| Message ID | 20190621095252.32307-11-vincenzo.frascino@arm.com (mailing list archive) |
|---|---|
| State | New |
| Headers | show |
| Series | Unify vDSOs across more architectures | expand |
On Fri, Jun 21, 2019 at 10:52:37AM +0100, Vincenzo Frascino wrote: > --- /dev/null > +++ b/arch/arm64/include/asm/vdso/compat_barrier.h > @@ -0,0 +1,51 @@ > +/* SPDX-License-Identifier: GPL-2.0 */ > +/* > + * Copyright (C) 2018 ARM Limited > + */ > +#ifndef __COMPAT_BARRIER_H > +#define __COMPAT_BARRIER_H > + > +#ifndef __ASSEMBLY__ > +/* > + * Warning: This code is meant to be used with > + * ENABLE_COMPAT_VDSO only. > + */ > +#ifndef ENABLE_COMPAT_VDSO > +#error This header is meant to be used with ENABLE_COMPAT_VDSO only > +#endif > + > +#ifdef dmb > +#undef dmb > +#endif > + > +#if __LINUX_ARM_ARCH__ >= 7 > +#define dmb(option) __asm__ __volatile__ ("dmb " #option : : : "memory") > +#elif __LINUX_ARM_ARCH__ == 6 > +#define dmb(x) __asm__ __volatile__ ("mcr p15, 0, %0, c7, c10, 5" \ > + : : "r" (0) : "memory") > +#else > +#define dmb(x) __asm__ __volatile__ ("" : : : "memory") > +#endif We don't need pre-ARMv7 barriers (they've been deprecated and the arm64 kernel actually traps and emulates them by default). Also your Makefile changes never define a __LINUX_ARM_ARCH__ lower than 7. Fix-up below: ------8<----------------------- From 5655a0313ce7bb731bfed6a19bcfe6b1100b542a Mon Sep 17 00:00:00 2001 From: Catalin Marinas <catalin.marinas@arm.com> Date: Mon, 24 Jun 2019 12:16:06 +0100 Subject: [PATCH] arm64: compat: No need for pre-ARMv7 barriers on an ARMv8 system This patch removes the deprecated (pre-ARMv7) compat barriers as they would not be used on an ARMv8 system. Fixes: a7f71a2c8903 ("arm64: compat: Add vDSO") Signed-off-by: Catalin Marinas <catalin.marinas@arm.com> --- arch/arm64/include/asm/vdso/compat_barrier.h | 7 ------- 1 file changed, 7 deletions(-) diff --git a/arch/arm64/include/asm/vdso/compat_barrier.h b/arch/arm64/include/asm/vdso/compat_barrier.h index ea24ea856b07..fb60a88b5ed4 100644 --- a/arch/arm64/include/asm/vdso/compat_barrier.h +++ b/arch/arm64/include/asm/vdso/compat_barrier.h @@ -18,14 +18,7 @@ #undef dmb #endif -#if __LINUX_ARM_ARCH__ >= 7 #define dmb(option) __asm__ __volatile__ ("dmb " #option : : : "memory") -#elif __LINUX_ARM_ARCH__ == 6 -#define dmb(x) __asm__ __volatile__ ("mcr p15, 0, %0, c7, c10, 5" \ - : : "r" (0) : "memory") -#else -#define dmb(x) __asm__ __volatile__ ("" : : : "memory") -#endif #if __LINUX_ARM_ARCH__ >= 8 #define aarch32_smp_mb() dmb(ish)
On Fri, Jun 21, 2019 at 3:18 AM Vincenzo Frascino <vincenzo.frascino@arm.com> wrote: > > Provide the arm64 compat (AArch32) vDSO in kernel/vdso32 in a similar > way to what happens in kernel/vdso. > > The compat vDSO leverages on an adaptation of the arm architecture code > with few changes: > - Use of lib/vdso for gettimeofday > - Implementation of syscall based fallback > - Introduction of clock_getres for the compat library > - Implementation of trampolines > - Implementation of elf note > > To build the compat vDSO a 32 bit compiler is required and needs to be > specified via CONFIG_CROSS_COMPILE_COMPAT_VDSO. > Hey Vincenzo! Congrats on getting this work merged, I know its been a long effort over a number of years! Though unfortunately, it seems the arm64 vdso code that just landed is breaking AOSP for me. I see a lot of the following errors: 01-01 01:22:14.097 755 755 F libc : Fatal signal 11 (SIGSEGV), code 1 (SEGV_MAPERR), fault addr 0x3cf2c96c in tid 755 (cameraserver), pid 755 (cameraserver) 01-01 01:22:14.112 759 759 F libc : Fatal signal 11 (SIGSEGV), code 1 (SEGV_MAPERR), fault addr 0x3cf2c96c in tid 759 (android.hardwar), pid 759 (android.hardwar) 01-01 01:22:14.120 756 756 F libc : Fatal signal 11 (SIGSEGV), code 1 (SEGV_MAPERR), fault addr 0x3cf2c96c in tid 756 (drmserver), pid 756 (drmserver) Which go away if I revert the vdso merge that went in via tip/timers. I tried to bisect things down a bit, but as some later fixes are required (at one point, date was returning the start epoch and never increasing), this hasn't worked too well. But I'm guessing since I see: "CROSS_COMPILE_COMPAT not defined or empty, the compat vDSO will not be built", and the system is half working, I'm guessing this is an issue with just the 32bit code failing. While I can try to sort out the proper CROSS_COMPILE_COMPAT in my build environment, I assume userland shouldn't be crashing if that value isn't set. Any chance this issue has already been raised? thanks -john
On Tue, 9 Jul 2019, John Stultz wrote: > Though unfortunately, it seems the arm64 vdso code that just landed is > breaking AOSP for me. > > I see a lot of the following errors: > 01-01 01:22:14.097 755 755 F libc : Fatal signal 11 (SIGSEGV), > code 1 (SEGV_MAPERR), fault addr 0x3cf2c96c in tid 755 (cameraserver), > pid 755 (cameraserver) > 01-01 01:22:14.112 759 759 F libc : Fatal signal 11 (SIGSEGV), > code 1 (SEGV_MAPERR), fault addr 0x3cf2c96c in tid 759 > (android.hardwar), pid 759 (android.hardwar) > 01-01 01:22:14.120 756 756 F libc : Fatal signal 11 (SIGSEGV), > code 1 (SEGV_MAPERR), fault addr 0x3cf2c96c in tid 756 (drmserver), > pid 756 (drmserver) > > Which go away if I revert the vdso merge that went in via tip/timers. > > I tried to bisect things down a bit, but as some later fixes are > required (at one point, date was returning the start epoch and never > increasing), this hasn't worked too well. But I'm guessing since I > see: "CROSS_COMPILE_COMPAT not defined or empty, the compat vDSO will > not be built", and the system is half working, I'm guessing this is an > issue with just the 32bit code failing. While I can try to sort out > the proper CROSS_COMPILE_COMPAT in my build environment, I assume > userland shouldn't be crashing if that value isn't set. The obvious question is whether the VDSO is mapped to the 32bit process in that case. It shouldn't... Thanks, tglx
Hi John, On Tue, Jul 09, 2019 at 09:02:54PM -0700, John Stultz wrote: > On Fri, Jun 21, 2019 at 3:18 AM Vincenzo Frascino > <vincenzo.frascino@arm.com> wrote: > > > > Provide the arm64 compat (AArch32) vDSO in kernel/vdso32 in a similar > > way to what happens in kernel/vdso. > > > > The compat vDSO leverages on an adaptation of the arm architecture code > > with few changes: > > - Use of lib/vdso for gettimeofday > > - Implementation of syscall based fallback > > - Introduction of clock_getres for the compat library > > - Implementation of trampolines > > - Implementation of elf note > > > > To build the compat vDSO a 32 bit compiler is required and needs to be > > specified via CONFIG_CROSS_COMPILE_COMPAT_VDSO. > > > > Hey Vincenzo! > Congrats on getting this work merged, I know its been a long effort > over a number of years! > > Though unfortunately, it seems the arm64 vdso code that just landed is > breaking AOSP for me. > > I see a lot of the following errors: > 01-01 01:22:14.097 755 755 F libc : Fatal signal 11 (SIGSEGV), > code 1 (SEGV_MAPERR), fault addr 0x3cf2c96c in tid 755 (cameraserver), > pid 755 (cameraserver) > 01-01 01:22:14.112 759 759 F libc : Fatal signal 11 (SIGSEGV), > code 1 (SEGV_MAPERR), fault addr 0x3cf2c96c in tid 759 > (android.hardwar), pid 759 (android.hardwar) > 01-01 01:22:14.120 756 756 F libc : Fatal signal 11 (SIGSEGV), > code 1 (SEGV_MAPERR), fault addr 0x3cf2c96c in tid 756 (drmserver), > pid 756 (drmserver) > > Which go away if I revert the vdso merge that went in via tip/timers. > > I tried to bisect things down a bit, but as some later fixes are > required (at one point, date was returning the start epoch and never > increasing), this hasn't worked too well. But I'm guessing since I > see: "CROSS_COMPILE_COMPAT not defined or empty, the compat vDSO will > not be built", and the system is half working, I'm guessing this is an > issue with just the 32bit code failing. While I can try to sort out > the proper CROSS_COMPILE_COMPAT in my build environment, I assume > userland shouldn't be crashing if that value isn't set. > > Any chance this issue has already been raised? First I've seen of it, although Vincenzo is likely to know better than me. In the meantime, please can you share your .config? Thanks, Will
On Wed, 10 Jul 2019, Will Deacon wrote: > On Tue, Jul 09, 2019 at 09:02:54PM -0700, John Stultz wrote: > > I tried to bisect things down a bit, but as some later fixes are > > required (at one point, date was returning the start epoch and never > > increasing), this hasn't worked too well. But I'm guessing since I > > see: "CROSS_COMPILE_COMPAT not defined or empty, the compat vDSO will > > not be built", and the system is half working, I'm guessing this is an > > issue with just the 32bit code failing. While I can try to sort out > > the proper CROSS_COMPILE_COMPAT in my build environment, I assume > > userland shouldn't be crashing if that value isn't set. > > > > Any chance this issue has already been raised? > > First I've seen of it, although Vincenzo is likely to know better than me. > In the meantime, please can you share your .config? I think the key is: CROSS_COMPILE_COMPAT not defined or empty. And then run 32bit userspace. Thanks, tglx
On Wed, Jul 10, 2019 at 10:58:25AM +0200, Thomas Gleixner wrote: > On Wed, 10 Jul 2019, Will Deacon wrote: > > On Tue, Jul 09, 2019 at 09:02:54PM -0700, John Stultz wrote: > > > I tried to bisect things down a bit, but as some later fixes are > > > required (at one point, date was returning the start epoch and never > > > increasing), this hasn't worked too well. But I'm guessing since I > > > see: "CROSS_COMPILE_COMPAT not defined or empty, the compat vDSO will > > > not be built", and the system is half working, I'm guessing this is an > > > issue with just the 32bit code failing. While I can try to sort out > > > the proper CROSS_COMPILE_COMPAT in my build environment, I assume > > > userland shouldn't be crashing if that value isn't set. > > > > > > Any chance this issue has already been raised? > > > > First I've seen of it, although Vincenzo is likely to know better than me. > > In the meantime, please can you share your .config? > > I think the key is: CROSS_COMPILE_COMPAT not defined or empty. And then run > 32bit userspace. The part I was wondering about is whether the vectors page has been disabled at the same time, since I'm fairly sure I've already been running with the above (but I can easily double-check). Will
Hi John, On 10/07/2019 05:02, John Stultz wrote: > On Fri, Jun 21, 2019 at 3:18 AM Vincenzo Frascino > <vincenzo.frascino@arm.com> wrote: >> >> Provide the arm64 compat (AArch32) vDSO in kernel/vdso32 in a similar >> way to what happens in kernel/vdso. >> >> The compat vDSO leverages on an adaptation of the arm architecture code >> with few changes: >> - Use of lib/vdso for gettimeofday >> - Implementation of syscall based fallback >> - Introduction of clock_getres for the compat library >> - Implementation of trampolines >> - Implementation of elf note >> >> To build the compat vDSO a 32 bit compiler is required and needs to be >> specified via CONFIG_CROSS_COMPILE_COMPAT_VDSO. >> > > Hey Vincenzo! > Congrats on getting this work merged, I know its been a long effort > over a number of years! > > Though unfortunately, it seems the arm64 vdso code that just landed is > breaking AOSP for me. > > I see a lot of the following errors: > 01-01 01:22:14.097 755 755 F libc : Fatal signal 11 (SIGSEGV), > code 1 (SEGV_MAPERR), fault addr 0x3cf2c96c in tid 755 (cameraserver), > pid 755 (cameraserver) > 01-01 01:22:14.112 759 759 F libc : Fatal signal 11 (SIGSEGV), > code 1 (SEGV_MAPERR), fault addr 0x3cf2c96c in tid 759 > (android.hardwar), pid 759 (android.hardwar) > 01-01 01:22:14.120 756 756 F libc : Fatal signal 11 (SIGSEGV), > code 1 (SEGV_MAPERR), fault addr 0x3cf2c96c in tid 756 (drmserver), > pid 756 (drmserver) > > Which go away if I revert the vdso merge that went in via tip/timers. > > I tried to bisect things down a bit, but as some later fixes are > required (at one point, date was returning the start epoch and never > increasing), this hasn't worked too well. But I'm guessing since I > see: "CROSS_COMPILE_COMPAT not defined or empty, the compat vDSO will > not be built", and the system is half working, I'm guessing this is an > issue with just the 32bit code failing. While I can try to sort out > the proper CROSS_COMPILE_COMPAT in my build environment, I assume > userland shouldn't be crashing if that value isn't set. > > Any chance this issue has already been raised? > I do not have Android (bionic/libc) as part of my testing environment hence I never saw this issue. Thanks for reporting it. I am investigating the problem and will post a fix as soon as it is ready. As Will suggested, .config would help the debugging and I would like to ask to you to test my fix once it is ready. Is that OK for you? > thanks > -john >
On 10/07/2019 07:12, Thomas Gleixner wrote: > On Tue, 9 Jul 2019, John Stultz wrote: >> Though unfortunately, it seems the arm64 vdso code that just landed is >> breaking AOSP for me. >> >> I see a lot of the following errors: >> 01-01 01:22:14.097 755 755 F libc : Fatal signal 11 (SIGSEGV), >> code 1 (SEGV_MAPERR), fault addr 0x3cf2c96c in tid 755 (cameraserver), >> pid 755 (cameraserver) >> 01-01 01:22:14.112 759 759 F libc : Fatal signal 11 (SIGSEGV), >> code 1 (SEGV_MAPERR), fault addr 0x3cf2c96c in tid 759 >> (android.hardwar), pid 759 (android.hardwar) >> 01-01 01:22:14.120 756 756 F libc : Fatal signal 11 (SIGSEGV), >> code 1 (SEGV_MAPERR), fault addr 0x3cf2c96c in tid 756 (drmserver), >> pid 756 (drmserver) >> >> Which go away if I revert the vdso merge that went in via tip/timers. >> >> I tried to bisect things down a bit, but as some later fixes are >> required (at one point, date was returning the start epoch and never >> increasing), this hasn't worked too well. But I'm guessing since I >> see: "CROSS_COMPILE_COMPAT not defined or empty, the compat vDSO will >> not be built", and the system is half working, I'm guessing this is an >> issue with just the 32bit code failing. While I can try to sort out >> the proper CROSS_COMPILE_COMPAT in my build environment, I assume >> userland shouldn't be crashing if that value isn't set. > > The obvious question is whether the VDSO is mapped to the 32bit process in > that case. It shouldn't... > Agreed. I am investing if/why this is happening and will post a fix accordingly. > Thanks, > > tglx >
Hi John, On 10/07/2019 10:47, Vincenzo Frascino wrote: > Hi John, > > On 10/07/2019 05:02, John Stultz wrote: >> On Fri, Jun 21, 2019 at 3:18 AM Vincenzo Frascino >> <vincenzo.frascino@arm.com> wrote: >>> >>> Provide the arm64 compat (AArch32) vDSO in kernel/vdso32 in a similar >>> way to what happens in kernel/vdso. >>> >>> The compat vDSO leverages on an adaptation of the arm architecture code >>> with few changes: >>> - Use of lib/vdso for gettimeofday >>> - Implementation of syscall based fallback >>> - Introduction of clock_getres for the compat library >>> - Implementation of trampolines >>> - Implementation of elf note >>> >>> To build the compat vDSO a 32 bit compiler is required and needs to be >>> specified via CONFIG_CROSS_COMPILE_COMPAT_VDSO. >>> >> >> Hey Vincenzo! >> Congrats on getting this work merged, I know its been a long effort >> over a number of years! >> >> Though unfortunately, it seems the arm64 vdso code that just landed is >> breaking AOSP for me. >> >> I see a lot of the following errors: >> 01-01 01:22:14.097 755 755 F libc : Fatal signal 11 (SIGSEGV), >> code 1 (SEGV_MAPERR), fault addr 0x3cf2c96c in tid 755 (cameraserver), >> pid 755 (cameraserver) >> 01-01 01:22:14.112 759 759 F libc : Fatal signal 11 (SIGSEGV), >> code 1 (SEGV_MAPERR), fault addr 0x3cf2c96c in tid 759 >> (android.hardwar), pid 759 (android.hardwar) >> 01-01 01:22:14.120 756 756 F libc : Fatal signal 11 (SIGSEGV), >> code 1 (SEGV_MAPERR), fault addr 0x3cf2c96c in tid 756 (drmserver), >> pid 756 (drmserver) >> >> Which go away if I revert the vdso merge that went in via tip/timers. >> >> I tried to bisect things down a bit, but as some later fixes are >> required (at one point, date was returning the start epoch and never >> increasing), this hasn't worked too well. But I'm guessing since I >> see: "CROSS_COMPILE_COMPAT not defined or empty, the compat vDSO will >> not be built", and the system is half working, I'm guessing this is an >> issue with just the 32bit code failing. While I can try to sort out >> the proper CROSS_COMPILE_COMPAT in my build environment, I assume >> userland shouldn't be crashing if that value isn't set. >> >> Any chance this issue has already been raised? >> > > I do not have Android (bionic/libc) as part of my testing environment hence I > never saw this issue. Thanks for reporting it. > > I am investigating the problem and will post a fix as soon as it is ready. > > As Will suggested, .config would help the debugging and I would like to ask to > you to test my fix once it is ready. Is that OK for you? > >> thanks >> -john >> > Seems that the problem you are experiencing is caused by an ABI regression for which I provided a fix in [1]. Could you please verify that works for you as well? [1] https://www.mail-archive.com/linux-kernel@vger.kernel.org/msg2046889.html Please find below the tests that I conducted to verify that when CROSS_COMPILE_COMPAT is not defined or empty the compat vdso library is not exposed. # cat main-arm32.c #include <stdio.h> #include <sys/auxv.h> int main() { uintptr_t vdso = (uintptr_t) getauxval(AT_SYSINFO_EHDR); printf("AT_SYSINFO_EHDR: %p\n", vdso); [...] return 0; } # file main-arm32 main-arm32: ELF 32-bit LSB executable, ARM, EABI5 version 1 (SYSV), dynamically linked, interpreter /lib/ld-, for GNU/Linux 3.2.0, with d When CROSS_COMPILE_COMPAT is _not_ defined: =========================================== # ./main-arm32 AT_SYSINFO_EHDR: (nil) Memory Map: ----------- 00008000-00010000 rw-p 00000000 00:00 0 00010000-00011000 r-xp 00000000 00:13 24905929 /home/vinfra01/bin/main-arm32 00011000-00020000 rw-p 00000000 00:00 0 00020000-00021000 r--p 00000000 00:13 24905929 /home/vinfra01/bin/main-arm32 00021000-00022000 rw-p 00001000 00:13 24905929 /home/vinfra01/bin/main-arm32 00022000-f773d000 rw-p 00000000 00:00 0 [heap] f773d000-f781f000 r-xp 00000000 00:13 24906912 /lib/arm-linux-gnueabihf/libc-2.27.so f781f000-f782f000 ---p 000e2000 00:13 24906912 /lib/arm-linux-gnueabihf/libc-2.27.so f782f000-f7831000 r--p 000e2000 00:13 24906912 /lib/arm-linux-gnueabihf/libc-2.27.so f7831000-f7832000 rw-p 000e4000 00:13 24906912 /lib/arm-linux-gnueabihf/libc-2.27.so f7832000-f7835000 rw-p 00000000 00:00 0 f7835000-f784b000 rw-p 00000000 00:00 0 f784b000-f7863000 r-xp 00000000 00:13 24906908 /lib/arm-linux-gnueabihf/ld-2.27.so f7863000-f7870000 rw-p 00000000 00:00 0 f7870000-f7872000 rw-p 00000000 00:00 0 f7872000-f7873000 r-xp 00000000 00:00 0 [sigpage] f7873000-f7874000 r--p 00018000 00:13 24906908 /lib/arm-linux-gnueabihf/ld-2.27.so f7874000-f7875000 rw-p 00019000 00:13 24906908 /lib/arm-linux-gnueabihf/ld-2.27.so f7875000-ff77a000 rw-p 00000000 00:00 0 ff87a000-ff89b000 rw-p 00000000 00:00 0 [stack] ff89b000-ffff0000 rw-p 00000000 00:00 0 ffff0000-ffff1000 r-xp 00000000 00:00 0 [vectors] ffff1000-fffff000 rw-p 00000000 00:00 0 vdsotest for compat [2]: ------------------------ Note: vDSO version of clock_gettime not found clock-gettime-monotonic: syscall: 796 nsec/call clock-gettime-monotonic: libc: 816 nsec/call clock-gettime-monotonic: vdso: not tested Note: vDSO version of clock_gettime not found Note: vDSO version of clock_gettime not found Note: vDSO version of clock_getres not found clock-getres-monotonic: syscall: 567 nsec/call clock-getres-monotonic: libc: 581 nsec/call clock-getres-monotonic: vdso: not tested Note: vDSO version of clock_getres not found Note: vDSO version of clock_getres not found Note: vDSO version of clock_gettime not found clock-gettime-monotonic-coarse: syscall: 617 nsec/call clock-gettime-monotonic-coarse: libc: 656 nsec/call clock-gettime-monotonic-coarse: vdso: not tested Note: vDSO version of clock_gettime not found Note: vDSO version of clock_gettime not found Note: vDSO version of clock_getres not found clock-getres-monotonic-coarse: syscall: 591 nsec/call clock-getres-monotonic-coarse: libc: 614 nsec/call clock-getres-monotonic-coarse: vdso: not tested Note: vDSO version of clock_getres not found Note: vDSO version of clock_getres not found Note: vDSO version of clock_gettime not found clock-gettime-realtime: syscall: 819 nsec/call clock-gettime-realtime: libc: 858 nsec/call clock-gettime-realtime: vdso: not tested Note: vDSO version of clock_gettime not found Note: vDSO version of clock_gettime not found Note: vDSO version of clock_getres not found clock-getres-realtime: syscall: 567 nsec/call clock-getres-realtime: libc: 583 nsec/call clock-getres-realtime: vdso: not tested Note: vDSO version of clock_getres not found Note: vDSO version of clock_getres not found Note: vDSO version of clock_gettime not found clock-gettime-realtime-coarse: syscall: 599 nsec/call clock-gettime-realtime-coarse: libc: 638 nsec/call clock-gettime-realtime-coarse: vdso: not tested Note: vDSO version of clock_gettime not found Note: vDSO version of clock_gettime not found Note: vDSO version of clock_getres not found clock-getres-realtime-coarse: syscall: 591 nsec/call clock-getres-realtime-coarse: libc: 610 nsec/call clock-getres-realtime-coarse: vdso: not tested Note: vDSO version of clock_getres not found Note: vDSO version of clock_getres not found Note: vDSO version of getcpu not found getcpu: syscall: 431 nsec/call getcpu: libc: 451 nsec/call getcpu: vdso: not tested Note: vDSO version of getcpu not found Note: vDSO version of getcpu not found Note: vDSO version of gettimeofday not found gettimeofday: syscall: 765 nsec/call gettimeofday: libc: 808 nsec/call gettimeofday: vdso: not tested Note: vDSO version of gettimeofday not found Note: vDSO version of gettimeofday not found When CROSS_COMPILE_COMPAT is defined: ===================================== # ./main-arm32 AT_SYSINFO_EHDR: 0xf7a67000 Memory map: ----------- 00008000-00010000 rw-p 00000000 00:00 0 00010000-00011000 r-xp 00000000 00:13 24905929 /home/vinfra01/bin/main-arm32 00011000-00020000 rw-p 00000000 00:00 0 00020000-00021000 r--p 00000000 00:13 24905929 /home/vinfra01/bin/main-arm32 00021000-00022000 rw-p 00001000 00:13 24905929 /home/vinfra01/bin/main-arm32 00022000-f7932000 rw-p 00000000 00:00 0 [heap] f7932000-f7a14000 r-xp 00000000 00:13 24906912 /lib/arm-linux-gnueabihf/libc-2.27.so f7a14000-f7a24000 ---p 000e2000 00:13 24906912 /lib/arm-linux-gnueabihf/libc-2.27.so f7a24000-f7a26000 r--p 000e2000 00:13 24906912 /lib/arm-linux-gnueabihf/libc-2.27.so f7a26000-f7a27000 rw-p 000e4000 00:13 24906912 /lib/arm-linux-gnueabihf/libc-2.27.so f7a27000-f7a2a000 rw-p 00000000 00:00 0 f7a2a000-f7a40000 rw-p 00000000 00:00 0 f7a40000-f7a58000 r-xp 00000000 00:13 24906908 /lib/arm-linux-gnueabihf/ld-2.27.so f7a58000-f7a64000 rw-p 00000000 00:00 0 f7a64000-f7a66000 rw-p 00000000 00:00 0 f7a66000-f7a67000 r--p 00000000 00:00 0 [vvar] f7a67000-f7a68000 r-xp 00000000 00:00 0 [vdso] f7a68000-f7a69000 r--p 00018000 00:13 24906908 /lib/arm-linux-gnueabihf/ld-2.27.so f7a69000-f7a6a000 rw-p 00019000 00:13 24906908 /lib/arm-linux-gnueabihf/ld-2.27.so f7a6a000-ff6d8000 rw-p 00000000 00:00 0 ff7d8000-ff7f9000 rw-p 00000000 00:00 0 [stack] ff7f9000-ffff0000 rw-p 00000000 00:00 0 ffff0000-ffff1000 r-xp 00000000 00:00 0 [vectors] ffff1000-fffff000 rw-p 00000000 00:00 0 vdsotest for compat [2]: ------------------------ clock-gettime-monotonic: syscall: 700 nsec/call clock-gettime-monotonic: libc: 226 nsec/call clock-gettime-monotonic: vdso: 189 nsec/call clock-getres-monotonic: syscall: 582 nsec/call clock-getres-monotonic: libc: 581 nsec/call clock-getres-monotonic: vdso: 40 nsec/call clock-gettime-monotonic-coarse: syscall: 618 nsec/call clock-gettime-monotonic-coarse: libc: 103 nsec/call clock-gettime-monotonic-coarse: vdso: 85 nsec/call clock-getres-monotonic-coarse: syscall: 602 nsec/call clock-getres-monotonic-coarse: libc: 610 nsec/call clock-getres-monotonic-coarse: vdso: 64 nsec/call clock-gettime-realtime: syscall: 741 nsec/call clock-gettime-realtime: libc: 209 nsec/call clock-gettime-realtime: vdso: 189 nsec/call clock-getres-realtime: syscall: 627 nsec/call clock-getres-realtime: libc: 604 nsec/call clock-getres-realtime: vdso: 41 nsec/call clock-gettime-realtime-coarse: syscall: 640 nsec/call clock-gettime-realtime-coarse: libc: 105 nsec/call clock-gettime-realtime-coarse: vdso: 84 nsec/call clock-getres-realtime-coarse: syscall: 597 nsec/call clock-getres-realtime-coarse: libc: 608 nsec/call clock-getres-realtime-coarse: vdso: 46 nsec/call Note: vDSO version of getcpu not found getcpu: syscall: 492 nsec/call getcpu: libc: 492 nsec/call getcpu: vdso: not tested Note: vDSO version of getcpu not found Note: vDSO version of getcpu not found gettimeofday: syscall: 664 nsec/call gettimeofday: libc: 224 nsec/call gettimeofday: vdso: 185 nsec/call [2] https://github.com/nathanlynch/vdsotest
diff --git a/arch/arm64/include/asm/vdso/compat_barrier.h b/arch/arm64/include/asm/vdso/compat_barrier.h new file mode 100644 index 000000000000..ea24ea856b07 --- /dev/null +++ b/arch/arm64/include/asm/vdso/compat_barrier.h @@ -0,0 +1,51 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* + * Copyright (C) 2018 ARM Limited + */ +#ifndef __COMPAT_BARRIER_H +#define __COMPAT_BARRIER_H + +#ifndef __ASSEMBLY__ +/* + * Warning: This code is meant to be used with + * ENABLE_COMPAT_VDSO only. + */ +#ifndef ENABLE_COMPAT_VDSO +#error This header is meant to be used with ENABLE_COMPAT_VDSO only +#endif + +#ifdef dmb +#undef dmb +#endif + +#if __LINUX_ARM_ARCH__ >= 7 +#define dmb(option) __asm__ __volatile__ ("dmb " #option : : : "memory") +#elif __LINUX_ARM_ARCH__ == 6 +#define dmb(x) __asm__ __volatile__ ("mcr p15, 0, %0, c7, c10, 5" \ + : : "r" (0) : "memory") +#else +#define dmb(x) __asm__ __volatile__ ("" : : : "memory") +#endif + +#if __LINUX_ARM_ARCH__ >= 8 +#define aarch32_smp_mb() dmb(ish) +#define aarch32_smp_rmb() dmb(ishld) +#define aarch32_smp_wmb() dmb(ishst) +#else +#define aarch32_smp_mb() dmb(ish) +#define aarch32_smp_rmb() aarch32_smp_mb() +#define aarch32_smp_wmb() dmb(ishst) +#endif + + +#undef smp_mb +#undef smp_rmb +#undef smp_wmb + +#define smp_mb() aarch32_smp_mb() +#define smp_rmb() aarch32_smp_rmb() +#define smp_wmb() aarch32_smp_wmb() + +#endif /* !__ASSEMBLY__ */ + +#endif /* __COMPAT_BARRIER_H */ diff --git a/arch/arm64/include/asm/vdso/compat_gettimeofday.h b/arch/arm64/include/asm/vdso/compat_gettimeofday.h new file mode 100644 index 000000000000..7a39c8340936 --- /dev/null +++ b/arch/arm64/include/asm/vdso/compat_gettimeofday.h @@ -0,0 +1,110 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* + * Copyright (C) 2018 ARM Limited + */ +#ifndef __ASM_VDSO_GETTIMEOFDAY_H +#define __ASM_VDSO_GETTIMEOFDAY_H + +#ifndef __ASSEMBLY__ + +#include <asm/unistd.h> +#include <uapi/linux/time.h> + +#include <asm/vdso/compat_barrier.h> + +#define VDSO_HAS_CLOCK_GETRES 1 + +static __always_inline int gettimeofday_fallback( + struct __kernel_old_timeval *_tv, + struct timezone *_tz) +{ + register struct timezone *tz asm("r1") = _tz; + register struct __kernel_old_timeval *tv asm("r0") = _tv; + register long ret asm ("r0"); + register long nr asm("r7") = __NR_compat_gettimeofday; + + asm volatile( + " swi #0\n" + : "=r" (ret) + : "r" (tv), "r" (tz), "r" (nr) + : "memory"); + + return ret; +} + +static __always_inline long clock_gettime_fallback( + clockid_t _clkid, + struct __kernel_timespec *_ts) +{ + register struct __kernel_timespec *ts asm("r1") = _ts; + register clockid_t clkid asm("r0") = _clkid; + register long ret asm ("r0"); + register long nr asm("r7") = __NR_compat_clock_gettime64; + + asm volatile( + " swi #0\n" + : "=r" (ret) + : "r" (clkid), "r" (ts), "r" (nr) + : "memory"); + + return ret; +} + +static __always_inline int clock_getres_fallback( + clockid_t _clkid, + struct __kernel_timespec *_ts) +{ + register struct __kernel_timespec *ts asm("r1") = _ts; + register clockid_t clkid asm("r0") = _clkid; + register long ret asm ("r0"); + register long nr asm("r7") = __NR_compat_clock_getres_time64; + + /* The checks below are required for ABI consistency with arm */ + if ((_clkid >= MAX_CLOCKS) && (_ts == NULL)) + return -EINVAL; + + asm volatile( + " swi #0\n" + : "=r" (ret) + : "r" (clkid), "r" (ts), "r" (nr) + : "memory"); + + return ret; +} + +static __always_inline u64 __arch_get_hw_counter(s32 clock_mode) +{ + u64 res; + + isb(); + asm volatile("mrrc p15, 1, %Q0, %R0, c14" : "=r" (res)); + + return res; +} + +static __always_inline const struct vdso_data *__arch_get_vdso_data(void) +{ + const struct vdso_data *ret; + + /* + * This simply puts &_vdso_data into ret. The reason why we don't use + * `ret = _vdso_data` is that the compiler tends to optimise this in a + * very suboptimal way: instead of keeping &_vdso_data in a register, + * it goes through a relocation almost every time _vdso_data must be + * accessed (even in subfunctions). This is both time and space + * consuming: each relocation uses a word in the code section, and it + * has to be loaded at runtime. + * + * This trick hides the assignment from the compiler. Since it cannot + * track where the pointer comes from, it will only use one relocation + * where __arch_get_vdso_data() is called, and then keep the result in + * a register. + */ + asm volatile("mov %0, %1" : "=r"(ret) : "r"(_vdso_data)); + + return ret; +} + +#endif /* !__ASSEMBLY__ */ + +#endif /* __ASM_VDSO_GETTIMEOFDAY_H */ diff --git a/arch/arm64/kernel/vdso32/.gitignore b/arch/arm64/kernel/vdso32/.gitignore new file mode 100644 index 000000000000..4fea950fa5ed --- /dev/null +++ b/arch/arm64/kernel/vdso32/.gitignore @@ -0,0 +1,2 @@ +vdso.lds +vdso.so.raw diff --git a/arch/arm64/kernel/vdso32/Makefile b/arch/arm64/kernel/vdso32/Makefile new file mode 100644 index 000000000000..288c14d30b45 --- /dev/null +++ b/arch/arm64/kernel/vdso32/Makefile @@ -0,0 +1,186 @@ +# SPDX-License-Identifier: GPL-2.0 +# +# Makefile for vdso32 +# + +# Absolute relocation type $(ARCH_REL_TYPE_ABS) needs to be defined before +# the inclusion of generic Makefile. +ARCH_REL_TYPE_ABS := R_ARM_JUMP_SLOT|R_ARM_GLOB_DAT|R_ARM_ABS32 +include $(srctree)/lib/vdso/Makefile + +COMPATCC := $(CROSS_COMPILE_COMPAT)gcc + +# Same as cc-*option, but using COMPATCC instead of CC +cc32-option = $(call try-run,\ + $(COMPATCC) $(1) -c -x c /dev/null -o "$$TMP",$(1),$(2)) +cc32-disable-warning = $(call try-run,\ + $(COMPATCC) -W$(strip $(1)) -c -x c /dev/null -o "$$TMP",-Wno-$(strip $(1))) +cc32-ldoption = $(call try-run,\ + $(COMPATCC) $(1) -nostdlib -x c /dev/null -o "$$TMP",$(1),$(2)) + +# We cannot use the global flags to compile the vDSO files, the main reason +# being that the 32-bit compiler may be older than the main (64-bit) compiler +# and therefore may not understand flags set using $(cc-option ...). Besides, +# arch-specific options should be taken from the arm Makefile instead of the +# arm64 one. +# As a result we set our own flags here. + +# From top-level Makefile +# NOSTDINC_FLAGS +VDSO_CPPFLAGS := -nostdinc -isystem $(shell $(COMPATCC) -print-file-name=include) +VDSO_CPPFLAGS += $(LINUXINCLUDE) +VDSO_CPPFLAGS += $(KBUILD_CPPFLAGS) + +# Common C and assembly flags +# From top-level Makefile +VDSO_CAFLAGS := $(VDSO_CPPFLAGS) +VDSO_CAFLAGS += $(call cc32-option,-fno-PIE) +ifdef CONFIG_DEBUG_INFO +VDSO_CAFLAGS += -g +endif +ifeq ($(shell $(CONFIG_SHELL) $(srctree)/scripts/gcc-goto.sh $(COMPATCC)), y) +VDSO_CAFLAGS += -DCC_HAVE_ASM_GOTO +endif + +# From arm Makefile +VDSO_CAFLAGS += $(call cc32-option,-fno-dwarf2-cfi-asm) +VDSO_CAFLAGS += -mabi=aapcs-linux -mfloat-abi=soft +ifeq ($(CONFIG_CPU_BIG_ENDIAN), y) +VDSO_CAFLAGS += -mbig-endian +else +VDSO_CAFLAGS += -mlittle-endian +endif + +# From arm vDSO Makefile +VDSO_CAFLAGS += -fPIC -fno-builtin -fno-stack-protector +VDSO_CAFLAGS += -DDISABLE_BRANCH_PROFILING + +# Try to compile for ARMv8. If the compiler is too old and doesn't support it, +# fall back to v7. There is no easy way to check for what architecture the code +# is being compiled, so define a macro specifying that (see arch/arm/Makefile). +VDSO_CAFLAGS += $(call cc32-option,-march=armv8-a -D__LINUX_ARM_ARCH__=8,\ + -march=armv7-a -D__LINUX_ARM_ARCH__=7) + +VDSO_CFLAGS := $(VDSO_CAFLAGS) +VDSO_CFLAGS += -DENABLE_COMPAT_VDSO=1 +# KBUILD_CFLAGS from top-level Makefile +VDSO_CFLAGS += -Wall -Wundef -Wstrict-prototypes -Wno-trigraphs \ + -fno-strict-aliasing -fno-common \ + -Werror-implicit-function-declaration \ + -Wno-format-security \ + -std=gnu89 +VDSO_CFLAGS += -O2 +# Some useful compiler-dependent flags from top-level Makefile +VDSO_CFLAGS += $(call cc32-option,-Wdeclaration-after-statement,) +VDSO_CFLAGS += $(call cc32-option,-Wno-pointer-sign) +VDSO_CFLAGS += $(call cc32-option,-fno-strict-overflow) +VDSO_CFLAGS += $(call cc32-option,-Werror=strict-prototypes) +VDSO_CFLAGS += $(call cc32-option,-Werror=date-time) +VDSO_CFLAGS += $(call cc32-option,-Werror=incompatible-pointer-types) + +# The 32-bit compiler does not provide 128-bit integers, which are used in +# some headers that are indirectly included from the vDSO code. +# This hack makes the compiler happy and should trigger a warning/error if +# variables of such type are referenced. +VDSO_CFLAGS += -D__uint128_t='void*' +# Silence some warnings coming from headers that operate on long's +# (on GCC 4.8 or older, there is unfortunately no way to silence this warning) +VDSO_CFLAGS += $(call cc32-disable-warning,shift-count-overflow) +VDSO_CFLAGS += -Wno-int-to-pointer-cast + +VDSO_AFLAGS := $(VDSO_CAFLAGS) +VDSO_AFLAGS += -D__ASSEMBLY__ + +VDSO_LDFLAGS := $(VDSO_CPPFLAGS) +# From arm vDSO Makefile +VDSO_LDFLAGS += -Wl,-Bsymbolic -Wl,--no-undefined -Wl,-soname=linux-vdso.so.1 +VDSO_LDFLAGS += -Wl,-z,max-page-size=4096 -Wl,-z,common-page-size=4096 +VDSO_LDFLAGS += -nostdlib -shared -mfloat-abi=soft +VDSO_LDFLAGS += $(call cc32-ldoption,-Wl$(comma)--hash-style=sysv) +VDSO_LDFLAGS += $(call cc32-ldoption,-Wl$(comma)--build-id) +VDSO_LDFLAGS += $(call cc32-ldoption,-fuse-ld=bfd) + + +# Borrow vdsomunge.c from the arm vDSO +# We have to use a relative path because scripts/Makefile.host prefixes +# $(hostprogs-y) with $(obj) +munge := ../../../arm/vdso/vdsomunge +hostprogs-y := $(munge) + +c-obj-vdso := note.o +c-obj-vdso-gettimeofday := vgettimeofday.o +asm-obj-vdso := sigreturn.o + +ifneq ($(c-gettimeofday-y),) +VDSO_CFLAGS_gettimeofday_o += -include $(c-gettimeofday-y) +endif + +VDSO_CFLAGS_REMOVE_vgettimeofday.o = $(CC_FLAGS_FTRACE) -Os + +# Build rules +targets := $(c-obj-vdso) $(c-obj-vdso-gettimeofday) $(asm-obj-vdso) vdso.so vdso.so.dbg vdso.so.raw +c-obj-vdso := $(addprefix $(obj)/, $(c-obj-vdso)) +c-obj-vdso-gettimeofday := $(addprefix $(obj)/, $(c-obj-vdso-gettimeofday)) +asm-obj-vdso := $(addprefix $(obj)/, $(asm-obj-vdso)) +obj-vdso := $(c-obj-vdso) $(c-obj-vdso-gettimeofday) $(asm-obj-vdso) + +obj-y += vdso.o +extra-y += vdso.lds +CPPFLAGS_vdso.lds += -P -C -U$(ARCH) + +# Force dependency (vdso.s includes vdso.so through incbin) +$(obj)/vdso.o: $(obj)/vdso.so + +include/generated/vdso32-offsets.h: $(obj)/vdso.so.dbg FORCE + $(call if_changed,vdsosym) + +# Strip rule for vdso.so +$(obj)/vdso.so: OBJCOPYFLAGS := -S +$(obj)/vdso.so: $(obj)/vdso.so.dbg FORCE + $(call if_changed,objcopy) + +$(obj)/vdso.so.dbg: $(obj)/vdso.so.raw $(obj)/$(munge) FORCE + $(call if_changed,vdsomunge) + +# Link rule for the .so file, .lds has to be first +$(obj)/vdso.so.raw: $(src)/vdso.lds $(obj-vdso) FORCE + $(call if_changed,vdsold) + $(call if_changed,vdso_check) + +# Compilation rules for the vDSO sources +$(c-obj-vdso): %.o: %.c FORCE + $(call if_changed_dep,vdsocc) +$(c-obj-vdso-gettimeofday): %.o: %.c FORCE + $(call if_changed_dep,vdsocc_gettimeofday) +$(asm-obj-vdso): %.o: %.S FORCE + $(call if_changed_dep,vdsoas) + +# Actual build commands +quiet_cmd_vdsold = VDSOL $@ + cmd_vdsold = $(COMPATCC) -Wp,-MD,$(depfile) $(VDSO_LDFLAGS) \ + -Wl,-T $(filter %.lds,$^) $(filter %.o,$^) -o $@ +quiet_cmd_vdsocc = VDSOC $@ + cmd_vdsocc = $(COMPATCC) -Wp,-MD,$(depfile) $(VDSO_CFLAGS) -c -o $@ $< +quiet_cmd_vdsocc_gettimeofday = VDSOC_GTD $@ + cmd_vdsocc_gettimeofday = $(COMPATCC) -Wp,-MD,$(depfile) $(VDSO_CFLAGS) $(VDSO_CFLAGS_gettimeofday_o) -c -o $@ $< +quiet_cmd_vdsoas = VDSOA $@ + cmd_vdsoas = $(COMPATCC) -Wp,-MD,$(depfile) $(VDSO_AFLAGS) -c -o $@ $< + +quiet_cmd_vdsomunge = MUNGE $@ + cmd_vdsomunge = $(obj)/$(munge) $< $@ + +# Generate vDSO offsets using helper script (borrowed from the 64-bit vDSO) +gen-vdsosym := $(srctree)/$(src)/../vdso/gen_vdso_offsets.sh +quiet_cmd_vdsosym = VDSOSYM $@ +# The AArch64 nm should be able to read an AArch32 binary + cmd_vdsosym = $(NM) $< | $(gen-vdsosym) | LC_ALL=C sort > $@ + +# Install commands for the unstripped file +quiet_cmd_vdso_install = INSTALL $@ + cmd_vdso_install = cp $(obj)/$@.dbg $(MODLIB)/vdso/vdso32.so + +vdso.so: $(obj)/vdso.so.dbg + @mkdir -p $(MODLIB)/vdso + $(call cmd,vdso_install) + +vdso_install: vdso.so diff --git a/arch/arm64/kernel/vdso32/note.c b/arch/arm64/kernel/vdso32/note.c new file mode 100644 index 000000000000..eff5bf9efb8b --- /dev/null +++ b/arch/arm64/kernel/vdso32/note.c @@ -0,0 +1,15 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Copyright (C) 2012-2018 ARM Limited + * + * This supplies .note.* sections to go into the PT_NOTE inside the vDSO text. + * Here we can supply some information useful to userland. + */ + +#include <linux/uts.h> +#include <linux/version.h> +#include <linux/elfnote.h> +#include <linux/build-salt.h> + +ELFNOTE32("Linux", 0, LINUX_VERSION_CODE); +BUILD_SALT; diff --git a/arch/arm64/kernel/vdso32/sigreturn.S b/arch/arm64/kernel/vdso32/sigreturn.S new file mode 100644 index 000000000000..1a81277c2d09 --- /dev/null +++ b/arch/arm64/kernel/vdso32/sigreturn.S @@ -0,0 +1,62 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* + * This file provides both A32 and T32 versions, in accordance with the + * arm sigreturn code. + * + * Copyright (C) 2018 ARM Limited + */ + +#include <linux/linkage.h> +#include <asm/asm-offsets.h> +#include <asm/unistd.h> + +#define ARM_ENTRY(name) \ + ENTRY(name) + +#define ARM_ENDPROC(name) \ + .type name, %function; \ + END(name) + + .text + + .arm + .fnstart + .save {r0-r15} + .pad #COMPAT_SIGFRAME_REGS_OFFSET + nop +ARM_ENTRY(__kernel_sigreturn_arm) + mov r7, #__NR_compat_sigreturn + svc #0 + .fnend +ARM_ENDPROC(__kernel_sigreturn_arm) + + .fnstart + .save {r0-r15} + .pad #COMPAT_RT_SIGFRAME_REGS_OFFSET + nop +ARM_ENTRY(__kernel_rt_sigreturn_arm) + mov r7, #__NR_compat_rt_sigreturn + svc #0 + .fnend +ARM_ENDPROC(__kernel_rt_sigreturn_arm) + + .thumb + .fnstart + .save {r0-r15} + .pad #COMPAT_SIGFRAME_REGS_OFFSET + nop +ARM_ENTRY(__kernel_sigreturn_thumb) + mov r7, #__NR_compat_sigreturn + svc #0 + .fnend +ARM_ENDPROC(__kernel_sigreturn_thumb) + + .fnstart + .save {r0-r15} + .pad #COMPAT_RT_SIGFRAME_REGS_OFFSET + nop +ARM_ENTRY(__kernel_rt_sigreturn_thumb) + mov r7, #__NR_compat_rt_sigreturn + svc #0 + .fnend +ARM_ENDPROC(__kernel_rt_sigreturn_thumb) diff --git a/arch/arm64/kernel/vdso32/vdso.S b/arch/arm64/kernel/vdso32/vdso.S new file mode 100644 index 000000000000..e72ac7bc4c04 --- /dev/null +++ b/arch/arm64/kernel/vdso32/vdso.S @@ -0,0 +1,19 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* + * Copyright (C) 2012 ARM Limited + */ + +#include <linux/init.h> +#include <linux/linkage.h> +#include <linux/const.h> +#include <asm/page.h> + + .globl vdso32_start, vdso32_end + .section .rodata + .balign PAGE_SIZE +vdso32_start: + .incbin "arch/arm64/kernel/vdso32/vdso.so" + .balign PAGE_SIZE +vdso32_end: + + .previous diff --git a/arch/arm64/kernel/vdso32/vdso.lds.S b/arch/arm64/kernel/vdso32/vdso.lds.S new file mode 100644 index 000000000000..a3944927eaeb --- /dev/null +++ b/arch/arm64/kernel/vdso32/vdso.lds.S @@ -0,0 +1,82 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* + * Adapted from arm64 version. + * + * GNU linker script for the VDSO library. + * Heavily based on the vDSO linker scripts for other archs. + * + * Copyright (C) 2012-2018 ARM Limited + */ + +#include <linux/const.h> +#include <asm/page.h> +#include <asm/vdso.h> + +OUTPUT_FORMAT("elf32-littlearm", "elf32-bigarm", "elf32-littlearm") +OUTPUT_ARCH(arm) + +SECTIONS +{ + PROVIDE_HIDDEN(_vdso_data = . - PAGE_SIZE); + . = VDSO_LBASE + SIZEOF_HEADERS; + + .hash : { *(.hash) } :text + .gnu.hash : { *(.gnu.hash) } + .dynsym : { *(.dynsym) } + .dynstr : { *(.dynstr) } + .gnu.version : { *(.gnu.version) } + .gnu.version_d : { *(.gnu.version_d) } + .gnu.version_r : { *(.gnu.version_r) } + + .note : { *(.note.*) } :text :note + + .dynamic : { *(.dynamic) } :text :dynamic + + .rodata : { *(.rodata*) } :text + + .text : { *(.text*) } :text =0xe7f001f2 + + .got : { *(.got) } + .rel.plt : { *(.rel.plt) } + + /DISCARD/ : { + *(.note.GNU-stack) + *(.data .data.* .gnu.linkonce.d.* .sdata*) + *(.bss .sbss .dynbss .dynsbss) + } +} + +/* + * We must supply the ELF program headers explicitly to get just one + * PT_LOAD segment, and set the flags explicitly to make segments read-only. + */ +PHDRS +{ + text PT_LOAD FLAGS(5) FILEHDR PHDRS; /* PF_R|PF_X */ + dynamic PT_DYNAMIC FLAGS(4); /* PF_R */ + note PT_NOTE FLAGS(4); /* PF_R */ +} + +VERSION +{ + LINUX_2.6 { + global: + __vdso_clock_gettime; + __vdso_gettimeofday; + __vdso_clock_getres; + __kernel_sigreturn_arm; + __kernel_sigreturn_thumb; + __kernel_rt_sigreturn_arm; + __kernel_rt_sigreturn_thumb; + __vdso_clock_gettime64; + local: *; + }; +} + +/* + * Make the sigreturn code visible to the kernel. + */ +VDSO_compat_sigreturn_arm = __kernel_sigreturn_arm; +VDSO_compat_sigreturn_thumb = __kernel_sigreturn_thumb; +VDSO_compat_rt_sigreturn_arm = __kernel_rt_sigreturn_arm; +VDSO_compat_rt_sigreturn_thumb = __kernel_rt_sigreturn_thumb; diff --git a/arch/arm64/kernel/vdso32/vgettimeofday.c b/arch/arm64/kernel/vdso32/vgettimeofday.c new file mode 100644 index 000000000000..54fc1c2ce93f --- /dev/null +++ b/arch/arm64/kernel/vdso32/vgettimeofday.c @@ -0,0 +1,59 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * ARM64 compat userspace implementations of gettimeofday() and similar. + * + * Copyright (C) 2018 ARM Limited + * + */ +#include <linux/time.h> +#include <linux/types.h> + +int __vdso_clock_gettime(clockid_t clock, + struct old_timespec32 *ts) +{ + /* The checks below are required for ABI consistency with arm */ + if ((u32)ts >= TASK_SIZE_32) + return -EFAULT; + + return __cvdso_clock_gettime32(clock, ts); +} + +int __vdso_clock_gettime64(clockid_t clock, + struct __kernel_timespec *ts) +{ + /* The checks below are required for ABI consistency with arm */ + if ((u32)ts >= TASK_SIZE_32) + return -EFAULT; + + return __cvdso_clock_gettime(clock, ts); +} + +int __vdso_gettimeofday(struct __kernel_old_timeval *tv, + struct timezone *tz) +{ + return __cvdso_gettimeofday(tv, tz); +} + +int __vdso_clock_getres(clockid_t clock_id, + struct old_timespec32 *res) +{ + /* The checks below are required for ABI consistency with arm */ + if ((u32)res >= TASK_SIZE_32) + return -EFAULT; + + return __cvdso_clock_getres_time32(clock_id, res); +} + +/* Avoid unresolved references emitted by GCC */ + +void __aeabi_unwind_cpp_pr0(void) +{ +} + +void __aeabi_unwind_cpp_pr1(void) +{ +} + +void __aeabi_unwind_cpp_pr2(void) +{ +}