eal: compute IOVA mode based on PA availability

Currently, if the bus selects IOVA as PA, the memory init can fail when
lacking access to physical addresses.
This can be quite hard for normal users to understand what is wrong
since this is the default behavior.

Catch this situation earlier in eal init by validating physical addresses
availability, or select IOVA when no clear preferrence had been expressed.

The bus code is changed so that it reports when it does not care about
the IOVA mode and let the eal init decide.

In Linux implementation, rework rte_eal_using_phys_addrs() so that it can
be called earlier but still avoid a circular dependency with
rte_mem_virt2phys().
In FreeBSD implementation, rte_eal_using_phys_addrs() always returns
false, so the detection part is left as is.

If librte_kni is compiled in and the KNI kmod is loaded,
- if the buses requested VA, force to PA if physical addresses are
  available as it was done before,
- else, keep iova as VA, KNI init will fail later.

Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Signed-off-by: David Marchand <david.marchand@redhat.com>
Acked-by: Anatoly Burakov <anatoly.burakov@intel.com>
This commit is contained in:
Ben Walker 2019-06-14 11:39:16 +02:00 committed by Thomas Monjalon
parent fe8fc6b0b7
commit c2361bab70
5 changed files with 53 additions and 51 deletions

View File

@ -237,10 +237,6 @@ rte_bus_get_iommu_class(void)
mode |= bus->get_iommu_class();
}
if (mode != RTE_IOVA_VA) {
/* Use default IOVA mode */
mode = RTE_IOVA_PA;
}
return mode;
}

View File

@ -392,7 +392,7 @@ struct rte_bus *rte_bus_find_by_name(const char *busname);
/**
* Get the common iommu class of devices bound on to buses available in the
* system. The default mode is PA.
* system. RTE_IOVA_DC means that no preferrence has been expressed.
*
* @return
* enum rte_iova_mode value.

View File

@ -765,13 +765,19 @@ rte_eal_init(int argc, char **argv)
/* if no EAL option "--iova-mode=<pa|va>", use bus IOVA scheme */
if (internal_config.iova_mode == RTE_IOVA_DC) {
/* autodetect the IOVA mapping mode (default is RTE_IOVA_PA) */
rte_eal_get_configuration()->iova_mode =
rte_bus_get_iommu_class();
enum rte_iova_mode iova_mode = rte_bus_get_iommu_class();
if (iova_mode == RTE_IOVA_DC)
iova_mode = RTE_IOVA_PA;
rte_eal_get_configuration()->iova_mode = iova_mode;
} else {
rte_eal_get_configuration()->iova_mode =
internal_config.iova_mode;
}
RTE_LOG(INFO, EAL, "Selected IOVA mode '%s'\n",
rte_eal_iova_mode() == RTE_IOVA_PA ? "PA" : "VA");
if (internal_config.no_hugetlbfs == 0) {
/* rte_config isn't initialized yet */
ret = internal_config.process_type == RTE_PROC_PRIMARY ?

