b4dd3d07861eaf35e72aaa2de555916fa3 Signed-off-by: Cyril Chemparathy --- lib/librte_eal/common/eal_common_memzone.c | 77 +- 1 file changed, 33 insertions(+), 44 deletions(-) diff --git a/lib

[dpdk-dev] [PATCH v2 05/12] config: remove RTE_LIBNAME definition.

The library name is now being pinned to "dpdk" instead of intel_dpdk, powerpc_dpdk, etc. As a result, we no longer need this config item. This patch removes it. Change-Id: I36f7cf6c18c3563c6f5ccdf01bb70579c7ccaa16 Signed-off-by: Cyril Chemparathy --- config/com

[dpdk-dev] [PATCH v2 04/12] eal: allow empty compile time flags

Signed-off-by: Cyril Chemparathy --- lib/librte_eal/common/eal_common_cpuflags.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/lib/librte_eal/common/eal_common_cpuflags.c b/lib/librte_eal/common/eal_common_cpuflags.c index 6fd360c..8ba7b30 100644 --- a/lib/librte_

[dpdk-dev] [PATCH v2 03/12] hash: check SSE flags only on x86 builds

This is necessary because the required CPU flags may not be defined on other architectures. Change-Id: I14d3f9f625b2e7567123f1c97095f8d06abd674b Signed-off-by: Cyril Chemparathy --- lib/librte_hash/rte_hash_crc.h | 2 ++ 1 file changed, 2 insertions(+) diff --git a/lib/librte_hash

[dpdk-dev] [PATCH v2 02/12] hash: fix compilation on non-X86 platforms

The "hash: remove duplicated code" change unfortunately broke the build for non-X86 platforms. This patch fixes this breakage. Change-Id: Ie109d67e681b75b45320fab1bf9de4eb9c9701bf Signed-off-by: Cyril Chemparathy --- lib/librte_hash/rte_jhash.h | 3 ++- 1 file changed, 2 insert

[dpdk-dev] [PATCH v2 01/12] test: limit x86 cpuflags checks to x86 builds

The original code mistakenly defaulted to X86 when RTE_ARCH_PPC_64 was left undefined. This did not accomodate other non-PPC/non-X86 architectures. This patch fixes this issue. Change-Id: I5e8cf33c2eb917f7f6583dc95ed0f336066a285e Signed-off-by: Cyril Chemparathy --- app/test/test_cpuflags.c

[dpdk-dev] [PATCH v2 00/12] Introducing the TILE-Gx platform

(cpuflags, SSE related, etc.), as well as minor extensions to to accomodate a wider range of hugepage sizes and configurable mempool element alignment boundaries. Changes in this series: v2: Removed RTE_LIBNAME per Thomas' feedback. Cyril Chemparathy (12): test: limit x86 cpuflags checks t

[dpdk-dev] [PATCH v4 9/9] librte_mbuf: Apply mtod-offset.cocci transform

simply merging in the change. Acked-by: Olivier Matz Change-Id: I1cdc4fe6f7954cebb520e495a9454cb586cace54 Signed-off-by: Cyril Chemparathy --- app/test-pmd/ieee1588fwd.c | 4 +- app/test-pmd/rxonly.c| 21 ++ app/test-pmd/txonly.c| 4

[dpdk-dev] [PATCH v4 8/9] librte_mbuf: Add transform for rte_pktmbuf_mtod_offset()

Signed-off-by: Cyril Chemparathy --- scripts/cocci.sh| 64 ++ scripts/cocci/mtod-offset.cocci | 76 + 2 files changed, 140 insertions(+) create mode 100755 scripts/cocci.sh create mode 100644 scripts/cocci/mtod

[dpdk-dev] [PATCH v4 7/9] librte_mbuf: Add rte_pktmbuf_mtod_offset()

: I4dec35a15e44a7b8c767559a8d3b294177f39552 Signed-off-by: Cyril Chemparathy --- lib/librte_mbuf/rte_mbuf.h | 24 +--- 1 file changed, 21 insertions(+), 3 deletions(-) diff --git a/lib/librte_mbuf/rte_mbuf.h b/lib/librte_mbuf/rte_mbuf.h index 6efd2b5..396963c 100644 --- a/lib

[dpdk-dev] [PATCH v4 6/9] app/test-pmd: pack simple_gre_hdr

Not packing this causes -Wcast-align breakage on machines that are strict on alignment. This patch fixes this bug. Acked-by: Olivier Matz Change-Id: I7feab9ea4fc33c4ebdf3503ca67477ca0809bb7e Signed-off-by: Cyril Chemparathy --- app/test-pmd/csumonly.c | 2 +- 1 file changed, 1 insertion(+), 1

[dpdk-dev] [PATCH v4 5/9] eal: add and use unaligned integer types

efine CONFIG_RTE_ARCH_STRICT_ALIGN in order to effect these new types. Change-Id: I6f7c429fc14233c991287d065ecf2a09dbd65ebb Signed-off-by: Cyril Chemparathy --- app/test-pmd/flowgen.c | 4 ++-- app/test-pmd/icmpecho.c| 2 +- app/test-pmd/txo

[dpdk-dev] [PATCH v4 4/9] hash: silence -Wcast-align on pointer arithmetic

(), thereby silencing the compiler by casting through (void *). Change-Id: Ia7102cf3f870752743cfe9f4443a3e53cd99bac1 Signed-off-by: Cyril Chemparathy --- lib/librte_hash/rte_hash.c | 10 +- 1 file changed, 5 insertions(+), 5 deletions(-) diff --git a/lib/librte_hash/rte_hash.c b/lib/librte_hash

[dpdk-dev] [PATCH v4 3/9] ethdev: silence -Wcast-align on pointer arithmetic

: I6460b4b84b89dce6ac0b7b20e7002d53bcbd22db Signed-off-by: Cyril Chemparathy --- lib/librte_ether/rte_ethdev.c | 24 1 file changed, 12 insertions(+), 12 deletions(-) diff --git a/lib/librte_ether/rte_ethdev.c b/lib/librte_ether/rte_ethdev.c index e13fde5..02cd07f 100644 --- a/lib

[dpdk-dev] [PATCH v4 2/9] mbuf: silence -Wcast-align on pointer arithmetic

: I60d9b7a9205ff0befb8dbb4cdcb1df6f9d9d0250 Signed-off-by: Cyril Chemparathy --- lib/librte_mbuf/rte_mbuf.h | 8 +--- 1 file changed, 1 insertion(+), 7 deletions(-) diff --git a/lib/librte_mbuf/rte_mbuf.h b/lib/librte_mbuf/rte_mbuf.h index 6c9cfd6..6efd2b5 100644 --- a/lib/librte_mbuf/rte_mbuf.h

[dpdk-dev] [PATCH v4 1/9] mempool: silence -Wcast-align on pointer arithmetic

: I4a67c9aa1bf012a0bf860d21bb8105db89062c76 Signed-off-by: Cyril Chemparathy --- lib/librte_mempool/rte_mempool.c | 2 +- lib/librte_mempool/rte_mempool.h | 6 ++ 2 files changed, 3 insertions(+), 5 deletions(-) diff --git a/lib/librte_mempool/rte_mempool.c b/lib/librte_mempool/rte_mempool.c

[dpdk-dev] [PATCH v4 0/9] Improve cast alignment for strict aligned machines

v3: Extends unaligned fixes to new code introduced on master. v4: Fixes based on Thomas' comments. Extends unaligned fixes to new code introduced on master. Cyril Chemparathy (9): mempool: silence -Wcast-align on pointer arithmetic mbuf: silence -Wcast-align on pointer arithmeti

[dpdk-dev] [PATCH 09/11] tile: initial TILE-Gx support.

On Mon, 22 Jun 2015 19:00:44 +0200 Thomas Monjalon wrote: > > > Not sure it makes sense to have different library names. > > > Why not renaming all to "dpdk"? > > > > This config doesn't seem to be getting used anywhere. Maybe just > > get rid of it globally? > > No it's used in mk/ direct

[dpdk-dev] [PATCH 09/11] tile: initial TILE-Gx support.

On Mon, 22 Jun 2015 11:39:07 +0200 Thomas Monjalon wrote: > > +CONFIG_RTE_LIBNAME="tile_dpdk" > > Not sure it makes sense to have different library names. > Why not renaming all to "dpdk"? This config doesn't seem to be getting used anywhere. Maybe just get rid of it globally?

[dpdk-dev] [PATCH v2 05/10] eal: add and use unaligned integer types

On Mon, 22 Jun 2015 11:36:20 +0200 Thomas Monjalon wrote: > 2015-06-19 10:34, Cyril Chemparathy: > > On machines that are strict on pointer alignment, current code > > breaks on GCC's -Wcast-align checks on casts from narrower to wider > > types. This patch introduces n

[dpdk-dev] [PATCH] mk: add support for gdb debug info generation

On Mon, 22 Jun 2015 08:44:41 +0100 "Gonzalez Monroy, Sergio" wrote: > I don't think you need to modify the makefiles and introduce a new > compile time option for this. > The same result can be easily achieved by setting EXTRA_CFLAGS in the > command line. ie: > $ make install T=x86_64-nat

[dpdk-dev] [PATCH v2 07/10] app/test: use struct ether_addr instead of a byte array cast

On Sun, 21 Jun 2015 21:56:24 -0400 Stephen Hemminger wrote: > On Fri, 19 Jun 2015 10:34:50 -0700 > Cyril Chemparathy wrote: > > > + static struct ether_addr src_mac = > > + { { 0x00, 0xFF, 0xAA, 0xFF, 0xAA, 0xFF } }; > > + stati

[dpdk-dev] [PATCH] examples/l2fwd: Add forward count limit.

