Changes in .mailmap

73f7ae1d721aa5c388123db11827937205985999 - net/mlx5: fix polling CQEs

Fri, 06 Dec 2024 00:58:11 +0000

net/mlx5: fix polling CQEsIn certain situations, the receive queue (rxq) fails to replenish itsinternal ring with memory buffers (mbufs) from the pool. This can happenwhen the pool has a limited number of mbufs allocated, and the userapplication holds incoming packets for an extended period, resulting in adelayed release of mbufs. Consequently, the pool becomes depleted,preventing the rxq from replenishing from it.There was a bug in the behavior of the vectorized rxq_cq_process_v routine,which handled completion queue entries (CQEs) in batches of four. Thisroutine consistently accessed four mbufs from the internal queue ring,regardless of whether they had been replenished. As a result, it couldaccess mbufs that no longer belonged to the poll mode driver (PMD).The fix involves checking if there are four replenished mbufs availablebefore allowing rxq_cq_process_v to handle the batch. Once replenishmentsucceeds during the polling process, the routine will resume its operation.Fixes: 1ded26239aa0 ("net/mlx5: refactor vectorized Rx")Cc: stable@dpdk.orgReported-by: Changqi Dingluo Signed-off-by: Gavin Hu Acked-by: Viacheslav Ovsiienko List of files: /dpdk/.mailmap

950820f10cb8a8c45998536546b5a00123fe40aa - net/ngbe: add probe and remove for VF device

Mon, 20 Jan 2025 09:42:27 +0000

net/ngbe: add probe and remove for VF deviceIntroduce virtual function driver in ngbe PMD, add simple init anduninit function to probe and remove the device.Signed-off-by: Zaiyu Wang List of files: /dpdk/.mailmap

7d73fa47f4ef9223cf39303bb731f071435228be - examples/flow_filtering: remove duplicate assignment

Wed, 15 Jan 2025 12:11:50 +0000

examples/flow_filtering: remove duplicate assignmentLast action in template was being assigned twice.Coverity issue: 451221Fixes: 16158f349000 ("examples/flow_filtering: introduce use cases snippets")Cc: stable@dpdk.orgSigned-off-by: Ariel Otilibili List of files: /dpdk/.mailmap

c4f16ab009ad53446157c6fcde85435b1ca8c780 - net/ena: fix missing default LLQ policy

Thu, 12 Dec 2024 15:24:55 +0000

net/ena: fix missing default LLQ policyThe driver failed to set a default llq_header_policy,causing LLQ to be disabled when the devarg value isnot specified by the application.Fixes: d7918d19d25e ("net/ena: restructure LLQ policy user setting")Cc: stable@dpdk.orgSigned-off-by: Yosef Raisman Signed-off-by: Shai Brandes Reviewed-by: Stephen Hemminger List of files: /dpdk/.mailmap

5b856206c74bbcf19e12cafa15382a7e15b0a1b5 - eal/x86: fix some intrinsics header include for Windows

Tue, 14 Jan 2025 16:21:13 +0000

eal/x86: fix some intrinsics header include for WindowsError reported:../lib/net/net_crc_sse.c:49:17: error: call to undeclared function'_mm_clmulepi64_si128'; ISO C99 and later do not support implicitfunction declarations [-Wimplicit-function-declaration]The fix is to remove the unnecessary ifdef around the inclusion ofheader file immintrin.h. This header also contains functions that donot require AVX instructions, so should not be included only when AVXis available.Bugzilla ID: 1595Fixes: da826b7135a4 ("eal: introduce ymm type for AVX 256-bit")Cc: stable@dpdk.orgReported-by: Pier Damouny Signed-off-by: Andre Muezerie Acked-by: Bruce Richardson List of files: /dpdk/.mailmap

1846fe767f00a9d0dade333e7838f81e6721b694 - vhost: clear ring addresses when getting vring base

Wed, 27 Nov 2024 02:03:49 +0000

vhost: clear ring addresses when getting vring baseClear ring addresses during vring base retrieval to handle guest rebootscenarios correctly. This is particularly important for vdpa-blk deviceswhere the following issue occurs:When a guest OS with vdpa-blk device reboots, during UEFI stage, onlyone vring is actually used and configured. However, QEMU still sendsenable messages for all configured queues. The remaining queues retaintheir addresses from before reboot, which reference invalid memorymappings in the rebooted guest.The issue manifests in vq_is_ready():static boolvq_is_ready(struct virtio_net *dev, struct vhost_virtqueue *vq){ /* Only checks pointer validity, not address freshness */ rings_ok = vq->desc && vq->avail && vq->used; ...}vq_is_ready() incorrectly considers these queues as ready because itonly checks if desc/avail/used pointers are non-NULL, but cannotdetect that these addresses are stale from the previous boot.Clear the ring addresses in vhost_user_get_vring_base() to forcethe guest driver to reconfigure them before use. This ensures thatvq_is_ready() will return false for queues with stale addressesuntil they are properly reconfigured by the guest driver.Fixes: 3ea7052f4b1b ("vhost: postpone rings addresses translation")Cc: stable@dpdk.orgSigned-off-by: Jianping Zhao Reviewed-by: Maxime Coquelin List of files: /dpdk/.mailmap

700989f512bbc2ee9758a8a9cb6973cfdeda6f27 - ring: make dump function more verbose

Fri, 06 Dec 2024 18:35:57 +0000

ring: make dump function more verboseThe current rte_ring_dump function uses the generic rte_ring_headtailstructure to access head/tail positions. This is incorrect for the RTScase where the head is stored in a different offset in the union ofstructs. Switching to a separate function for each sync type allowsto dump correct head/tail values and extra metadata.Signed-off-by: Eimear Morrissey Signed-off-by: Konstantin Ananyev Acked-by: Morten Brørup Acked-by: Stephen Hemminger List of files: /dpdk/.mailmap

8750576fb2a9a067ffbcce4bab6481f3bfa47097 - doc: reword some sample app guides

Mon, 14 Oct 2024 17:03:01 +0000

doc: reword some sample app guidesI have reviewed these sections for grammar/clarityand made small modifications to the formatting of sectionsto adhere to a template which will create uniformalityin the sample application user guides overall.Signed-off-by: Nandini Persad Acked-by: Chengwen Feng List of files: /dpdk/.mailmap

f5418c97fc19a20234ec0cb72ef0ffb5e93fdf2e - dts: add l2fwd test suite

Wed, 13 Nov 2024 16:31:28 +0000

dts: add l2fwd test suiteAdd a basic L2 forwarding test suite which tests the correctfunctionality of the forwarding facility built-in in the DPDK.The tests are performed with different queues numbers per port.Signed-off-by: Luca Vizzarro Signed-off-by: Thomas Wilks Reviewed-by: Paul Szczepanek Reviewed-by: Dean Marx Reviewed-by: Patrick Robb Tested-by: Patrick Robb List of files: /dpdk/.mailmap

ecaff610f53d7f4771150a99ffb54e27159f6029 - dts: rename build target to DPDK build

Fri, 27 Sep 2024 16:08:48 +0000

dts: rename build target to DPDK buildSince the DPDK may already be built, some more general nameis needed that includes both the DPDK location and the buildconfig (if we are going to build).Signed-off-by: Tomáš Ďurovec Signed-off-by: Luca Vizzarro List of files: /dpdk/.mailmap

30495f54583d93d50df02ec5e15cbaad6f6dcc6a - raw/gdtc: introduce driver skeleton

Thu, 14 Nov 2024 09:20:08 +0000

raw/gdtc: introduce driver skeletonIntroduce rawdev driver support for GDTC whichcan help to connect two separate hosts with each other.Signed-off-by: Yong Zhang List of files: /dpdk/.mailmap

c77887ad87931526768ae501596fcba15c21db82 - usertools/devbind: support all VFIO no-IOMMU sysfs values

Tue, 12 Nov 2024 14:07:19 +0000

usertools/devbind: support all VFIO no-IOMMU sysfs valuesThis patch adds support for 'y' and 'Y' values when readingvfio-pci unsafe_noiommu_mode flag.Possible values were taken from linux kernel(sysfs__read_bool() in tools/lib/api/fs/fs.c)Signed-off-by: Rogelio Domínguez Hernández Reviewed-by: Robin Jarry List of files: /dpdk/.mailmap

5d5dcd8492f8b2863a2de0f5bad223df27749c09 - doc: modernize build instructions on Windows

Mon, 11 Nov 2024 16:14:08 +0000

doc: modernize build instructions on WindowsMeson 0.57 was an unstable version and is now outdated.The referenced bug in Meson 0.58 is fixed in stable releases.Recommend the first version containing the fix: 1.5.2.Building DPDK applications that run on 32-bit Windows iscurrently not supported. However, some Visual Studio environmentsdefault to producing 32-bit binaries.Recommend instructing the developer prompt to produce 64-bit binarieswhen that is the case.Signed-off-by: Andre Muezerie Acked-by: Dmitry Kozlyuk List of files: /dpdk/.mailmap

9b170cfc6303a9a9a7279149ac6800a72239ad4e - net/r8169: add driver skeleton

Wed, 13 Nov 2024 09:28:38 +0000

net/r8169: add driver skeletonMeson build infrastructure, r8169_ethdev minimal skeleton,header with Realtek NIC device and vendor IDs.Signed-off-by: Howard Wang Reviewed-by: Ferruh Yigit List of files: /dpdk/.mailmap

019c181687371f24196f437b648dc80f67519b72 - net/bnxt: fix LRO offload capability

Tue, 02 Apr 2024 19:20:18 +0000

net/bnxt: fix LRO offload capabilityFix LRO offload capability for P7 devices.Export the capability to the application only if compressedRx CQE mode is not enabled.LRO aka TPA is not supported when compressed CQE mode is set.Fixes: 3b56c3ffc182 ("net/bnxt: refactor code to support P7 devices")Cc: stable@dpdk.orgSigned-off-by: Ajit Khaparde Reviewed-by: Vasuthevan Maheswaran List of files: /dpdk/.mailmap

7b8400464f14637ed2669dbf732c256bf2447de6 - net/bnxt: fix reading SFF-8436 SFP EEPROMs

Mon, 12 Aug 2024 10:34:05 +0000

net/bnxt: fix reading SFF-8436 SFP EEPROMsIf a SFP which supports SFF-8436 is present thencurrently the DDM information present in the eepromis not read. Furthermore bnxt_get_module_eeprom()will return -EINVAL for these eeproms since thelength of these eeproms is 512 bytes but we areonly ever selecting 2 pages (256 bytes) to read.Fixes: 6253a23491a4 ("net/bnxt: dump SFP module info")Cc: stable@dpdk.orgSigned-off-by: Peter Morrow Reviewed-by: Ajit Khaparde List of files: /dpdk/.mailmap

1190f2f8d5abf82c843ad071ad4c7d0aea202cce - net/bnxt/tf_core: fix slice count in case of HA entry move

Thu, 07 Nov 2024 13:52:14 +0000

net/bnxt/tf_core: fix slice count in case of HA entry moveWhen entries are moved during HA, a shared move function transfersTCAM entries by using get/set message APIs, and the slice number of theentry is required to accomplish the movement. The slice number iscalculated as the product of row_slice and entry size. Before callingget/set message APIs, the source entry size should be updated with thedestination entry size; otherwise, it might corrupt the slice number field,which may result in writing an incorrect entry. A fix is made which nowcopies the entry size from the source to the destination before callingget/set message APIs, ensuring the correct slice number is modified.Fixes: 97435d7906d7 ("net/bnxt: update Truflow core")Cc: stable@dpdk.orgSigned-off-by: Sangtani Parag Satishbhai Reviewed-by: Sriharsha Basavapatna Reviewed-by: Ajit Khaparde List of files: /dpdk/.mailmap

19da63ccfc7e83f06e1bb14d830d4e0de99b13cf - net/bonding: promote experimental API stable

Wed, 30 Oct 2024 15:42:54 +0000

net/bonding: promote experimental API stableRemove rte_experimental macros from the stable functionsas they have been part of the stable API since 23.11.Signed-off-by: Lukas Sismis Acked-by: Huisong Li Acked-by: Stephen Hemminger List of files: /dpdk/.mailmap

29e89288ff14f153c981bd6658ef80939bc16a05 - net/zxdh: add driver skeleton

Mon, 04 Nov 2024 11:58:44 +0000

net/zxdh: add driver skeletonAdd basic zxdh ethdev init and register PCI probe functions.Update doc files.Signed-off-by: Junlong Wang Acked-by: Ferruh Yigit List of files: /dpdk/.mailmap

ebab0e8b2257aa049dd35dedc7efd230b0f45b88 - examples/l3fwd: fix read beyond boundaries

Thu, 07 Nov 2024 18:50:51 +0000

examples/l3fwd: fix read beyond boundariesASAN report:ERROR: AddressSanitizer: unknown-crash on address 0x7ffffef92e32 at pc 0x00000053d1e9 bp 0x7ffffef92c00 sp 0x7ffffef92bf8READ of size 16 at 0x7ffffef92e32 thread T0 #0 0x53d1e8 in _mm_loadu_si128 /usr/lib64/gcc/x86_64-suse-linux/11/include/emmintrin.h:703 #1 0x53d1e8 in send_packets_multi ../examples/l3fwd/l3fwd_sse.h:125 #2 0x53d1e8 in acl_send_packets ../examples/l3fwd/l3fwd_acl.c:1048 #3 0x53ec18 in acl_main_loop ../examples/l3fwd/l3fwd_acl.c:1127 #4 0x12151eb in rte_eal_mp_remote_launch ../lib/eal/common/eal_common_launch.c:83 #5 0x5bf2df in main ../examples/l3fwd/main.c:1647 #6 0x7f6d42a0d2bc in __libc_start_main (/lib64/libc.so.6+0x352bc) #7 0x527499 in _start (/home/kananyev/dpdk-l3fwd-acl/x86_64-native-linuxapp-gcc-dbg-b1/examples/dpdk-l3fwd+0x527499)Reason for that is that send_packets_multi() uses 16B loads to accessinput dst_port[]and might read beyond array boundaries.Right now, it doesn't cause any real issue - junk values are ignored, alsoinside l3fwd we always allocate dst_port[] array on the stack, somemory beyond it is always available.Anyway, it probably need to be fixed.The patch below simply allocates extra space for dst_port[], sosend_packets_multi() will never read beyond its boundaries.Probably a better fix would be to change send_packets_multi()itself to avoid access beyond 'nb_rx' entries.Bugzilla ID: 1502Fixes: 94c54b4158d5 ("examples/l3fwd: rework exact-match")Cc: stable@dpdk.orgSigned-off-by: Konstantin Ananyev Acked-by: Stephen Hemminger List of files: /dpdk/.mailmap