[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Buffer/window/frame redisplay bit

From: Dmitry Antipov
Subject: Buffer/window/frame redisplay bit
Date: Mon, 20 Jan 2014 15:42:00 +0400
User-agent: Mozilla/5.0 (X11; Linux x86_64; rv:24.0) Gecko/20100101 Thunderbird/24.2.0

I've noticed that windows are created with zero redisplay bit, new
frames has this bit set, but buffer texts leaves them uninitialized.
This is sometimes shown by valgrind, for example:

==15146== Conditional jump or move depends on uninitialised value(s)
==15146==    at 0x45D7A0: propagate_buffer_redisplay (xdisp.c:13153)
==15146==    by 0x45EC14: redisplay_internal (xdisp.c:13608)
==15146==    by 0x45CB74: redisplay (xdisp.c:12919)
==15146==    by 0x56BECA: read_char (keyboard.c:2563)
==15146==    by 0x57B32F: read_key_sequence (keyboard.c:9071)
==15146==    by 0x568EF0: command_loop_1 (keyboard.c:1445)
==15146==    by 0x608602: internal_condition_case (eval.c:1345)
==15146==    by 0x5687CC: command_loop_2 (keyboard.c:1170)
==15146==    by 0x607A85: internal_catch (eval.c:1109)
==15146==    by 0x568780: command_loop (keyboard.c:1149)
==15146==    by 0x567E30: recursive_edit_1 (keyboard.c:777)
==15146==    by 0x568000: Frecursive_edit (keyboard.c:841)

IMO b->text->redisplay should be initialized in Fget_buffer_create, but
I have no representative test case to decide whether it should be 0 or 1.


reply via email to

[Prev in Thread] Current Thread [Next in Thread]