| Message ID | BANLkTinzZeiiOChBDDymhCsuOw57fTP2Ng@mail.gmail.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
On Thu, May 26, 2011 at 11:48 PM, Steve French <smfrench@gmail.com> wrote: > Shirish - below is a patch to fix a sparse warning in the cifsacl > code. The change around line 458 is cosmetic (doesn't look like the > original code caused a problem but at least this quiets sparse) but > doesn't sid_authusers have an endian error. How far back does this > bug go - does it affect much? Yes, sid_authusers have an endian error. > > diff --git a/fs/cifs/cifsacl.c b/fs/cifs/cifsacl.c > index 076b69c..5f02b4e 100644 > --- a/fs/cifs/cifsacl.c > +++ b/fs/cifs/cifsacl.c > @@ -38,7 +38,7 @@ static const struct cifs_sid sid_everyone = { > 1, 1, {0, 0, 0, 0, 0, 1}, {0} }; > /* security id for Authenticated Users system group */ > static const struct cifs_sid sid_authusers = { > - 1, 1, {0, 0, 0, 0, 0, 5}, {11} }; This variable was addedd in commit 2fbc2f1729e785a7b2faf9d8d60926bb1ff62af0 > + 1, 1, {0, 0, 0, 0, 0, 5}, {cpu_to_le32(11)} }; > /* group users */ > static const struct cifs_sid sid_user = {1, 2 , {0, 0, 0, 0, 0, 5}, {} }; > > @@ -458,7 +458,8 @@ int compare_sids(const struct cifs_sid *ctsid, > const struct cifs_sid *cwsid) > if (num_subauth) { > for (i = 0; i < num_subauth; ++i) { > if (ctsid->sub_auth[i] != cwsid->sub_auth[i]) { > - if (ctsid->sub_auth[i] > cwsid->sub_auth[i]) > + if (le32_to_cpu(ctsid->sub_auth[i]) > > + le32_to_cpu(cwsid->sub_auth[i])) This code has been around for quite some time. > return 1; > else > return -1; > > > -- > Thanks, > > Steve > -- To unsubscribe from this list: send the line "unsubscribe linux-cifs" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
On Thu, 26 May 2011 23:48:45 -0500 Steve French <smfrench@gmail.com> wrote: > Shirish - below is a patch to fix a sparse warning in the cifsacl > code. The change around line 458 is cosmetic (doesn't look like the > original code caused a problem but at least this quiets sparse) but > doesn't sid_authusers have an endian error. How far back does this > bug go - does it affect much? > > diff --git a/fs/cifs/cifsacl.c b/fs/cifs/cifsacl.c > index 076b69c..5f02b4e 100644 > --- a/fs/cifs/cifsacl.c > +++ b/fs/cifs/cifsacl.c > @@ -38,7 +38,7 @@ static const struct cifs_sid sid_everyone = { > 1, 1, {0, 0, 0, 0, 0, 1}, {0} }; > /* security id for Authenticated Users system group */ > static const struct cifs_sid sid_authusers = { > - 1, 1, {0, 0, 0, 0, 0, 5}, {11} }; > + 1, 1, {0, 0, 0, 0, 0, 5}, {cpu_to_le32(11)} }; ^^^^^^^^^ This will break on big-endian arches. You need to use __constant_cpu_to_le32() instead. > /* group users */ > static const struct cifs_sid sid_user = {1, 2 , {0, 0, 0, 0, 0, 5}, {} }; > > @@ -458,7 +458,8 @@ int compare_sids(const struct cifs_sid *ctsid, > const struct cifs_sid *cwsid) > if (num_subauth) { > for (i = 0; i < num_subauth; ++i) { > if (ctsid->sub_auth[i] != cwsid->sub_auth[i]) { > - if (ctsid->sub_auth[i] > cwsid->sub_auth[i]) > + if (le32_to_cpu(ctsid->sub_auth[i]) > > + le32_to_cpu(cwsid->sub_auth[i])) > return 1; > else > return -1; > >
diff --git a/fs/cifs/cifsacl.c b/fs/cifs/cifsacl.c index 076b69c..5f02b4e 100644 --- a/fs/cifs/cifsacl.c +++ b/fs/cifs/cifsacl.c @@ -38,7 +38,7 @@ static const struct cifs_sid sid_everyone = { 1, 1, {0, 0, 0, 0, 0, 1}, {0} }; /* security id for Authenticated Users system group */ static const struct cifs_sid sid_authusers = { - 1, 1, {0, 0, 0, 0, 0, 5}, {11} }; + 1, 1, {0, 0, 0, 0, 0, 5}, {cpu_to_le32(11)} }; /* group users */ static const struct cifs_sid sid_user = {1, 2 , {0, 0, 0, 0, 0, 5}, {} }; @@ -458,7 +458,8 @@ int compare_sids(const struct cifs_sid *ctsid, const struct cifs_sid *cwsid) if (num_subauth) { for (i = 0; i < num_subauth; ++i) { if (ctsid->sub_auth[i] != cwsid->sub_auth[i]) { - if (ctsid->sub_auth[i] > cwsid->sub_auth[i]) + if (le32_to_cpu(ctsid->sub_auth[i]) > + le32_to_cpu(cwsid->sub_auth[i])) return 1; else