| Message ID | 20200521152308.33096-1-steven.price@arm.com (mailing list archive) |
|---|---|
| Headers | show |
| Series | Fix W+X debug feature on x86 | expand |
On Thu, 21 May 2020 16:23:06 +0100 Steven Price <steven.price@arm.com> wrote: > Jan alert me[1] that the W+X detection debug feature was broken in x86 > by my change[2] to switch x86 to use the generic ptdump infrastructure. > > Fundamentally the approach of trying to move the calculation of > effective permissions into note_page() was broken because note_page() is > only called for 'leaf' entries and the effective permissions are passed > down via the internal nodes of the page tree. The solution I've taken > here is to create a new (optional) callback which is called for all > nodes of the page tree and therefore can calculate the effective > permissions. > > Secondly on some configurations (32 bit with PAE) "unsigned long" is not > large enough to store the table entries. The fix here is simple - let's > just use a u64. I assumed that a cc:stable was appropriate on both of these(?). > I'd welcome testing (and other comments), especially if you have a > configuration which previously triggered W+X warnings as I don't have > such a setup. I'll wait a while for such testing. If nothing happens then I guess we merge it up and see what then happens.
On 21/05/2020 20:08, Andrew Morton wrote: > On Thu, 21 May 2020 16:23:06 +0100 Steven Price <steven.price@arm.com> wrote: > >> Jan alert me[1] that the W+X detection debug feature was broken in x86 >> by my change[2] to switch x86 to use the generic ptdump infrastructure. >> >> Fundamentally the approach of trying to move the calculation of >> effective permissions into note_page() was broken because note_page() is >> only called for 'leaf' entries and the effective permissions are passed >> down via the internal nodes of the page tree. The solution I've taken >> here is to create a new (optional) callback which is called for all >> nodes of the page tree and therefore can calculate the effective >> permissions. >> >> Secondly on some configurations (32 bit with PAE) "unsigned long" is not >> large enough to store the table entries. The fix here is simple - let's >> just use a u64. > > I assumed that a cc:stable was appropriate on both of these(?). Yes thanks. >> I'd welcome testing (and other comments), especially if you have a >> configuration which previously triggered W+X warnings as I don't have >> such a setup. > > I'll wait a while for such testing. If nothing happens then I guess we > merge it up and see what then happens. > Thanks, Steve
Jan alert me[1] that the W+X detection debug feature was broken in x86 by my change[2] to switch x86 to use the generic ptdump infrastructure. Fundamentally the approach of trying to move the calculation of effective permissions into note_page() was broken because note_page() is only called for 'leaf' entries and the effective permissions are passed down via the internal nodes of the page tree. The solution I've taken here is to create a new (optional) callback which is called for all nodes of the page tree and therefore can calculate the effective permissions. Secondly on some configurations (32 bit with PAE) "unsigned long" is not large enough to store the table entries. The fix here is simple - let's just use a u64. I'd welcome testing (and other comments), especially if you have a configuration which previously triggered W+X warnings as I don't have such a setup. [1] https://lore.kernel.org/lkml/d573dc7e-e742-84de-473d-f971142fa319@suse.com/ [2] 2ae27137b2db ("x86: mm: convert dump_pagetables to use walk_page_range") Steven Price (2): x86: mm: ptdump: Calculate effective permissions correctly mm: ptdump: Expand type of 'val' in note_page() arch/arm64/mm/dump.c | 2 +- arch/x86/mm/dump_pagetables.c | 33 ++++++++++++++++++++------------- include/linux/ptdump.h | 3 ++- mm/ptdump.c | 17 ++++++++++++++++- 4 files changed, 39 insertions(+), 16 deletions(-)