Revert "py/nlr: Factor out common NLR code to generic functions."

This reverts commit 6a3a742a6c9caaa2be0fd0aac7a5df4ac816081c.

The above commit has number of faults starting from the motivation down
to the actual implementation.

1. Faulty implementation.

The original code contained functions like:

NORETURN void nlr_jump(void *val) {
    nlr_buf_t **top_ptr = &MP_STATE_THREAD(nlr_top);
    nlr_buf_t *top = *top_ptr;
...
     __asm volatile (
    "mov    %0, %%edx           \n" // %edx points to nlr_buf
    "mov    28(%%edx), %%esi    \n" // load saved %esi
    "mov    24(%%edx), %%edi    \n" // load saved %edi
    "mov    20(%%edx), %%ebx    \n" // load saved %ebx
    "mov    16(%%edx), %%esp    \n" // load saved %esp
    "mov    12(%%edx), %%ebp    \n" // load saved %ebp
    "mov    8(%%edx), %%eax     \n" // load saved %eip
    "mov    %%eax, (%%esp)      \n" // store saved %eip to stack
    "xor    %%eax, %%eax        \n" // clear return register
    "inc    %%al                \n" // increase to make 1, non-local return
     "ret                        \n" // return
    :                               // output operands
    : "r"(top)                      // input operands
    :                               // clobbered registers
     );
}

Which clearly stated that C-level variable should be a parameter of the
assembly, whcih then moved it into correct register.

Whereas now it's:

NORETURN void nlr_jump_tail(nlr_buf_t *top) {
    (void)top;

    __asm volatile (
    "mov    28(%edx), %esi      \n" // load saved %esi
    "mov    24(%edx), %edi      \n" // load saved %edi
    "mov    20(%edx), %ebx      \n" // load saved %ebx
    "mov    16(%edx), %esp      \n" // load saved %esp
    "mov    12(%edx), %ebp      \n" // load saved %ebp
    "mov    8(%edx), %eax       \n" // load saved %eip
    "mov    %eax, (%esp)        \n" // store saved %eip to stack
    "xor    %eax, %eax          \n" // clear return register
    "inc    %al                 \n" // increase to make 1, non-local return
    "ret                        \n" // return
    );

    for (;;); // needed to silence compiler warning
}

Which just tries to perform operations on a completely random register (edx
in this case). The outcome is the expected: saving the pure random luck of
the compiler putting the right value in the random register above, there's
a crash.

2. Non-critical assessment.

The original commit message says "There is a small overhead introduced
(typically 1 machine instruction)". That machine instruction is a call
if a compiler doesn't perform tail optimization (happens regularly), and
it's 1 instruction only with the broken code shown above, fixing it
requires adding more. With inefficiencies already presented in the NLR
code, the overhead becomes "considerable" (several times more than 1%),
not "small".

The commit message also says "This eliminates duplicated code.". An
obvious way to eliminate duplication would be to factor out common code
to macros, not introduce overhead and breakage like above.

3. Faulty motivation.

All this started with a report of warnings/errors happening for a niche
compiler. It could have been solved in one the direct ways: a) fixing it
just for affected compiler(s); b) rewriting it in proper assembly (like
it was before BTW); c) by not doing anything at all, MICROPY_NLR_SETJMP
exists exactly to address minor-impact cases like thar (where a) or b) are
not applicable). Instead, a backwards "solution" was put forward, leading
to all the issues above.

The best action thus appears to be revert and rework, not trying to work
around what went haywire in the first place.

1 files changed, 37 insertions, 34 deletions

diff --git a/py/nlr.h b/py/nlr.h
index 012a73c31..1235f1460 100644
--- a/py/nlr.h
+++ b/py/nlr.h
@@ -30,29 +30,29 @@
 // exception handling, basically a stack of setjmp/longjmp buffers
 
 #include <limits.h>
+#include <setjmp.h>
 #include <assert.h>
 
 #include "py/mpconfig.h"
 
