You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
It has been reported to me in private that it is possible for Peanut-GB to call NULL as a function if a spurious write to 0xFF50 happens. If gb_bootrom_read is not set by the frontend implementation, Peanut-GB could end up calling NULL as a function point when reading the first 0x100 bytes of ROM bank 0.
Thank you to the anonymous reporter.
The text was updated successfully, but these errors were encountered:
Fixed issue #110 whereby a spurious write could enable the bootrom.
Previously, the bootrom could be enabled by writing 0 to 0xFF50. But
this is not possible on real hardware after the bootrom writes 1 to
0xFF50. No game is expected to attempt to re-enable the bootrom, but a
bug in a game or the emulator could cause it to happen.
If the bootrom was enabled by the game, and the gb_bootrom_read function
was set to NULL, then a NULL pointer exception was possible.
Signed-off-by: Mahyar Koshkouei <mk@deltabeard.com>
It has been reported to me in private that it is possible for Peanut-GB to call NULL as a function if a spurious write to 0xFF50 happens. If gb_bootrom_read is not set by the frontend implementation, Peanut-GB could end up calling NULL as a function point when reading the first 0x100 bytes of ROM bank 0.
Thank you to the anonymous reporter.
The text was updated successfully, but these errors were encountered: