| Message ID | 20220118185354.464517-1-yury.norov@gmail.com (mailing list archive) |
|---|---|
| State | New |
| Headers | show |
| Series | [RFC] arm64: don't vmap() invalid page | expand |
On Tue, Jan 18, 2022 at 10:53:54AM -0800, Yury Norov wrote: > vmap() takes struct page *pages as one of arguments, and user may provide > an invalid pointer, which would lead to DABT at address translation later. > Currently, kernel checks the pages against NULL. In my case, however, the > address was not NULL, and was big enough so that the hardware generated > Address Size Abort. > > Interestingly, this abort happens even if copy_from_kernel_nofault() is used, > which is quite inconvenient for debugging purposes. > > This patch adds an arch_vmap_page_valid() helper into vmap() path, so that > architectures may add arch-specific checks of the pointer passed into vmap. > > For arm64, if the page passed to vmap() corresponds to a physical address > greater than maximum possible value as described in TCR_EL1.IPS register, the > following table walk would generate Address Size Abort. Instead of creating > the invalid mapping, kernel will return ERANGE in such situation. This seems like a very elaborate way of spelling: pfn_valid(page_to_pfn(page)); which doesn't require any architecture hook. No?
On Tue, Jan 18, 2022 at 10:56 AM Matthew Wilcox <willy@infradead.org> wrote: > > On Tue, Jan 18, 2022 at 10:53:54AM -0800, Yury Norov wrote: > > vmap() takes struct page *pages as one of arguments, and user may provide > > an invalid pointer, which would lead to DABT at address translation later. > > Currently, kernel checks the pages against NULL. In my case, however, the > > address was not NULL, and was big enough so that the hardware generated > > Address Size Abort. > > > > Interestingly, this abort happens even if copy_from_kernel_nofault() is used, > > which is quite inconvenient for debugging purposes. > > > > This patch adds an arch_vmap_page_valid() helper into vmap() path, so that > > architectures may add arch-specific checks of the pointer passed into vmap. > > > > For arm64, if the page passed to vmap() corresponds to a physical address > > greater than maximum possible value as described in TCR_EL1.IPS register, the > > following table walk would generate Address Size Abort. Instead of creating > > the invalid mapping, kernel will return ERANGE in such situation. > > This seems like a very elaborate way of spelling: > > pfn_valid(page_to_pfn(page)); > > which doesn't require any architecture hook. No? Looks like yes. I'll resend later today if there are no other comments. Thank you for the hint.
On Tue, Jan 18, 2022 at 10:53:54AM -0800, Yury Norov wrote: > vmap() takes struct page *pages as one of arguments, and user may provide > an invalid pointer, which would lead to DABT at address translation later. > Currently, kernel checks the pages against NULL. In my case, however, the > address was not NULL, and was big enough so that the hardware generated > Address Size Abort. Can you give an example of when this might happen? It sounds like you're actually hitting this, so a backtrace would be nice. I'm a bit confused as to when why we'd try to vmap() pages that we didn't have a legitimate struct page for -- where did these addresses come from? It sounds like this is going wrong at a higher level, and we're passing entirely bogus struct page pointers around. This seems like the sort of thing DEBUG_VIRTUAL or similar should check when we initially generate the struct page pointer. Thanks, Mark. > Interestingly, this abort happens even if copy_from_kernel_nofault() is used, > which is quite inconvenient for debugging purposes. > > This patch adds an arch_vmap_page_valid() helper into vmap() path, so that > architectures may add arch-specific checks of the pointer passed into vmap. > > For arm64, if the page passed to vmap() corresponds to a physical address > greater than maximum possible value as described in TCR_EL1.IPS register, the > following table walk would generate Address Size Abort. Instead of creating > the invalid mapping, kernel will return ERANGE in such situation. > > Signed-off-by: Yury Norov <yury.norov@gmail.com> > --- > arch/arm64/include/asm/vmalloc.h | 41 ++++++++++++++++++++++++++++++++ > include/linux/vmalloc.h | 7 ++++++ > mm/vmalloc.c | 8 +++++-- > 3 files changed, 54 insertions(+), 2 deletions(-) > > diff --git a/arch/arm64/include/asm/vmalloc.h b/arch/arm64/include/asm/vmalloc.h > index b9185503feae..e9d43ee019ad 100644 > --- a/arch/arm64/include/asm/vmalloc.h > +++ b/arch/arm64/include/asm/vmalloc.h > @@ -4,6 +4,47 @@ > #include <asm/page.h> > #include <asm/pgtable.h> > > +static inline u64 pa_size(u64 ips) > +{ > + switch (ips) { > + case 0b000: > + return 1UL << 32; > + case 0b001: > + return 1UL << 36; > + case 0b010: > + return 1UL << 40; > + case 0b011: > + return 1UL << 42; > + case 0b100: > + return 1UL << 44; > + case 0b101: > + return 1UL << 48; > + case 0b110: > + return 1UL << 52; > + /* All other values */ > + default: > + return 1UL << 52; > + } > +} > + > +#define arch_vmap_page_valid arch_vmap_page_valid > +static inline int arch_vmap_page_valid(struct page *page) > +{ > + u64 tcr, ips, paddr_size; > + > + if (!page) > + return -ENOMEM; > + > + tcr = read_sysreg_s(SYS_TCR_EL1); > + ips = (tcr & TCR_IPS_MASK) >> TCR_IPS_SHIFT; > + > + paddr_size = pa_size(ips); > + if (page_to_phys(page) >= paddr_size) > + return -ERANGE; > + > + return 0; > +} > + > #ifdef CONFIG_HAVE_ARCH_HUGE_VMAP > > #define arch_vmap_pud_supported arch_vmap_pud_supported > diff --git a/include/linux/vmalloc.h b/include/linux/vmalloc.h > index 6e022cc712e6..08b567d8bafc 100644 > --- a/include/linux/vmalloc.h > +++ b/include/linux/vmalloc.h > @@ -119,6 +119,13 @@ static inline int arch_vmap_pte_supported_shift(unsigned long size) > } > #endif > > +#ifndef arch_vmap_page_valid > +static inline int arch_vmap_page_valid(struct page *page) > +{ > + return page ? 0 : -ENOMEM; > +} > +#endif > + > /* > * Highlevel APIs for driver use > */ > diff --git a/mm/vmalloc.c b/mm/vmalloc.c > index d2a00ad4e1dd..ee0384405cdd 100644 > --- a/mm/vmalloc.c > +++ b/mm/vmalloc.c > @@ -472,11 +472,15 @@ static int vmap_pages_pte_range(pmd_t *pmd, unsigned long addr, > return -ENOMEM; > do { > struct page *page = pages[*nr]; > + int ret; > > if (WARN_ON(!pte_none(*pte))) > return -EBUSY; > - if (WARN_ON(!page)) > - return -ENOMEM; > + > + ret = arch_vmap_page_valid(page); > + if (WARN_ON(ret)) > + return ret; > + > set_pte_at(&init_mm, addr, pte, mk_pte(page, prot)); > (*nr)++; > } while (pte++, addr += PAGE_SIZE, addr != end); > -- > 2.30.2 >
diff --git a/arch/arm64/include/asm/vmalloc.h b/arch/arm64/include/asm/vmalloc.h index b9185503feae..e9d43ee019ad 100644 --- a/arch/arm64/include/asm/vmalloc.h +++ b/arch/arm64/include/asm/vmalloc.h @@ -4,6 +4,47 @@ #include <asm/page.h> #include <asm/pgtable.h> +static inline u64 pa_size(u64 ips) +{ + switch (ips) { + case 0b000: + return 1UL << 32; + case 0b001: + return 1UL << 36; + case 0b010: + return 1UL << 40; + case 0b011: + return 1UL << 42; + case 0b100: + return 1UL << 44; + case 0b101: + return 1UL << 48; + case 0b110: + return 1UL << 52; + /* All other values */ + default: + return 1UL << 52; + } +} + +#define arch_vmap_page_valid arch_vmap_page_valid +static inline int arch_vmap_page_valid(struct page *page) +{ + u64 tcr, ips, paddr_size; + + if (!page) + return -ENOMEM; + + tcr = read_sysreg_s(SYS_TCR_EL1); + ips = (tcr & TCR_IPS_MASK) >> TCR_IPS_SHIFT; + + paddr_size = pa_size(ips); + if (page_to_phys(page) >= paddr_size) + return -ERANGE; + + return 0; +} + #ifdef CONFIG_HAVE_ARCH_HUGE_VMAP #define arch_vmap_pud_supported arch_vmap_pud_supported diff --git a/include/linux/vmalloc.h b/include/linux/vmalloc.h index 6e022cc712e6..08b567d8bafc 100644 --- a/include/linux/vmalloc.h +++ b/include/linux/vmalloc.h @@ -119,6 +119,13 @@ static inline int arch_vmap_pte_supported_shift(unsigned long size) } #endif +#ifndef arch_vmap_page_valid +static inline int arch_vmap_page_valid(struct page *page) +{ + return page ? 0 : -ENOMEM; +} +#endif + /* * Highlevel APIs for driver use */ diff --git a/mm/vmalloc.c b/mm/vmalloc.c index d2a00ad4e1dd..ee0384405cdd 100644 --- a/mm/vmalloc.c +++ b/mm/vmalloc.c @@ -472,11 +472,15 @@ static int vmap_pages_pte_range(pmd_t *pmd, unsigned long addr, return -ENOMEM; do { struct page *page = pages[*nr]; + int ret; if (WARN_ON(!pte_none(*pte))) return -EBUSY; - if (WARN_ON(!page)) - return -ENOMEM; + + ret = arch_vmap_page_valid(page); + if (WARN_ON(ret)) + return ret; + set_pte_at(&init_mm, addr, pte, mk_pte(page, prot)); (*nr)++; } while (pte++, addr += PAGE_SIZE, addr != end);
vmap() takes struct page *pages as one of arguments, and user may provide an invalid pointer, which would lead to DABT at address translation later. Currently, kernel checks the pages against NULL. In my case, however, the address was not NULL, and was big enough so that the hardware generated Address Size Abort. Interestingly, this abort happens even if copy_from_kernel_nofault() is used, which is quite inconvenient for debugging purposes. This patch adds an arch_vmap_page_valid() helper into vmap() path, so that architectures may add arch-specific checks of the pointer passed into vmap. For arm64, if the page passed to vmap() corresponds to a physical address greater than maximum possible value as described in TCR_EL1.IPS register, the following table walk would generate Address Size Abort. Instead of creating the invalid mapping, kernel will return ERANGE in such situation. Signed-off-by: Yury Norov <yury.norov@gmail.com> --- arch/arm64/include/asm/vmalloc.h | 41 ++++++++++++++++++++++++++++++++ include/linux/vmalloc.h | 7 ++++++ mm/vmalloc.c | 8 +++++-- 3 files changed, 54 insertions(+), 2 deletions(-)