Skip to content

Harden integer ranges in parallel BC compression #744

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
Roland Shum (ShumWengSang) wants to merge 4 commits into
base: main
Choose a base branch
from
Open

Harden integer ranges in parallel BC compression #744

Changes from all commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
9f1e448
Harden integer ranges in parallel BC compression
Sep 23, 2026
a497ca6
Apply repository formatting to block extent declarations
Sep 23, 2026
a04f4e5
Use uint64_t intermediates for checked parallel block counts
Sep 24, 2026
386480d
Narrow checked parallel block count to size_t
Sep 25, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
21 changes: 19 additions & 2 deletions DirectXTex/DirectXTexCompress.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -297,7 +297,24 @@ namespace
return HRESULT_E_NOT_SUPPORTED;

// Refactored version of loop to support parallel independance
const size_t nBlocks = std::max<size_t>(1, (image.width + 3) / 4) * std::max<size_t>(1, (image.height + 3) / 4);
// The existing loop iterator and coordinate calculations use int. Limit
// their input dimensions before calculating the block count.
// Direct3D texture size limits are below INT32_MAX, so this should not pose a problem in practice.
if (image.width > static_cast<size_t>(INT32_MAX) || image.height > static_cast<size_t>(INT32_MAX))
{
Comment thread
ShumWengSang marked this conversation as resolved.
return HRESULT_E_ARITHMETIC_OVERFLOW;
}

// The dimension check keeps the rounded counts and their product within uint64_t.
const uint64_t nbWidthBlocks = std::max<uint64_t>(1, (uint64_t(image.width) + 3) / 4);
const uint64_t nbHeightBlocks = std::max<uint64_t>(1, (uint64_t(image.height) + 3) / 4);
const uint64_t blockCount = nbWidthBlocks * nbHeightBlocks;
if (blockCount > INT32_MAX)
{
return HRESULT_E_ARITHMETIC_OVERFLOW;
}

const auto nBlocks = static_cast<size_t>(blockCount);

bool fail = false;

Expand All @@ -317,7 +334,7 @@ namespace
continue;
}

const int nbWidth = std::max<int>(1, int((image.width + 3) / 4));
const int nbWidth = static_cast<int>(nbWidthBlocks);

int y = nb / nbWidth;
const int x = (nb - (y * nbWidth)) * 4;
Expand Down
Loading