mirror of
https://github.com/reactos/reactos.git
synced 2025-02-28 19:32:59 +00:00
7791a36fc2
Yes, simply muting those assertions is a super-*dirrty*-workaround https://www.youtube.com/watch?v=4Rg3sAb8Id8 It sweeps under the carpet a mild form of FS-corruption without actually preventing it from happening. And I will happily revert that workaround and replace it with something better, once Timo or anyone else will find something better to do on master head. For now it helps me to increase the chance of succeeding the testbot runs without hitting those assertions, and it also increases the chances to succeed a git clone from within ros. And it will decrease the chance of actual ros users running into those endless cascades of assertions which - when run without a debugger - will make ros halt and enforce the user to reboot, and even when run with the debugger can never be escaped, because you have to 'i'gnore endless amounts of them. I tested this hackfix for a week, and have not seen any *severe* side-effects of muting them. Not nice to hack around like that, for sure not. But it is the best I can come up with for now. Something like that is not needed for any older release, because the older releases did still use old MM/CC and old non-MS-fastfat. The real bug is most likely to find within our MM/CC, at least when assuming that MS PL-licensed fastfat wasn't hacked too much by us when it was introduced during 0.4.15-dev'ing. |
||
|---|---|---|
| .. | ||
| base | ||
| battery | ||
| bluetooth | ||
| bus | ||
| crypto | ||
| filesystems | ||
| filters | ||
| hid | ||
| input | ||
| ksfilter | ||
| multimedia | ||
| network | ||
| parallel | ||
| processor | ||
| sac | ||
| serial | ||
| setup | ||
| storage | ||
| usb | ||
| wdm | ||
| wmi | ||
| CMakeLists.txt | ||