Hi David, Le mer. 12 nov. 2025 à 10:11, David Marchand <[email protected]> a écrit : > > Calling rte_cryptodev_pmd_get_named_dev() is enough to retrieve the > crypto device object. > > This leaves no user of the cryptodev field in the fslmc device object. > > Signed-off-by: David Marchand <[email protected]> > --- > drivers/bus/fslmc/bus_fslmc_driver.h | 1 - > drivers/crypto/dpaa2_sec/dpaa2_sec_dpseci.c | 8 +++++--- > 2 files changed, 5 insertions(+), 4 deletions(-) > > diff --git a/drivers/bus/fslmc/bus_fslmc_driver.h > b/drivers/bus/fslmc/bus_fslmc_driver.h > index b6a6238a01..45c7a519f9 100644 > --- a/drivers/bus/fslmc/bus_fslmc_driver.h > +++ b/drivers/bus/fslmc/bus_fslmc_driver.h > @@ -99,7 +99,6 @@ struct rte_dpaa2_device { > struct rte_device device; /**< Inherit core device */ > union { > struct rte_eth_dev *eth_dev; /**< ethernet device */ > - struct rte_cryptodev *cryptodev; /**< Crypto Device */ > struct rte_dma_dev *dmadev; /**< DMA Device */ > struct rte_rawdev *rawdev; /**< Raw Device */ > }; > diff --git a/drivers/crypto/dpaa2_sec/dpaa2_sec_dpseci.c > b/drivers/crypto/dpaa2_sec/dpaa2_sec_dpseci.c > index ca10d88da7..698548e6ea 100644 > --- a/drivers/crypto/dpaa2_sec/dpaa2_sec_dpseci.c > +++ b/drivers/crypto/dpaa2_sec/dpaa2_sec_dpseci.c > @@ -4503,8 +4503,6 @@ cryptodev_dpaa2_sec_probe(struct rte_dpaa2_driver > *dpaa2_drv __rte_unused, > return -ENOMEM; > } > > - dpaa2_dev->cryptodev = cryptodev; > - > if (dpaa2_svr_family == SVR_LX2160A) > rta_set_sec_era(RTA_SEC_ERA_10); > else > @@ -4526,10 +4524,14 @@ cryptodev_dpaa2_sec_probe(struct rte_dpaa2_driver > *dpaa2_drv __rte_unused, > static int > cryptodev_dpaa2_sec_remove(struct rte_dpaa2_device *dpaa2_dev) > { > + char cryptodev_name[RTE_CRYPTODEV_NAME_MAX_LEN]; > struct rte_cryptodev *cryptodev; > int ret; > > - cryptodev = dpaa2_dev->cryptodev; > + snprintf(cryptodev_name, sizeof(cryptodev_name), "dpsec-%d", > + dpaa2_dev->object_id); > + > + cryptodev = rte_cryptodev_pmd_get_named_dev(cryptodev_name); > if (cryptodev == NULL) > return -ENODEV; > > -- > 2.51.0 > Should we add a function to generate the cryptodev name instead of duplicating the code, similar to how it’s done for PCI devices (i.e., rte_pci_device_name)?
Maxime Leroy
