Re: [PATCH v3 3/7] x86/xen: drop tests for highmem in pv code
From: kernel test robot
Date: Fri Aug 07 2020 - 08:13:49 EST
Hi Juergen,
I love your patch! Perhaps something to improve:
[auto build test WARNING on tip/x86/core]
[also build test WARNING on tip/x86/asm v5.8 next-20200806]
[cannot apply to xen-tip/linux-next tip/x86/vdso]
[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]
url: https://github.com/0day-ci/linux/commits/Juergen-Gross/Remove-32-bit-Xen-PV-guest-support/20200807-164058
base: https://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git ef2ff0f5d6008d325c9a068e20981c0d0acc4d6b
config: x86_64-allyesconfig (attached as .config)
compiler: gcc-9 (Debian 9.3.0-15) 9.3.0
reproduce (this is a W=1 build):
# save the attached .config to linux build tree
make W=1 ARCH=x86_64
If you fix the issue, kindly add following tag as appropriate
Reported-by: kernel test robot <lkp@xxxxxxxxx>
All warnings (new ones prefixed by >>):
arch/x86/xen/enlighten_pv.c: In function 'set_aliased_prot':
>> arch/x86/xen/enlighten_pv.c:348:15: warning: variable 'page' set but not used [-Wunused-but-set-variable]
348 | struct page *page;
| ^~~~
arch/x86/xen/enlighten_pv.c: At top level:
arch/x86/xen/enlighten_pv.c:1198:34: warning: no previous prototype for 'xen_start_kernel' [-Wmissing-prototypes]
1198 | asmlinkage __visible void __init xen_start_kernel(void)
| ^~~~~~~~~~~~~~~~
vim +/page +348 arch/x86/xen/enlighten_pv.c
e1dab14cf68d1e0 Vitaly Kuznetsov 2017-03-14 335
e1dab14cf68d1e0 Vitaly Kuznetsov 2017-03-14 336 /*
e1dab14cf68d1e0 Vitaly Kuznetsov 2017-03-14 337 * Set the page permissions for a particular virtual address. If the
e1dab14cf68d1e0 Vitaly Kuznetsov 2017-03-14 338 * address is a vmalloc mapping (or other non-linear mapping), then
e1dab14cf68d1e0 Vitaly Kuznetsov 2017-03-14 339 * find the linear mapping of the page and also set its protections to
e1dab14cf68d1e0 Vitaly Kuznetsov 2017-03-14 340 * match.
e1dab14cf68d1e0 Vitaly Kuznetsov 2017-03-14 341 */
e1dab14cf68d1e0 Vitaly Kuznetsov 2017-03-14 342 static void set_aliased_prot(void *v, pgprot_t prot)
e1dab14cf68d1e0 Vitaly Kuznetsov 2017-03-14 343 {
e1dab14cf68d1e0 Vitaly Kuznetsov 2017-03-14 344 int level;
e1dab14cf68d1e0 Vitaly Kuznetsov 2017-03-14 345 pte_t *ptep;
e1dab14cf68d1e0 Vitaly Kuznetsov 2017-03-14 346 pte_t pte;
e1dab14cf68d1e0 Vitaly Kuznetsov 2017-03-14 347 unsigned long pfn;
e1dab14cf68d1e0 Vitaly Kuznetsov 2017-03-14 @348 struct page *page;
e1dab14cf68d1e0 Vitaly Kuznetsov 2017-03-14 349 unsigned char dummy;
64aef3716eab524 Juergen Gross 2020-08-07 350 void *av;
e1dab14cf68d1e0 Vitaly Kuznetsov 2017-03-14 351
e1dab14cf68d1e0 Vitaly Kuznetsov 2017-03-14 352 ptep = lookup_address((unsigned long)v, &level);
e1dab14cf68d1e0 Vitaly Kuznetsov 2017-03-14 353 BUG_ON(ptep == NULL);
e1dab14cf68d1e0 Vitaly Kuznetsov 2017-03-14 354
e1dab14cf68d1e0 Vitaly Kuznetsov 2017-03-14 355 pfn = pte_pfn(*ptep);
e1dab14cf68d1e0 Vitaly Kuznetsov 2017-03-14 356 page = pfn_to_page(pfn);
e1dab14cf68d1e0 Vitaly Kuznetsov 2017-03-14 357
e1dab14cf68d1e0 Vitaly Kuznetsov 2017-03-14 358 pte = pfn_pte(pfn, prot);
e1dab14cf68d1e0 Vitaly Kuznetsov 2017-03-14 359
e1dab14cf68d1e0 Vitaly Kuznetsov 2017-03-14 360 /*
e1dab14cf68d1e0 Vitaly Kuznetsov 2017-03-14 361 * Careful: update_va_mapping() will fail if the virtual address
e1dab14cf68d1e0 Vitaly Kuznetsov 2017-03-14 362 * we're poking isn't populated in the page tables. We don't
e1dab14cf68d1e0 Vitaly Kuznetsov 2017-03-14 363 * need to worry about the direct map (that's always in the page
e1dab14cf68d1e0 Vitaly Kuznetsov 2017-03-14 364 * tables), but we need to be careful about vmap space. In
e1dab14cf68d1e0 Vitaly Kuznetsov 2017-03-14 365 * particular, the top level page table can lazily propagate
e1dab14cf68d1e0 Vitaly Kuznetsov 2017-03-14 366 * entries between processes, so if we've switched mms since we
e1dab14cf68d1e0 Vitaly Kuznetsov 2017-03-14 367 * vmapped the target in the first place, we might not have the
e1dab14cf68d1e0 Vitaly Kuznetsov 2017-03-14 368 * top-level page table entry populated.
