On 08/19/2013 06:55 PM, Dave Hansen wrote: > On 08/19/2013 08:17 AM, Jerome Marchand wrote: >> Some applications that run on HPC clusters are designed around the >> availability of RAM and the overcommit ratio is fine tuned to get the >> maximum usage of memory without swapping. With growing memory, the >> 1%-of-all-RAM grain provided by overcommit_ratio has become too coarse >> for these workload (on a 2TB machine it represents no less than >> 20GB). >> >> This patch adds the new overcommit_kbytes sysctl variable that allow a >> much finer grain. > > Instead of introducing yet another tunable, why don't we just make the > ratio that comes in from the user more fine-grained? > > sysctl overcommit_ratio=0.2 > > We change the internal 'sysctl_overcommit_ratio' to store tenths or > hundreths of a percent (or whatever), then parse the input as two > integers. I don't think we need fully correct floating point parsing > and rounding here, so it shouldn't be too much of a chore. It'd > probably end up being less code than you have as it stands. >
Now that I think about it, that could break user space. Sure write access wouldn't be a problem (one can still write a plain integer), but a script that reads a fractional value when it expects an integer might not be able to cope with it. -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
