From: kernel test robot <lkp@intel.com>
To: "Matthew Wilcox (Oracle)" <willy@infradead.org>, linux-mm@kvack.org
Cc: llvm@lists.linux.dev, kbuild-all@lists.01.org,
"Matthew Wilcox (Oracle)" <willy@infradead.org>,
Muchun Song <songmuchun@bytedance.com>
Subject: Re: [PATCH 2/2] mm: Convert page_vma_mapped_walk to work on PFNs
Date: Fri, 4 Feb 2022 09:30:24 +0800 [thread overview]
Message-ID: <202202040601.7rVVBoVe-lkp@intel.com> (raw)
In-Reply-To: <20220203171904.609984-2-willy@infradead.org>
Hi "Matthew,
Thank you for the patch! Yet something to improve:
[auto build test ERROR on linus/master]
[also build test ERROR on tip/perf/core v5.17-rc2 next-20220203]
[cannot apply to hnaz-mm/master]
[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/Matthew-Wilcox-Oracle/mm-Add-pvmw_set_page/20220204-012111
base: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git c36c04c2e132fc39f6b658bf607aed4425427fd7
config: hexagon-randconfig-r045-20220130 (https://download.01.org/0day-ci/archive/20220204/202202040601.7rVVBoVe-lkp@intel.com/config)
compiler: clang version 15.0.0 (https://github.com/llvm/llvm-project a73e4ce6a59b01f0e37037761c1e6889d539d233)
reproduce (this is a W=1 build):
wget https://raw.githubusercontent.com/intel/lkp-tests/master/sbin/make.cross -O ~/bin/make.cross
chmod +x ~/bin/make.cross
# https://github.com/0day-ci/linux/commit/4a5a2cece4c5d9ac56322c6828efbea7fcd2e480
git remote add linux-review https://github.com/0day-ci/linux
git fetch --no-tags linux-review Matthew-Wilcox-Oracle/mm-Add-pvmw_set_page/20220204-012111
git checkout 4a5a2cece4c5d9ac56322c6828efbea7fcd2e480
# save the config file to linux build tree
mkdir build_dir
COMPILER_INSTALL_PATH=$HOME/0day COMPILER=clang make.cross W=1 O=build_dir ARCH=hexagon SHELL=/bin/bash
If you fix the issue, kindly add following tag as appropriate
Reported-by: kernel test robot <lkp@intel.com>
All errors (new ones prefixed by >>):
>> mm/page_vma_mapped.c:219:20: error: implicit declaration of function 'pmd_pfn' [-Werror,-Wimplicit-function-declaration]
if (!check_pmd(pmd_pfn(pmde), pvmw))
^
mm/page_vma_mapped.c:219:20: note: did you mean 'pmd_off'?
include/linux/pgtable.h:149:22: note: 'pmd_off' declared here
static inline pmd_t *pmd_off(struct mm_struct *mm, unsigned long va)
^
1 error generated.
vim +/pmd_pfn +219 mm/page_vma_mapped.c
126
127 /**
128 * page_vma_mapped_walk - check if @pvmw->pfn is mapped in @pvmw->vma at
129 * @pvmw->address
130 * @pvmw: pointer to struct page_vma_mapped_walk. page, vma, address and flags
131 * must be set. pmd, pte and ptl must be NULL.
132 *
133 * Returns true if the page is mapped in the vma. @pvmw->pmd and @pvmw->pte point
134 * to relevant page table entries. @pvmw->ptl is locked. @pvmw->address is
135 * adjusted if needed (for PTE-mapped THPs).
136 *
137 * If @pvmw->pmd is set but @pvmw->pte is not, you have found PMD-mapped page
138 * (usually THP). For PTE-mapped THP, you should run page_vma_mapped_walk() in
139 * a loop to find all PTEs that map the THP.
140 *
141 * For HugeTLB pages, @pvmw->pte is set to the relevant page table entry
142 * regardless of which page table level the page is mapped at. @pvmw->pmd is
143 * NULL.
144 *
145 * Returns false if there are no more page table entries for the page in
146 * the vma. @pvmw->ptl is unlocked and @pvmw->pte is unmapped.
147 *
148 * If you need to stop the walk before page_vma_mapped_walk() returned false,
149 * use page_vma_mapped_walk_done(). It will do the housekeeping.
