[dpdk-dev] [PATCH] bus/pci: check if 5-level paging is enabled when testing IOMMU address width

[Drocula]

Thanks, will refine in v2.

On Thu, Aug 9, 2018, 18:49 Burakov, Anatoly <anatoly.burakov at intel.com>
wrote:

> On 05-Aug-18 7:41 PM, Drocula wrote:
> > The kernel version 4.14 released with the support of 5-level paging.
> > When PML5 enabled, user-space virtual addresses uses up to 56 bits.
> > see kernel's Documentation/x86/x86_64/mm.txt.
> >
> > Signed-off-by: Drocula <quzeyao at gmail.com>
> > ---
> >   drivers/bus/pci/linux/pci.c | 27 +++++++++++++++++++++++++--
> >   1 file changed, 25 insertions(+), 2 deletions(-)
> >
> > diff --git a/drivers/bus/pci/linux/pci.c b/drivers/bus/pci/linux/pci.c
> > index 004600f..8913d6d 100644
> > --- a/drivers/bus/pci/linux/pci.c
> > +++ b/drivers/bus/pci/linux/pci.c
> > @@ -4,6 +4,7 @@
> >
> >   #include <string.h>
> >   #include <dirent.h>
> > +#include <sys/mman.h>
> >
> >   #include <rte_log.h>
> >   #include <rte_bus.h>
> > @@ -553,12 +554,34 @@
> >   }
> >
> >   #if defined(RTE_ARCH_X86)
> > +/*
> > + * Try to detect whether the system uses 5-level page table.
> > + */
> > +static bool
> > +system_uses_PML5(void)
> > +{
> > +     void *page_4k, *mask = (void *)0xf0000000000000;
> > +     page_4k = mmap(mask, 4096, PROT_READ | PROT_WRITE,
> > +             MAP_PRIVATE | MAP_ANONYMOUS, -1, 0);
> > +
> > +     if (page_4k == (void *) -1)
> > +             return false;
>
> Shouldn't this be MAP_FAILED?
>
> > +     munmap(page_4k, 4096);
> > +
> > +     if ((unsigned long)page_4k & (unsigned long)mask)
> > +             return true;
> > +     return false;
> > +}
> > +
> >   static bool
> >   pci_one_device_iommu_support_va(struct rte_pci_device *dev)
> >   {
> >   #define VTD_CAP_MGAW_SHIFT  16
> >   #define VTD_CAP_MGAW_MASK   (0x3fULL << VTD_CAP_MGAW_SHIFT)
> > -#define X86_VA_WIDTH 47 /* From Documentation/x86/x86_64/mm.txt */
> > +/*  From Documentation/x86/x86_64/mm.txt */
> > +#define X86_VA_WIDTH_PML4 47
> > +#define X86_VA_WIDTH_PML5 56
> > +
> >       struct rte_pci_addr *addr = &dev->addr;
> >       char filename[PATH_MAX];
> >       FILE *fp;
> > @@ -589,7 +612,7 @@
> >       fclose(fp);
> >
> >       mgaw = ((vtd_cap_reg & VTD_CAP_MGAW_MASK) >> VTD_CAP_MGAW_SHIFT) +
> 1;
> > -     if (mgaw < X86_VA_WIDTH)
> > +     if (mgaw < (system_uses_PML5() ? X86_VA_WIDTH_PML5 :
> X86_VA_WIDTH_PML4))
>
> This is perhaps nitpicking and a question of personal preferences, but i
> think storing this in a var would be more readable than doing ternary
> operator inside of an if statement.
>
> >               return false;
> >
> >       return true;
> >
>
>
> --
> Thanks,
> Anatoly
>