Update perlhack.pod - fix link to mirror git repo #1

MartinMcGrath · 2013-12-18T14:28:09Z

Change dead link http://github.com/mirrors/perl to https://github.com/Perl/perl5

to avoid splain output like this on 80-column terminals: rewinddir() attempted on invalid dirhandle foo at -e line 1 (#1) (W io) The dirhandle you tried to do a rewinddir() on is either closed or no t really a dirhandle. Check your control flow.

to the other part of the message. diagnostics.pm won’t find it otherwise: $ perl -Mdiagnostics -we '"foo"->bar' Can't locate object method "bar" via package "foo" (perhaps you forgot to load "foo"?) at -e line 1 (#1) Uncaught exception from user code: Can't locate object method "bar" via package "foo" (perhaps you forgot to load "foo"?) at -e line 1. Now we have this: Can't locate object method "bar" via package "foo" (perhaps you forgot to load "foo"?) at -e line 1 (#1) (F) You called a method on a class that did not exist, and the method could not be found in UNIVERSAL. This often means that a method requires a package that has not been loaded. Uncaught exception from user code: Can't locate object method "bar" via package "foo" (perhaps you forgot to load "foo"?) at -e line 1.

In the op tree, a statement consists of a nextstate/dbstate op (of class cop) followed by the contents of the statement. This cop is created after the statement has been parsed. So if you have nested statements, the outermost statement has the highest sequence number (cop_seq). Every sub (including BEGIN blocks) has a sequence number indicating where it occurs in its containing sub. So BEGIN { } #1 # seq 2 { # seq 1 ... } is indistinguishable from # seq 2 { BEGIN { } #1 # seq 1 ... } because the sequence number of the BEGIN block is 1 in both examples. By reserving a sequence number at the start of every block and using it once the block has finished parsing, we can do this: BEGIN { } #1 # seq 1 { # seq 2 ... } # seq 1 { BEGIN { } #2 # seq 2 ... } and now B::Deparse can tell where to put the blocks. PL_compiling.cop_seq was unused, so this is where I am stashing the pending sequence number.

$#1 is a syntax error.

to the other part of the message. diagnostics.pm wonâ€™t find it otherwise: $ perl -Mdiagnostics -we '"foo"->bar' Can't locate object method "bar" via package "foo" (perhaps you forgot to load "foo"?) at -e line 1 (#1) Uncaught exception from user code: Can't locate object method "bar" via package "foo" (perhaps you forgot to load "foo"?) at -e line 1. Now we have this: Can't locate object method "bar" via package "foo" (perhaps you forgot to load "foo"?) at -e line 1 (#1) (F) You called a method on a class that did not exist, and the method could not be found in UNIVERSAL. This often means that a method requires a package that has not been loaded. Uncaught exception from user code: Can't locate object method "bar" via package "foo" (perhaps you forgot to load "foo"?) at -e line 1. (cherry picked from commit 8af56b9)

8c2e27d changed B::Deparse to deparse $#{1} as that instead of as $#1 which is a syntax error, but also changed $#_ to deparse as $#{_} which isn't necessary, and broke GFUJI/macro-0.06.tar.gz This could be considered a bug in macro.pm, but since we don't need to deparse $#_ as $#{_} let's not do that.

Coverity CID 135025 (#1 of 1): Out-of-bounds read (OVERRUN) 29. overrun-local: Overrunning array addr.sun_path of 108 bytes at byte offset 108 using index addr_len (which evaluates to 108). 864 for (addr_len = 0; addr.sun_path[addr_len] 28. incr: Incrementing addr_len. The value of addr_len may now be up to 108. 865 && addr_len < maxlen; addr_len++); Reported upstream as https://rt.cpan.org/Ticket/Display.html?id=111707

Coverity CID 135025 (Perl#1 of 1): Out-of-bounds read (OVERRUN) 29. overrun-local: Overrunning array addr.sun_path of 108 bytes at byte offset 108 using index addr_len (which evaluates to 108). 864 for (addr_len = 0; addr.sun_path[addr_len] 28. incr: Incrementing addr_len. The value of addr_len may now be up to 108. 865 && addr_len < maxlen; addr_len++); Reported upstream as https://rt.cpan.org/Ticket/Display.html?id=111707

In an eighty-column terminal, this is what it looked like before this commit: $ ./perl -Ilib -Mdiagnostics -E 'state ($x)=1' Initialization of state variables in list context currently forbidden at -e line 1, at EOF Execution of -e aborted due to compilation errors (#1) (F) state only permits initializing a single scalar variable, in scalar context. So state $a = 42 is allowed, but not state ($a) = 42. To apply state semantics to a hash or array, store a hash or array reference in a sca lar variable.