This commit adds a forward count argument, which is used to terminate the test once a certain number of packets have been forwarded. Change-Id: Ia3e7ff5d41c3e947509b0653d53271b882fc04de Signed-off-by: Cyril Chemparathy --- examples/l2fwd/main.c | 46

[dpdk-dev] [PATCH] mk: add support for gdb debug info generation

From: Cyril Chemparathy It is often useful to build with debug enabled, we add a config (CONFIG_RTE_TOOLCHAIN_DEBUG) to do so. Note: This patch does not include corresponding changes for ICC. The author pleads abject ignorance in this regard, and welcomes recommendations. :-) Change-Id

[dpdk-dev] [PATCH 11/11] maintainers: claim responsibility for TILE-Gx platform

Change-Id: I6491108ff86c1249bf4ffa4d4624c01b4594805e Signed-off-by: Cyril Chemparathy --- MAINTAINERS | 4 1 file changed, 4 insertions(+) diff --git a/MAINTAINERS b/MAINTAINERS index 9362c19..bffc1e6 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -113,6 +113,10 @@ M: Bruce Richardson M

[dpdk-dev] [PATCH 10/11] tile: Add TILE-Gx mPIPE poll mode driver.

This commit adds a poll mode driver for the mPIPE hardware present on TILE-Gx SoCs. Change-Id: I1b9a9ef2c9f1c96810ec58b4d2ae77b870a6ec94 Signed-off-by: Cyril Chemparathy --- config/defconfig_tile-tilegx-linuxapp-gcc |1 + drivers/net/Makefile |1 + drivers/net/mpipe

[dpdk-dev] [PATCH 09/11] tile: initial TILE-Gx support.

This commit adds support for the TILE-Gx platform, as well as the TILE CPU architecture. This architecture port is fairly simple due to its reliance on generics for most arch stuff. Change-Id: I809fcf740e25ba5976a6b7736c1673515338cf80 Signed-off-by: Cyril Chemparathy --- config/defconfig_tile

[dpdk-dev] [PATCH 08/11] tile: add page sizes for TILE-Gx/Mx platforms

This patch adds a few new page sizes that are supported on the TILE-Gx and TILE-Mx platforms. Change-Id: I7da783703149c4f96b84d0017fa4f1191fc029c6 Signed-off-by: Cyril Chemparathy --- lib/librte_eal/common/eal_common_memzone.c | 8 + lib/librte_eal/common/include/rte_memory.h | 16

[dpdk-dev] [PATCH 07/11] mempool: allow config override on element alignment

: I9cd789d92b0bc9c8f44a633de59bb04d45d927a7 Signed-off-by: Cyril Chemparathy --- lib/librte_mempool/rte_mempool.c | 16 +--- lib/librte_mempool/rte_mempool.h | 6 ++ 2 files changed, 15 insertions(+), 7 deletions(-) diff --git a/lib/librte_mempool/rte_mempool.c b/lib/librte_mempool/rte_mempool.c index 190cfd9

[dpdk-dev] [PATCH 06/11] memzone: allow multiple pagesizes to be requested

on PPC), thereby allowing for a broader range of hugepages on architectures that support it. Change-Id: Ic3713f61da49629a570fe4de34a8aaf5e2e0a19b Signed-off-by: Cyril Chemparathy --- lib/librte_eal/common/eal_common_memzone.c | 58 ++ 1 file changed, 27 insertions

[dpdk-dev] [PATCH 05/11] memzone: refactor rte_memzone_reserve() variants

then called by all three variants of rte_memzone_reserve(). Change-Id: Id26c25b4dd3d07861eaf35e72aaa2de555916fa3 Signed-off-by: Cyril Chemparathy --- lib/librte_eal/common/eal_common_memzone.c | 77 +- 1 file changed, 33 insertions(+), 44 deletions(-) diff --git a/lib

[dpdk-dev] [PATCH 04/11] eal: allow empty compile time flags

Signed-off-by: Cyril Chemparathy --- lib/librte_eal/common/eal_common_cpuflags.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/lib/librte_eal/common/eal_common_cpuflags.c b/lib/librte_eal/common/eal_common_cpuflags.c index 6fd360c..8ba7b30 100644 --- a/lib/librte_

[dpdk-dev] [PATCH 03/11] hash: check SSE flags only on x86 builds

This is necessary because the required CPU flags may not be defined on other architectures. Change-Id: I14d3f9f625b2e7567123f1c97095f8d06abd674b Signed-off-by: Cyril Chemparathy --- lib/librte_hash/rte_hash_crc.h | 2 ++ 1 file changed, 2 insertions(+) diff --git a/lib/librte_hash

