Message ID | 20200826162413.21637-1-andriy.shevchenko@linux.intel.com (mailing list archive) |
---|---|
State | Superseded |
Headers | show |
Series | [v1] USB: gadget: u_f: Unbreak offset calculation in VLAs | expand |
On Wed, Aug 26, 2020 at 07:24:13PM +0300, Andy Shevchenko wrote: > Inadvertently the commit b1cd1b65afba ("USB: gadget: u_f: add overflow checks > to VLA macros") makes VLA macros to always return 0 due to different scope of > two variables of the same name. Obviously we need to have only one. > > Fixes: b1cd1b65afba ("USB: gadget: u_f: add overflow checks to VLA macros") > Cc: stable <stable@kernel.org> > Reported-by: Marek Szyprowski <m.szyprowski@samsung.com> > Tested-by: Marek Szyprowski <m.szyprowski@samsung.com> > Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> > --- > drivers/usb/gadget/u_f.h | 8 ++++---- > 1 file changed, 4 insertions(+), 4 deletions(-) > > diff --git a/drivers/usb/gadget/u_f.h b/drivers/usb/gadget/u_f.h > index df4e1dcb357d..0b5c5f0dd073 100644 > --- a/drivers/usb/gadget/u_f.h > +++ b/drivers/usb/gadget/u_f.h > @@ -25,8 +25,8 @@ > size_t offset = 0; \ > if (groupname##__next != SIZE_MAX) { \ > size_t align_mask = __alignof__(type) - 1; \ > - size_t offset = (groupname##__next + align_mask) \ > - & ~align_mask; \ > + offset = (groupname##__next + align_mask) & \ > + ~align_mask; \ > size_t size = array_size(n, sizeof(type)); \ > if (check_add_overflow(offset, size, \ > &groupname##__next)) { \ > @@ -43,8 +43,8 @@ > size_t offset = 0; \ > if (groupname##__next != SIZE_MAX) { \ > size_t align_mask = __alignof__(type) - 1; \ > - size_t offset = (groupname##__next + align_mask) \ > - & ~align_mask; \ > + offset = (groupname##__next + align_mask) & \ > + ~align_mask; \ > if (check_add_overflow(offset, groupname##_##name##__sz,\ > &groupname##__next)) { \ > groupname##__next = SIZE_MAX; \ This patch, while maybe fixing the issue, has a bunch of build warnings so I can't take it as-is: $ make drivers/usb/gadget/ CALL scripts/checksyscalls.sh CALL scripts/atomic/check-atomics.sh DESCEND objtool CC [M] drivers/usb/gadget/configfs.o In file included from drivers/usb/gadget/configfs.c:10: drivers/usb/gadget/configfs.c: In function ‘ext_prop_make’: drivers/usb/gadget/u_f.h:30:4: warning: ISO C90 forbids mixed declarations and code [-Wdeclaration-after-statement] 30 | size_t size = array_size(n, sizeof(type)); \ | ^~~~~~ drivers/usb/gadget/configfs.c:1066:2: note: in expansion of macro ‘vla_item’ 1066 | vla_item(data_chunk, struct usb_os_desc_ext_prop, ext_prop, 1); | ^~~~~~~~ drivers/usb/gadget/u_f.h:30:4: warning: ISO C90 forbids mixed declarations and code [-Wdeclaration-after-statement] 30 | size_t size = array_size(n, sizeof(type)); \ | ^~~~~~ drivers/usb/gadget/configfs.c:1067:2: note: in expansion of macro ‘vla_item’ 1067 | vla_item(data_chunk, struct config_item_type, ext_prop_type, 1); | ^~~~~~~~ drivers/usb/gadget/configfs.c: In function ‘usb_os_desc_prepare_interf_dir’: drivers/usb/gadget/u_f.h:30:4: warning: ISO C90 forbids mixed declarations and code [-Wdeclaration-after-statement] 30 | size_t size = array_size(n, sizeof(type)); \ | ^~~~~~ drivers/usb/gadget/configfs.c:1194:2: note: in expansion of macro ‘vla_item’ 1194 | vla_item(data_chunk, struct config_group, os_desc_group, 1); | ^~~~~~~~ drivers/usb/gadget/u_f.h:30:4: warning: ISO C90 forbids mixed declarations and code [-Wdeclaration-after-statement] 30 | size_t size = array_size(n, sizeof(type)); \ | ^~~~~~ drivers/usb/gadget/configfs.c:1195:2: note: in expansion of macro ‘vla_item’ 1195 | vla_item(data_chunk, struct config_item_type, os_desc_type, 1); | ^~~~~~~~ drivers/usb/gadget/u_f.h:30:4: warning: ISO C90 forbids mixed declarations and code [-Wdeclaration-after-statement] 30 | size_t size = array_size(n, sizeof(type)); \ | ^~~~~~ drivers/usb/gadget/configfs.c:1196:2: note: in expansion of macro ‘vla_item’ 1196 | vla_item(data_chunk, struct config_item_type, interface_type, 1); | ^~~~~~~~ Care to try it again? thanks, greg k-h
On 26.08.2020 18:28, Greg Kroah-Hartman wrote: > On Wed, Aug 26, 2020 at 07:24:13PM +0300, Andy Shevchenko wrote: >> Inadvertently the commit b1cd1b65afba ("USB: gadget: u_f: add overflow checks >> to VLA macros") makes VLA macros to always return 0 due to different scope of >> two variables of the same name. Obviously we need to have only one. >> >> Fixes: b1cd1b65afba ("USB: gadget: u_f: add overflow checks to VLA macros") >> Cc: stable <stable@kernel.org> >> Reported-by: Marek Szyprowski <m.szyprowski@samsung.com> >> Tested-by: Marek Szyprowski <m.szyprowski@samsung.com> >> Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> >> --- >> drivers/usb/gadget/u_f.h | 8 ++++---- >> 1 file changed, 4 insertions(+), 4 deletions(-) >> >> diff --git a/drivers/usb/gadget/u_f.h b/drivers/usb/gadget/u_f.h >> index df4e1dcb357d..0b5c5f0dd073 100644 >> --- a/drivers/usb/gadget/u_f.h >> +++ b/drivers/usb/gadget/u_f.h >> @@ -25,8 +25,8 @@ >> size_t offset = 0; \ >> if (groupname##__next != SIZE_MAX) { \ >> size_t align_mask = __alignof__(type) - 1; \ >> - size_t offset = (groupname##__next + align_mask) \ >> - & ~align_mask; \ >> + offset = (groupname##__next + align_mask) & \ >> + ~align_mask; \ >> size_t size = array_size(n, sizeof(type)); \ >> if (check_add_overflow(offset, size, \ >> &groupname##__next)) { \ >> @@ -43,8 +43,8 @@ >> size_t offset = 0; \ >> if (groupname##__next != SIZE_MAX) { \ >> size_t align_mask = __alignof__(type) - 1; \ >> - size_t offset = (groupname##__next + align_mask) \ >> - & ~align_mask; \ >> + offset = (groupname##__next + align_mask) & \ >> + ~align_mask; \ >> if (check_add_overflow(offset, groupname##_##name##__sz,\ >> &groupname##__next)) { \ >> groupname##__next = SIZE_MAX; \ > This patch, while maybe fixing the issue, has a bunch of build warnings > so I can't take it as-is: That's why I said that I've moved offset assignment in the first part of the patch after size_t size = array_size(n, sizeof(type)); Best regards
On Wed, Aug 26, 2020 at 06:31:01PM +0200, Marek Szyprowski wrote: > On 26.08.2020 18:28, Greg Kroah-Hartman wrote: > > On Wed, Aug 26, 2020 at 07:24:13PM +0300, Andy Shevchenko wrote: > That's why I said that I've moved offset assignment in the first part of > the patch after size_t size = array_size(n, sizeof(type)); v2 has been sent.
diff --git a/drivers/usb/gadget/u_f.h b/drivers/usb/gadget/u_f.h index df4e1dcb357d..0b5c5f0dd073 100644 --- a/drivers/usb/gadget/u_f.h +++ b/drivers/usb/gadget/u_f.h @@ -25,8 +25,8 @@ size_t offset = 0; \ if (groupname##__next != SIZE_MAX) { \ size_t align_mask = __alignof__(type) - 1; \ - size_t offset = (groupname##__next + align_mask) \ - & ~align_mask; \ + offset = (groupname##__next + align_mask) & \ + ~align_mask; \ size_t size = array_size(n, sizeof(type)); \ if (check_add_overflow(offset, size, \ &groupname##__next)) { \ @@ -43,8 +43,8 @@ size_t offset = 0; \ if (groupname##__next != SIZE_MAX) { \ size_t align_mask = __alignof__(type) - 1; \ - size_t offset = (groupname##__next + align_mask) \ - & ~align_mask; \ + offset = (groupname##__next + align_mask) & \ + ~align_mask; \ if (check_add_overflow(offset, groupname##_##name##__sz,\ &groupname##__next)) { \ groupname##__next = SIZE_MAX; \