From: <eagostini@nvidia.com>
To: <dev@dpdk.org>
Cc: Elena Agostini <eagostini@nvidia.com>
Subject: [PATCH v2 1/3] gpudev: mem alloc aligned memory
Date: Sat, 8 Jan 2022 00:20:01 +0000 [thread overview]
Message-ID: <20220108002003.21153-1-eagostini@nvidia.com> (raw)
In-Reply-To: <20220104014721.1799-1-eagostini@nvidia.com>
From: Elena Agostini <eagostini@nvidia.com>
Similarly to rte_malloc, rte_gpu_mem_alloc accept as
input the memory alignment size.
GPU driver should return GPU memory address aligned
with the input value.
Changelog:
- rte_gpu_mem_alloc parameters order
Signed-off-by: Elena Agostini <eagostini@nvidia.com>
---
lib/gpudev/gpudev.c | 10 ++++++++--
lib/gpudev/gpudev_driver.h | 2 +-
lib/gpudev/rte_gpudev.h | 10 +++++++---
3 files changed, 16 insertions(+), 6 deletions(-)
diff --git a/lib/gpudev/gpudev.c b/lib/gpudev/gpudev.c
index 9ae36dbae9..59e2169292 100644
--- a/lib/gpudev/gpudev.c
+++ b/lib/gpudev/gpudev.c
@@ -527,7 +527,7 @@ rte_gpu_info_get(int16_t dev_id, struct rte_gpu_info *info)
}
void *
-rte_gpu_mem_alloc(int16_t dev_id, size_t size)
+rte_gpu_mem_alloc(int16_t dev_id, size_t size, unsigned int align)
{
struct rte_gpu *dev;
void *ptr;
@@ -549,7 +549,13 @@ rte_gpu_mem_alloc(int16_t dev_id, size_t size)
if (size == 0) /* dry-run */
return NULL;
- ret = dev->ops.mem_alloc(dev, size, &ptr);
+ if (align && !rte_is_power_of_2(align)) {
+ GPU_LOG(ERR, "requested alignment is not a power of two %u", align);
+ rte_errno = EINVAL;
+ return NULL;
+ }
+
+ ret = dev->ops.mem_alloc(dev, size, align, &ptr);
switch (ret) {
case 0:
diff --git a/lib/gpudev/gpudev_driver.h b/lib/gpudev/gpudev_driver.h
index cb7b101f2f..0ed7478e9b 100644
--- a/lib/gpudev/gpudev_driver.h
+++ b/lib/gpudev/gpudev_driver.h
@@ -27,7 +27,7 @@ enum rte_gpu_state {
struct rte_gpu;
typedef int (rte_gpu_close_t)(struct rte_gpu *dev);
typedef int (rte_gpu_info_get_t)(struct rte_gpu *dev, struct rte_gpu_info *info);
-typedef int (rte_gpu_mem_alloc_t)(struct rte_gpu *dev, size_t size, void **ptr);
+typedef int (rte_gpu_mem_alloc_t)(struct rte_gpu *dev, size_t size, unsigned int align, void **ptr);
typedef int (rte_gpu_mem_free_t)(struct rte_gpu *dev, void *ptr);
typedef int (rte_gpu_mem_register_t)(struct rte_gpu *dev, size_t size, void *ptr);
typedef int (rte_gpu_mem_unregister_t)(struct rte_gpu *dev, void *ptr);
diff --git a/lib/gpudev/rte_gpudev.h b/lib/gpudev/rte_gpudev.h
index fa3f3aad4f..9e2e2c5dce 100644
--- a/lib/gpudev/rte_gpudev.h
+++ b/lib/gpudev/rte_gpudev.h
@@ -364,18 +364,22 @@ int rte_gpu_info_get(int16_t dev_id, struct rte_gpu_info *info);
* @param size
* Number of bytes to allocate.
* Requesting 0 will do nothing.
- *
+ * @param align
+ * If 0, the return is a pointer that is suitably aligned for any kind of
+ * variable (in the same manner as malloc()).
+ * Otherwise, the return is a pointer that is a multiple of *align*. In
+ * this case, it must obviously be a power of two.
* @return
* A pointer to the allocated memory, otherwise NULL and rte_errno is set:
* - ENODEV if invalid dev_id
- * - EINVAL if reserved flags
+ * - EINVAL if align is not a power of two
* - ENOTSUP if operation not supported by the driver
* - E2BIG if size is higher than limit
* - ENOMEM if out of space
* - EPERM if driver error
*/
__rte_experimental
-void *rte_gpu_mem_alloc(int16_t dev_id, size_t size)
+void *rte_gpu_mem_alloc(int16_t dev_id, size_t size, unsigned int align)
__rte_alloc_size(2);
/**
--
2.17.1
next prev parent reply other threads:[~2022-01-07 16:10 UTC|newest]
Thread overview: 12+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-01-04 1:47 [PATCH v1 0/3] GPU memory aligned eagostini
2022-01-04 1:47 ` [PATCH v1 1/3] gpudev: mem alloc aligned memory eagostini
2022-01-04 1:47 ` [PATCH v1 2/3] app/test-gpudev: test aligned memory allocation eagostini
2022-01-04 1:47 ` [PATCH v1 3/3] gpu/cuda: mem alloc aligned memory eagostini
2022-01-03 18:05 ` Stephen Hemminger
2022-01-03 18:15 ` Elena Agostini
2022-01-03 18:17 ` Stephen Hemminger
2022-01-03 18:22 ` Elena Agostini
2022-01-08 0:20 ` eagostini [this message]
2022-01-08 0:20 ` [PATCH v2 2/3] app/test-gpudev: test aligned memory allocation eagostini
2022-01-08 0:20 ` [PATCH v2 3/3] gpu/cuda: mem alloc aligned memory eagostini
2022-01-21 10:34 ` [PATCH v2 1/3] gpudev: " Thomas Monjalon
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20220108002003.21153-1-eagostini@nvidia.com \
--to=eagostini@nvidia.com \
--cc=dev@dpdk.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.