Thomas Glanzmann wrote: [Mon Mar 13 2006, 01:54:55PM EST] > Hello Aron, Hi Thomas, > short answer: yes, it does. Long answer on debian sarge with > self-compiled qdbm libs: Okay, thanks for verifying this. My build and testing process has been very similar to yours, with the exception that I've been testing with gdbm instead of db4. Just to make sure, I tested with db4 and found it makes no difference for me. > # MAILDIR: > sithglan@faui02:/var/tmp/sithglan/mutt$ time ./mutt -nF~/null -f > ~/Maildir/.lists.uml -e 'exec exit' > > real 2m59.359s > user 0m2.220s > sys 0m5.354s > sithglan@faui02:/var/tmp/sithglan/mutt$ time ./mutt -nF~/null -f > ~/Maildir/.lists.uml -e 'exec exit' > > real 0m1.039s > user 0m0.579s > sys 0m0.395s Thanks for the idea to test with maildir. Indeed header caching works fine for me with maildir. I used a similarly spare configuration file. I did 3 runs: (1) warm OS fs cache, then removed the mutt header cache; (2) populate mutt header cache and take a time measurement; (3) third to test with populated header cache. The second run took 5 seconds, third run took 0.5 seconds, clearly using the cache. > # IMAP: > > sithglan@faui02:/var/tmp/sithglan/mutt$ time ./mutt -nF~/null -f > imaps://glanzman@xxxxxxxxxxxxxxxxx/ -e 'exec exit' > > real 0m24.834s > user 0m0.069s > sys 0m0.038s > sithglan@faui02:/var/tmp/sithglan/mutt$ !! > time ./mutt -nF~/null -f imaps://glanzman@xxxxxxxxxxxxxxxxx/ -e 'exec > exit' > > real 0m5.490s > user 0m0.065s > sys 0m0.026s Running the same test using another IMAP server shows it doesn't work here. Thank you for the testing ideas. Regards, Aron
Attachment:
pgpTeabGPrv7X.pgp
Description: PGP signature