- * gnutls.c (emacs_gnutls_handshake): Don't return a garbage value.
- Expect the caller to check whether GNUTLS_STAGE_HANDSHAKE_CANDO <=
- proc->gnutls_initstage, if the check is needed. The check isn't
- needed for one caller, Fgnutls_boot. (Bug#8556)
- (emacs_gnutls_read): Do that check. This is the other caller.
- (emacs_gnutls_handle_error): Remove unused local.
+2011-04-26 Paul Eggert <eggert@cs.ucla.edu>
+
- supposed to be handshaking.
++ * gnutls.c (emacs_gnutls_handle_error): Remove unused local.
+ (Fgnutls_boot): gnutls_certificate_verify_peers2 wants unsigned *.
+ Remove unused local.
+ (emacs_gnutls_write): Don't use uninitialized rtnval if nbyte <= 0.
+
+ lisp.h: Fix a problem with aliasing and vector headers. (Bug#8546)
+ GCC 4.6.0 optimizes based on type-based alias analysis. For
+ example, if b is of type struct buffer * and v of type struct
+ Lisp_Vector *, then gcc -O2 was incorrectly assuming that &b->size
+ != &v->size, and therefore "v->size = 1; b->size = 2; return
+ v->size;" must therefore return 1. This assumption is incorrect
+ for Emacs, since it type-puns struct Lisp_Vector * with many other
+ types. To fix this problem, this patch adds a new type struct
+ vectorlike_header that documents the constraints on layout of vectors
+ and pseudovectors, and helps optimizing compilers not get fooled
+ by Emacs's type punning. It also adds the macros XSETTYPED_PVECTYPE
+ XSETTYPED_PSEUDOVECTOR, TYPED_PSEUDOVECTORP, for similar reasons.
+ * lisp.h (XSETTYPED_PVECTYPE): New macro, specifying the name of
+ the size member.
+ (XSETPVECTYPE): Rewrite in terms of new macro.
+ (XSETPVECTYPESIZE): New macro, specifying both type and size.
+ This is a bit clearer, and further avoids the possibility of
+ undesirable aliasing.
+ (XSETTYPED_PSEUDOVECTOR): New macro, specifying the size.
+ (XSETPSEUDOVECTOR): Rewrite in terms of XSETTYPED_PSEUDOVECTOR.
+ (XSETSUBR): Rewrite in terms of XSETTYPED_PSEUDOVECTOR and XSIZE,
+ since Lisp_Subr is a special case (no "next" field).
+ (ASIZE): Now uses header.size rather than size. All
+ previous uses of XVECTOR (foo)->size replaced to use this macro,
+ to avoid the hassle of writing XVECTOR (foo)->header.size.
+ (struct vectorlike_header): New type.
+ (TYPED_PSEUDOVECTORP): New macro, also specifying the C type of the
+ object, to help avoid aliasing.
+ (PSEUDOVECTORP): Rewrite in terms of TYPED_PSEUDOVECTORP.
+ (SUBRP): Likewise, since Lisp_Subr is a special case.
+ * lisp.h (struct Lisp_Vector, struct Lisp_Char_Table):
+ (struct Lisp_Sub_Char_Table, struct Lisp_Bool_Vector):
+ (struct Lisp_Hash_Table): Combine first two members into a single
+ struct vectorlike_header member. All uses of "size" and "next" members
+ changed to be "header.size" and "header.next".
+ * buffer.h (struct buffer): Likewise.
+ * font.h (struct font_spec, struct font_entity, struct font): Likewise.
+ * frame.h (struct frame): Likewise.
+ * process.h (struct Lisp_Process): Likewise.
+ * termhooks.h (struct terminal): Likewise.
+ * window.c (struct save_window_data, struct saved_window): Likewise.
+ * window.h (struct window): Likewise.
+ * alloc.c (allocate_buffer, Fmake_bool_vector, allocate_pseudovector):
+ Use XSETPVECTYPESIZE, not XSETPVECTYPE, to avoid aliasing problems.
+ * buffer.c (init_buffer_once): Likewise.
+ * lread.c (defsubr): Use XSETTYPED_PVECTYPE, since Lisp_Subr is a
+ special case.
+ * process.c (Fformat_network_address): Use local var for size,
+ for brevity.
+
+ * bytecode.c (exec_byte_code): Don't use XVECTOR before CHECK_VECTOR.