…sv.c -svfix.pl is quick throwaway garbage done in 20 mins and probably doesnt regen sv_inline.h properly and copy pasting/replace regexps were was used anyway to fix up the code, it can be rewritten correctly tho and put in as a official regen.pl script how to finish this fix, options set and unset sv_type as a CPP macro then -#include a header 17 times that contains ONLY only Perl_newSV_type() and its mortal() sister creating 17*2 static inline fns (basically what I did here), code is stepable, extra ms'es of I/O build times perf degrade debate may or may not come up, I hope the CC has a sane in memory cache for .h files and doesn't go back to the kernel or put the entire Perl_newSV_type() fnc in a #define "#define PETS blah(foo(myarg)) + \ cat(dog(fur)) + \ laser(ball(toy)) " then execute that macro 17 times or write 17 Perl_newSV_type() copies into sv_inline.h with /regen.pl infrastructure (fastest for build speed core and build speed CPAN and code is c dbg stepable) OR is a dedicated "sv_newg.h" for regen.pl needed? does the master Perl_newSV_type() template live in a .pl or a .h? i dont have an opinion or against concept of sv_inline.h just have 5-10 hand written versions sv type specific of Perl_newSV_type(), its a cheap gimick fix to keep all 17 types together mashed with if/else/switch in 1 func and expecting bug free perfection from LTO engines of various C compilers, and expecting perfection from an single vendor LTO engine is very against the spirit of portable code -todo ideas, turn those super long #define ==?:==?:==?: into char array/struct initializers, stored in macros, one faux-string per each column of struct body_details, use that macro as c auto stk rw array initializer, then do the U32 len [3] = "\x01\x02\x03"[sv_type]; or U8 sizes [3] = {1,2,3}; U32 len = sizes[sv_type]; which in perl core would look U32 arena_size = SVDB_AR_SZ_DECL; U32 len = arena_size[sv_type]; maybe VC will optimize those since no global memory is used. Only Perl_newSV_typeX() needs this. in this commit static inline Perl_newSV_typeX(pTHX_ const svtype type) which is the ONLY Perl_newSV_type*() variant that take an arbitrary svtype arg, this is the fallback for gv_pvn_add_by() since I couldn't "const" that call to newSV_type() cuz gv_pvn_add_by() is only place in the whole core that takes a random SV type number. Internals of Perl_newSV_typeX() are trashy, here is an example, MSVC DID not turn this into a jump table but instead 17 test/cond_jump ops. v5 = (char *)S_new_body(v2); v6 = 40i64; if ( v2 == 15 ) v6 = 136i64; if ( v2 == 14 ) v6 = 104i64; if ( v2 == 13 ) v6 = 104i64; if ( v2 == 12 ) v6 = 32i64; if ( v2 == 11 ) v6 = 40i64; if ( v2 == 10 ) v6 = 80i64; if ( v2 == 9 ) v6 = 48i64; if ( v2 == 8 ) v6 = 224i64; if ( v2 == 7 ) v6 = 48i64; if ( v2 == 6 ) v6 = 32i64; if ( v2 == 5 ) v6 = 24i64; if ( v2 == 4 ) v6 = 40i64; if ( v2 == 3 ) v6 = 16i64; if ( v2 == 2 ) v6 = 0i64; if ( v2 == 1 ) v6 = 0i64; memset(v5, 0, v6 & -(signed __int64)(v2 != 0)); Solution is move Perl_newSV_typeX() to sv.c, and let it be struct body_details driven. Cuz it only purpose is when newSV_type() absolutly CAN NOT be constant folded (random number input). it only has 1 caller in core. S_new_body() properly const folded away in 99% of cases except for TWO callers Perl_newSV_typeX() and Perl_make_trie(). Perl_make_trie() failure to inline is bizzare, since Perl_make_trie() internally does "v9 = S_new_body(SVt_PVAV);" and DID inline Perl_newSV_typeSVt_PVAV() !!! and therefore Perl_make_trie() has the AV field initing/nulling code. Here is the "optimized" contents of S_new_body(), its junk performance/design wise (but runtime