At 03/16/2012 11:23 AM, HATAYAMA Daisuke Wrote:
> From: Wen Congyang <[email protected]>
> Subject: [RFC][PATCH 11/14 v9] introduce a new monitor command 'dump' to dump
> guest's memory
> Date: Wed, 14 Mar 2012 10:11:35 +0800
>
>> +/*
>> + * QEMU dump
>> + *
>> + * Copyright Fujitsu, Corp. 2011
>> + *
>
> Now 2012.
On, I forgot to update it.
>
>> + /*
>> + * calculate phdr_num
>> + *
>> + * the type of phdr->num is uint16_t, so we should avoid overflow
>
> e_phnum is correct.
Yes
>
>> + */
>> + s->phdr_num = 1; /* PT_NOTE */
>> + if (s->list.num < (1 << 16) - 2) {
>
> s->list.num < UINT16_MAX is better.
>
>> + s->phdr_num += s->list.num;
>> + s->have_section = false;
>> + } else {
>> + s->have_section = true;
>> + s->phdr_num = PN_XNUM;
>> +
>> + /* the type of shdr->sh_info is uint32_t, so we should avoid
>> overflow */
>> + if (s->list.num > (1ULL << 32) - 2) {
>
> s->list.num < UINT32_MAX is better.
>
>> + s->sh_info = 0xffffffff;
>
> UINT32_MAX is better. Is it rough around here?
>
>> + } else {
>> + s->sh_info += s->list.num;
>> + }
>> + }
>
> Now orders of processings in positive and negative cases for e_phnum
> and sh_info are different. It's better to make them sorted in the same
> order.
>
> if (phdr_num not overflow?) {
> not overflow case;
> } else {
> overflow case;
> if (sh_info not overflow?) {
> not overflow case;
> } else {
> overflow case;
> }
> }
>
> is better.
OK
Thanks
Wen Congyang
>
> Thanks.
> HATAYAMA, Daisuke
>
>