150 */
151 bool page_vma_mapped_walk(struct page_vma_mapped_walk *pvmw)
152 {
153 struct vm_area_struct *vma = pvmw->vma;
154 struct mm_struct *mm = vma->vm_mm;
155 unsigned long end;
156 pgd_t *pgd;
157 p4d_t *p4d;
158 pud_t *pud;
159 pmd_t pmde;
160
161 /* The only possible pmd mapping has been handled on last iteration */
162 if (pvmw->pmd && !pvmw->pte)
163 return not_found(pvmw);
164
165 if (unlikely(is_vm_hugetlb_page(vma))) {
166 unsigned long size = pvmw->nr_pages * PAGE_SIZE;
167 /* The only possible mapping was handled on last iteration */
168 if (pvmw->pte)
169 return not_found(pvmw);
170
171 /* when pud is not present, pte will be NULL */
172 pvmw->pte = huge_pte_offset(mm, pvmw->address, size);
173 if (!pvmw->pte)
174 return false;
175
176 pvmw->ptl = huge_pte_lockptr(size_to_hstate(size), mm,
177 pvmw->pte);
178 spin_lock(pvmw->ptl);
179 if (!check_pte(pvmw))
180 return not_found(pvmw);
181 return true;
182 }
183
184 end = vma_address_end(pvmw);
185 if (pvmw->pte)
186 goto next_pte;
187 restart:
188 do {
189 pgd = pgd_offset(mm, pvmw->address);
190 if (!pgd_present(*pgd)) {
191 step_forward(pvmw, PGDIR_SIZE);
192 continue;
193 }
194 p4d = p4d_offset(pgd, pvmw->address);
195 if (!p4d_present(*p4d)) {
196 step_forward(pvmw, P4D_SIZE);
197 continue;
198 }
199 pud = pud_offset(p4d, pvmw->address);
200 if (!pud_present(*pud)) {
201 step_forward(pvmw, PUD_SIZE);
202 continue;
203 }
204
205 pvmw->pmd = pmd_offset(pud, pvmw->address);
206 /*
207 * Make sure the pmd value isn't cached in a register by the
208 * compiler and used as a stale value after we've observed a
209 * subsequent update.
210 */
211 pmde = READ_ONCE(*pvmw->pmd);
212
213 if (pmd_trans_huge(pmde) || is_pmd_migration_entry(pmde)) {
214 pvmw->ptl = pmd_lock(mm, pvmw->pmd);
215 pmde = *pvmw->pmd;
216 if (likely(pmd_trans_huge(pmde))) {
217 if (pvmw->flags & PVMW_MIGRATION)
218 return not_found(pvmw);
> 219 if (!check_pmd(pmd_pfn(pmde), pvmw))
220 return not_found(pvmw);
221 return true;
222 }
223 if (!pmd_present(pmde)) {
224 swp_entry_t entry;
225
226 if (!thp_migration_supported() ||
227 !(pvmw->flags & PVMW_MIGRATION))
228 return not_found(pvmw);
229 entry = pmd_to_swp_entry(pmde);
230 if (!is_migration_entry(entry) ||
231 !check_pmd(swp_offset(entry), pvmw))
232 return not_found(pvmw);
233 return true;
234 }
235 /* THP pmd was split under us: handle on pte level */
236 spin_unlock(pvmw->ptl);
237 pvmw->ptl = NULL;
238 } else if (!pmd_present(pmde)) {
239 /*
240 * If PVMW_SYNC, take and drop THP pmd lock so that we
241 * cannot return prematurely, while zap_huge_pmd() has
242 * cleared *pmd but not decremented compound_mapcount().
243 */
244 if ((pvmw->flags & PVMW_SYNC) &&
245 (pvmw->nr_pages >= HPAGE_PMD_NR)) {
246 spinlock_t *ptl = pmd_lock(mm, pvmw->pmd);
247
248 spin_unlock(ptl);
249 }
250 step_forward(pvmw, PMD_SIZE);
251 continue;
252 }
253 if (!map_pte(pvmw))
254 goto next_pte;
255 this_pte:
256 if (check_pte(pvmw))
257 return true;
258 next_pte:
259 do {
260 pvmw->address += PAGE_SIZE;
261 if (pvmw->address >= end)
262 return not_found(pvmw);
263 /* Did we cross page table boundary? */
264 if ((pvmw->address & (PMD_SIZE - PAGE_SIZE)) == 0) {
265 if (pvmw->ptl) {
266 spin_unlock(pvmw->ptl);
267 pvmw->ptl = NULL;
268 }
269 pte_unmap(pvmw->pte);
270 pvmw->pte = NULL;
271 goto restart;
272 }
273 pvmw->pte++;
274 if ((pvmw->flags & PVMW_SYNC) && !pvmw->ptl) {
275 pvmw->ptl = pte_lockptr(mm, pvmw->pmd);
276 spin_lock(pvmw->ptl);
277 }
278 } while (pte_none(*pvmw->pte));
279
280 if (!pvmw->ptl) {
281 pvmw->ptl = pte_lockptr(mm, pvmw->pmd);
282 spin_lock(pvmw->ptl);
283 }
284 goto this_pte;
285 } while (pvmw->address < end);
286
287 return false;
288 }
289
---
0-DAY CI Kernel Test Service, Intel Corporation
https://lists.01.org/hyperkitty/list/kbuild-all@lists.01.org
next prev parent reply other threads:[~2022-02-04 1:30 UTC|newest]
Thread overview: 7+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-02-03 17:19 [PATCH 1/2] mm: Add pvmw_set_page() Matthew Wilcox (Oracle)
2022-02-03 17:19 ` [PATCH 2/2] mm: Convert page_vma_mapped_walk to work on PFNs Matthew Wilcox (Oracle)
2022-02-04 1:30 ` kernel test robot
2022-02-04 1:30 ` kernel test robot [this message]
2022-02-08 8:04 ` [PATCH 1/2] mm: Add pvmw_set_page() Muchun Song
2022-02-08 14:49 ` Matthew Wilcox
2022-02-09 3:11 ` Muchun Song
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=202202040601.7rVVBoVe-lkp@intel.com \
--to=lkp@intel.com \
--cc=kbuild-all@lists.01.org \
--cc=linux-mm@kvack.org \
--cc=llvm@lists.linux.dev \
--cc=songmuchun@bytedance.com \
--cc=willy@infradead.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox