[dpdk-dev,v2] eal: don't advertise a physical address when no hugepages

  When populating a mempool with a virtual memory area, the mempool
library expects to be able to get the physical address of each page.

When started with --no-huge, the physical addresses may not be available
because the pages are not locked in memory. It sometimes returns
RTE_BAD_PHYS_ADDR, which makes the mempool_populate() function to fail.

This was working before the commit cdc242f260e7 ("eal/linux: support
running as unprivileged user"), because rte_mem_virt2phy() was returning
0 instead of RTE_BAD_PHYS_ADDR, which was seen as a valid physical
address.

Since --no-huge is a debug function that breaks the support of physical
drivers, always set physical addresses to RTE_BAD_PHYS_ADDR in memzones
or in rte_mem_virt2phy(), and ensure that mempool won't complain in that
case.

Fixes: cdc242f260e7 ("eal/linux: support running as unprivileged user")

CC: stable@dpdk.org
Signed-off-by: Olivier Matz <olivier.matz@6wind.com>
---

v1 -> v2:
- initialize memory segments (rte_memseg) with RTE_BAD_PHYS_ADDR
  if started with --no-huge
- return RTE_BAD_PHYS_ADDR when rte_malloc_virt2phy() if the
  memory segment has no physical address
- drop change in eal_common_memzone.c: the physical address is
  now set to RTE_BAD_PHYS_ADDR as per return value of
  rte_malloc_virt2phy()


 lib/librte_eal/common/include/rte_malloc.h | 2 +-
 lib/librte_eal/common/rte_malloc.c         | 4 +++-
 lib/librte_eal/linuxapp/eal/eal_memory.c   | 9 ++++++++-
 lib/librte_mempool/rte_mempool.c           | 2 +-
 4 files changed, 13 insertions(+), 4 deletions(-)
  

On Mon, Jul 3, 2017 at 12:04 PM, Olivier Matz <olivier.matz@6wind.com> wrote:
> When populating a mempool with a virtual memory area, the mempool
> library expects to be able to get the physical address of each page.
>
> When started with --no-huge, the physical addresses may not be available
> because the pages are not locked in memory. It sometimes returns
> RTE_BAD_PHYS_ADDR, which makes the mempool_populate() function to fail.
>
> This was working before the commit cdc242f260e7 ("eal/linux: support
> running as unprivileged user"), because rte_mem_virt2phy() was returning
> 0 instead of RTE_BAD_PHYS_ADDR, which was seen as a valid physical
> address.
>
> Since --no-huge is a debug function that breaks the support of physical
> drivers, always set physical addresses to RTE_BAD_PHYS_ADDR in memzones
> or in rte_mem_virt2phy(), and ensure that mempool won't complain in that
> case.
>
> Fixes: cdc242f260e7 ("eal/linux: support running as unprivileged user")
>
> CC: stable@dpdk.org
> Signed-off-by: Olivier Matz <olivier.matz@6wind.com>

