java - How can the offset for a JVM jump instruction be 32768? -


while writing answer question jvm byte code offsets, noticed in behavior of javac , resulting class files can not explain:

when compiling class this

class farjump {     public static void main(string args[])     {         call(0, 1);     }      public static void call(int x, int y)     {         if (x < y)         {             y++;             y++;              // ... (10921 times - code post here!)              y++;             y++;         }         system.out.println(y);     }  }

then resulting byte code contain following if_icmpge instruction:

public static void call(int, int);     code:        0: iload_0        1: iload_1        2: if_icmpge     32768        5: iinc          1, 1        8: iinc          1, 1        ...

according documentation of jump instructions, offset (which 32768 in case) computed follows:

if comparison succeeds, unsigned branchbyte1 , branchbyte2 used construct signed 16-bit offset, offset calculated (branchbyte1 << 8) | branchbyte2.

so offset said signed 16 bit value. however, maximum value signed 16 bit value can hold 32767, , not 32768.

the resulting class file still seems valid, , can executed normally.

i had @ bytecode checking in openjdk, , seems (to me) valid due parentheses being misplaced:

int jump = (((signed char)(code[offset+1])) << 8) + code[offset+2];

it cast first byte signed char. then apply shift, , add second byte. have expected be

int jump = (((signed char)(code[offset+1]) << 8)) + code[offset+2];

or maybe even

int jump = (signed char)((code[offset+1]) << 8) + code[offset+2]);

but i'm not familiar type promotions , possible compiler-specific caveats of shifting signed , unsigned types, i'm not sure whether there deeper meaning behind cast...

so jump offset of 32768 comply specification? , jump computation code in openjdk make sense in regard?

the argument if_icmpge offset, javap shows jump target absolute position. is, javap should show getstatic @ 32768: , not 32770: (i.e., 2 + 32768).


Comments

Post a Comment

Popular posts from this blog

Email notification in google apps script -

https://support.google.com/docs/answer/91588?hl=en i have sheet the cells filled in data pulled via apps script, website. pull in exact same data, not. on old sheets version use able emails when data "changes" different value. new sheets, email when new data pasted cells, though exact same data......and, can't set @ ranges anymore, entire document. there easy way watch cells when data changes, , email myself? this looks lot you're trying do. for checking whether value has changed after edit, this gives necessary information it. hence, know whether new value same 1 or not. implementing "workaround" talked in second link, able compare , send email myself if value changed. initiate onedit trigger goes off onedit function uses these lines after checks in place have passed. var email = "xyz@whatever.com"; mailapp.sendemail(email, "triggermail", "a change has occurred in sheet"); note: sends "notification...
Read more

c++ - Difference between pre and post decrement in recursive function argument -

i have following sample code used pre-decrement void function(int c) { if(c == 0) { return; } else { cout << "dp" << c << endl; function(--c); cout << c << endl; return; } } this function give output input 4 : dp3 dp2 dp1 dp0 0 1 2 3 but when use post decrement void function(int c) { if(c == 0) { return; } else { cout << "dp" << c << endl; function(c--); cout << c << endl; return; } } output goes in infinite loop dp4 can explain me in detail why happens ? this happens because function(c--) called same value of c , when finishes c decremented. recursively called, hence called same value no chance of return , hit stack overflow error. assume this, int x = 1, y = 0; y = x++ result y == 1 , x == 2. if y = ++x , bot x , y 2.
Read more

javascript - IE11 incompatibility with jQuery's 'readonly'? -

the code i'm running has no issues in firefox, or chrome. user typing 1 textbox, tabs next , can continue typing that. when run webpage on ie11, user tabs next textbox , can't type in. textbox has double-clicked before user can insert anything. i've googled , seems ie give problems when using older jquery (i.e. prefers 'prop' 'attr' ). in code below jquery's readonly keyword being used in way ie11 no longer recognise/accept? i assume issue readonly , removeclass('ignore') working expected. //when focus on textbox 1 $("#field1").focus(function() { //remove readonly property textbox 1 $('#field1').prop('readonly', false).removeclass('ignore'); //add readonly property other textboxes $('#field2, #field3, #field4') .prop('readonly', 'readonly').addclass('ignore').val(''); }); i'm using jquery 2.1.3 ...
Read more