On 24.06.2008 04:17, Peter Stuge wrote:
Please see attached fr.cacheprobe_goto.patch.
For those who fiercly advocate against goto; please compare fr.cacheprobe_goto.patch (which I propose to commit) with the mess in fr.cacheprobe_if.patch (which is not signed off and only included for reference) - goto ftw!
Timing:
flashrom r3386: 1.099s flashrom r3387 (10ms patch): 1.880s flashrom r3387 + this patch: 0.882s
May be considered petty to optimize away this one second, but it is quite noticeable, and not neccessary.
flashrom: Cache probed id:s
Best case it is merely pointless to repeat a probe sequence. Worst case it causes system instability, as we saw with the AMIC A49LF040A.
This problem is not fixed by this patch.
This also shortens flashrom's execution time roughly one second.
Signed-off-by: Peter Stuge peter@stuge.se
NACK. This patch randomly (depending on flashchips.c order) breaks probing for ~80% of the chips we currently support. If it works for you. you are just lucky.
Regards, Carl-Daniel