> -----Original Message----- > From: Joe Perches [mailto:joe@xxxxxxxxxxx] > Sent: Tuesday, July 9, 2019 9:40 PM > To: Gote, Nitin R <nitin.r.gote@xxxxxxxxx>; corbet@xxxxxxx > Cc: akpm@xxxxxxxxxxxxxxxxxxxx; apw@xxxxxxxxxxxxx; > keescook@xxxxxxxxxxxx; linux-doc@xxxxxxxxxxxxxxx; linux- > kernel@xxxxxxxxxxxxxxx; kernel-hardening@xxxxxxxxxxxxxxxxxx > Subject: Re: [PATCH v4] Added warnings in checkpatch.pl script to : > > On Tue, 2019-07-09 at 21:18 +0530, NitinGote wrote: > > From: Nitin Gote <nitin.r.gote@xxxxxxxxx> > > > > 1. Deprecate strcpy() in favor of strscpy(). > > 2. Deprecate strlcpy() in favor of strscpy(). > > 3. Deprecate strncpy() in favor of strscpy() or strscpy_pad(). > > > > Updated strncpy() section in Documentation/process/deprecated.rst > > to cover strscpy_pad() case. > > Please slow down your patch submission rate for this instance and respond > appropriately to the comments you've been given. Sure, I will explore this things more. And sorry, I missed to incorporate one comment. I will take care of such things. > > This stuff is not critical bug fixing. > Noted. > The subject could be something like: > > Subject: [PATCH v#] Documentation/checkpatch: Prefer strscpy over > strcpy/strlcpy > How about this : Subject: [PATCH v#] Doc/checkpatch: Prefer strscpy/strscpy_pad over strcpy/strlcpy/strncpy > > diff --git a/scripts/checkpatch.pl b/scripts/checkpatch.pl > [] > > @@ -605,6 +605,20 @@ foreach my $entry (keys %deprecated_apis) { } > > $deprecated_apis_search = "(?:${deprecated_apis_search})"; > > > > +our %deprecated_string_apis = ( > > + "strcpy" => "strscpy", > > + "strlcpy" => "strscpy", > > + "strncpy" => "strscpy, strscpy_pad or > for non-NUL-terminated strings, strncpy() can still be used, but destinations > should be marked with the __nonstring", > > 'the' is not necessary. Noted. > > There could likely also be a strscat created for strcat, strlcat and strncat. > I have not found reference for strscat in kernel. Could you please give any reference for strscat ? > btw: > > There were several defects in the kernel for misuses of strlcpy. > > Did you or anyone else have an opinion on stracpy to avoid duplicating the > first argument in a sizeof()? > > strlcpy(foo, bar, sizeof(foo)) > to > stracpy(foo, bar) > > where foo must be char array compatible ? > > https://lore.kernel.org/lkml/d1524130f91d7cfd61bc736623409693d2895f57. > camel@xxxxxxxxxxx/ > > As I understood, your trying to give new interface like stracpy(), to avoid duplication of first argument in a sizeof(), we can also make it more robust for users by adding check or warn in checkpatch.pl to prefer stracpy(). Did you or anyone has opinion on this ? Thanks, Nitin Gote
