On 1/23/2024 12:07 PM, Power, Ciara wrote: > Hi Ferruh, > >> -----Original Message----- >> From: Ferruh Yigit <ferruh.yi...@amd.com> >> Sent: Friday, January 19, 2024 2:51 PM >> To: Sivaramakrishnan, VenkatX <venkatx.sivaramakrish...@intel.com>; Igor >> Russkikh <irussk...@marvell.com>; Selwin Sebastian >> <selwin.sebast...@amd.com>; Ajit Khaparde >> <ajit.khapa...@broadcom.com>; Somnath Kotur >> <somnath.ko...@broadcom.com>; Nithin Dabilpuram >> <ndabilpu...@marvell.com>; Kiran Kumar K <kirankum...@marvell.com>; >> Sunil Kumar Kori <sk...@marvell.com>; Satha Rao >> <skotesh...@marvell.com>; Zhang, Yuying <yuying.zh...@intel.com>; Xing, >> Beilei <beilei.x...@intel.com>; Rahul Lakkireddy >> <rahul.lakkire...@chelsio.com>; Hemant Agrawal >> <hemant.agra...@nxp.com>; Sachin Saxena <sachin.sax...@nxp.com>; Su, >> Simei <simei...@intel.com>; Wu, Wenjun1 <wenjun1...@intel.com>; >> Gagandeep Singh <g.si...@nxp.com>; John Daley <johnd...@cisco.com>; >> Hyong Youb Kim <hyon...@cisco.com>; Gaetan Rivet <gr...@u256.net>; >> Zhang, Qi Z <qi.z.zh...@intel.com>; Wang, Xiao W <xiao.w.w...@intel.com>; >> Jie Hai <haij...@huawei.com>; Yisen Zhuang <yisen.zhu...@huawei.com>; >> Wu, Jingjing <jingjing...@intel.com>; Yang, Qiming >> <qiming.y...@intel.com>; Guo, Junfeng <junfeng....@intel.com>; Andrew >> Boyer <andrew.bo...@amd.com>; Long Li <lon...@microsoft.com>; Matan >> Azrad <ma...@nvidia.com>; Viacheslav Ovsiienko <viachesl...@nvidia.com>; >> Dariusz Sosnowski <dsosnow...@nvidia.com>; Ori Kam <or...@nvidia.com>; >> Suanming Mou <suanmi...@nvidia.com>; Chaoyong He >> <chaoyong...@corigine.com>; Jiawen Wu <jiawe...@trustnetic.com>; >> Harman Kalra <hka...@marvell.com>; Devendra Singh Rawat >> <dsinghra...@marvell.com>; Alok Prasad <pa...@marvell.com>; Andrew >> Rybchenko <andrew.rybche...@oktetlabs.ru>; Jerin Jacob >> <jer...@marvell.com>; Maciej Czekaj <mcze...@marvell.com>; Jian Wang >> <jianw...@trustnetic.com>; Behrens, Jochen <jbehr...@vmware.com>; >> Thomas Monjalon <tho...@monjalon.net> >> Cc: dev@dpdk.org; Power, Ciara <ciara.po...@intel.com> >> Subject: Re: [PATCH v5 2/2] drivers/net: return number of types in get >> supported types >> >> On 1/18/2024 12:07 PM, Sivaramakrishnan Venkat wrote: >>> Missing "RTE_PTYPE_UNKNOWN" ptype causes buffer overflow. >>> Enhance code such that the dev_supported_ptypes_get() function pointer >>> now returns the number of elements to eliminate the need for >>> "RTE_PTYPE_UNKNOWN" as the last item. >>> >>> Signed-off-by: Sivaramakrishnan Venkat >>> <venkatx.sivaramakrish...@intel.com> >>> >>> -- >>> v5: >>> - modified commit message. >>> - tidied formatting of code. >>> - added doxygen comment. >>> v4: >>> - split into two patches, one for backporting and another one for >>> upstream rework. >>> v3: >>> - reworked the function to return number of elements and remove the >>> need for RTE_PTYPE_UNKNOWN in list. >>> v2: >>> - extended fix for multiple drivers. >>> --- >> >> <...> >> >>> 59 files changed, 188 insertions(+), 141 deletions(-) >>> >> >> Some driver still have the flag: >> - drivers/net/mvneta/mvneta_ethdev.c >> - drivers/net/mvpp2/mrvl_ethdev.c >> - pfe >> - dpaa >> - drivers/net/thunderx/nicvf_ethdev.c >> - drivers/net/nfp/nfp_net_common.c >> >> Above seems the ones updated in previous patch, flags added in previous >> patch should be removed in this one. >> >> >> And following seems missed and still has the flag: >> >> - drivers/net/ngbe/ngbe_ptypes.c >> >> <...> >> >>> @@ -3971,9 +3975,6 @@ rte_eth_dev_set_ptypes(uint16_t port_id, >> uint32_t ptype_mask, >>> } >>> } >>> >>> - if (set_ptypes != NULL && j < num) >>> - set_ptypes[j] = RTE_PTYPE_UNKNOWN; >>> - >>> >> >> This change is new in this version and not mentioned in the changelog. >> >> 'rte_eth_dev_set_ptypes()' returns 'set_ptypes' that terminated with >> 'RTE_PTYPE_UNKNOWN', this is how that API works. >> Why changing it in this patch? > > Apologies, yes, we missed this in the changelog. > > For the change itself, if we are removing the need for RTE_PTYPE_UNKNOWN in > the supported ptypes lists to mark the last element, do we still need to add > it here when setting ptypes list? > Maybe a misunderstanding on my part - I thought it would be the same for both > cases. > >
They are two different APIs, and 'rte_eth_dev_set_ptypes()' returns ptypes to user that will be terminated by RTE_PTYPE_UNKNOWN, so to not break the user code we can't change it.