correct/no bugs) void **__fastcall S_new_body(svtype sv_type) { svtype v1; // er9 __int64 v2; // rbx void **result; // rax signed int v4; // ecx signed __int64 v5; // rax v1 = sv_type; v2 = sv_type; result = (void **)PL_body_roots[sv_type]; if ( !result ) { v4 = 4080; if ( v1 == 15 ) v4 = 3264; if ( v1 == 14 ) v4 = 2080; if ( v1 == 13 ) v4 = 4056; if ( v1 == 12 ) v4 = 4064; if ( v1 == 11 ) v4 = 4080; if ( v1 == 10 ) v4 = 4080; if ( v1 == 9 ) v4 = 4080; if ( v1 == 8 ) v4 = 4032; if ( v1 == 7 ) v4 = 4080; if ( v1 == 6 ) v4 = 3296; if ( v1 == 5 ) v4 = 3424; if ( v1 == 4 ) v4 = 3120; if ( v1 == 3 ) v4 = 3536; if ( v1 == 2 ) v4 = 0; if ( v1 == 1 ) v4 = 0; v5 = 40i64; if ( v1 == 15 ) v5 = 136i64; if ( v1 == 14 ) v5 = 104i64; if ( v1 == 13 ) v5 = 104i64; if ( v1 == 12 ) v5 = 32i64; if ( v1 == 11 ) v5 = 40i64; if ( v1 == 10 ) v5 = 80i64; if ( v1 == 9 ) v5 = 48i64; if ( v1 == 8 ) v5 = 224i64; if ( v1 == 7 ) v5 = 48i64; if ( v1 == 6 ) v5 = 32i64; if ( v1 == 5 ) v5 = 24i64; if ( v1 == 4 ) v5 = 40i64; if ( v1 == 3 ) v5 = 16i64; if ( v1 == 2 ) v5 = 0i64; if ( v1 == 1 ) v5 = 0i64; result = (void **)Perl_more_bodies(v1, v5 & -(signed __int64)(v1 != 0), v4 & (unsigned int)-(v1 != 0)); } PL_body_roots[v2] = *result; return result; } ------------ disassembly view of S_new_body() ------------ cmp r9d, 0Fh lea edi, [rbp+28h] mov r8d, 0FF0h lea r11d, [rbp+20h] mov edx, 0CC0h lea r10d, [rbp+30h] mov ecx, r8d mov eax, r9d cmovz ecx, edx cmp r9d, 0Eh mov edx, 820h cmovz ecx, edx cmp r9d, 0Dh lea edx, [r8-18h] cmovz ecx, edx cmp r9d, 0Ch lea edx, [r8-10h] cmovz ecx, edx cmp r9d, 0Bh lea edx, [r8-30h] cmovz ecx, r8d cmp r9d, 0Ah cmovz ecx, r8d cmp r9d, 9 cmovz ecx, r8d cmp r9d, 8 cmovz ecx, edx cmp r9d, 7 mov edx, 0CE0h cmovz ecx, r8d cmp r9d, 6 cmovz ecx, edx cmp r9d, 5 mov edx, 0D60h cmovz ecx, edx cmp r9d, 4 mov edx, 0C30h cmovz ecx, edx cmp r9d, 3 mov edx, 0DD0h cmovz ecx, edx cmp r9d, 2 lea edx, [rdi+60h] cmovz ecx, ebp cmp r9d, 1 cmovz ecx, ebp neg eax sbb eax, eax and eax, ecx mov ecx, r9d mov r8d, eax cmp r9d, 0Fh mov eax, edi cmovz eax, edx cmp r9d, 0Eh lea edx, [rbp+68h] cmovz eax, edx cmp r9d, 0Dh cmovz eax, edx cmp r9d, 0Ch lea edx, [rbp+50h] cmovz eax, r11d cmp r9d, 0Bh cmovz eax, edi cmp r9d, 0Ah cmovz eax, edx cmp r9d, 9 mov edx, 0E0h cmovz eax, r10d cmp r9d, 8 cmovz eax, edx cmp r9d, 7 lea edx, [rbp+18h] cmovz eax, r10d cmp r9d, 6 cmovz eax, r11d cmp r9d, 5 cmovz eax, edx cmp r9d, 4 lea edx, [rbp+10h] cmovz eax, edi cmp r9d, 3 cmovz eax, edx cmp r9d, 2 cmovz eax, ebp cmp r9d, 1 cmovz eax, ebp neg ecx mov ecx, r9d sbb rdx, rdx and rdx, rax call Perl_more_bodies ---------------------- 17 test ops and 17 conditional_move_constant_8_bits ops solution, turn S_new_body() back into a macro so no CC ever tries to ref-inline it. It was a macro before sv_inline.h branch was merged TODO add XSApitest.xs that worlds longest macros are identical to the master correct copy (struct body_details). byte size drops from before these 3 commits to this "success commit" mp.exe 0x1241AC-0x1224EC=7360 0x19D3D8-0x19B8E8=6896 p541.dll 0x154886-0x1532A6=5600 0x1AA19E-0x1A862E=7024 BEFORE Dump of file ..\miniperl.exe SECTION HEADER Perl#1 .text name 1241AC virtual size SECTION HEADER Perl#2 .rdata name 19D3D8 virtual size Dump of file ..\perl541.dll SECTION HEADER Perl#1 .text name 154886 virtual size SECTION HEADER Perl#2 .rdata name 1AA19E virtual size AFTER Dump of file ..\perl541.dll SECTION HEADER Perl#1 .text name 1532A6 virtual size SECTION HEADER Perl#2 .rdata name 1A862E virtual size Dump of file ..\miniperl.exe SECTION HEADER Perl#1 .text name 1224EC virtual size SECTION HEADER Perl#2 .rdata name 19B8E8 virtual size

