On Thu, Jan 09, 2025 at 12:23:01AM -0800, Christoph Hellwig wrote: > On Thu, Jan 09, 2025 at 09:12:03AM +0100, Greg Kroah-Hartman wrote: > > > Hey, when I duplicated the method to convert sysfs over to a proper > > > seq_file based approach that avoids buffer overflows you basically > > > came up with the same line that Alexei had here. > > > > I did? Sorry about that, I don't remember that. > > It's been a while.. > > > As for seq_file for sysfs, is that for binary attributes only, or for > > all? I can't recall that at all. > > Non-binary ones. Ah, yeah, well the churn for "one single value" sysfs files would be rough and seq_file doesn't really do much, if anything, for them as they should be all simple strings that never overflow or are complex. Yes, there are exceptions, so maybe for just them? I don't want to make it easier to abuse sysfs files, but if you feel it would really help out, I'm willing to reconsider it. thanks, greg k-h >