+
+ Make the Lisp reader and string-to-float more consistent (Bug#8525)
+ * data.c (atof): Remove decl; no longer used or needed.
+ (digit_to_number): Move to lread.c.
+ (Fstring_to_number): Use new string_to_number function, to be
+ consistent with how the Lisp reader treats infinities and NaNs.
+ Do not assume that floating-point numbers represent EMACS_INT
+ without losing information; this is not true on most 64-bit hosts.
+ Avoid double-rounding errors, by insisting on integers when
+ parsing non-base-10 numbers, as the documentation specifies.
+ * lisp.h (string_to_number): New decl, replacing ...
+ (isfloat_string): Remove.
+ * lread.c: Include <inttypes.h>, for uintmax_t and strtoumax.
+ (read1): Do not accept +. and -. as integers; this
+ appears to have been a coding error. Similarly, do not accept
+ strings like +-1e0 as floating point numbers. Do not report
+ overflow for integer overflows unless the base is not 10 which
+ means we have no simple and reliable way to continue.
+ Break out the floating-point parsing into a new
+ function string_to_number, so that Fstring_to_number parses
+ floating point numbers consistently with the Lisp reader.
+ (digit_to_number): Moved here from data.c. Make it static inline.
+ (E_CHAR, EXP_INT): Remove, replacing with ...
+ (E_EXP): New macro, to solve the "1.0e+" problem mentioned below.
+ (string_to_number): New function, replacing isfloat_string.
+ This function checks for valid syntax and produces the resulting
+ Lisp float number too. Rework it so that string-to-number
+ no longer mishandles examples like "1.0e+". Use strtoumax,
+ so that overflow for non-base-10 numbers is reported only when
+ there's no portable and simple way to convert to floating point.
+
+ * textprop.c (set_text_properties_1): Rewrite for clarity,
+ and to avoid GCC warning about integer overflow.
+
+ * intervals.h (struct interval): Use EMACS_INT for members
+ where EMACS_UINT might cause problems. See
+ <http://lists.gnu.org/archive/html/emacs-devel/2011-04/msg00514.html>.
+ (CHECK_TOTAL_LENGTH): Remove cast to EMACS_INT; no longer needed.
+ * intervals.c (interval_deletion_adjustment): Now returns EMACS_INT.
+ All uses changed.
+ (offset_intervals): Tell GCC not to worry about length overflow
+ when negating a negative length.
+
+ * alloc.c (overrun_check_malloc, overrun_check_realloc): Now static.
+ (overrun_check_free): Likewise.
+
+ * alloc.c (SDATA_SIZE) [!GC_CHECK_STRING_BYTES]: Avoid runtime check
+ in the common case where SDATA_DATA_OFFSET is a multiple of Emacs
+ word size.
+
+ * gnutls.c: Fix problems found by GCC 4.6.0 on Ubuntu 10.10.
+ (gnutls_make_error): Rename local to avoid shadowing.
+ (gnutls_emacs_global_deinit): ifdef out; not used.
+ (Fgnutls_boot): Use const for pointer to readonly storage.
+ Comment out unused local. Fix pointer signedness problems.
+
+ * lread.c (openp): Don't stuff size_t into an 'int'.
+ Use <= on length, not < on length + 1, to avoid GCC 4.6.0 warning
+ about possible signed overflow.
+
+ * gtkutil.c: Fix problems found by GCC 4.6.0 on Ubuntu 10.10.
+ (GDK_KEY_g): Don't define if already defined.
+ (xg_prepare_tooltip): Avoid pointer signedness problem.
+ (xg_set_toolkit_scroll_bar_thumb): Redo to avoid two casts.
+
+ * process.c (Fnetwork_interface_info): Avoid left-shift undefined
+ behavior with 1 << 31. GCC 4.6.0 warns about this on 32-bit hosts.
+
+ * xfns.c (Fx_window_property): Simplify a bit,
+ to make a bit faster and to avoid GCC 4.6.0 warning.
+ * xselect.c (x_get_window_property, x_handle_dnd_message): Likewise.
+
+ * fns.c (internal_equal): Don't assume size_t fits in int.
+
+ * alloc.c (compact_small_strings): Tighten assertion a little.
+
+ Replace pEd with more-general pI, and fix some printf arg casts.