View File

@ -985,6 +985,7 @@ rte_eal_init(int argc, char **argv)
static char logid[PATH_MAX];
char cpuset[RTE_CPU_AFFINITY_STR_LEN];
char thread_name[RTE_MAX_THREAD_NAME_LEN];
bool phys_addrs;
/* checks if the machine is adequate */
if (!rte_cpu_is_supported()) {
@ -1075,25 +1076,46 @@ rte_eal_init(int argc, char **argv)
return -1;
}
phys_addrs = rte_eal_using_phys_addrs() != 0;
/* if no EAL option "--iova-mode=<pa|va>", use bus IOVA scheme */
if (internal_config.iova_mode == RTE_IOVA_DC) {
/* autodetect the IOVA mapping mode (default is RTE_IOVA_PA) */
rte_eal_get_configuration()->iova_mode =
rte_bus_get_iommu_class();
/* autodetect the IOVA mapping mode */
enum rte_iova_mode iova_mode = rte_bus_get_iommu_class();
/* Workaround for KNI which requires physical address to work */
if (rte_eal_get_configuration()->iova_mode == RTE_IOVA_VA &&
rte_eal_check_module("rte_kni") == 1) {
rte_eal_get_configuration()->iova_mode = RTE_IOVA_PA;
RTE_LOG(WARNING, EAL,
"Some devices want IOVA as VA but PA will be used because.. "
"KNI module inserted\n");
if (iova_mode == RTE_IOVA_DC) {
iova_mode = phys_addrs ? RTE_IOVA_PA : RTE_IOVA_VA;
RTE_LOG(DEBUG, EAL,
"Buses did not request a specific IOVA mode, using '%s' based on physical addresses availability.\n",
phys_addrs ? "PA" : "VA");
}
#ifdef RTE_LIBRTE_KNI
/* Workaround for KNI which requires physical address to work */
if (iova_mode == RTE_IOVA_VA &&
rte_eal_check_module("rte_kni") == 1) {
if (phys_addrs) {
iova_mode = RTE_IOVA_PA;
RTE_LOG(WARNING, EAL, "Forcing IOVA as 'PA' because KNI module is loaded\n");
} else {
RTE_LOG(DEBUG, EAL, "KNI can not work since physical addresses are unavailable\n");
}
}
#endif
rte_eal_get_configuration()->iova_mode = iova_mode;
} else {
rte_eal_get_configuration()->iova_mode =
internal_config.iova_mode;
}
if (rte_eal_iova_mode() == RTE_IOVA_PA && !phys_addrs) {
rte_eal_init_alert("Cannot use IOVA as 'PA' since physical addresses are not available");
rte_errno = EINVAL;
return -1;
}
RTE_LOG(INFO, EAL, "Selected IOVA mode '%s'\n",
rte_eal_iova_mode() == RTE_IOVA_PA ? "PA" : "VA");
if (internal_config.no_hugetlbfs == 0) {
/* rte_config isn't initialized yet */
ret = internal_config.process_type == RTE_PROC_PRIMARY ?

View File

@ -65,34 +65,10 @@
* zone as well as a physical contiguous zone.
*/
static bool phys_addrs_available = true;
static int phys_addrs_available = -1;
#define RANDOMIZE_VA_SPACE_FILE "/proc/sys/kernel/randomize_va_space"
static void
test_phys_addrs_available(void)
{
uint64_t tmp = 0;
phys_addr_t physaddr;
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) {
if (rte_eal_iova_mode() == RTE_IOVA_PA)
RTE_LOG(ERR, EAL,
"Cannot obtain physical addresses: %s. "
"Only vfio will function.\n",
strerror(errno));
phys_addrs_available = false;
}
}
/*
* Get physical address of any mapped virtual address in the current process.
*/
@ -105,8 +81,7 @@ rte_mem_virt2phy(const void *virtaddr)
int page_size;
off_t offset;
/* Cannot parse /proc/self/pagemap, no need to log errors everywhere */
if (!phys_addrs_available)
if (phys_addrs_available == 0)
return RTE_BAD_IOVA;
/* standard page size */
@ -1336,8 +1311,6 @@ eal_legacy_hugepage_init(void)
int nr_hugefiles, nr_hugepages = 0;
void *addr;
test_phys_addrs_available();
memset(used_hp, 0, sizeof(used_hp));
/* get pointer to global configuration */
@ -1516,7 +1489,7 @@ eal_legacy_hugepage_init(void)
continue;
}
if (phys_addrs_available &&
if (rte_eal_using_phys_addrs() &&
rte_eal_iova_mode() != RTE_IOVA_VA) {
/* find physical addresses for each hugepage */
if (find_physaddrs(&tmp_hp[hp_offset], hpi) < 0) {
@ -1735,8 +1708,6 @@ eal_hugepage_init(void)
uint64_t memory[RTE_MAX_NUMA_NODES];
int hp_sz_idx, socket_id;
test_phys_addrs_available();
memset(used_hp, 0, sizeof(used_hp));
for (hp_sz_idx = 0;
@ -1879,8 +1850,6 @@ eal_legacy_hugepage_attach(void)
"into secondary processes\n");
}
test_phys_addrs_available();
fd_hugepage = open(eal_hugepage_data_path(), O_RDONLY);
if (fd_hugepage < 0) {
RTE_LOG(ERR, EAL, "Could not open %s\n",
@ -2020,6 +1989,15 @@ rte_eal_hugepage_attach(void)
int
rte_eal_using_phys_addrs(void)
{
if (phys_addrs_available == -1) {
uint64_t tmp = 0;
if (rte_eal_has_hugepages() != 0 &&
rte_mem_virt2phy(&tmp) != RTE_BAD_PHYS_ADDR)
phys_addrs_available = 1;
else
phys_addrs_available = 0;
}
return phys_addrs_available;
}