e1dab14cf68d1e0 Vitaly Kuznetsov 2017-03-14 369 *
e1dab14cf68d1e0 Vitaly Kuznetsov 2017-03-14 370 * We disable preemption because we want the same mm active when
e1dab14cf68d1e0 Vitaly Kuznetsov 2017-03-14 371 * we probe the target and when we issue the hypercall. We'll
e1dab14cf68d1e0 Vitaly Kuznetsov 2017-03-14 372 * have the same nominal mm, but if we're a kernel thread, lazy
e1dab14cf68d1e0 Vitaly Kuznetsov 2017-03-14 373 * mm dropping could change our pgd.
e1dab14cf68d1e0 Vitaly Kuznetsov 2017-03-14 374 *
e1dab14cf68d1e0 Vitaly Kuznetsov 2017-03-14 375 * Out of an abundance of caution, this uses __get_user() to fault
e1dab14cf68d1e0 Vitaly Kuznetsov 2017-03-14 376 * in the target address just in case there's some obscure case
e1dab14cf68d1e0 Vitaly Kuznetsov 2017-03-14 377 * in which the target address isn't readable.
e1dab14cf68d1e0 Vitaly Kuznetsov 2017-03-14 378 */
e1dab14cf68d1e0 Vitaly Kuznetsov 2017-03-14 379
e1dab14cf68d1e0 Vitaly Kuznetsov 2017-03-14 380 preempt_disable();
e1dab14cf68d1e0 Vitaly Kuznetsov 2017-03-14 381
fe557319aa06c23 Christoph Hellwig 2020-06-17 382 copy_from_kernel_nofault(&dummy, v, 1);
e1dab14cf68d1e0 Vitaly Kuznetsov 2017-03-14 383
e1dab14cf68d1e0 Vitaly Kuznetsov 2017-03-14 384 if (HYPERVISOR_update_va_mapping((unsigned long)v, pte, 0))
e1dab14cf68d1e0 Vitaly Kuznetsov 2017-03-14 385 BUG();
e1dab14cf68d1e0 Vitaly Kuznetsov 2017-03-14 386
64aef3716eab524 Juergen Gross 2020-08-07 387 av = __va(PFN_PHYS(pfn));
e1dab14cf68d1e0 Vitaly Kuznetsov 2017-03-14 388
64aef3716eab524 Juergen Gross 2020-08-07 389 if (av != v && HYPERVISOR_update_va_mapping((unsigned long)av, pte, 0))
e1dab14cf68d1e0 Vitaly Kuznetsov 2017-03-14 390 BUG();
e1dab14cf68d1e0 Vitaly Kuznetsov 2017-03-14 391
e1dab14cf68d1e0 Vitaly Kuznetsov 2017-03-14 392 preempt_enable();
e1dab14cf68d1e0 Vitaly Kuznetsov 2017-03-14 393 }
e1dab14cf68d1e0 Vitaly Kuznetsov 2017-03-14 394
---
0-DAY CI Kernel Test Service, Intel Corporation
https://lists.01.org/hyperkitty/list/kbuild-all@xxxxxxxxxxxx
Attachment:
.config.gz
Description: application/gzip