[dpdk-dev] [PATCH 02/11] hash: fix compilation on non-X86 platforms

The "hash: remove duplicated code" change unfortunately broke the build for non-X86 platforms. This patch fixes this breakage. Change-Id: Ie109d67e681b75b45320fab1bf9de4eb9c9701bf Signed-off-by: Cyril Chemparathy --- lib/librte_hash/rte_jhash.h | 3 ++- 1 file changed, 2 insert

[dpdk-dev] [PATCH 01/11] test: limit x86 cpuflags checks to x86 builds

The original code mistakenly defaulted to X86 when RTE_ARCH_PPC_64 was left undefined. This did not accomodate other non-PPC/non-X86 architectures. This patch fixes this issue. Change-Id: I5e8cf33c2eb917f7f6583dc95ed0f336066a285e Signed-off-by: Cyril Chemparathy --- app/test/test_cpuflags.c

[dpdk-dev] [PATCH 00/11] Introducing the TILE-Gx platform

(cpuflags, SSE related, etc.), as well as minor extensions to to accomodate a wider range of hugepage sizes and configurable mempool element alignment boundaries. Cyril Chemparathy (11): test: limit x86 cpuflags checks to x86 builds hash: fix compilation on non-X86 platforms hash: check SSE flags

[dpdk-dev] [PATCH v3 10/10] librte_mbuf: Apply mtod-offset.cocci transform

simply merging in the change. Acked-by: Olivier Matz Signed-off-by: Cyril Chemparathy --- app/test-pmd/ieee1588fwd.c | 4 +- app/test-pmd/rxonly.c| 21 ++ app/test-pmd/txonly.c| 4 +- app/test/packet_burst_generator.c| 5

[dpdk-dev] [PATCH v3 09/10] librte_mbuf: Add transform for rte_pktmbuf_mtod_offset()

This patch adds a coccinelle (see http://coccinelle.lip6.fr/) transform to use the newly added rte_pktmbuf_mtod_offset() helper. In addition, we add a simple script to apply all available transforms to a codebase. Acked-by: Olivier Matz Signed-off-by: Cyril Chemparathy --- scripts/cocci.sh

[dpdk-dev] [PATCH v3 08/10] librte_mbuf: Add rte_pktmbuf_mtod_offset()

-by: Cyril Chemparathy --- lib/librte_mbuf/rte_mbuf.h | 24 +--- 1 file changed, 21 insertions(+), 3 deletions(-) diff --git a/lib/librte_mbuf/rte_mbuf.h b/lib/librte_mbuf/rte_mbuf.h index 6efd2b5..396963c 100644 --- a/lib/librte_mbuf/rte_mbuf.h +++ b/lib/librte_mbuf/rte_mbuf.h

[dpdk-dev] [PATCH v3 07/10] app/test: use struct ether_addr instead of a byte array cast

This patch instantiates struct ether_addr instead of type casting a uint8_t pointer to the same structure. The type cast breaks on machines that enforce strict alignment. Acked-by: Olivier Matz Signed-off-by: Cyril Chemparathy --- app/test/test_pmd_perf.c | 10 +- 1 file changed, 5

[dpdk-dev] [PATCH v3 06/10] app/test-pmd: pack simple_gre_hdr

Not packing this causes -Wcast-align breakage on machines that are strict on alignment. This patch fixes this bug. Acked-by: Olivier Matz Signed-off-by: Cyril Chemparathy --- app/test-pmd/csumonly.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/app/test-pmd/csumonly.c b

[dpdk-dev] [PATCH v3 05/10] eal: add and use unaligned integer types

efine CONFIG_RTE_ARCH_STRICT_ALIGN in order to effect these new types. Signed-off-by: Cyril Chemparathy --- app/test-pmd/flowgen.c | 4 ++-- app/test-pmd/icmpecho.c| 2 +- app/test-pmd/txonly.c | 2 +- app/test/packet_burst_genera

[dpdk-dev] [PATCH v3 04/10] hash: silence -Wcast-align on pointer arithmetic

(), thereby silencing the compiler by casting through (void *). Acked-by: Olivier Matz Signed-off-by: Cyril Chemparathy --- lib/librte_hash/rte_hash.c | 10 +- 1 file changed, 5 insertions(+), 5 deletions(-) diff --git a/lib/librte_hash/rte_hash.c b/lib/librte_hash/rte_hash.c index 9245716

[dpdk-dev] [PATCH v3 03/10] ethdev: silence -Wcast-align on pointer arithmetic

-off-by: Cyril Chemparathy --- lib/librte_ether/rte_ethdev.c | 24 1 file changed, 12 insertions(+), 12 deletions(-) diff --git a/lib/librte_ether/rte_ethdev.c b/lib/librte_ether/rte_ethdev.c index e13fde5..02cd07f 100644 --- a/lib/librte_ether/rte_ethdev.c +++ b/lib