-// If MICROPY_NLR_SETJMP is not enabled then auto-detect the machine arch
-// Allow a port to set MICROPY_NLR_NUM_REGS to define their own implementation
-#if !MICROPY_NLR_SETJMP && !defined(MICROPY_NLR_NUM_REGS)
+typedef struct _nlr_buf_t nlr_buf_t;
+struct _nlr_buf_t {
+    // the entries here must all be machine word size
+    nlr_buf_t *prev;
+    void *ret_val; // always a concrete object (an exception instance)
+#if !defined(MICROPY_NLR_SETJMP) || !MICROPY_NLR_SETJMP
 #if defined(__i386__)
-    #define MICROPY_NLR_X86 (1)
-    #define MICROPY_NLR_NUM_REGS (6)
+    void *regs[6];
 #elif defined(__x86_64__)
-    #define MICROPY_NLR_X64 (1)
-    #if defined(__CYGWIN__)
-        #define MICROPY_NLR_NUM_REGS (12)
-    #else
-        #define MICROPY_NLR_NUM_REGS (8)
-    #endif
+  #if defined(__CYGWIN__)
+    void *regs[12];
+  #else
+    void *regs[8];
+  #endif
 #elif defined(__thumb2__) || defined(__thumb__) || defined(__arm__)
-    #define MICROPY_NLR_THUMB (1)
-    #define MICROPY_NLR_NUM_REGS (10)
+    void *regs[10];
 #elif defined(__xtensa__)
-    #define MICROPY_NLR_XTENSA (1)
-    #define MICROPY_NLR_NUM_REGS (10)
+    void *regs[10];
 #else
     #define MICROPY_NLR_SETJMP (1)
     //#warning "No native NLR support for this arch, using setjmp implementation"
@@ -60,39 +60,41 @@
 #endif
 
 #if MICROPY_NLR_SETJMP
-#include <setjmp.h>
-#endif
-
-typedef struct _nlr_buf_t nlr_buf_t;
-struct _nlr_buf_t {
-    // the entries here must all be machine word size
-    nlr_buf_t *prev;
-    void *ret_val; // always a concrete object (an exception instance)
-
-    #if MICROPY_NLR_SETJMP
     jmp_buf jmpbuf;
-    #else
-    void *regs[MICROPY_NLR_NUM_REGS];
-    #endif
+#endif
 
     #if MICROPY_ENABLE_PYSTACK
     void *pystack;
     #endif
 };
 
+// Helper macros to save/restore the pystack state
+#if MICROPY_ENABLE_PYSTACK
+#define MP_NLR_SAVE_PYSTACK(nlr_buf) (nlr_buf)->pystack = MP_STATE_THREAD(pystack_cur)
+#define MP_NLR_RESTORE_PYSTACK(nlr_buf) MP_STATE_THREAD(pystack_cur) = (nlr_buf)->pystack
+#else
+#define MP_NLR_SAVE_PYSTACK(nlr_buf) (void)nlr_buf
+#define MP_NLR_RESTORE_PYSTACK(nlr_buf) (void)nlr_buf
+#endif
+
 #if MICROPY_NLR_SETJMP
+#include "py/mpstate.h"
+
+NORETURN void nlr_setjmp_jump(void *val);
 // nlr_push() must be defined as a macro, because "The stack context will be
 // invalidated if the function which called setjmp() returns."
-// For this case it is safe to call nlr_push_tail() first.
-#define nlr_push(buf) (nlr_push_tail(buf), setjmp((buf)->jmpbuf))
+#define nlr_push(buf) ( \
+    (buf)->prev = MP_STATE_THREAD(nlr_top), \
+    MP_NLR_SAVE_PYSTACK(buf), \
+    MP_STATE_THREAD(nlr_top) = (buf), \
+    setjmp((buf)->jmpbuf))
+#define nlr_pop() { MP_STATE_THREAD(nlr_top) = MP_STATE_THREAD(nlr_top)->prev; }
+#define nlr_jump(val) nlr_setjmp_jump(val)
 #else
 unsigned int nlr_push(nlr_buf_t *);
-#endif
-
-unsigned int nlr_push_tail(nlr_buf_t *top);
 void nlr_pop(void);
 NORETURN void nlr_jump(void *val);
-NORETURN void nlr_jump_tail(nlr_buf_t *top);
+#endif
 
 // This must be implemented by a port.  It's called by nlr_jump
 // if no nlr buf has been pushed.  It must not return, but rather
@@ -121,6 +123,7 @@ NORETURN void nlr_jump_fail(void *val);
 /*
 #define nlr_push(val) \
     printf("nlr_push: before: nlr_top=%p, val=%p\n", MP_STATE_THREAD(nlr_top), val),assert(MP_STATE_THREAD(nlr_top) != val),nlr_push(val)
+#endif
 */
 #endif