
26 Oct
2010
26 Oct
'10
11:12 p.m.
Dear Alessandro Rubini,
In message 20101026205756.GA2139@morgana.i.gnudd.com you wrote:
Instead of:
DEFINE(GENERATED_GBL_DATA_SIZE, (sizeof(struct global_data)+15) & ~15);
I'd use:
DEFINE(GENERATED_GBL_DATA_SIZE, (sizeof(struct global_data)),
leaving the alignment requirement in the structure itself (include/asm/global_data.h for each architecture).
Is it guaranteed (I mean by the C standard) that the alignment of a struct (which affects only the possible start address) also has effect on the sizeof() for that struct, in the sense that sizeof() is guaranteed to be a multiple of that alignment requirement?
Best regards,
Wolfgang Denk
--
DENX Software Engineering GmbH, MD: Wolfgang Denk & Detlev Zundel
HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany
Phone: (+49)-8142-66989-10 Fax: (+49)-8142-66989-80 Email: wd@denx.de
panic: kernel trap (ignored)