stable.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
To: linux-kernel@vger.kernel.org
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
	stable@vger.kernel.org, John Sperbeck <jsperbeck@google.com>,
	Benjamin Herrenschmidt <benh@kernel.crashing.org>,
	Michael Ellerman <mpe@ellerman.id.au>
Subject: [PATCH 4.14 30/38] powerpc/mm: Fix SEGV on mapped region to return SEGV_ACCERR
Date: Mon,  8 Jan 2018 13:59:23 +0100	[thread overview]
Message-ID: <20180108125917.877101029@linuxfoundation.org> (raw)
In-Reply-To: <20180108125915.951963528@linuxfoundation.org>

4.14-stable review patch.  If anyone has any objections, please let me know.

------------------

From: John Sperbeck <jsperbeck@google.com>

commit ecb101aed86156ec7cd71e5dca668e09146e6994 upstream.

The recent refactoring of the powerpc page fault handler in commit
c3350602e876 ("powerpc/mm: Make bad_area* helper functions") caused
access to protected memory regions to indicate SEGV_MAPERR instead of
the traditional SEGV_ACCERR in the si_code field of a user-space
signal handler. This can confuse debug libraries that temporarily
change the protection of memory regions, and expect to use SEGV_ACCERR
as an indication to restore access to a region.

This commit restores the previous behavior. The following program
exhibits the issue:

    $ ./repro read  || echo "FAILED"
    $ ./repro write || echo "FAILED"
    $ ./repro exec  || echo "FAILED"

    #include <stdio.h>
    #include <stdlib.h>
    #include <string.h>
    #include <unistd.h>
    #include <signal.h>
    #include <sys/mman.h>
    #include <assert.h>

    static void segv_handler(int n, siginfo_t *info, void *arg) {
            _exit(info->si_code == SEGV_ACCERR ? 0 : 1);
    }

    int main(int argc, char **argv)
    {
            void *p = NULL;
            struct sigaction act = {
                    .sa_sigaction = segv_handler,
                    .sa_flags = SA_SIGINFO,
            };

            assert(argc == 2);
            p = mmap(NULL, getpagesize(),
                    (strcmp(argv[1], "write") == 0) ? PROT_READ : 0,
                    MAP_PRIVATE|MAP_ANONYMOUS, -1, 0);
            assert(p != MAP_FAILED);

            assert(sigaction(SIGSEGV, &act, NULL) == 0);
            if (strcmp(argv[1], "read") == 0)
                    printf("%c", *(unsigned char *)p);
            else if (strcmp(argv[1], "write") == 0)
                    *(unsigned char *)p = 0;
            else if (strcmp(argv[1], "exec") == 0)
                    ((void (*)(void))p)();
            return 1;  /* failed to generate SEGV */
    }

Fixes: c3350602e876 ("powerpc/mm: Make bad_area* helper functions")
Signed-off-by: John Sperbeck <jsperbeck@google.com>
Acked-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
[mpe: Add commit references in change log]
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

---
 arch/powerpc/mm/fault.c |    7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

--- a/arch/powerpc/mm/fault.c
+++ b/arch/powerpc/mm/fault.c
@@ -145,6 +145,11 @@ static noinline int bad_area(struct pt_r
 	return __bad_area(regs, address, SEGV_MAPERR);
 }
 
+static noinline int bad_access(struct pt_regs *regs, unsigned long address)
+{
+	return __bad_area(regs, address, SEGV_ACCERR);
+}
+
 static int do_sigbus(struct pt_regs *regs, unsigned long address,
 		     unsigned int fault)
 {
@@ -490,7 +495,7 @@ retry:
 
 good_area:
 	if (unlikely(access_error(is_write, is_exec, vma)))
-		return bad_area(regs, address);
+		return bad_access(regs, address);
 
 	/*
 	 * If for any reason at all we couldn't handle the fault,

  parent reply	other threads:[~2018-01-08 12:59 UTC|newest]