…ly in sv.c -svfix.pl is quick throwaway garbage done in 20 mins and probably doesnt regen sv_inline.h properly and copy pasting/replace regexps were was used anyway to fix up the code, it can be rewritten correctly tho and put in as a official regen.pl script how to finish this fix, options set and unset sv_type as a CPP macro then -#include a header 17 times that contains ONLY only Perl_newSV_type() and its mortal() sister creating 17*2 static inline fns (basically what I did here), code is stepable, extra ms'es of I/O build times perf degrade debate may or may not come up, I hope the CC has a sane in memory cache for .h files and doesn't go back to the kernel or put the entire Perl_newSV_type() fnc in a #define "#define PETS blah(foo(myarg)) + \ cat(dog(fur)) + \ laser(ball(toy)) " then execute that macro 17 times or write 17 Perl_newSV_type() copies into sv_inline.h with /regen.pl infrastructure (fastest for build speed core and build speed CPAN and code is c dbg stepable) OR is a dedicated "sv_newg.h" for regen.pl needed? does the master Perl_newSV_type() template live in a .pl or a .h? i dont have an opinion or against concept of sv_inline.h just have 5-10 hand written versions sv type specific of Perl_newSV_type(), its a cheap gimick fix to keep all 17 types together mashed with if/else/switch in 1 func and expecting bug free perfection from LTO engines of various C compilers, and expecting perfection from an single vendor LTO engine is very against the spirit of portable code -todo ideas, turn those super long #define ==?:==?:==?: into char array/struct initializers, stored in macros, one faux-string per each column of struct body_details, use that macro as c auto stk rw array initializer, then do the U32 len [3] = "\x01\x02\x03"[sv_type]; or U8 sizes [3] = {1,2,3}; U32 len = sizes[sv_type]; which in perl core would look U32 arena_size = SVDB_AR_SZ_DECL; U32 len = arena_size[sv_type]; maybe VC will optimize those since no global memory is used. Only Perl_newSV_typeX() needs this. in this commit static inline Perl_newSV_typeX(pTHX_ const svtype type) which is the ONLY Perl_newSV_type*() variant that take an arbitrary svtype arg, this is the fallback for gv_pvn_add_by() since I couldn't "const" that call to newSV_type() cuz gv_pvn_add_by() is only place in the whole core that takes a random SV type number. Internals of Perl_newSV_typeX() are trashy, here is an example, MSVC DID not turn this into a jump table but instead 17 test/cond_jump ops. v5 = (char *)S_new_body(v2); v6 = 40i64; if ( v2 == 15 ) v6 = 136i64; if ( v2 == 14 ) v6 = 104i64; if ( v2 == 13 ) v6 = 104i64; if ( v2 == 12 ) v6 = 32i64; if ( v2 == 11 ) v6 = 40i64; if ( v2 == 10 ) v6 = 80i64; if ( v2 == 9 ) v6 = 48i64; if ( v2 == 8 ) v6 = 224i64; if ( v2 == 7 ) v6 = 48i64; if ( v2 == 6 ) v6 = 32i64; if ( v2 == 5 ) v6 = 24i64; if ( v2 == 4 ) v6 = 40i64; if ( v2 == 3 ) v6 = 16i64; if ( v2 == 2 ) v6 = 0i64; if ( v2 == 1 ) v6 = 0i64; memset(v5, 0, v6 & -(signed __int64)(v2 != 0)); Solution is move Perl_newSV_typeX() to sv.c, and let it be struct body_details driven. Cuz it only purpose is when newSV_type() absolutly CAN NOT be constant folded (random number input). it only has 1 caller in core. S_new_body() properly const folded away in 99% of cases except for TWO callers Perl_newSV_typeX() and Perl_make_trie(). Perl_make_trie() failure to inline is bizzare, since Perl_make_trie() internally does "v9 = S_new_body(SVt_PVAV);" and DID inline Perl_newSV_typeSVt_PVAV() !!! and therefore Perl_make_trie() has the AV field initing/nulling code. Here is the "optimized" contents of S_new_body(), its junk performance/design wise (but runtime correct/no bugs) void **__fastcall S_new_body(svtype sv_type) { svtype v1; // er9 __int64 v2; // rbx void **result; // rax signed int v4; // ecx signed __int64 v5; // rax v1 = sv_type; v2 = sv_type; result = (void **)PL_body_roots[sv_type]; if ( !result ) { v4 = 4080; if ( v1 == 15 ) v4 = 3264; if ( v1 == 14 ) v4 = 2080; if ( v1 == 13 ) v4 = 4056; if ( v1 == 12 ) v4 = 4064; if ( v1 == 11 ) v4 = 4080; if ( v1 == 10 ) v4 = 4080; if ( v1 == 9 ) v4 = 4080; if ( v1 == 8 ) v4 = 4032; if ( v1 == 7 ) v4 = 4080; if ( v1 == 6 ) v4 = 3296; if ( v1 == 5 ) v4 = 3424; if ( v1 == 4 ) v4 = 3120; if ( v1 == 3 ) v4 = 3536; if ( v1 == 2 ) v4 = 0; if ( v1 == 1 ) v4 = 0; v5 = 40i64; if ( v1 == 15 ) v5 = 136i64; if ( v1 == 14 ) v5 = 104i64; if ( v1 == 13 ) v5 = 104i64; if ( v1 == 12 ) v5 = 32i64; if ( v1 == 11 ) v5 = 40i64; if ( v1 == 10 ) v5 = 80i64; if ( v1 == 9 ) v5 = 48i64; if ( v1 == 8 ) v5 = 224i64; if ( v1 == 7 ) v5 = 48i64; if ( v1 == 6 ) v5 = 32i64; if ( v1 == 5 ) v5 = 24i64; if ( v1 == 4 ) v5 = 40i64; if ( v1 == 3 ) v5 = 16i64; if ( v1 == 2 ) v5 = 0i64; if ( v1 == 1 ) v5 = 0i64; result = (void **)Perl_more_bodies(v1, v5 & -(signed __int64)(v1 != 0), v4 & (unsigned int)-(v1 != 0)); } PL_body_roots[v2] = *result; return result; } ------------ disassembly view of S_new_body() ------------ cmp r9d, 0Fh lea edi, [rbp+28h] mov r8d, 0FF0h lea r11d, [rbp+20h] mov edx, 0CC0h lea r10d, [rbp+30h] mov ecx, r8d mov eax, r9d cmovz ecx, edx cmp r9d, 0Eh mov edx, 820h cmovz ecx, edx cmp r9d, 0Dh lea edx, [r8-18h] cmovz ecx, edx cmp r9d, 0Ch lea edx, [r8-10h] cmovz ecx, edx cmp r9d, 0Bh lea edx, [r8-30h] cmovz ecx, r8d cmp r9d, 0Ah cmovz ecx, r8d cmp r9d, 9 cmovz ecx, r8d cmp r9d, 8 cmovz ecx, edx cmp r9d, 7 mov edx, 0CE0h cmovz ecx, r8d cmp r9d, 6 cmovz ecx, edx cmp r9d, 5 mov edx, 0D60h cmovz ecx, edx cmp r9d, 4 mov edx, 0C30h cmovz ecx, edx cmp r9d, 3 mov edx, 0DD0h cmovz ecx, edx cmp r9d, 2 lea edx, [rdi+60h] cmovz ecx, ebp cmp r9d, 1 cmovz ecx, ebp neg eax sbb eax, eax and eax, ecx mov ecx, r9d mov r8d, eax cmp r9d, 0Fh mov eax, edi cmovz eax, edx cmp r9d, 0Eh lea edx, [rbp+68h] cmovz eax, edx cmp r9d, 0Dh cmovz eax, edx cmp r9d, 0Ch lea edx, [rbp+50h] cmovz eax, r11d cmp r9d, 0Bh cmovz eax, edi cmp r9d, 0Ah cmovz eax, edx cmp r9d, 9 mov edx, 0E0h cmovz eax, r10d cmp r9d, 8 cmovz eax, edx cmp r9d, 7 lea edx, [rbp+18h] cmovz eax, r10d cmp r9d, 6 cmovz eax, r11d cmp r9d, 5 cmovz eax, edx cmp r9d, 4 lea edx, [rbp+10h] cmovz eax, edi cmp r9d, 3 cmovz eax, edx cmp r9d, 2 cmovz eax, ebp cmp r9d, 1 cmovz eax, ebp neg ecx mov ecx, r9d sbb rdx, rdx and rdx, rax call Perl_more_bodies ---------------------- 17 test ops and 17 conditional_move_constant_8_bits ops solution, turn S_new_body() back into a macro so no CC ever tries to ref-inline it. It was a macro before sv_inline.h branch was merged TODO add XSApitest.xs that worlds longest macros are identical to the master correct copy (struct body_details). byte size drops from before these 3 commits to this "success commit" mp.exe 0x1241AC-0x1224EC=7360 0x19D3D8-0x19B8E8=6896 p541.dll 0x154886-0x1532A6=5600 0x1AA19E-0x1A862E=7024 BEFORE Dump of file ..\miniperl.exe SECTION HEADER Perl#1 .text name 1241AC virtual size SECTION HEADER Perl#2 .rdata name 19D3D8 virtual size Dump of file ..\perl541.dll SECTION HEADER Perl#1 .text name 154886 virtual size SECTION HEADER Perl#2 .rdata name 1AA19E virtual size AFTER Dump of file ..\perl541.dll SECTION HEADER Perl#1 .text name 1532A6 virtual size SECTION HEADER Perl#2 .rdata name 1A862E virtual size Dump of file ..\miniperl.exe SECTION HEADER Perl#1 .text name 1224EC virtual size SECTION HEADER Perl#2 .rdata name 19B8E8 virtual size

Add builtin::getcwd Perl#3 gcc syntax fixes -win32_get_childdir() skip the strlen() because GetCurrentDirectoryA() gives it to us, handle 32KB paths if encountered. It is an infinite retry loop since its been reported, in multithreading, GCD()/CWD can change and get longer on our OS thread's between overflow Perl#1 and correct-size attempt Perl#2 because CWD val is a race cond. So all overflow conditions must trigger realloc. If the whole C stack is used up with alloca() and infinite retry. A SEGV is good. Win API/UNICODE_STRING struct is hard coded to USHORT/SHORT. If GetCurrentDirectoryA() returns above 65KB a SEGV is good. If GetCurrentDirectoryA()'s impl is doing {return buflen+1;} which is an API violation, OS is damaged, a SEGV is good. The race retry will never realistically trigger more than 1x ever, 2x rounds through retry loop might happen after a few centuries semi-guess. -CPerlHost::GetChildDir(void) is TODO now that m_pvDir->GetCurrentDirectoryA/W have correct retvals. -perllib.c silence warnings, return debug code accidentally removed in Merge WinCE and Win32 directories -- Initial patch 7bd379e 4/27/2006 7:30:00 PM

All the code in gv.c, is very old and has gotten zero optimizing since 5.000 alpha. SV*s are instantly turned into PVNs on the front end instantly loosing and chance of [future] SVPV COW Shared HEK key string optimization. HEK*s are unknown to gv_* API. All inputs are continously parsed for ' and :: without exception, even if they are read only (SEGV) C literals or PP SvREADONLY() SvPROTECT() read only literals or API contract read only HEK* PV buffers. Returned from hv_store*() hv_fetch*(), HE*s, aren't exploited to pass the shared HEK* onto gv_init_*() or gv_name_set(), and gv_name_set() on front end only understands PVNs, but on backend, in the GP struct and GV body struct, ONLY understands HEK*s. Therefore no RC++, and looking up the ShHEK again in PL_strtab. The large amount of tiny extern exported symbols wrapper funs added over the years also causes C dbg call stacks even at -O1/-O2, to be 2-5 call frames deep of 3 line shims/stub functions before reaching the main logic. I can't tell what is a mathom and what isn't. So to lay provisions needed for future commits, that add proper SV*/HEK*/U32 hash precalculation, not to mention the memcmp() in hv_common() is skipped if left and right ptr addr are equal. The front end of gv_* needs cleanup. -move U32 flags to the start of the the func, so flags can encode details what void * Perl#1 means, and if vararg void * Perl#2 exists (PVN with N as size_t is only 2nd arg user right now). Since gv_stashpvs() is very common on core and CPAN, and called over and over in 1 proc, since most interp core and CPAN XS devs don't know GV*s have an RC that can be ++ed and stored in a MY_CXT struct. Also nobody knows "stashes" are HV*s or PP packages/classes are implimented with HV*s. So there is reason to pay extra attention to gv_stashpvs() b/c of its high usage/call sites per library. So if the STRLEN can be CC constant folded, and fits in a U8, store the length in the flags arg. Saves on CPU ops in all the callers to push 2 args, vs 3. Public API gv_stashpvs(str, create)'s create arg [flags in reality] can't be optimized away or removed, so combine the 2 CC time constant args, so they fold/optimize into 1 cpu op. -at some point perl core needs to cache/create/move around C level arrays of RC++ed ShHEKs to pass to the gv_*() APIs. SVPVs aren't exactly the right format for storing sanitized (no */::/'/SUPER/main/UNIVERSAL) and pre-parsed/splitted "package tokens", since SVs easily wind up or escape into PP-state, and SV RO flags/COW flags aren't the most honored and respected parts of the API by CPAN XS/maybe core. ShHEKs escaping into PP-state is rarer than "generic SVs" escaping into PP-state or CPAN XS state. All legacy XS code any quality and entry/beginner XS people, will pick "char *" getter macros vs an unknown opaque "HEK" type (and newSVpvn() to capture/move those char *s). Users who know what a HEK* is and how to RC++ it, know not to write to it. Also a bad write to a ShHEK will cause more PP or SEGV breakage/panics or proc exits, alot faster than a bad write to a SVfRO "SVPV" buffer. Hash doesn't match char string in a ShHEK will term the prc faster. So vararg on gv_*() is a provision for a future prototype, that accepts 1, 2, 3 or more HEK*s passed array style, that already were sanitized to not have ::s. 0xFF length was picked b/c there was bitfield space, shaving to 32/64/128 chars for gv_stashpvs(str, create) is possible if the bits are needed b/c a terminal is 80 chars, would fit almost all absolute ("::") C string package names, and everthing in core and CPAN. -the stubs remain as exported stub funcs, on purpose for now, it makes certain diag tools I use slightly easier to use vs optimized out inlines or macros. In 5.43 or 5.45 the exported stub funcs can be converted to macros no static inline, which is intent of this commit. The vararg is the 1 and only entry point to all of gv_stash* logic. -flipping I32 flags to the front requires "_p" suffixes for private for ABI reasons, public API still thinks I32 flags is always the last arg -since all front end wrappers, are 1-away from instead of multiple frames away, they are more likely to LTO inline away inside of libperl (not XS) on any CC. CCs have cost/benefit/wall time cut offs for scoring potential inlines opportunities. Going 2 layers, or 3+ layers of small inlines, is asking alot from a CC, that has to traverse a tree of nodes to do each inline, and the cut off could be as low as 1 inline fn and no more unrolling or folding.

This is #1 of a small series of commits to refactor the INPUT_handler() method and turn it into a Node subclass method. This commit changes the main loop from using $_ to hold the current line, to using the variable $line instead.

This is #1 of a small series of commits to refactor the OUTPUT_handler() method and turn it into a Node subclass method. This series is very similar to the one earlier in this branch which did the same for INPUT_handler(). This commit changes the main loop from using $_ to hold the current line, to using the variable $line instead.

This is #1 of a small series of commits to refactor the INPUT_handler() method and turn it into a Node subclass method. This commit changes the main loop from using $_ to hold the current line, to using the variable $line instead.

This is #1 of a small series of commits to refactor the OUTPUT_handler() method and turn it into a Node subclass method. This series is very similar to the one earlier in this branch which did the same for INPUT_handler(). This commit changes the main loop from using $_ to hold the current line, to using the variable $line instead.

This is #1 of a small series of commits to refactor the INPUT_handler() method and turn it into a Node subclass method. This commit changes the main loop from using $_ to hold the current line, to using the variable $line instead.

This is #1 of a small series of commits to refactor the OUTPUT_handler() method and turn it into a Node subclass method. This series is very similar to the one earlier in this branch which did the same for INPUT_handler(). This commit changes the main loop from using $_ to hold the current line, to using the variable $line instead.

Fixes a Coverity issue: >>> function_return: Function Perl_delimcpy_no_escape(tmpbuf, tmpbuf + 4096UL, s, bufend, 58, &len) modifies its argument, assigning 2147483647 to len. 3553 s = delimcpy_no_escape(tmpbuf, tmpbuf + sizeof tmpbuf, s, bufend, 3554 ':', &len); >>> CID 583353: (Perl#1 of 1): Overflowed constant (INTEGER_OVERFLOW) >>> overflow_const: Expression len + 1, where len is known to be equal to 2147483647, overflows the type of len + 1, which is type int. 3558 if (len + 1 + strlen(scriptname) + MAX_EXT_LEN >= sizeof tmpbuf) 3559 continue; /* don't search dir with too-long name */ If there is not enough available space in tmpbuf, delimcpy_no_escape sets len to I32_MAX, but the following code does not check for this. (I believe this case is reachable simply by setting PATH to a huge string.) Avoid the potential overflow by rewriting A + B >= C as A >= C - B (Also, make 'len' unsigned (specifically, size_t) to match the type of sizeof/strlen() and avoid warnings about comparisons between signed and unsigned integers.)

