search for: reloadedbufferorerr

...y link will read from the VFS cache // (hopefully!) or from disk if the memory pressure wasn't too high. <- I don’t understand why we are reading from the cache file (via tryLoadingBuffer), though we already have the same content in OutputBuffer? auto ReloadedBufferOrErr = CacheEntry.tryLoadingBuffer(); <-The explanation says this is done to reduce the memory pressure. But it’s not clear to me why this helps to reduce memory pressure, since we still have to keep if (auto EC = ReloadedBufferOrErr.getError()) {...

...> > // (hopefully!) or from disk if the memory pressure wasn't too > high. <- I don’t understand why we are reading from the > cache file (via tryLoadingBuffer), though we already have the same content > in OutputBuffer? > > auto ReloadedBufferOrErr = CacheEntry.tryLoadingBuffer(); > <-The explanation says this is done to reduce the memory pressure. But it’s > not clear to me why this helps to reduce memory pressure, since we still > have to keep > > if (auto EC = ReloadedBufferOrErr.getError()) > {...

...y link will read from the VFS cache // (hopefully!) or from disk if the memory pressure wasn't too high. <- I don’t understand why we are reading from the cache file (via tryLoadingBuffer), though we already have the same content in OutputBuffer? auto ReloadedBufferOrErr = CacheEntry.tryLoadingBuffer(); <-The explanation says this is done to reduce the memory pressure. But it’s not clear to me why this helps to reduce memory pressure, since we still have to keep if (auto EC = ReloadedBufferOrErr.getError()) {...

...l read from the VFS cache > // (hopefully!) or from disk if the memory pressure wasn't too high. <- I don’t understand why we are reading from the cache file (via tryLoadingBuffer), though we already have the same content in OutputBuffer? > auto ReloadedBufferOrErr = CacheEntry.tryLoadingBuffer(); <-The explanation says this is done to reduce the memory pressure. But it’s not clear to me why this helps to reduce memory pressure, since we still have to keep > if (auto EC = ReloadedBufferOrErr.getError()) {...

...> > // (hopefully!) or from disk if the memory pressure wasn't too > high. <- I don’t understand why we are reading from the > cache file (via tryLoadingBuffer), though we already have the same content > in OutputBuffer? > > auto ReloadedBufferOrErr = CacheEntry.tryLoadingBuffer(); > <-The explanation says this is done to reduce the memory pressure. But it’s > not clear to me why this helps to reduce memory pressure, since we still > have to keep > > if (auto EC = ReloadedBufferOrErr.getError()) > {...

...y link will read from the VFS cache // (hopefully!) or from disk if the memory pressure wasn't too high. <- I don’t understand why we are reading from the cache file (via tryLoadingBuffer), though we already have the same content in OutputBuffer? auto ReloadedBufferOrErr = CacheEntry.tryLoadingBuffer(); <-The explanation says this is done to reduce the memory pressure. But it’s not clear to me why this helps to reduce memory pressure, since we still have to keep if (auto EC = ReloadedBufferOrErr.getError()) {...

...it will trigger a further discussion. > > > Solution #0 (it won’t require much modifications): > (a) If ‘rename’ fails, do not write into the cache directly (we don’t want to have non-atomic writes!). > (b) If ‘rename’ fails, try to read from the cache. > auto ReloadedBufferOrErr = CacheEntry.tryLoadingBuffer(); > (c) If reading from the cache fails simply use the file that you just compiled directly (bypassing the cache entirely). > > This solution might cause cache misses (causing recompilations), but at least it should prevent data races when two ThinLTO pro...

...below. Hopefully, it will trigger a further discussion. Solution #0 (it won't require much modifications): (a) If 'rename' fails, do not write into the cache directly (we don't want to have non-atomic writes!). (b) If 'rename' fails, try to read from the cache. auto ReloadedBufferOrErr = CacheEntry.tryLoadingBuffer(); (c) If reading from the cache fails simply use the file that you just compiled directly (bypassing the cache entirely). This solution might cause cache misses (causing recompilations), but at least it should prevent data races when two ThinLTO processes write...

...ly, it will trigger a > further discussion. > > > *Solution #0 (it won’t require much modifications):* > (a) If ‘rename’ fails, do not write into the cache directly (we don’t > want to have non-atomic writes!). > (b) If ‘rename’ fails, try to read from the cache. > *auto ReloadedBufferOrErr = CacheEntry.tryLoadingBuffer();* > *(c)* If reading from the cache fails simply use the file that you just > compiled directly (bypassing the cache entirely). > > This solution might cause cache misses (causing recompilations), but at > least it should prevent data races when two...

...er discussion. >> >> >> Solution #0 (it won’t require much modifications): >> (a) If ‘rename’ fails, do not write into the cache directly (we don’t want to have non-atomic writes!). >> (b) If ‘rename’ fails, try to read from the cache. >> auto ReloadedBufferOrErr = CacheEntry.tryLoadingBuffer(); >> (c) If reading from the cache fails simply use the file that you just compiled directly (bypassing the cache entirely). >> >> This solution might cause cache misses (causing recompilations), but at least it should prevent data races when two...

...see with the two solutions below. Hopefully, it will trigger a further discussion. Solution #0 (it won’t require much modifications): (a) If ‘rename’ fails, do not write into the cache directly (we don’t want to have non-atomic writes!). (b) If ‘rename’ fails, try to read from the cache. auto ReloadedBufferOrErr = CacheEntry.tryLoadingBuffer(); (c) If reading from the cache fails simply use the file that you just compiled directly (bypassing the cache entirely). This solution might cause cache misses (causing recompilations), but at least it should prevent data races when two ThinLTO processes write to...