linux-kernel - Re: [PATCH v3 25/30] gpu: nova-core: Hopper/Blackwell: larger WPR2 (GSP) heap

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <71e90bb8-b7de-48e7-a43a-68d1e73bbe6a@nvidia.com>
Date: Sun, 8 Feb 2026 18:05:43 -0800
From: John Hubbard <jhubbard@...dia.com>
To: Timur Tabi <ttabi@...dia.com>, Alexandre Courbot <acourbot@...dia.com>,
 "dakr@...nel.org" <dakr@...nel.org>
Cc: "aliceryhl@...gle.com" <aliceryhl@...gle.com>,
 "lossin@...nel.org" <lossin@...nel.org>,
 "a.hindborg@...nel.org" <a.hindborg@...nel.org>,
 "boqun.feng@...il.com" <boqun.feng@...il.com>, Zhi Wang <zhiw@...dia.com>,
 "simona@...ll.ch" <simona@...ll.ch>,
 "alex.gaynor@...il.com" <alex.gaynor@...il.com>,
 "ojeda@...nel.org" <ojeda@...nel.org>, "tmgross@...ch.edu"
 <tmgross@...ch.edu>,
 "nouveau@...ts.freedesktop.org" <nouveau@...ts.freedesktop.org>,
 "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
 "rust-for-linux@...r.kernel.org" <rust-for-linux@...r.kernel.org>,
 "bjorn3_gh@...tonmail.com" <bjorn3_gh@...tonmail.com>,
 Eliot Courtney <ecourtney@...dia.com>, "airlied@...il.com"
 <airlied@...il.com>, Joel Fernandes <joelagnelf@...dia.com>,
 "bhelgaas@...gle.com" <bhelgaas@...gle.com>,
 "gary@...yguo.net" <gary@...yguo.net>, Alistair Popple <apopple@...dia.com>
Subject: Re: [PATCH v3 25/30] gpu: nova-core: Hopper/Blackwell: larger WPR2
 (GSP) heap

On 2/6/26 11:03 AM, Timur Tabi wrote:
> On Thu, 2026-02-05 at 20:21 -0800, John Hubbard wrote:
>> +    fn client_alloc_size(chipset: Chipset) -> u64 {
>> +        if crate::fb::hal::fb_hal(chipset)
>> +            .non_wpr_heap_size()
>> +            .is_some()
>> +        {
>> +            GSP_FW_HEAP_PARAM_CLIENT_ALLOC_SIZE_GH100
>> +        } else {
>> +            u64::from(bindings::GSP_FW_HEAP_PARAM_CLIENT_ALLOC_SIZE)
>> +        }
>> +        .align_up(GSP_HEAP_ALIGNMENT)
>> +        .unwrap_or(u64::MAX)
> 
> I understand not wanting to return a Result<> for impossible situations, but I don't know if this is
> the answer.  For one thing, there is no way we can allocate a heap that large.
> 
> Maybe we need to have some kind of infallible align_up() variant, but in this case, I would say a
> panic is warranted.

You are right. And that agrees with Miguel Ojeda's overall philosophy
about handling this type of error, too.

I'll change it to this for the next version:

diff --git a/drivers/gpu/nova-core/gsp/fw.rs b/drivers/gpu/nova-core/gsp/fw.rs
index 0d1d9eecd77c..927bcee6a5a5 100644
--- a/drivers/gpu/nova-core/gsp/fw.rs
+++ b/drivers/gpu/nova-core/gsp/fw.rs
@@ -83,7 +83,7 @@ fn client_alloc_size(chipset: Chipset) -> u64 {
              u64::from(bindings::GSP_FW_HEAP_PARAM_CLIENT_ALLOC_SIZE)
          }
          .align_up(GSP_HEAP_ALIGNMENT)
-        .unwrap_or(u64::MAX)
+        .expect("client_alloc_size alignment overflow")
      }
  
      /// Returns the amount of memory to reserve for management purposes for a framebuffer of size
@@ -94,7 +94,7 @@ fn management_overhead(fb_size: u64) -> u64 {
          u64::from(bindings::GSP_FW_HEAP_PARAM_SIZE_PER_GB_FB)
              .saturating_mul(fb_size_gb)
              .align_up(GSP_HEAP_ALIGNMENT)
-            .unwrap_or(u64::MAX)
+            .expect("management_overhead alignment overflow")
      }
  }


thanks,
-- 
John Hubbard