From: Oded Gabbay <ogabbay@kernel.org>
To: linux-kernel@vger.kernel.org
Cc: Yuri Nudelman <ynudelman@habana.ai>
Subject: [PATCH 2/6] habanalabs: fix debugfs device memory MMU VA translation
Date: Mon, 4 Oct 2021 13:08:26 +0300 [thread overview]
Message-ID: <20211004100830.27573-2-ogabbay@kernel.org> (raw)
In-Reply-To: <20211004100830.27573-1-ogabbay@kernel.org>
From: Yuri Nudelman <ynudelman@habana.ai>
The translation in debugfs of device memory MMU virtual addresses was
wrong as it did not take into consideration the fact that the page
sizes there can be not power of 2.
Signed-off-by: Yuri Nudelman <ynudelman@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
---
drivers/misc/habanalabs/common/mmu/mmu.c | 30 +++++++++++++-----------
1 file changed, 16 insertions(+), 14 deletions(-)
diff --git a/drivers/misc/habanalabs/common/mmu/mmu.c b/drivers/misc/habanalabs/common/mmu/mmu.c
index 792d25b79ea6..aa96917f62e5 100644
--- a/drivers/misc/habanalabs/common/mmu/mmu.c
+++ b/drivers/misc/habanalabs/common/mmu/mmu.c
@@ -501,23 +501,25 @@ static void hl_mmu_pa_page_with_offset(struct hl_ctx *ctx, u64 virt_addr,
if ((hops->range_type == HL_VA_RANGE_TYPE_DRAM) &&
!is_power_of_2(prop->dram_page_size)) {
- unsigned long dram_page_size = prop->dram_page_size;
- u64 page_offset_mask;
- u64 phys_addr_mask;
- u32 bit;
+ u64 dram_page_size, dram_base, abs_phys_addr, abs_virt_addr,
+ page_id, page_start;
+ u32 page_off;
/*
- * find last set bit in page_size to cover all bits of page
- * offset. note that 1 has to be added to bit index.
- * note that the internal ulong variable is used to avoid
- * alignment issue.
+ * Bit arithmetics cannot be used for non power of two page
+ * sizes. In addition, since bit arithmetics is not used,
+ * we cannot ignore dram base. All that shall be considerd.
*/
- bit = find_last_bit(&dram_page_size,
- sizeof(dram_page_size) * BITS_PER_BYTE) + 1;
- page_offset_mask = (BIT_ULL(bit) - 1);
- phys_addr_mask = ~page_offset_mask;
- *phys_addr = (tmp_phys_addr & phys_addr_mask) |
- (virt_addr & page_offset_mask);
+
+ dram_page_size = prop->dram_page_size;
+ dram_base = prop->dram_base_address;
+ abs_phys_addr = tmp_phys_addr - dram_base;
+ abs_virt_addr = virt_addr - dram_base;
+ page_id = DIV_ROUND_DOWN_ULL(abs_phys_addr, dram_page_size);
+ page_start = page_id * dram_page_size;
+ div_u64_rem(abs_virt_addr, dram_page_size, &page_off);
+
+ *phys_addr = page_start + page_off + dram_base;
} else {
/*
* find the correct hop shift field in hl_mmu_properties
--
2.17.1
next prev parent reply other threads:[~2021-10-04 10:08 UTC|newest]
Thread overview: 6+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-10-04 10:08 [PATCH 1/6] habanalabs: add support for a long interrupt target value Oded Gabbay
2021-10-04 10:08 ` Oded Gabbay [this message]
2021-10-04 10:08 ` [PATCH 3/6] habanalabs: define soft-reset as inference op Oded Gabbay
2021-10-04 10:08 ` [PATCH 4/6] habanalabs: refactor reset log message Oded Gabbay
2021-10-04 10:08 ` [PATCH 5/6] habanalabs: prevent race between fd close/open Oded Gabbay
2021-10-04 10:08 ` [PATCH 6/6] habanalabs: take timestamp on wait for interrupt Oded Gabbay
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=20211004100830.27573-2-ogabbay@kernel.org \
--to=ogabbay@kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=ynudelman@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