From: Oded Gabbay <ogabbay@kernel.org>
To: linux-kernel@vger.kernel.org
Cc: SW_Drivers@habana.ai, Ofir Bitton <obitton@habana.ai>
Subject: [PATCH] habanalabs: support reserving aligned va block
Date: Sat, 14 Nov 2020 11:24:14 +0200 [thread overview]
Message-ID: <20201114092414.12244-4-ogabbay@kernel.org> (raw)
In-Reply-To: <20201114092414.12244-1-ogabbay@kernel.org>
From: Ofir Bitton <obitton@habana.ai>
Add support for reserving va block with alignment different than
page size. This is a pre-requisite for allocations needed in future
ASICs
Signed-off-by: Ofir Bitton <obitton@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
---
drivers/misc/habanalabs/common/habanalabs.h | 4 +++-
drivers/misc/habanalabs/common/memory.c | 8 +++++---
drivers/misc/habanalabs/gaudi/gaudi.c | 3 ++-
3 files changed, 10 insertions(+), 5 deletions(-)
diff --git a/drivers/misc/habanalabs/common/habanalabs.h b/drivers/misc/habanalabs/common/habanalabs.h
index eeb78381177b..76ab9741895d 100644
--- a/drivers/misc/habanalabs/common/habanalabs.h
+++ b/drivers/misc/habanalabs/common/habanalabs.h
@@ -303,6 +303,8 @@ enum hl_device_hw_state {
HL_DEVICE_HW_STATE_DIRTY
};
+#define HL_MMU_VA_ALIGNMENT_NOT_NEEDED 0
+
/**
* struct hl_mmu_properties - ASIC specific MMU address translation properties.
* @start_addr: virtual start address of the memory region.
@@ -2112,7 +2114,7 @@ int hl_vm_init(struct hl_device *hdev);
void hl_vm_fini(struct hl_device *hdev);
u64 hl_reserve_va_block(struct hl_device *hdev, struct hl_ctx *ctx,
- enum hl_va_range_type type, u32 size);
+ enum hl_va_range_type type, u32 size, u32 alignment);
int hl_unreserve_va_block(struct hl_device *hdev, struct hl_ctx *ctx,
u64 start_addr, u64 size);
int hl_pin_host_memory(struct hl_device *hdev, u64 addr, u64 size,
diff --git a/drivers/misc/habanalabs/common/memory.c b/drivers/misc/habanalabs/common/memory.c
index f27ca80d3c3c..351c9927151f 100644
--- a/drivers/misc/habanalabs/common/memory.c
+++ b/drivers/misc/habanalabs/common/memory.c
@@ -626,18 +626,20 @@ static u64 get_va_block(struct hl_device *hdev, struct hl_va_range *va_range,
* @ctx: current context
* @type: virtual addresses range type.
* @size: requested block size.
+ * @alignment: required alignment in bytes of the virtual block start address,
+ * 0 means no alignment.
*
* This function does the following:
* - Iterate on the virtual block list to find a suitable virtual block for the
- * given size.
+ * given size and alignment.
* - Reserve the requested block and update the list.
* - Return the start address of the virtual block.
*/
u64 hl_reserve_va_block(struct hl_device *hdev, struct hl_ctx *ctx,
- enum hl_va_range_type type, u32 size)
+ enum hl_va_range_type type, u32 size, u32 alignment)
{
return get_va_block(hdev, ctx->va_range[type], size, 0,
- ctx->va_range[type]->page_size);
+ max(alignment, ctx->va_range[type]->page_size));
}
/**
diff --git a/drivers/misc/habanalabs/gaudi/gaudi.c b/drivers/misc/habanalabs/gaudi/gaudi.c
index bf34ca29e42b..f1b8d20cf2ce 100644
--- a/drivers/misc/habanalabs/gaudi/gaudi.c
+++ b/drivers/misc/habanalabs/gaudi/gaudi.c
@@ -7802,7 +7802,8 @@ static int gaudi_internal_cb_pool_init(struct hl_device *hdev,
}
hdev->internal_cb_va_base = hl_reserve_va_block(hdev, ctx,
- HL_VA_RANGE_TYPE_HOST, HOST_SPACE_INTERNAL_CB_SZ);
+ HL_VA_RANGE_TYPE_HOST, HOST_SPACE_INTERNAL_CB_SZ,
+ HL_MMU_VA_ALIGNMENT_NOT_NEEDED);
if (!hdev->internal_cb_va_base)
goto destroy_internal_cb_pool;
--
2.17.1
prev parent reply other threads:[~2020-11-14 9:25 UTC|newest]
Thread overview: 4+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-11-14 9:24 [PATCH] habanalabs: add boot errors prints Oded Gabbay
2020-11-14 9:24 ` [PATCH] habanalabs: fix MMU debugfs operations Oded Gabbay
2020-11-14 9:24 ` [PATCH] habanalabs: share a single ctx-mutex between all MMUs Oded Gabbay
2020-11-14 9:24 ` Oded Gabbay [this message]
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=20201114092414.12244-4-ogabbay@kernel.org \
--to=ogabbay@kernel.org \
--cc=SW_Drivers@habana.ai \
--cc=linux-kernel@vger.kernel.org \
--cc=obitton@habana.ai \
/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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox