| Message ID | 1411775208-14280-1-git-send-email-nathan_lynch@mentor.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
On 27 September 2014 01:46, Nathan Lynch <nathan_lynch@mentor.com> wrote: > Joachim Eastwood reports that commit fbfb872f5f41 "ARM: 8148/1: flush > TLS and thumbee register state during exec" causes a boot-time crash > on a Cortex-M4 nommu system: > > Freeing unused kernel memory: 68K (281e5000 - 281f6000) > Unhandled exception: IPSR = 00000005 LR = fffffff1 > CPU: 0 PID: 1 Comm: swapper Not tainted 3.17.0-rc6-00313-gd2205fa30aa7 #191 > task: 29834000 ti: 29832000 task.ti: 29832000 > PC is at flush_thread+0x2e/0x40 > LR is at flush_thread+0x21/0x40 > pc : [<2800954a>] lr : [<2800953d>] psr: 4100000b > sp : 29833d60 ip : 00000000 fp : 00000001 > r10: 00003cf8 r9 : 29b1f000 r8 : 00000000 > r7 : 29b0bc00 r6 : 29834000 r5 : 29832000 r4 : 29832000 > r3 : ffff0ff0 r2 : 29832000 r1 : 00000000 r0 : 282121f0 > xPSR: 4100000b > CPU: 0 PID: 1 Comm: swapper Not tainted 3.17.0-rc6-00313-gd2205fa30aa7 #191 > [<2800afa5>] (unwind_backtrace) from [<2800a327>] (show_stack+0xb/0xc) > [<2800a327>] (show_stack) from [<2800a963>] (__invalid_entry+0x4b/0x4c) > > The problem is that set_tls is attempting to clear the TLS location in > the kernel-user helper page, which isn't set up on V7M. > > Fix this by guarding the write to the kuser helper page with > a CONFIG_KUSER_HELPERS ifdef. > > Reported-by: Joachim Eastwood <manabian@gmail.com> > Fixes: fbfb872f5f41 ARM: 8148/1: flush TLS and thumbee register state during exec > Cc: stable@vger.kernel.org > Signed-off-by: Nathan Lynch <nathan_lynch@mentor.com> > --- Works for me. Tested-by: Joachim Eastwood <manabian@gmail.com> Thanks! regards, Joachim Eastwood
On Sat, Sep 27, 2014 at 11:30:26AM +0100, Joachim Eastwood wrote: > On 27 September 2014 01:46, Nathan Lynch <nathan_lynch@mentor.com> wrote: > > Joachim Eastwood reports that commit fbfb872f5f41 "ARM: 8148/1: flush > > TLS and thumbee register state during exec" causes a boot-time crash > > on a Cortex-M4 nommu system: > > > > Freeing unused kernel memory: 68K (281e5000 - 281f6000) > > Unhandled exception: IPSR = 00000005 LR = fffffff1 > > CPU: 0 PID: 1 Comm: swapper Not tainted 3.17.0-rc6-00313-gd2205fa30aa7 #191 > > task: 29834000 ti: 29832000 task.ti: 29832000 > > PC is at flush_thread+0x2e/0x40 > > LR is at flush_thread+0x21/0x40 > > pc : [<2800954a>] lr : [<2800953d>] psr: 4100000b > > sp : 29833d60 ip : 00000000 fp : 00000001 > > r10: 00003cf8 r9 : 29b1f000 r8 : 00000000 > > r7 : 29b0bc00 r6 : 29834000 r5 : 29832000 r4 : 29832000 > > r3 : ffff0ff0 r2 : 29832000 r1 : 00000000 r0 : 282121f0 > > xPSR: 4100000b > > CPU: 0 PID: 1 Comm: swapper Not tainted 3.17.0-rc6-00313-gd2205fa30aa7 #191 > > [<2800afa5>] (unwind_backtrace) from [<2800a327>] (show_stack+0xb/0xc) > > [<2800a327>] (show_stack) from [<2800a963>] (__invalid_entry+0x4b/0x4c) > > > > The problem is that set_tls is attempting to clear the TLS location in > > the kernel-user helper page, which isn't set up on V7M. > > > > Fix this by guarding the write to the kuser helper page with > > a CONFIG_KUSER_HELPERS ifdef. > > > > Reported-by: Joachim Eastwood <manabian@gmail.com> > > Fixes: fbfb872f5f41 ARM: 8148/1: flush TLS and thumbee register state during exec > > Cc: stable@vger.kernel.org > > Signed-off-by: Nathan Lynch <nathan_lynch@mentor.com> > > --- > > Works for me. > > Tested-by: Joachim Eastwood <manabian@gmail.com> Out of interest, which M4 board are you using to run Linux? Are they generally available? Will
On 29 September 2014 11:03, Will Deacon <will.deacon@arm.com> wrote: > On Sat, Sep 27, 2014 at 11:30:26AM +0100, Joachim Eastwood wrote: >> On 27 September 2014 01:46, Nathan Lynch <nathan_lynch@mentor.com> wrote: >> > Joachim Eastwood reports that commit fbfb872f5f41 "ARM: 8148/1: flush >> > TLS and thumbee register state during exec" causes a boot-time crash >> > on a Cortex-M4 nommu system: >> > >> > Freeing unused kernel memory: 68K (281e5000 - 281f6000) >> > Unhandled exception: IPSR = 00000005 LR = fffffff1 >> > CPU: 0 PID: 1 Comm: swapper Not tainted 3.17.0-rc6-00313-gd2205fa30aa7 #191 >> > task: 29834000 ti: 29832000 task.ti: 29832000 >> > PC is at flush_thread+0x2e/0x40 >> > LR is at flush_thread+0x21/0x40 >> > pc : [<2800954a>] lr : [<2800953d>] psr: 4100000b >> > sp : 29833d60 ip : 00000000 fp : 00000001 >> > r10: 00003cf8 r9 : 29b1f000 r8 : 00000000 >> > r7 : 29b0bc00 r6 : 29834000 r5 : 29832000 r4 : 29832000 >> > r3 : ffff0ff0 r2 : 29832000 r1 : 00000000 r0 : 282121f0 >> > xPSR: 4100000b >> > CPU: 0 PID: 1 Comm: swapper Not tainted 3.17.0-rc6-00313-gd2205fa30aa7 #191 >> > [<2800afa5>] (unwind_backtrace) from [<2800a327>] (show_stack+0xb/0xc) >> > [<2800a327>] (show_stack) from [<2800a963>] (__invalid_entry+0x4b/0x4c) >> > >> > The problem is that set_tls is attempting to clear the TLS location in >> > the kernel-user helper page, which isn't set up on V7M. >> > >> > Fix this by guarding the write to the kuser helper page with >> > a CONFIG_KUSER_HELPERS ifdef. >> > >> > Reported-by: Joachim Eastwood <manabian@gmail.com> >> > Fixes: fbfb872f5f41 ARM: 8148/1: flush TLS and thumbee register state during exec >> > Cc: stable@vger.kernel.org >> > Signed-off-by: Nathan Lynch <nathan_lynch@mentor.com> >> > --- >> >> Works for me. >> >> Tested-by: Joachim Eastwood <manabian@gmail.com> > > Out of interest, which M4 board are you using to run Linux? I am using Embedded Artists' LPC4357 Developer's Kit. http://www.embeddedartists.com/products/kits/lpc4357_kit.php >Are they generally available? They have a webshop so you should be able to buy one. Seems to be out of stock now, though. I have a github repo with all my code for LPC4357 here: https://github.com/manabian/linux-lpc regards Joachim Eastwood
diff --git a/arch/arm/include/asm/tls.h b/arch/arm/include/asm/tls.h index 36172adda9d0..5f833f7adba1 100644 --- a/arch/arm/include/asm/tls.h +++ b/arch/arm/include/asm/tls.h @@ -81,6 +81,7 @@ static inline void set_tls(unsigned long val) asm("mcr p15, 0, %0, c13, c0, 3" : : "r" (val)); } else { +#ifdef CONFIG_KUSER_HELPERS /* * User space must never try to access this * directly. Expect your app to break @@ -89,6 +90,7 @@ static inline void set_tls(unsigned long val) * entry-armv.S for details) */ *((unsigned int *)0xffff0ff0) = val; +#endif } }
Joachim Eastwood reports that commit fbfb872f5f41 "ARM: 8148/1: flush TLS and thumbee register state during exec" causes a boot-time crash on a Cortex-M4 nommu system: Freeing unused kernel memory: 68K (281e5000 - 281f6000) Unhandled exception: IPSR = 00000005 LR = fffffff1 CPU: 0 PID: 1 Comm: swapper Not tainted 3.17.0-rc6-00313-gd2205fa30aa7 #191 task: 29834000 ti: 29832000 task.ti: 29832000 PC is at flush_thread+0x2e/0x40 LR is at flush_thread+0x21/0x40 pc : [<2800954a>] lr : [<2800953d>] psr: 4100000b sp : 29833d60 ip : 00000000 fp : 00000001 r10: 00003cf8 r9 : 29b1f000 r8 : 00000000 r7 : 29b0bc00 r6 : 29834000 r5 : 29832000 r4 : 29832000 r3 : ffff0ff0 r2 : 29832000 r1 : 00000000 r0 : 282121f0 xPSR: 4100000b CPU: 0 PID: 1 Comm: swapper Not tainted 3.17.0-rc6-00313-gd2205fa30aa7 #191 [<2800afa5>] (unwind_backtrace) from [<2800a327>] (show_stack+0xb/0xc) [<2800a327>] (show_stack) from [<2800a963>] (__invalid_entry+0x4b/0x4c) The problem is that set_tls is attempting to clear the TLS location in the kernel-user helper page, which isn't set up on V7M. Fix this by guarding the write to the kuser helper page with a CONFIG_KUSER_HELPERS ifdef. Reported-by: Joachim Eastwood <manabian@gmail.com> Fixes: fbfb872f5f41 ARM: 8148/1: flush TLS and thumbee register state during exec Cc: stable@vger.kernel.org Signed-off-by: Nathan Lynch <nathan_lynch@mentor.com> --- arch/arm/include/asm/tls.h | 2 ++ 1 file changed, 2 insertions(+)