* Re: [PATCH -fixes] riscv: Fix wrong usage of lm_alias() when splitting a huge linear mapping
2023-12-11 14:19 [PATCH -fixes] riscv: Fix wrong usage of lm_alias() when splitting a huge linear mapping Alexandre Ghiti
@ 2023-12-11 22:51 ` Conor Dooley
2023-12-12 0:48 ` kernel test robot
1 sibling, 0 replies; 3+ messages in thread
From: Conor Dooley @ 2023-12-11 22:51 UTC (permalink / raw)
To: Alexandre Ghiti
Cc: Paul Walmsley, Palmer Dabbelt, Albert Ou, linux-riscv,
linux-kernel, syzbot+afb726d49f84c8d95ee1
[-- Attachment #1.1: Type: text/plain, Size: 1866 bytes --]
On Mon, Dec 11, 2023 at 03:19:29PM +0100, Alexandre Ghiti wrote:
> lm_alias() can only be used on kernel mappings since it explicitly uses
> __pa_symbol(), so simply fix this by checking where the address belongs
> to before.
>
> Fixes: 311cd2f6e253 ("riscv: Fix set_memory_XX() and set_direct_map_XX() by splitting huge linear mappings")
> Reported-by: syzbot+afb726d49f84c8d95ee1@syzkaller.appspotmail.com
> Closes: https://lore.kernel.org/linux-riscv/000000000000620dd0060c02c5e1@google.com/
> Signed-off-by: Alexandre Ghiti <alexghiti@rivosinc.com>
../arch/riscv/mm/pageattr.c:311:14: warning: variable 'ret' is used uninitialized whenever '&&' condition is false [-Wsometimes-uninitialized]
> ---
> arch/riscv/mm/pageattr.c | 13 ++++++++++---
> 1 file changed, 10 insertions(+), 3 deletions(-)
>
> diff --git a/arch/riscv/mm/pageattr.c b/arch/riscv/mm/pageattr.c
> index fc5fc4f785c4..f5f8aa1d38d6 100644
> --- a/arch/riscv/mm/pageattr.c
> +++ b/arch/riscv/mm/pageattr.c
> @@ -304,9 +304,16 @@ static int __set_memory(unsigned long addr, int numpages, pgprot_t set_mask,
> if (ret)
> goto unlock;
> }
> - } else if (is_kernel_mapping(start) || is_linear_mapping(start)) {
> - lm_start = (unsigned long)lm_alias(start);
> - lm_end = (unsigned long)lm_alias(end);
> + } else {
> + if (is_kernel_mapping(start)) {
> + lm_start = (unsigned long)lm_alias(start);
> + lm_end = (unsigned long)lm_alias(end);
> + } else if (is_linear_mapping(start)) {
> + lm_start = start;
> + lm_end = end;
> + } else {
> + goto unlock;
> + }
>
> ret = split_linear_mapping(lm_start, lm_end);
> if (ret)
> --
> 2.39.2
>
>
> _______________________________________________
> linux-riscv mailing list
> linux-riscv@lists.infradead.org
> http://lists.infradead.org/mailman/listinfo/linux-riscv
[-- Attachment #1.2: signature.asc --]
[-- Type: application/pgp-signature, Size: 228 bytes --]
[-- Attachment #2: Type: text/plain, Size: 161 bytes --]
_______________________________________________
linux-riscv mailing list
linux-riscv@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-riscv
^ permalink raw reply [flat|nested] 3+ messages in thread* Re: [PATCH -fixes] riscv: Fix wrong usage of lm_alias() when splitting a huge linear mapping
2023-12-11 14:19 [PATCH -fixes] riscv: Fix wrong usage of lm_alias() when splitting a huge linear mapping Alexandre Ghiti
2023-12-11 22:51 ` Conor Dooley
@ 2023-12-12 0:48 ` kernel test robot
1 sibling, 0 replies; 3+ messages in thread
From: kernel test robot @ 2023-12-12 0:48 UTC (permalink / raw)
To: Alexandre Ghiti, Paul Walmsley, Palmer Dabbelt, Albert Ou,
linux-riscv, linux-kernel
Cc: llvm, oe-kbuild-all, syzbot+afb726d49f84c8d95ee1
Hi Alexandre,
kernel test robot noticed the following build warnings:
[auto build test WARNING on linus/master]
[also build test WARNING on v6.7-rc5 next-20231211]
[If your patch is applied to the wrong git tree, kindly drop us a note.
And when submitting patch, we suggest to use '--base' as documented in
https://git-scm.com/docs/git-format-patch#_base_tree_information]
url: https://github.com/intel-lab-lkp/linux/commits/Alexandre-Ghiti/riscv-Fix-wrong-usage-of-lm_alias-when-splitting-a-huge-linear-mapping/20231211-222156
base: linus/master
patch link: https://lore.kernel.org/r/20231211141929.74027-1-alexghiti%40rivosinc.com
patch subject: [PATCH -fixes] riscv: Fix wrong usage of lm_alias() when splitting a huge linear mapping
config: riscv-alldefconfig (https://download.01.org/0day-ci/archive/20231212/202312120825.UkTJCa1g-lkp@intel.com/config)
compiler: clang version 17.0.0 (https://github.com/llvm/llvm-project.git 4a5ac14ee968ff0ad5d2cc1ffa0299048db4c88a)
reproduce (this is a W=1 build): (https://download.01.org/0day-ci/archive/20231212/202312120825.UkTJCa1g-lkp@intel.com/reproduce)
If you fix the issue in a separate patch/commit (i.e. not just a new version of
the same patch/commit), kindly add following tags
| Reported-by: kernel test robot <lkp@intel.com>
| Closes: https://lore.kernel.org/oe-kbuild-all/202312120825.UkTJCa1g-lkp@intel.com/
All warnings (new ones prefixed by >>):
>> arch/riscv/mm/pageattr.c:311:14: warning: variable 'ret' is used uninitialized whenever 'if' condition is false [-Wsometimes-uninitialized]
311 | } else if (is_linear_mapping(start)) {
| ^~~~~~~~~~~~~~~~~~~~~~~~
arch/riscv/include/asm/page.h:130:2: note: expanded from macro 'is_linear_mapping'
130 | ((x) >= PAGE_OFFSET && (!IS_ENABLED(CONFIG_64BIT) || (x) < PAGE_OFFSET + KERN_VIRT_SIZE))
| ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
arch/riscv/mm/pageattr.c:348:9: note: uninitialized use occurs here
348 | return ret;
| ^~~
arch/riscv/mm/pageattr.c:311:10: note: remove the 'if' if its condition is always true
311 | } else if (is_linear_mapping(start)) {
| ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
312 | lm_start = start;
313 | lm_end = end;
314 | } else {
| ~~~~~~~
315 | goto unlock;
| ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
316 | }
| ~~~~~~~~~~~~~~~~~
>> arch/riscv/mm/pageattr.c:311:14: warning: variable 'ret' is used uninitialized whenever '&&' condition is false [-Wsometimes-uninitialized]
311 | } else if (is_linear_mapping(start)) {
| ^~~~~~~~~~~~~~~~~~~~~~~~
arch/riscv/include/asm/page.h:130:3: note: expanded from macro 'is_linear_mapping'
130 | ((x) >= PAGE_OFFSET && (!IS_ENABLED(CONFIG_64BIT) || (x) < PAGE_OFFSET + KERN_VIRT_SIZE))
| ^~~~~~~~~~~~~~~~~~
arch/riscv/mm/pageattr.c:348:9: note: uninitialized use occurs here
348 | return ret;
| ^~~
arch/riscv/mm/pageattr.c:311:14: note: remove the '&&' if its condition is always true
311 | } else if (is_linear_mapping(start)) {
| ^
arch/riscv/include/asm/page.h:130:3: note: expanded from macro 'is_linear_mapping'
130 | ((x) >= PAGE_OFFSET && (!IS_ENABLED(CONFIG_64BIT) || (x) < PAGE_OFFSET + KERN_VIRT_SIZE))
| ^
arch/riscv/mm/pageattr.c:265:9: note: initialize the variable 'ret' to silence this warning
265 | int ret;
| ^
| = 0
2 warnings generated.
vim +311 arch/riscv/mm/pageattr.c
261
262 static int __set_memory(unsigned long addr, int numpages, pgprot_t set_mask,
263 pgprot_t clear_mask)
264 {
265 int ret;
266 unsigned long start = addr;
267 unsigned long end = start + PAGE_SIZE * numpages;
268 unsigned long __maybe_unused lm_start;
269 unsigned long __maybe_unused lm_end;
270 struct pageattr_masks masks = {
271 .set_mask = set_mask,
272 .clear_mask = clear_mask
273 };
274
275 if (!numpages)
276 return 0;
277
278 mmap_write_lock(&init_mm);
279
280 #ifdef CONFIG_64BIT
281 /*
282 * We are about to change the permissions of a kernel mapping, we must
283 * apply the same changes to its linear mapping alias, which may imply
284 * splitting a huge mapping.
285 */
286
287 if (is_vmalloc_or_module_addr((void *)start)) {
288 struct vm_struct *area = NULL;
289 int i, page_start;
290
291 area = find_vm_area((void *)start);
292 page_start = (start - (unsigned long)area->addr) >> PAGE_SHIFT;
293
294 for (i = page_start; i < page_start + numpages; ++i) {
295 lm_start = (unsigned long)page_address(area->pages[i]);
296 lm_end = lm_start + PAGE_SIZE;
297
298 ret = split_linear_mapping(lm_start, lm_end);
299 if (ret)
300 goto unlock;
301
302 ret = walk_page_range_novma(&init_mm, lm_start, lm_end,
303 &pageattr_ops, NULL, &masks);
304 if (ret)
305 goto unlock;
306 }
307 } else {
308 if (is_kernel_mapping(start)) {
309 lm_start = (unsigned long)lm_alias(start);
310 lm_end = (unsigned long)lm_alias(end);
> 311 } else if (is_linear_mapping(start)) {
312 lm_start = start;
313 lm_end = end;
314 } else {
315 goto unlock;
316 }
317
318 ret = split_linear_mapping(lm_start, lm_end);
319 if (ret)
320 goto unlock;
321
322 ret = walk_page_range_novma(&init_mm, lm_start, lm_end,
323 &pageattr_ops, NULL, &masks);
324 if (ret)
325 goto unlock;
326 }
327
328 ret = walk_page_range_novma(&init_mm, start, end, &pageattr_ops, NULL,
329 &masks);
330
331 unlock:
332 mmap_write_unlock(&init_mm);
333
334 /*
335 * We can't use flush_tlb_kernel_range() here as we may have split a
336 * hugepage that is larger than that, so let's flush everything.
337 */
338 flush_tlb_all();
339 #else
340 ret = walk_page_range_novma(&init_mm, start, end, &pageattr_ops, NULL,
341 &masks);
342
343 mmap_write_unlock(&init_mm);
344
345 flush_tlb_kernel_range(start, end);
346 #endif
347
348 return ret;
349 }
350
--
0-DAY CI Kernel Test Service
https://github.com/intel/lkp-tests/wiki
_______________________________________________
linux-riscv mailing list
linux-riscv@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-riscv
^ permalink raw reply [flat|nested] 3+ messages in thread