diff options
| author | Robin Getz <robin.getz@analog.com> | 2007-10-09 17:24:30 +0800 |
|---|---|---|
| committer | Bryan Wu <bryan.wu@analog.com> | 2007-10-09 17:24:30 +0800 |
| commit | 2ebcade590dcf822dcdadcc4f8f68efd3ff2e217 (patch) | |
| tree | 38532077386784de87ac65aba73e22c399c96ecb /arch/blackfin/mach-common | |
| parent | 0ae53640b54f2c30e52044f7102ba08915b988a7 (diff) | |
Blackfin arch: fix endless loop bug when a double fault happens
Today when a double fault happens (exception during an exception
handling event), we go into an endless loop, with nothing comming out
the UART. With this patch, we actually see that we have commited a
double fault event
Signed-off-by: Robin Getz <robin.getz@analog.com>
Signed-off-by: Bryan Wu <bryan.wu@analog.com>
Diffstat (limited to 'arch/blackfin/mach-common')
| -rw-r--r-- | arch/blackfin/mach-common/entry.S | 69 |
1 files changed, 54 insertions, 15 deletions
diff --git a/arch/blackfin/mach-common/entry.S b/arch/blackfin/mach-common/entry.S index 3feca05694f..e2239361cac 100644 --- a/arch/blackfin/mach-common/entry.S +++ b/arch/blackfin/mach-common/entry.S @@ -29,21 +29,7 @@ * 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ -/* - * 25-Dec-2004 - LG Soft India - * 1. Fix in return_from_int, to make sure any pending - * system call in ILAT for this process to get - * executed, otherwise in case context switch happens, - * system call of first process (i.e in ILAT) will be - * carried forward to the switched process. - * 2. Removed Constant references for the following - * a. IPEND - * b. EXCAUSE mask - * c. PAGE Mask - */ - -/* - * NOTE: This code handles signal-recognition, which happens every time +/* NOTE: This code handles signal-recognition, which happens every time * after a timer-interrupt and after each system call. */ @@ -175,6 +161,13 @@ ENTRY(_ex_replaceable) nop; ENTRY(_ex_trap_c) + /* Make sure we are not in a double fault */ + p4.l = lo(IPEND); + p4.h = hi(IPEND); + r7 = [p4]; + CC = BITTST (r7, 5); + if CC jump _double_fault; + /* Call C code (trap_c) to handle the exception, which most * likely involves sending a signal to the current process. * To avoid double faults, lower our priority to IRQ5 first. @@ -220,6 +213,52 @@ ENTRY(_ex_trap_c) rtx; ENDPROC(_ex_trap_c) +/* We just realized we got an exception, while we were processing a different + * exception. This is a unrecoverable event, so crash + */ +ENTRY(_double_fault) + /* Turn caches & protection off, to ensure we don't get any more + * double exceptions + */ + + P4.L = LO(IMEM_CONTROL); + P4.H = HI(IMEM_CONTROL); + + R5 = [P4]; /* Control Register*/ + BITCLR(R5,ENICPLB_P); + SSYNC; /* SSYNC required before writing to IMEM_CONTROL. */ + .align 8; + [P4] = R5; + SSYNC; + + P4.L = LO(DMEM_CONTROL); + P4.H = HI(DMEM_CONTROL); + R5 = [P4]; + BITCLR(R5,ENDCPLB_P); + SSYNC; /* SSYNC required before writing to DMEM_CONTROL. */ + .align 8; + [P4] = R5; + SSYNC; + + /* Fix up the stack */ + (R7:6,P5:4) = [sp++]; + ASTAT = [sp++]; + SP = EX_SCRATCH_REG; + + /* We should be out of the exception stack, and back down into + * kernel or user space stack + */ + SAVE_ALL_SYS + + r0 = sp; /* stack frame pt_regs pointer argument ==> r0 */ + SP += -12; + call _double_fault_c; + SP += 12; +.L_double_fault_panic: + JUMP .L_double_fault_panic + +ENDPROC(_double_fault) + ENTRY(_exception_to_level5) SAVE_ALL_SYS |