m68k: clean up and merge MMU and non-MMU versions of entry.S
Subsystems (2)
cover m68k: clean up and merge MMU and non-MMU versions of entry.S
m68k: clean up and merge the MMU and non-MMU versions of entry.S
With a little cleanup we can use much of the entry.S code on both the MMU and non-MMU m68k targets.
--- arch/m68k/kernel/entry.S | 452 +++++++++++++++++++++++++++++++++++++++++++- arch/m68k/kernel/entry_mm.S | 422 ----------------------------------------- arch/m68k/kernel/entry_no.S | 130 ------------ 3 files changed, 451 insertions(+), 553 deletions(-)
PATCH 1/3 m68k: consistently call ret_from_exception in entry code
From: Greg Ungerer <[email protected]>
The ret_from_excption code is referenced by its function name, or by a label set at the start of its code. Modify all uses to consistently just use the function name "ret_from_exception".
This will help when we come to merge the MMU and non-MMU entry code.
--- arch/m68k/kernel/entry_mm.S | 13 ++++++------- 1 files changed, 6 insertions(+), 7 deletions(-)
arch/m68k/kernel/entry_mm.S +6 -7
@@ -57,7 +57,7 @@ ENTRY(buserr) movel %sp,%sp@- | stack frame pointer argument bsrl buserr_c addql #4,%sp - jra .Lret_from_exception + jra ret_from_exception ENTRY(trap) SAVE_ALL_INT
@@ -65,7 +65,7 @@ ENTRY(trap) movel %sp,%sp@- | stack frame pointer argument bsrl trap_c addql #4,%sp - jra .Lret_from_exception + jra ret_from_exception | After a fork we jump here directly from resume, | so that %d1 contains the previous task
@@ -74,7 +74,7 @@ ENTRY(ret_from_fork) movel %d1,%sp@- jsr schedule_tail addql #4,%sp - jra .Lret_from_exception + jra ret_from_exception do_trace_entry: movel #-ENOSYS,%sp@(PT_OFF_D0)| needed for strace
@@ -96,7 +96,7 @@ do_trace_exit: jbsr syscall_trace RESTORE_SWITCH_STACK addql #4,%sp - jra .Lret_from_exception + jra ret_from_exception ENTRY(ret_from_signal) movel %curptr@(TASK_STACK),%a1
@@ -115,7 +115,7 @@ ENTRY(ret_from_signal) addql #4,%sp 1: #endif - jra .Lret_from_exception + jra ret_from_exception ENTRY(system_call) SAVE_ALL_SYS
@@ -154,7 +154,6 @@ syscall_exit_work: ENTRY(ret_from_exception) -.Lret_from_exception: btst #5,%sp@(PT_OFF_SR) | check if returning to kernel bnes 1f | if so, skip resched, signals | only allow interrupts when we are really the last one on the
@@ -230,7 +229,7 @@ ret_from_last_interrupt: /* check if we need to do software interrupts */ tstl irq_stat+CPUSTAT_SOFTIRQ_PENDING - jeq .Lret_from_exception + jeq ret_from_exception pea ret_from_exception jra do_softirq
Signature · 1 line
1.7.0.4
Re: [PATCH 1/3] m68k: consistently call ret_from_exception in entry code
Quoted 3 lines · <[email protected]> writes:
The
ret_from_excptioncode is referenced by its function name, or by a label set at the start of its code. Modify all uses to consistently just use the function name "ret_from_exception".
This will cause all branches to use the long variant, because ret_from_exception is a global symbol (and thus overridable in the ELF model). The point of the additional local label is to allow the assembler to relax the branches.
Andreas.
Signature · 3 lines
Andreas Schwab, [email protected] GPG Key fingerprint = 58CA 54C7 6D53 942B 1756 01D3 44D5 214B 8276 4ED5 "And now for something completely different."
Re: [PATCH 1/3] m68k: consistently call ret_from_exception in entry code
Hi Andreas,
Quoted 10 lines · Andreas Schwab
Quoted 3 lines · <[email protected]> writes:
The
ret_from_excptioncode is referenced by its function name, or by a label set at the start of its code. Modify all uses to consistently just use the function name "ret_from_exception".This will cause all branches to use the long variant, because
ret_from_exceptionis a global symbol (and thus overridable in the ELF model). The point of the additional local label is to allow the assembler to relax the branches.
Ah, ok. I was wondering why it was done this way.
The only problem cases for me in merging entry.S are the uses in:
buserr
trap
ret_from_fork
buserr and trap are not hot paths, so do you see any issue making them the long variant?
I don't know that even the ret_from_fork case is too much to worry about either. We have done an awfull lot of work up to this point, a couple of bytes more and a long branch won't make any real difference to performance here.
Regards
Greg
------------------------------------------------------------------------
Greg Ungerer -- Principal Engineer EMAIL: [email protected] SnapGear Group, McAfee PHONE: +61 7 3435 2888 8 Gardner Close FAX: +61 7 3217 5323 Milton, QLD, 4064, Australia WEB: http://www.SnapGear.com
PATCH 2/3 m68k: use jbsr to call functions instead of bsrl
In reply to [email protected]From: Greg Ungerer <[email protected]>
There is a few places that the m68k entry code uses the bsrl instruction to call other functions. That instruction is only supported on 68020 and higher CPU types. If we use jbsr instead the code will be clean for all 68k and ColdFire CPU types.
