dpaa_sec needs translations between physical and virtual addresses. V to P translation is relatively fast, as memory is managed in contiguous segments. The result of each V to P translation is used to update the DPAA iova table, which should be updated by a Mem event callback, but is not. Then the DPAA iova table has entries for all needed memory ranges. With this patch, dpaa_mem_ptov will always use dpaax_iova_table_get_va, which ensures optimal performance. Fixes: 5a7dbb934d75 ('dpaa: enable dpaax library') Cc: stable@dpdk.org Cc: akhil.goyal@nxp.com Cc: hemant.agrawal@nxp.com Signed-off-by: Thierry Herbelot <thierry.herbelot@6wind.com> --- drivers/crypto/dpaa_sec/dpaa_sec.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/drivers/crypto/dpaa_sec/dpaa_sec.c b/drivers/crypto/dpaa_sec/dpaa_sec.c index 122c80a072ff..22b8b1d63ce0 100644 --- a/drivers/crypto/dpaa_sec/dpaa_sec.c +++ b/drivers/crypto/dpaa_sec/dpaa_sec.c @@ -38,6 +38,7 @@ #include <rte_dpaa_bus.h> #include <dpaa_sec.h> #include <dpaa_sec_log.h> +#include <dpaax_iova_table.h> enum rta_sec_era rta_sec_era; @@ -100,8 +101,10 @@ dpaa_mem_vtop(void *vaddr) const struct rte_memseg *ms; ms = rte_mem_virt2memseg(vaddr, NULL); - if (ms) + if (ms) { + dpaax_iova_table_update(ms->iova, (void *)ms->addr_64, ms->len); return ms->iova + RTE_PTR_DIFF(vaddr, ms->addr); + } return (size_t)NULL; } -- 2.20.1
Hi Thierry, > -----Original Message----- > From: Thierry Herbelot <thierry.herbelot@6wind.com> > Sent: Wednesday, September 11, 2019 9:37 PM > To: dev@dpdk.org > Cc: Thomas Monjalon <thomas@monjalon.net>; stable@dpdk.org; Akhil > Goyal <akhil.goyal@nxp.com>; Hemant Agrawal > <hemant.agrawal@nxp.com> > Subject: [PATCH] drivers/crypto/dpaa_sec: update DPAA iova table in > dpaa_mem_vtop > Importance: High > > dpaa_sec needs translations between physical and virtual addresses. > V to P translation is relatively fast, as memory is managed in contiguous > segments. > > The result of each V to P translation is used to update the DPAA iova table, > which should be updated by a Mem event callback, but is not. > Then the DPAA iova table has entries for all needed memory ranges. > > With this patch, dpaa_mem_ptov will always use dpaax_iova_table_get_va, > which ensures optimal performance. > > Fixes: 5a7dbb934d75 ('dpaa: enable dpaax library') > Cc: stable@dpdk.org > Cc: akhil.goyal@nxp.com > Cc: hemant.agrawal@nxp.com > > Signed-off-by: Thierry Herbelot <thierry.herbelot@6wind.com> > --- > drivers/crypto/dpaa_sec/dpaa_sec.c | 5 ++++- > 1 file changed, 4 insertions(+), 1 deletion(-) > > diff --git a/drivers/crypto/dpaa_sec/dpaa_sec.c > b/drivers/crypto/dpaa_sec/dpaa_sec.c > index 122c80a072ff..22b8b1d63ce0 100644 > --- a/drivers/crypto/dpaa_sec/dpaa_sec.c > +++ b/drivers/crypto/dpaa_sec/dpaa_sec.c > @@ -38,6 +38,7 @@ > #include <rte_dpaa_bus.h> > #include <dpaa_sec.h> > #include <dpaa_sec_log.h> > +#include <dpaax_iova_table.h> > > enum rta_sec_era rta_sec_era; > > @@ -100,8 +101,10 @@ dpaa_mem_vtop(void *vaddr) > const struct rte_memseg *ms; > > ms = rte_mem_virt2memseg(vaddr, NULL); > - if (ms) > + if (ms) { > + dpaax_iova_table_update(ms->iova, (void *)ms->addr_64, > ms->len); [Hemant] This fails 32bit (i686) compilation You can use "ms->addr" instead of "(void *)ms->addr_64" > return ms->iova + RTE_PTR_DIFF(vaddr, ms->addr); > + } > return (size_t)NULL; > } > > -- > 2.20.1
dpaa_sec needs translations between physical and virtual addresses. V to P translation is relatively fast, as memory is managed in contiguous segments. The result of each V to P translation is used to update the DPAA iova table, which should be updated by a Mem event callback, but is not. Then the DPAA iova table has entries for all needed memory ranges. With this patch, dpaa_mem_ptov will always use dpaax_iova_table_get_va, which ensures optimal performance. Fixes: 5a7dbb934d75 ('dpaa: enable dpaax library') Cc: stable@dpdk.org Cc: akhil.goyal@nxp.com Cc: hemant.agrawal@nxp.com V2: use "ms->addr" instead of "(void *)ms->addr_64" to fix compilation on 32bit (i686) Signed-off-by: Thierry Herbelot <thierry.herbelot@6wind.com> --- drivers/crypto/dpaa_sec/dpaa_sec.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/drivers/crypto/dpaa_sec/dpaa_sec.c b/drivers/crypto/dpaa_sec/dpaa_sec.c index 122c80a072ff..b8d142107d2a 100644 --- a/drivers/crypto/dpaa_sec/dpaa_sec.c +++ b/drivers/crypto/dpaa_sec/dpaa_sec.c @@ -38,6 +38,7 @@ #include <rte_dpaa_bus.h> #include <dpaa_sec.h> #include <dpaa_sec_log.h> +#include <dpaax_iova_table.h> enum rta_sec_era rta_sec_era; @@ -100,8 +101,10 @@ dpaa_mem_vtop(void *vaddr) const struct rte_memseg *ms; ms = rte_mem_virt2memseg(vaddr, NULL); - if (ms) + if (ms) { + dpaax_iova_table_update(ms->iova, ms->addr, ms->len); return ms->iova + RTE_PTR_DIFF(vaddr, ms->addr); + } return (size_t)NULL; } -- 2.20.1
Acked-by: Hemant Agrawal <hemant.agrawal@nxp.com>
> > dpaa_sec needs translations between physical and virtual addresses. > V to P translation is relatively fast, as memory is managed in > contiguous segments. > > The result of each V to P translation is used to update the DPAA iova > table, which should be updated by a Mem event callback, but is not. > Then the DPAA iova table has entries for all needed memory ranges. > > With this patch, dpaa_mem_ptov will always use dpaax_iova_table_get_va, > which ensures optimal performance. > > Fixes: 5a7dbb934d75 ('dpaa: enable dpaax library') > Cc: stable@dpdk.org > Cc: akhil.goyal@nxp.com > Cc: hemant.agrawal@nxp.com > > V2: use "ms->addr" instead of "(void *)ms->addr_64" to fix > compilation on 32bit (i686) Removed this line > > Signed-off-by: Thierry Herbelot <thierry.herbelot@6wind.com> > --- Applied to dpdk-next-crypto Thanks.