Fixes a Coverity issue: >>> function_return: Function Perl_delimcpy_no_escape(tmpbuf, tmpbuf + 4096UL, s, bufend, 58, &len) modifies its argument, assigning 2147483647 to len. 3553 s = delimcpy_no_escape(tmpbuf, tmpbuf + sizeof tmpbuf, s, bufend, 3554 ':', &len); >>> CID 583353: (#1 of 1): Overflowed constant (INTEGER_OVERFLOW) >>> overflow_const: Expression len + 1, where len is known to be equal to 2147483647, overflows the type of len + 1, which is type int. 3558 if (len + 1 + strlen(scriptname) + MAX_EXT_LEN >= sizeof tmpbuf) 3559 continue; /* don't search dir with too-long name */ If there is not enough available space in tmpbuf, delimcpy_no_escape sets len to I32_MAX, but the following code does not check for this. (I believe this case is reachable simply by setting PATH to a huge string.) Avoid the potential overflow by rewriting A + B >= C as A >= C - B (Also, make 'len' unsigned (specifically, size_t) to match the type of sizeof/strlen() and avoid warnings about comparisons between signed and unsigned integers.)

Update perlhack.pod - fix link to mirror git repo

d1b2ed9

Change dead link http://github.com/mirrors/perl to https://github.com/Perl/perl5

MartinMcGrath closed this Dec 18, 2013

p5p pushed a commit that referenced this pull request Dec 13, 2014

Deparse $#{1} with braces

8c2e27d

$#1 is a syntax error.

This was referenced Oct 14, 2024

Fix huge inline bloat msvc sv inline h #22667

Open

csighandler3: try a fallback interpreter if none in TLS #22530

Closed

This was referenced Dec 1, 2024

Perl_sv_clear: unrolled SvREFCNT_dec and sv_free2....isn't that?! #22798

Closed

[doc] perlcall/perlguts encourage obsolete call_method() / call_pv() and deny most of call_sv()'s API #22804

Open

bulk88 mentioned this pull request Jan 9, 2025

perlembed demos has SEGV risks, ithreads + setjmp + PERL_SET_CONTEXT are too dangerous to use #22901

Open

Tux mentioned this pull request Jan 23, 2025

perl -V being mockable is a security problem re: #21135 Errno archname/osvers too strict. #22939

Open

bulk88 mentioned this pull request Feb 27, 2025

add compact vararg impl gv_stashsvpvn_cached_p, add gv_stashhek #23041

Open

bulk88 mentioned this pull request Apr 12, 2025

Try to force inlining of newSV_type (i -> I in embed.fnc) #23190

Closed

bulk88 mentioned this pull request May 2, 2025

Perl_repeatcpy() arg 4 "IV count" has overflow on 32b CPUs #23243

Open

Flamefire mentioned this pull request May 7, 2025

[doc] Changing libpth for building extensions #23267

Closed

lillianokok mentioned this pull request Jun 6, 2025

executing the test command dist/ExtUtils-CBuilder/t/03-cplusplusplus fails, #23355

Closed

jkeenan mentioned this pull request Jun 9, 2025

perldoc does not generate doc properly if run as root and file is not world readable #17701

Closed

bulk88 mentioned this pull request Jun 16, 2025

pp_reverse - chunk-at-a-time string reversal #23371

Open

bulk88 mentioned this pull request Jul 6, 2025

t/DynaLoader.t and DynaLoader::dl_unload_file() are SEGVing, are they right, or libperl is right and they are broken? #23406

Open

djerius mentioned this pull request Jul 28, 2025

perldb improperly handles blessed coderefs #23486

Open

bulk88 mentioned this pull request Aug 9, 2025

Add IO::Socket::SSL and Net::SSLeay to the Perl core #20739

Open

endocrantz mentioned this pull request Sep 16, 2025

pp.c build problem with Dusethreads 5.42 (perl5-5.43.2 as well) #23713

Closed

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

Update perlhack.pod - fix link to mirror git repo #1

Update perlhack.pod - fix link to mirror git repo #1

Uh oh!

MartinMcGrath commented Dec 18, 2013

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

1 participant

Update perlhack.pod - fix link to mirror git repo #1

Update perlhack.pod - fix link to mirror git repo #1

Uh oh!

Conversation

MartinMcGrath commented Dec 18, 2013

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

1 participant