--- arch/m68k/kernel/entry_mm.S | 4 ++-- 1 files changed, 2 insertions(+), 2 deletions(-)
arch/m68k/kernel/entry_mm.S +2 -2
@@ -55,7 +55,7 @@ ENTRY(buserr) SAVE_ALL_INT GET_CURRENT(%d0) movel %sp,%sp@- | stack frame pointer argument - bsrl buserr_c + jbsr buserr_c addql #4,%sp jra ret_from_exception
@@ -63,7 +63,7 @@ ENTRY(trap) SAVE_ALL_INT GET_CURRENT(%d0) movel %sp,%sp@- | stack frame pointer argument - bsrl trap_c + jbsr trap_c addql #4,%sp jra ret_from_exception
Signature · 9 lines
1.7.0.4 _______________________________________________ uClinux-dev mailing list [email protected] http://mailman.uclinux.org/mailman/listinfo/uclinux-dev This message was resent by [email protected] To unsubscribe see: http://mailman.uclinux.org/mailman/options/uclinux-dev
PATCH 3/3 m68k: merge the MMU and non-MMU versions of the entry.S code
In reply to [email protected]From: Greg Ungerer <[email protected]>
Some of the entry.S code is common to both MMU and non-MMU builds. So merge the entry_no.S and entry_mm.S files back into a single file. With a little code movement we only need a single #ifdef.
---
arch/m68k/kernel/{entry_mm.S => entry.S} | 105 ++++++++++++++++--------
arch/m68k/kernel/entry_no.S | 130 ------------------------------
2 files changed, 69 insertions(+), 166 deletions(-)
rename arch/m68k/kernel/{entry_mm.S => entry.S} (94%)
delete mode 100644 arch/m68k/kernel/entry_no.Sarch/m68k/kernel/entry.S renamed from arch/m68k/kernel/entry_mm.S +69 -36
@@ -51,6 +51,42 @@ .globl user_irqvec_fixup .text +ENTRY(sys_fork) + SAVE_SWITCH_STACK + pea %sp@(SWITCH_STACK_SIZE) + jbsr m68k_fork + addql #4,%sp + RESTORE_SWITCH_STACK + rts + +ENTRY(sys_clone) + SAVE_SWITCH_STACK + pea %sp@(SWITCH_STACK_SIZE) + jbsr m68k_clone + addql #4,%sp + RESTORE_SWITCH_STACK + rts + +ENTRY(sys_vfork) + SAVE_SWITCH_STACK + pea %sp@(SWITCH_STACK_SIZE) + jbsr m68k_vfork + addql #4,%sp + RESTORE_SWITCH_STACK + rts + +ENTRY(sys_sigreturn) + SAVE_SWITCH_STACK + jbsr do_sigreturn + RESTORE_SWITCH_STACK + rts + +ENTRY(sys_rt_sigreturn) + SAVE_SWITCH_STACK + jbsr do_rt_sigreturn + RESTORE_SWITCH_STACK + rts + ENTRY(buserr) SAVE_ALL_INT GET_CURRENT(%d0)
@@ -76,6 +112,38 @@ ENTRY(ret_from_fork) addql #4,%sp jra ret_from_exception +#if defined(CONFIG_COLDFIRE) || !defined(CONFIG_MMU) + +#ifdef TRAP_DBG_INTERRUPT + +.globl dbginterrupt +ENTRY(dbginterrupt) + SAVE_ALL_INT + GET_CURRENT(%d0) + movel %sp,%sp@- /* stack frame pointer argument */ + jsr dbginterrupt_c + addql #4,%sp + jra ret_from_exception +#endif + +ENTRY(reschedule) + /* save top of frame */ + pea %sp@ + jbsr set_esp0 + addql #4,%sp + pea ret_from_exception + jmp schedule + +ENTRY(ret_from_user_signal) + moveq #__NR_sigreturn,%d0 + trap #0 + +ENTRY(ret_from_user_rt_signal) + movel #__NR_rt_sigreturn,%d0 + trap #0 + +#else + do_trace_entry: movel #-ENOSYS,%sp@(PT_OFF_D0)| needed for strace subql #4,%sp
@@ -273,42 +341,6 @@ ENTRY(bad_inthandler) RESTORE_ALL -ENTRY(sys_fork) - SAVE_SWITCH_STACK - pea %sp@(SWITCH_STACK_SIZE) - jbsr m68k_fork - addql #4,%sp - RESTORE_SWITCH_STACK - rts - -ENTRY(sys_clone) - SAVE_SWITCH_STACK - pea %sp@(SWITCH_STACK_SIZE) - jbsr m68k_clone - addql #4,%sp - RESTORE_SWITCH_STACK - rts - -ENTRY(sys_vfork) - SAVE_SWITCH_STACK - pea %sp@(SWITCH_STACK_SIZE) - jbsr m68k_vfork - addql #4,%sp - RESTORE_SWITCH_STACK - rts - -ENTRY(sys_sigreturn) - SAVE_SWITCH_STACK - jbsr do_sigreturn - RESTORE_SWITCH_STACK - rts - -ENTRY(sys_rt_sigreturn) - SAVE_SWITCH_STACK - jbsr do_rt_sigreturn - RESTORE_SWITCH_STACK - rts - resume: /* * Beware - when entering resume, prev (the current task) is
@@ -416,3 +448,4 @@ resume: rts +#endif /* CONFIG_MMU && !CONFIG_COLDFIRE */
arch/m68k/kernel/entry_no.S deleted +0 -130
130 lines not on this page · raw message
Signature · 1 line
1.7.0.4