Thread overview: 40+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-01-08 12:58 [PATCH 4.14 00/38] 4.14.13-stable review Greg Kroah-Hartman
2018-01-08 12:58 ` [PATCH 4.14 01/38] x86/mm: Set MODULES_END to 0xffffffffff000000 Greg Kroah-Hartman
2018-01-08 12:58 ` [PATCH 4.14 02/38] x86/mm: Map cpu_entry_area at the same place on 4/5 level Greg Kroah-Hartman
2018-01-08 12:58 ` [PATCH 4.14 03/38] x86/kaslr: Fix the vaddr_end mess Greg Kroah-Hartman
2018-01-08 12:58 ` [PATCH 4.14 04/38] x86/events/intel/ds: Use the proper cache flush method for mapping ds buffers Greg Kroah-Hartman
2018-01-08 12:58 ` [PATCH 4.14 05/38] x86/tlb: Drop the _GPL from the cpu_tlbstate export Greg Kroah-Hartman
2018-01-08 12:58 ` [PATCH 4.14 06/38] x86/alternatives: Add missing \n at end of ALTERNATIVE inline asm Greg Kroah-Hartman
2018-01-08 12:59 ` [PATCH 4.14 07/38] x86/pti: Rename BUG_CPU_INSECURE to BUG_CPU_MELTDOWN Greg Kroah-Hartman
2018-01-08 12:59 ` [PATCH 4.14 08/38] kernel/acct.c: fix the acct->needcheck check in check_free_space() Greg Kroah-Hartman
2018-01-08 12:59 ` [PATCH 4.14 09/38] mm/mprotect: add a cond_resched() inside change_pmd_range() Greg Kroah-Hartman
2018-01-08 12:59 ` [PATCH 4.14 10/38] mm/sparse.c: wrong allocation for mem_section Greg Kroah-Hartman
2018-01-08 12:59 ` [PATCH 4.14 11/38] userfaultfd: clear the vma->vm_userfaultfd_ctx if UFFD_EVENT_FORK fails Greg Kroah-Hartman
2018-01-08 12:59 ` [PATCH 4.14 12/38] btrfs: fix refcount_t usage when deleting btrfs_delayed_nodes Greg Kroah-Hartman
2018-01-08 12:59 ` [PATCH 4.14 13/38] efi/capsule-loader: Reinstate virtual capsule mapping Greg Kroah-Hartman
2018-01-08 12:59 ` [PATCH 4.14 14/38] crypto: n2 - cure use after free Greg Kroah-Hartman
2018-01-08 12:59 ` [PATCH 4.14 15/38] crypto: chacha20poly1305 - validate the digest size Greg Kroah-Hartman
2018-01-08 12:59 ` [PATCH 4.14 16/38] crypto: pcrypt - fix freeing pcrypt instances Greg Kroah-Hartman
2018-01-08 12:59 ` [PATCH 4.14 17/38] crypto: chelsio - select CRYPTO_GF128MUL Greg Kroah-Hartman
2018-01-08 12:59 ` [PATCH 4.14 21/38] fscache: Fix the default for fscache_maybe_release_page() Greg Kroah-Hartman
2018-01-08 12:59 ` [PATCH 4.14 22/38] x86 / CPU: Avoid unnecessary IPIs in arch_freq_get_on_cpu() Greg Kroah-Hartman
2018-01-08 12:59 ` [PATCH 4.14 23/38] x86 / CPU: Always show current CPU frequency in /proc/cpuinfo Greg Kroah-Hartman
2018-01-08 12:59 ` [PATCH 4.14 24/38] kernel/signal.c: protect the traced SIGNAL_UNKILLABLE tasks from SIGKILL Greg Kroah-Hartman
2018-01-08 12:59 ` [PATCH 4.14 25/38] kernel/signal.c: protect the SIGNAL_UNKILLABLE tasks from !sig_kernel_only() signals Greg Kroah-Hartman
2018-01-08 12:59 ` [PATCH 4.14 26/38] kernel/signal.c: remove the no longer needed SIGNAL_UNKILLABLE check in complete_signal() Greg Kroah-Hartman
2018-01-08 12:59 ` [PATCH 4.14 27/38] iommu/arm-smmu-v3: Dont free page table ops twice Greg Kroah-Hartman
2018-01-08 12:59 ` [PATCH 4.14 28/38] iommu/arm-smmu-v3: Cope with duplicated Stream IDs Greg Kroah-Hartman
2018-01-08 12:59 ` [PATCH 4.14 29/38] ARC: uaccess: dont use "l" gcc inline asm constraint modifier Greg Kroah-Hartman
2018-01-08 12:59 ` Greg Kroah-Hartman [this message]
2018-01-08 12:59 ` [PATCH 4.14 31/38] Input: elantech - add new icbody type 15 Greg Kroah-Hartman
2018-01-08 12:59 ` [PATCH 4.14 32/38] x86/microcode/AMD: Add support for fam17h microcode loading Greg Kroah-Hartman
2018-01-08 12:59 ` [PATCH 4.14 34/38] parisc: Fix alignment of pa_tlb_lock in assembly on 32-bit SMP kernel Greg Kroah-Hartman
2018-01-08 12:59 ` [PATCH 4.14 35/38] parisc: qemu idle sleep support Greg Kroah-Hartman
2018-01-08 12:59 ` [PATCH 4.14 37/38] KVM: s390: fix cmma migration for multiple memory slots Greg Kroah-Hartman
2018-01-08 12:59 ` [PATCH 4.14 38/38] KVM: s390: prevent buffer overrun on memory hotplug during migration Greg Kroah-Hartman
2018-01-08 20:59 ` [PATCH 4.14 00/38] 4.14.13-stable review Shuah Khan
2018-01-09  9:11   ` Greg Kroah-Hartman
2018-01-09  8:45 ` Naresh Kamboju
2018-01-09  9:27   ` Greg Kroah-Hartman
2018-01-09 13:45 ` Guenter Roeck
2018-01-09 19:07   ` Greg Kroah-Hartman

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=20180108125917.877101029@linuxfoundation.org \
    --to=gregkh@linuxfoundation.org \
    --cc=benh@kernel.crashing.org \
    --cc=jsperbeck@google.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mpe@ellerman.id.au \
    --cc=stable@vger.kernel.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;
as well as URLs for NNTP newsgroup(s).