+ * lisp.h (pI): New macro, generalizing old pEd macro to other
+ conversion specifiers. For example, use "...%"pI"d..." rather
+ than "...%"pEd"...".
+ (pEd): Remove. All uses replaced with similar uses of pI.
+ * src/m/amdx86-64.h, src/m/ia64.h, src/m/ibms390x.h: Likewise.
+ * alloc.c (check_pure_size): Don't overflow by converting size to int.
+ * bidi.c (bidi_dump_cached_states): Use pI to avoid cast.
+ * data.c (Fnumber_to_string): Use pI instead of if-then-else-abort.
+ * dbusbind.c (xd_append_arg): Use pI to avoid cast.
+ (Fdbus_method_return_internal, Fdbus_method_error_internal): Likewise.
+ * font.c (font_unparse_xlfd): Avoid potential buffer overrun on
+ 64-bit hosts.
+ (font_unparse_xlfd, font_unparse_fcname): Use pI to avoid casts.
+ * keyboard.c (record_char, modify_event_symbol): Use pI to avoid casts.
+ * print.c (safe_debug_print, print_object): Likewise.
+ (print_object): Don't overflow by converting EMACS_INT or EMACS_UINT
+ to int.
+ Use pI instead of if-then-else-abort. Use %p to avoid casts,
+ avoiding the 0 flag, which is not portable.
+ * process.c (Fmake_network_process): Use pI to avoid cast.
+ * region-cache.c (pp_cache): Likewise.
+ * xdisp.c (decode_mode_spec): Likewise.
+ * xrdb.c (x_load_resources) [USE_MOTIF]: Use pI to avoid undefined
+ behavior on 64-bit hosts with printf arg.
+ * xselect.c (x_queue_event): Use %p to avoid casts, avoiding 0 flag.
+ (x_stop_queuing_selection_requests): Likewise.
+ (x_get_window_property): Don't truncate byte count to an 'int'
+ when tracing.
+
+ * frame.c (frame_name_fnn_p): Get rid of strtol, which isn't right
+ here, since it parses constructs like leading '-' and spaces,
+ which are not wanted; and it overflows with large numbers.
+ Instead, simply match F[0-9]+, which is what is wanted anyway.
+
+ * alloc.c: Remove unportable assumptions about struct layout.
+ (SDATA_SELECTOR, SDATA_DATA_OFFSET): New macros.
+ (SDATA_OF_STRING, SDATA_SIZE, allocate_string_data):
+ (allocate_vectorlike, make_pure_vector): Use the new macros,
+ plus offsetof, to remove unportable assumptions about struct layout.
+ These assumptions hold on all porting targets that I know of, but
+ they are not guaranteed, they're easy to remove, and removing them
+ makes further changes easier.
+
+ * alloc.c (BLOCK BYTES): Fix typo by changing "ablock" to "ablocks".
+ This doesn't fix a bug but makes the code clearer.
+ (string_overrun_cookie): Now const. Use initializers that
+ don't formally overflow signed char, to avoid warnings.
+ (allocate_string_data) [GC_CHECK_STRING_OVERRUN]: Fix typo that
+ can cause Emacs to crash when string overrun checking is enabled.
+ (allocate_buffer): Don't assume sizeof (struct buffer) is a
+ multiple of sizeof (EMACS_INT); it need not be, if
+ alignof(EMACS_INT) < sizeof (EMACS_INT).
+ (check_sblock, check_string_bytes, check_string_free_list): Protoize.
+
+ 2011-04-26 Juanma Barranquero <lekktu@gmail.com>
+
+ * keyboard.c (QCrtl): Rename from Qrtl. All uses changed.
+
+ 2011-04-26 Teodor Zlatanov <tzz@lifelogs.com>
+
+ * gnutls.c (emacs_gnutls_handshake): Return an error if we're not
++ supposed to be handshaking. (Bug#8556)
+ Reported by Paul Eggert <eggert@cs.ucla.edu>.
+
+ 2011-04-26 Daniel Colascione <dan.colascione@gmail.com>
+
+ * lisp.h (Qdebug): List symbol.
+ * eval.c (Qdebug): Restore global linkage.
+ * keyboard.c (debug-on-event): New variable.
+ (handle_user_signal): Break into debugger when debug-on-event
+ matches the current signal symbol.
+
2011-04-25 Dan Nicolaescu <dann@ics.uci.edu>
* alloc.c (check_sblock, check_string_bytes)