From: Joel Fernandes <joelagnelf@nvidia.com>
To: linux-kernel@vger.kernel.org, Danilo Krummrich <dakr@kernel.org>,
David Airlie <airlied@gmail.com>, Simona Vetter <simona@ffwll.ch>
Cc: nouveau@lists.freedesktop.org, dri-devel@lists.freedesktop.org,
Alexandre Courbot <acourbot@nvidia.com>,
John Hubbard <jhubbard@nvidia.com>,
Shirish Baskaran <sbaskaran@nvidia.com>,
Alistair Popple <apopple@nvidia.com>,
Timur Tabi <ttabi@nvidia.com>, Ben Skeggs <bskeggs@nvidia.com>,
Joel Fernandes <joelagnelf@nvidia.com>
Subject: [PATCH 2/6] nova-core: doc: Clarify sysmembar operations
Date: Wed, 23 Apr 2025 18:53:58 -0400 [thread overview]
Message-ID: <20250423225405.139613-3-joelagnelf@nvidia.com> (raw)
In-Reply-To: <20250423225405.139613-1-joelagnelf@nvidia.com>
sysmembar is a critical operation that the GSP falcon needs to perform
in the reset sequence. Add some code comments to clarify.
Signed-off-by: Joel Fernandes <joelagnelf@nvidia.com>
---
drivers/gpu/nova-core/gpu.rs | 11 ++++++++++-
drivers/gpu/nova-core/regs.rs | 2 ++
2 files changed, 12 insertions(+), 1 deletion(-)
diff --git a/drivers/gpu/nova-core/gpu.rs b/drivers/gpu/nova-core/gpu.rs
index 4d03a0b11b64..61031bccb7d3 100644
--- a/drivers/gpu/nova-core/gpu.rs
+++ b/drivers/gpu/nova-core/gpu.rs
@@ -158,6 +158,9 @@ pub(crate) struct Gpu {
/// MMIO mapping of PCI BAR 0
bar: Devres<Bar0>,
fw: Firmware,
+ /// A system memory page for sysmembar (A GPU-initiated hardware memory-barrier
+ /// operation that flushes all pending GPU-side memory writes that were done
+ /// through PCIE, to system memory).
sysmem_flush: DmaObject,
timer: Timer,
bios: Vbios,
@@ -204,7 +207,13 @@ pub(crate) fn new(
devinit::wait_gfw_boot_completion(&bar)
.inspect_err(|_| pr_err!("GFW boot did not complete"))?;
- // System memory page required for sysmembar to properly flush into system memory.
+ // System memory page required for sysmembar which is a GPU-initiated hardware
+ // memory-barrier operation that flushes all pending GPU-side memory writes that
+ // were done through PCIE, to system memory. It is required for Falcon to be reset
+ // as the reset operation involves a reset handshake. When the falcon acks the
+ // reset, it writes its acknowledgement into system memory, but for this write to
+ // be visible to the host, the falcon needs to do sysmembar to flush
+ // its writes and prevent the driver from timing out.
let sysmem_flush = {
let page = DmaObject::new(
pdev.as_ref(),
diff --git a/drivers/gpu/nova-core/regs.rs b/drivers/gpu/nova-core/regs.rs
index f4a6a382e83f..22906ab1a43a 100644
--- a/drivers/gpu/nova-core/regs.rs
+++ b/drivers/gpu/nova-core/regs.rs
@@ -37,6 +37,8 @@
/* PFB */
+/// These two registers together hold the physical system memory address
+/// that is used by the GPU for perform sysmembar operation (see gpu.rs).
register!(PfbNisoFlushSysmemAddr@0x00100c10;
31:0 adr_39_08 => as u32
);
--
2.43.0
next prev parent reply other threads:[~2025-04-23 22:54 UTC|newest]
Thread overview: 14+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-04-23 22:53 [PATCH 0/6] Additional documentation for nova-core Joel Fernandes
2025-04-23 22:53 ` [PATCH 1/6] nova-core: doc: Add code comments related to devinit Joel Fernandes
2025-04-23 22:53 ` Joel Fernandes [this message]
2025-04-23 22:53 ` [PATCH 3/6] nova-core: docs: Document vbios layout Joel Fernandes
2025-04-23 23:12 ` Timur Tabi
2025-04-23 23:30 ` Joel Fernandes
2025-04-23 22:54 ` [PATCH 4/6] nova-core: docs: Document fwsec operation and layout Joel Fernandes
2025-04-23 22:54 ` [PATCH 5/6] gpu: nova-core: Clarify fields in FalconAppifHdrV1 Joel Fernandes
2025-04-24 1:18 ` Alexandre Courbot
[not found] ` <174546399398.876.3516508778193165894@patchwork.local>
2025-04-24 7:45 ` [5/6] " Alexandre Courbot
2025-04-24 13:03 ` Joel Fernandes
2025-04-23 22:54 ` [PATCH 6/6] nova-core: docs: Document devinit process Joel Fernandes
2025-04-24 1:25 ` [PATCH 0/6] Additional documentation for nova-core Alexandre Courbot
2025-04-24 8:10 ` Danilo Krummrich
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=20250423225405.139613-3-joelagnelf@nvidia.com \
--to=joelagnelf@nvidia.com \
--cc=acourbot@nvidia.com \
--cc=airlied@gmail.com \
--cc=apopple@nvidia.com \
--cc=bskeggs@nvidia.com \
--cc=dakr@kernel.org \
--cc=dri-devel@lists.freedesktop.org \
--cc=jhubbard@nvidia.com \
--cc=linux-kernel@vger.kernel.org \
--cc=nouveau@lists.freedesktop.org \
--cc=sbaskaran@nvidia.com \
--cc=simona@ffwll.ch \
--cc=ttabi@nvidia.com \
/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