[dpdk-dev] [PATCH v3 02/10] mbuf: silence -Wcast-align on pointer arithmetic

: Olivier Matz Signed-off-by: Cyril Chemparathy --- lib/librte_mbuf/rte_mbuf.h | 8 +--- 1 file changed, 1 insertion(+), 7 deletions(-) diff --git a/lib/librte_mbuf/rte_mbuf.h b/lib/librte_mbuf/rte_mbuf.h index 6c9cfd6..6efd2b5 100644 --- a/lib/librte_mbuf/rte_mbuf.h +++ b/lib/librte_mbuf/rte_mbuf.h

[dpdk-dev] [PATCH v3 01/10] mempool: silence -Wcast-align on pointer arithmetic

: Olivier Matz Signed-off-by: Cyril Chemparathy --- lib/librte_mempool/rte_mempool.h | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) diff --git a/lib/librte_mempool/rte_mempool.h b/lib/librte_mempool/rte_mempool.h index bc2bae0..8be040b 100644 --- a/lib/librte_mempool/rte_mempool.h +++ b

[dpdk-dev] [PATCH v3 00/10] Improve cast alignment for strict aligned machines

v3: Extends unaligned fixes to new code introduced on master. Cyril Chemparathy (10): mempool: silence -Wcast-align on pointer arithmetic mbuf: silence -Wcast-align on pointer arithmetic ethdev: silence -Wcast-align on pointer arithmetic hash: silence -Wcast-align on pointer arithmetic

[dpdk-dev] [PATCH v2 10/10] librte_mbuf: Apply mtod-offset.cocci transform

simply merging in the change. Acked-by: Olivier Matz Signed-off-by: Cyril Chemparathy --- app/test-pmd/ieee1588fwd.c | 4 +- app/test-pmd/rxonly.c| 21 ++ app/test-pmd/txonly.c| 4 +- app/test/packet_burst_generator.c| 5

[dpdk-dev] [PATCH v2 09/10] librte_mbuf: Add transform for rte_pktmbuf_mtod_offset()

This patch adds a coccinelle (see http://coccinelle.lip6.fr/) transform to use the newly added rte_pktmbuf_mtod_offset() helper. In addition, we add a simple script to apply all available transforms to a codebase. Acked-by: Olivier Matz Signed-off-by: Cyril Chemparathy --- scripts/cocci.sh

[dpdk-dev] [PATCH v2 08/10] librte_mbuf: Add rte_pktmbuf_mtod_offset()

-by: Cyril Chemparathy --- lib/librte_mbuf/rte_mbuf.h | 24 +--- 1 file changed, 21 insertions(+), 3 deletions(-) diff --git a/lib/librte_mbuf/rte_mbuf.h b/lib/librte_mbuf/rte_mbuf.h index 6efd2b5..396963c 100644 --- a/lib/librte_mbuf/rte_mbuf.h +++ b/lib/librte_mbuf/rte_mbuf.h

[dpdk-dev] [PATCH v2 07/10] app/test: use struct ether_addr instead of a byte array cast

This patch instantiates struct ether_addr instead of type casting a uint8_t pointer to the same structure. The type cast breaks on machines that enforce strict alignment. Acked-by: Olivier Matz Signed-off-by: Cyril Chemparathy --- app/test/test_pmd_perf.c | 10 +- 1 file changed, 5

[dpdk-dev] [PATCH v2 06/10] app/test-pmd: pack simple_gre_hdr

Not packing this causes -Wcast-align breakage on machines that are strict on alignment. This patch fixes this bug. Acked-by: Olivier Matz Signed-off-by: Cyril Chemparathy --- app/test-pmd/csumonly.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/app/test-pmd/csumonly.c b

[dpdk-dev] [PATCH v2 05/10] eal: add and use unaligned integer types

clang, and equivalents will need to be plugged in. Signed-off-by: Cyril Chemparathy --- app/test-pmd/flowgen.c | 4 ++-- app/test-pmd/txonly.c | 2 +- app/test/packet_burst_generator.c | 4 ++-- app/test/test_mbuf.c

[dpdk-dev] [PATCH v2 04/10] hash: silence -Wcast-align on pointer arithmetic

(), thereby silencing the compiler by casting through (void *). Acked-by: Olivier Matz Signed-off-by: Cyril Chemparathy --- lib/librte_hash/rte_hash.c | 10 +- 1 file changed, 5 insertions(+), 5 deletions(-) diff --git a/lib/librte_hash/rte_hash.c b/lib/librte_hash/rte_hash.c index 9245716

[dpdk-dev] [PATCH v2 03/10] ethdev: silence -Wcast-align on pointer arithmetic

-off-by: Cyril Chemparathy --- lib/librte_ether/rte_ethdev.c | 24 1 file changed, 12 insertions(+), 12 deletions(-) diff --git a/lib/librte_ether/rte_ethdev.c b/lib/librte_ether/rte_ethdev.c index e13fde5..02cd07f 100644 --- a/lib/librte_ether/rte_ethdev.c +++ b/lib

[dpdk-dev] [PATCH v2 02/10] mbuf: silence -Wcast-align on pointer arithmetic

: Olivier Matz Signed-off-by: Cyril Chemparathy --- lib/librte_mbuf/rte_mbuf.h | 8 +--- 1 file changed, 1 insertion(+), 7 deletions(-) diff --git a/lib/librte_mbuf/rte_mbuf.h b/lib/librte_mbuf/rte_mbuf.h index 6c9cfd6..6efd2b5 100644 --- a/lib/librte_mbuf/rte_mbuf.h +++ b/lib/librte_mbuf/rte_mbuf.h

[dpdk-dev] [PATCH v2 01/10] mempool: silence -Wcast-align on pointer arithmetic

: Olivier Matz Signed-off-by: Cyril Chemparathy --- lib/librte_mempool/rte_mempool.h | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) diff --git a/lib/librte_mempool/rte_mempool.h b/lib/librte_mempool/rte_mempool.h index bc2bae0..8be040b 100644 --- a/lib/librte_mempool/rte_mempool.h +++ b

[dpdk-dev] [PATCH v2 00/10] Improve cast alignment for strict aligned machines

*)(rte_pktmbuf_mtod(m, char *) + offset) with: rte_pktmbuf_mtod_offset(m, struct foo *, offset) To ensure consistency, the above transform was applied throughout the code base using the coccinelle semantic patching tool. Cyril Chemparathy (10): mempool: silence -Wcast-align on pointer

[dpdk-dev] [PATCH 08/10] librte_mbuf: Add rte_pktmbuf_mtod_offset()

On Fri, 19 Jun 2015 17:58:59 +0200 Olivier MATZ wrote: > Small comment here: m_headlen(m) should be replaced by "the length of > the first segment". Indeed, there is no m_headlen() function (I see > that it is coming from a copy/paste of the mtod() comment). Could you > also fix it by the way? W

[dpdk-dev] [PATCH 01/10] eal: add and use unaligned integer types

e pushed > for 2.1. > > On 04/29/2015 06:15 PM, Cyril Chemparathy wrote: > > On machines that are strict on pointer alignment, current code > > breaks on GCC's -Wcast-align checks on casts from narrower to wider > > types. This patch introduces new unaligned_uint(16|32

[dpdk-dev] mempool with custom alignment

On Wed, 3 Jun 2015 11:30:17 -0700 Hi Dax, Dax Rawal wrote: > Hi, > I use mempool APIs to allocate DMA-able descriptor ring and buffers > so that I get both physical and virtual addresses of allocated > memory. I buffers I get from the mempool APIs are 64 byte aligned. My > requirement is 256byte

[dpdk-dev] [PATCH v2 12/12] mk: introduce Tilera Tile architecture

On Fri, 22 May 2015 18:03:04 +0200 Thomas Monjalon wrote: > Hi, > > 2015-01-06 14:53, Zhigang Lu: > > Add defconfig and rte.vars.mk files for Tile architecture. > > > > Signed-off-by: Zhigang Lu > > Signed-off-by: Cyril Chemparathy > > This series

[dpdk-dev] [PATCH 00/10] Improve cast alignment for strict aligned machines

On Mon, 4 May 2015 11:50:09 +0200 Olivier MATZ wrote: > Hi Cyril, > > On 04/29/2015 06:15 PM, Cyril Chemparathy wrote: > > This series contains a few improvements that allow the DPDK code > > base to build properly on machines that enforce strict pointer cast >

[dpdk-dev] [PATCH 10/10] librte_mbuf: Apply mtod-offset.cocci transform

simply merging in the change. Signed-off-by: Cyril Chemparathy --- app/test-pmd/ieee1588fwd.c | 4 +- app/test-pmd/rxonly.c| 21 ++ app/test-pmd/txonly.c| 4 +- app/test/packet_burst_generator.c| 5 ++- examples/dpdk_qat

[dpdk-dev] [PATCH 09/10] librte_mbuf: Add transform for rte_pktmbuf_mtod_offset()

This patch adds a coccinelle (see http://coccinelle.lip6.fr/) transform to use the newly added rte_pktmbuf_mtod_offset() helper. In addition, we add a simple script to apply all available transforms to a codebase. Signed-off-by: Cyril Chemparathy --- scripts/cocci.sh| 64

[dpdk-dev] [PATCH 08/10] librte_mbuf: Add rte_pktmbuf_mtod_offset()