Reviewed-by: Jan Blunck <jblunck@infradead.org>

> ---
>
> v1 -> v2:
> - initialize memory segments (rte_memseg) with RTE_BAD_PHYS_ADDR
>   if started with --no-huge
> - return RTE_BAD_PHYS_ADDR when rte_malloc_virt2phy() if the
>   memory segment has no physical address
> - drop change in eal_common_memzone.c: the physical address is
>   now set to RTE_BAD_PHYS_ADDR as per return value of
>   rte_malloc_virt2phy()
>
>
>  lib/librte_eal/common/include/rte_malloc.h | 2 +-
>  lib/librte_eal/common/rte_malloc.c         | 4 +++-
>  lib/librte_eal/linuxapp/eal/eal_memory.c   | 9 ++++++++-
>  lib/librte_mempool/rte_mempool.c           | 2 +-
>  4 files changed, 13 insertions(+), 4 deletions(-)
>
> diff --git a/lib/librte_eal/common/include/rte_malloc.h b/lib/librte_eal/common/include/rte_malloc.h
> index 61aac322b..3d37f79b8 100644
> --- a/lib/librte_eal/common/include/rte_malloc.h
> +++ b/lib/librte_eal/common/include/rte_malloc.h
> @@ -329,7 +329,7 @@ rte_malloc_set_limit(const char *type, size_t max);
>   * @param addr
>   *   Address obtained from a previous rte_malloc call
>   * @return
> - *   NULL on error
> + *   RTE_BAD_PHYS_ADDR on error
>   *   otherwise return physical address of the buffer
>   */
>  phys_addr_t
> diff --git a/lib/librte_eal/common/rte_malloc.c b/lib/librte_eal/common/rte_malloc.c
> index f4a883529..5c0627bf4 100644
> --- a/lib/librte_eal/common/rte_malloc.c
> +++ b/lib/librte_eal/common/rte_malloc.c
> @@ -253,6 +253,8 @@ rte_malloc_virt2phy(const void *addr)
>  {
>         const struct malloc_elem *elem = malloc_elem_from_data(addr);
>         if (elem == NULL)
> -               return 0;
> +               return RTE_BAD_PHYS_ADDR;
> +       if (elem->ms->phys_addr == RTE_BAD_PHYS_ADDR)
> +               return RTE_BAD_PHYS_ADDR;
>         return elem->ms->phys_addr + ((uintptr_t)addr - (uintptr_t)elem->ms->addr);
>  }
> diff --git a/lib/librte_eal/linuxapp/eal/eal_memory.c b/lib/librte_eal/linuxapp/eal/eal_memory.c
> index 647d89c58..040f24a43 100644
> --- a/lib/librte_eal/linuxapp/eal/eal_memory.c
> +++ b/lib/librte_eal/linuxapp/eal/eal_memory.c
> @@ -112,6 +112,13 @@ test_phys_addrs_available(void)
>         if (rte_xen_dom0_supported())
>                 return;
>
> +       if (!rte_eal_has_hugepages()) {
> +               RTE_LOG(ERR, EAL,
> +                       "Started without hugepages support, physical addresses not available\n");
> +               phys_addrs_available = false;
> +               return;
> +       }
> +
>         physaddr = rte_mem_virt2phy(&tmp);
>         if (physaddr == RTE_BAD_PHYS_ADDR) {
>                 RTE_LOG(ERR, EAL,
> @@ -1054,7 +1061,7 @@ rte_eal_hugepage_init(void)
>                                         strerror(errno));
>                         return -1;
>                 }
> -               mcfg->memseg[0].phys_addr = (phys_addr_t)(uintptr_t)addr;
> +               mcfg->memseg[0].phys_addr = RTE_BAD_PHYS_ADDR;
>                 mcfg->memseg[0].addr = addr;
>                 mcfg->memseg[0].hugepage_sz = RTE_PGSIZE_4K;
>                 mcfg->memseg[0].len = internal_config.memory;
> diff --git a/lib/librte_mempool/rte_mempool.c b/lib/librte_mempool/rte_mempool.c
> index f65310f60..6fc3c9c7c 100644
> --- a/lib/librte_mempool/rte_mempool.c
> +++ b/lib/librte_mempool/rte_mempool.c
> @@ -476,7 +476,7 @@ rte_mempool_populate_virt(struct rte_mempool *mp, char *addr,
>                 /* required for xen_dom0 to get the machine address */
>                 paddr = rte_mem_phy2mch(-1, paddr);
>
> -               if (paddr == RTE_BAD_PHYS_ADDR) {
> +               if (paddr == RTE_BAD_PHYS_ADDR && rte_eal_has_hugepages()) {
>                         ret = -EINVAL;
>                         goto fail;
>                 }
> --
> 2.11.0
>
  

03/07/2017 12:17, Jan Blunck:
> On Mon, Jul 3, 2017 at 12:04 PM, Olivier Matz <olivier.matz@6wind.com> wrote:
> > When populating a mempool with a virtual memory area, the mempool
> > library expects to be able to get the physical address of each page.
> >
> > When started with --no-huge, the physical addresses may not be available
> > because the pages are not locked in memory. It sometimes returns
> > RTE_BAD_PHYS_ADDR, which makes the mempool_populate() function to fail.
> >
> > This was working before the commit cdc242f260e7 ("eal/linux: support
> > running as unprivileged user"), because rte_mem_virt2phy() was returning
> > 0 instead of RTE_BAD_PHYS_ADDR, which was seen as a valid physical
> > address.
> >
> > Since --no-huge is a debug function that breaks the support of physical
> > drivers, always set physical addresses to RTE_BAD_PHYS_ADDR in memzones
> > or in rte_mem_virt2phy(), and ensure that mempool won't complain in that
> > case.
> >
> > Fixes: cdc242f260e7 ("eal/linux: support running as unprivileged user")
> >
> > CC: stable@dpdk.org
> > Signed-off-by: Olivier Matz <olivier.matz@6wind.com>
> 
> Reviewed-by: Jan Blunck <jblunck@infradead.org>

Applied, thanks