Advisory

GHSA-qp29-wxp5-wh82

Also known as: CVE-2025-55154

Summary

imagemagick: integer overflows in MNG magnification

Details

Vulnerability Details

The magnified size calculations in ReadOneMNGIMage (in coders/png.c) are unsafe and can overflow, leading to memory corruption.

The source snippet below is heavily abbreviated due to the size of the function, but hopefully the important points are captured.

static Image *ReadOneMNGImage(MngReadInfo* mng_info,
  const ImageInfo *image_info,ExceptionInfo *exception)
{

// Lots of stuff, this is effectively a state machine for the MNG rendering commands,
// skip to the point where we start processing the "MAGN" command.

        if (memcmp(type,mng_MAGN,4) == 0)
          {
            png_uint_16
              magn_first,
              magn_last,
              magn_mb,
              magn_ml,
              magn_mr,
              magn_mt,
              magn_mx,
              magn_my,
              magn_methx,
              magn_methy;

// Details unimportant, but each of the `magn_xxx` variables is read from the file.

            if (magn_first == 0 || magn_last == 0)
              {
                /* Save the magnification factors for object 0 */
                mng_info->magn_mb=magn_mb;
                mng_info->magn_ml=magn_ml;
                mng_info->magn_mr=magn_mr;
                mng_info->magn_mt=magn_mt;
                mng_info->magn_mx=magn_mx;
                mng_info->magn_my=magn_my;
                mng_info->magn_methx=magn_methx;
                mng_info->magn_methy=magn_methy;
              }
          }

// Details unimportant, we load the image to be scaled and store it in `image`

    if (mng_type)
      {
        MngBox
          crop_box;

        if (((mng_info->magn_methx > 0) && (mng_info->magn_methx <= 5)) &&
            ((mng_info->magn_methy > 0) && (mng_info->magn_methy <= 5)))
          {
            png_uint_32
               magnified_height,
               magnified_width;

            if (logging != MagickFalse)
              (void) LogMagickEvent(CoderEvent,GetMagickModule(),
                "  Processing MNG MAGN chunk");

            if (image->columns == 1)
              mng_info->magn_methx = 1;
            if (image->rows == 1)
              mng_info->magn_methy = 1;
            if (mng_info->magn_methx == 1)
              {
                magnified_width=mng_info->magn_ml; // [0]
                
                if (image->columns > 1)
                   magnified_width += mng_info->magn_mr; // [1]

                if (image->columns > 2)
                   magnified_width += (png_uint_32)
                      ((image->columns-2)*(mng_info->magn_mx)); // [2]
               }

// Different cases handle available scaling kinds, all of which have similar issues...

// We now check whether the output image is larger than the input image in either
// dimension, and if so, we will allocate a new image buffer of size
// `magnified_width * magnified_height`.

            if (magnified_height > image->rows ||
                magnified_width > image->columns)
              {
                Image
                  *large_image;

// Snip...

                large_image->columns=magnified_width;
                large_image->rows=magnified_height;

                magn_methx=mng_info->magn_methx;
                magn_methy=mng_info->magn_methy;

// In between here, we allocate the pixel buffer for `large_image`.

                /* magnify the rows into the right side of the large image */

                if (logging != MagickFalse)
                  (void) LogMagickEvent(CoderEvent,GetMagickModule(),
                    "    Magnify the rows to %.20g",
                    (double) large_image->rows);
                m=(ssize_t) mng_info->magn_mt;
                yy=0;
                length=(size_t) GetPixelChannels(image)*image->columns;
                next=(Quantum *) AcquireQuantumMemory(length,sizeof(*next));
                prev=(Quantum *) AcquireQuantumMemory(length,sizeof(*prev));

                if ((prev == (Quantum *) NULL) ||
                    (next == (Quantum *) NULL))
                  {
                    if (prev != (Quantum *) NULL)
                      prev=(Quantum *) RelinquishMagickMemory(prev);
                    if (next != (Quantum *) NULL)
                      next=(Quantum *) RelinquishMagickMemory(next);
                    image=DestroyImageList(image);
                    ThrowReaderException(ResourceLimitError,
                      "MemoryAllocationFailed");
                  }

                n=GetAuthenticPixels(image,0,0,image->columns,1,exception);
                (void) memcpy(next,n,length);

                for (y=0; y < (ssize_t) image->rows; y++)
                {
                  if (y == 0)
                    m=(ssize_t) mng_info->magn_mt;

                  else if (magn_methy > 1 && y == (ssize_t) image->rows-2)
                    m=(ssize_t) mng_info->magn_mb;

                  else if (magn_methy <= 1 && y == (ssize_t) image->rows-1)
                    m=(ssize_t) mng_info->magn_mb;

                  else if (magn_methy > 1 && y == (ssize_t) image->rows-1)
                    m=1;

                  else
                    m=(ssize_t) mng_info->magn_my;

                  n=prev;
                  prev=next;
                  next=n;

                  if (y < (ssize_t) image->rows-1)
                    {
                      n=GetAuthenticPixels(image,0,y+1,image->columns,1,
                          exception);
                      (void) memcpy(next,n,length);
                    }

                  for (i=0; i < m; i++, yy++)
                  {
                    Quantum
                      *pixels;

                    assert(yy < (ssize_t) large_image->rows);
                    pixels=prev;
                    n=next;
                    q=GetAuthenticPixels(large_image,0,yy,large_image->columns,
                      1,exception);
                    if (q == (Quantum *) NULL)
                      break;
                    q+=(ptrdiff_t) (large_image->columns-image->columns)*
                      GetPixelChannels(large_image); // [3]

If we look at the calculation for magnified_width, we can see that we are storing the results in a png_uint32. The operations at \[0\] and \[1\] are safe, since mng_info->magn_ml and mng_info->magn_mx are both 16-bit unsigned integers, but both the multiplication at \[2\] and the addition of the result of that multiplication to magnified_width can overflow, leading to a value of magnified_width that is smaller than required.

When we then operate on the pixel buffers, we use the original parameters for the magnification, and we assume (reasonably?) that the output buffer is larger than the input buffer when calculating where to write the upsampled/magnified pixel values. Unfortunately, after the overflow has happened, this assumption is no longer true, and the calculation at \[3\] will end up with a q pointer outside the buffer bounds.

This issue leads to an out-of-bounds write of controlled data beyond the bounds of a heap allocation.

Severity

High Basis: The source's own band: HIGH

Dates

Published
2025-08-25 16:23 UTC
Modified
2025-11-04 16:41 UTC

Affected packages and fixes

18 affected packages; all NuGet · All systems; affected: earlier versions
PackageFix
Magick.NET-Q8-x86The export has no generic command for this ecosystem.
Magick.NET-Q8-x64The export has no generic command for this ecosystem.
Magick.NET-Q8-arm64The export has no generic command for this ecosystem.
Magick.NET-Q8-OpenMP-x64The export has no generic command for this ecosystem.
Magick.NET-Q8-OpenMP-arm64The export has no generic command for this ecosystem.
Magick.NET-Q8-AnyCPUThe export has no generic command for this ecosystem.
Magick.NET-Q16-x86The export has no generic command for this ecosystem.
Magick.NET-Q16-x64The export has no generic command for this ecosystem.
Magick.NET-Q16-arm64The export has no generic command for this ecosystem.
Magick.NET-Q16-OpenMP-x64The export has no generic command for this ecosystem.
Magick.NET-Q16-OpenMP-arm64The export has no generic command for this ecosystem.
Magick.NET-Q16-HDRI-x86The export has no generic command for this ecosystem.
Magick.NET-Q16-HDRI-x64The export has no generic command for this ecosystem.
Magick.NET-Q16-HDRI-arm64The export has no generic command for this ecosystem.
Magick.NET-Q16-HDRI-OpenMP-x64The export has no generic command for this ecosystem.
Magick.NET-Q16-HDRI-OpenMP-arm64The export has no generic command for this ecosystem.
Magick.NET-Q16-HDRI-AnyCPUThe export has no generic command for this ecosystem.
Magick.NET-Q16-AnyCPUThe export has no generic command for this ecosystem.

Fix commands are the export's own generic commands. Ojuju never runs them. Check them against your setup first.

Check your own computer: guard packages check

View this record on OSV.dev

Source

Source: GitHub Advisory Database (reviewed advisories) · CC BY 4.0

Sources & licences