-by: Cyril Chemparathy --- lib/librte_mbuf/rte_mbuf.h | 20 +++- 1 file changed, 19 insertions(+), 1 deletion(-) diff --git a/lib/librte_mbuf/rte_mbuf.h b/lib/librte_mbuf/rte_mbuf.h index 5ddd8dd..e8f1bf4 100644 --- a/lib/librte_mbuf/rte_mbuf.h +++ b/lib/librte_mbuf/rte_mbuf.h

[dpdk-dev] [PATCH 07/10] app/test: use struct ether_addr instead of a byte array cast

This patch instantiates struct ether_addr instead of type casting a uint8_t pointer to the same structure. The type cast breaks on machines that enforce strict alignment. Signed-off-by: Cyril Chemparathy --- app/test/test_pmd_perf.c | 10 +- 1 file changed, 5 insertions(+), 5 deletions

[dpdk-dev] [PATCH 06/10] app/test-pmd: pack simple_gre_hdr

Not packing this causes -Wcast-align breakage on machines that are strict on alignment. This patch fixes this bug. Signed-off-by: Cyril Chemparathy --- app/test-pmd/csumonly.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/app/test-pmd/csumonly.c b/app/test-pmd/csumonly.c

[dpdk-dev] [PATCH 05/10] ethdev: silence -Wcast-align on pointer arithmetic

Statistics offsets in the rte_stats_strings[] array are always 64-bit aligned. However, the compiler is unaware of this fact and complains on -Wcast-align. This patch modifies the code to use RTE_PTR_ADD(), thereby silencing the compiler by casting through (void *). Signed-off-by: Cyril

[dpdk-dev] [PATCH 04/10] mbuf: silence -Wcast-align on pointer arithmetic

: Cyril Chemparathy --- lib/librte_mbuf/rte_mbuf.h | 8 +--- 1 file changed, 1 insertion(+), 7 deletions(-) diff --git a/lib/librte_mbuf/rte_mbuf.h b/lib/librte_mbuf/rte_mbuf.h index 70b0987..5ddd8dd 100644 --- a/lib/librte_mbuf/rte_mbuf.h +++ b/lib/librte_mbuf/rte_mbuf.h @@ -337,13 +337,7

[dpdk-dev] [PATCH 03/10] hash: silence -Wcast-align on pointer arithmetic

(), thereby silencing the compiler by casting through (void *). Signed-off-by: Cyril Chemparathy --- lib/librte_hash/rte_hash.c | 10 +- 1 file changed, 5 insertions(+), 5 deletions(-) diff --git a/lib/librte_hash/rte_hash.c b/lib/librte_hash/rte_hash.c index 9245716..67dff5b 100644 --- a/lib

[dpdk-dev] [PATCH 02/10] mempool: silence -Wcast-align on pointer arithmetic

-by: Cyril Chemparathy --- lib/librte_mempool/rte_mempool.h | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) diff --git a/lib/librte_mempool/rte_mempool.h b/lib/librte_mempool/rte_mempool.h index 9001312..1fb4184 100644 --- a/lib/librte_mempool/rte_mempool.h +++ b/lib/librte_mempool

[dpdk-dev] [PATCH 01/10] eal: add and use unaligned integer types

clang, and equivalents will need to be plugged in. Signed-off-by: Cyril Chemparathy --- app/test-pmd/flowgen.c | 4 ++-- app/test-pmd/txonly.c | 2 +- app/test/packet_burst_generator.c | 4 ++-- app/test/test_mbuf.c

[dpdk-dev] [PATCH 00/10] Improve cast alignment for strict aligned machines

*)(rte_pktmbuf_mtod(m, char *) + offset) with: rte_pktmbuf_mtod_offset(m, struct foo *, offset) To ensure consistency, the above transform was applied throughout the code base using the coccinelle semantic patching tool. Cyril Chemparathy (10): eal: add and use unaligned integer types

[dpdk-dev] [PATCH 01/15] mk: introduce Tilera Tile architecture

On 12/8/2014 6:25 AM, Neil Horman wrote: > On Mon, Dec 08, 2014 at 11:09:13AM +, Bruce Richardson wrote: >> On Mon, Dec 08, 2014 at 04:59:24PM +0800, Zhigang Lu wrote: >>> Add defconfig and rte.vars.mk files for Tile architecture. >>> >>> Signed-off-by:

[dpdk-dev] [PATCH 11/15] eal/tile: add EAL support for global mPIPE initialization

Hi Neil, On 12/8/2014 12:03 PM, Neil Horman wrote: > On Mon, Dec 08, 2014 at 04:59:34PM +0800, Zhigang Lu wrote: >> The TileGx mPIPE hardware provides Ethernet connectivity, >> packet classification, and packet load balancing services. >> >> Signed-off-by: Zhigang

[dpdk-dev] [PATCH 02/15] eal/tile: add atomic operations for TileGx

On 12/8/2014 6:28 AM, Neil Horman wrote: > On Mon, Dec 08, 2014 at 04:59:25PM +0800, Zhigang Lu wrote: >> This patch adds architecture specific memory barrier operations for >> TileGx. >> >> Signed-off-by: Zhigang Lu >> Signed-off-by: Cyril Chemparathy >&g

[dpdk-dev] [PATCH 09/12] Remove iopl operation for IBM Power architecture

On 9/26/2014 2:36 AM, Chao Zhu wrote: > iopl() call is mostly for the i386 architecture. In Power architecture. > It doesn't exist. This patch modified rte_eal_iopl_init() and make it > return -1 on Power. This means rte_config.flags will not contain > EAL_FLG_HIGH_IOPL flag on IBM Power architectu

[dpdk-dev] [PATCH 0/7] Patches to split architecture specific operations from DPDK

On 9/26/2014 2:33 AM, Chao Zhu wrote: > The set of patches split x86 architecture specific operations from DPDK and > put them to the > arch directories of i686 and x86_64 architecture. This will make the adpotion > of DPDK much easier > on other computer architecture. For a new architecture, jus

[dpdk-dev] next releases

Hi Thomas, On 8/25/2014 10:15 AM, Thomas Monjalon wrote: > Hello all, > > I am back from holidays; thanks for all the > patches/reviews/comments done during last weeks. > > I'd like to have a version 1.7.1, ideally at the end of this week. > > For the coming days, > - first priority is to inte

[dpdk-dev] [PATCH 0/6] Extensions to test-pmd

On 5/16/2014 7:22 AM, Thomas Monjalon wrote: > Hi Cyril, > > 2014-04-03 10:30, Cyril Chemparathy: >> This patch series contains a few minor extensions to test-pmd. These >> changes have been added primarily for convenience while testing out various >> scenarios with DP

[dpdk-dev] [PATCH] mk: add support for gdb debug info generation

It is often useful to build with debug enabled, we add a config (CONFIG_RTE_TOOLCHAIN_DEBUG) to do so. Note: This patch does not include corresponding changes for ICC. The author pleads abject ignorance in this regard, and welcomes recommendations. :-) Signed-off-by: Cyril Chemparathy

[dpdk-dev] [PATCH 6/6] test-pmd: add flowgen forwarding engine

This commit adds a multi-flow traffic generator forwarding engine. In effect, this forwarding mode functions very similar to the txonly mode, with the difference that it generates multiple L4 flows. Signed-off-by: Cyril Chemparathy --- app/test-pmd/Makefile | 1 + app/test-pmd/flowgen.c

[dpdk-dev] [PATCH 5/6] test-pmd: add mac swap forwarding mode

This commit adds a new forwarding mode, in which the source and destination MAC addresses of packets are swapped before forwarding according to the port-topology in effect. This has been designed for operation primarily with --port-topology=loopback. Signed-off-by: Cyril Chemparathy --- app

[dpdk-dev] [PATCH 4/6] test-pmd: allow txpkts to be setup via command line

We allow a new --txpkts command-line parameter to configure segment sizes when in txonly or flowgen forwarding modes. Signed-off-by: Cyril Chemparathy --- app/test-pmd/parameters.c | 57 +++ 1 file changed, 57 insertions(+) diff --git a/app/test-pmd

[dpdk-dev] [PATCH 3/6] test-pmd: allow command line selection of forwarding mode

This commit allows testpmd forwarding mode selection through command line parameters instead of having to set this up in interactive mode. Signed-off-by: Cyril Chemparathy --- app/test-pmd/parameters.c | 4 1 file changed, 4 insertions(+) diff --git a/app/test-pmd/parameters.c b/app/test

[dpdk-dev] [PATCH 2/6] test-pmd: add support for auto-start when interactive

This patch adds support for a command-line argument --auto-start (-a). When running in interactive mode, this allows us to start traffic without user intervention before dropping to the prompt. Signed-off-by: Cyril Chemparathy --- app/test-pmd/parameters.c | 20 app/test

[dpdk-dev] [PATCH 1/6] test-pmd: add support for single port loopback topology

This commits adds a new "loop" option to the --port-topology argument. With the loop option specified, ingress traffic is simply transmitted back on the same interface. Signed-off-by: Cyril Chemparathy --- app/test-pmd/config.c | 16 ++-- app/test-pmd/parameters.c |

[dpdk-dev] [PATCH 0/6] Extensions to test-pmd

Hi, This patch series contains a few minor extensions to test-pmd. These changes have been added primarily for convenience while testing out various scenarios with DPDK. Thanks -- Cyril. Cyril Chemparathy (6): test-pmd: add support for single port loopback topology test-pmd: add support

  1   2   >