Re: [PATCH] eal: allow worker lcore stacks to be allocated from hugepage memory

[Don Wallwork]

On 5/3/2022 9:08 AM, Wang, Haiyue wrote:
-----Original Message-----
From: Don Wallwork <d...@xsightlabs.com>
Sent: Monday, May 2, 2022 22:11
To: dev@dpdk.org
Cc: d...@xsightlabs.com; step...@networkplumber.org; 
m...@smartsharesystems.com; Burakov, Anatoly
<anatoly.bura...@intel.com>; dmitry.kozl...@gmail.com; Richardson, Bruce 
<bruce.richard...@intel.com>;
honnappa.nagaraha...@arm.com; n...@arm.com
Subject: [PATCH] eal: allow worker lcore stacks to be allocated from hugepage 
memory
Add support for using hugepages for worker lcore stack memory.  The
intent is to improve performance by reducing stack memory related TLB
misses and also by using memory local to the NUMA node of each lcore.

EAL option '--huge-worker-stack [stack-size-kbytes]' is added to allow
the feature to be enabled at runtime.  If the size is not specified,
the system pthread stack size will be used.

Signed-off-by: Don Wallwork <d...@xsightlabs.com>
---
  lib/eal/common/eal_common_options.c | 31 ++++++++++++++
  lib/eal/common/eal_internal_cfg.h   |  4 ++
  lib/eal/common/eal_options.h        |  2 +
  lib/eal/linux/eal.c                 | 65 ++++++++++++++++++++++++++++-
  4 files changed, 100 insertions(+), 2 deletions(-)

diff --git a/lib/eal/common/eal_common_options.c 
b/lib/eal/common/eal_common_options.c
index f247a42455..be9db9ee37 100644
--- a/lib/eal/common/eal_common_options.c
+++ b/lib/eal/common/eal_common_options.c
@@ -103,6 +103,7 @@ eal_long_options[] = {
        {OPT_TELEMETRY,         0, NULL, OPT_TELEMETRY_NUM        },
        {OPT_NO_TELEMETRY,      0, NULL, OPT_NO_TELEMETRY_NUM     },
        {OPT_FORCE_MAX_SIMD_BITWIDTH, 1, NULL, OPT_FORCE_MAX_SIMD_BITWIDTH_NUM},
+       {OPT_HUGE_WORKER_STACK, 2, NULL, OPT_HUGE_WORKER_STACK_NUM     },

        {0,                     0, NULL, 0                        }
  };
@@ -1618,6 +1619,22 @@ eal_parse_huge_unlink(const char *arg, struct 
hugepage_file_discipline *out)
        return -1;
  }

+static int
+eal_parse_huge_worker_stack(const char *arg, size_t *huge_worker_stack_size)
+{
+       size_t worker_stack_size;
+       if (arg == NULL) {
+               *huge_worker_stack_size = USE_OS_STACK_SIZE;
+               return 0;
+       }
+       worker_stack_size = atoi(arg);
+       if (worker_stack_size == 0)
+               return -1;
Should we also to check "worker_stack_size *1024  < PTHREAD_STACK_MIN" ?
This may be too restrictive in certain environments.  For example, 
memory constrained platforms may require a smaller worker stack size 
than this limit would allow.
+
+       *huge_worker_stack_size = worker_stack_size * 1024;
+       return 0;
+}
+

--
2.17.1