Hello Michael, You're welcome and thanks. HATAYAMA Daisuke From: Michael Kerrisk <mtk.manpages@xxxxxxxxx> Subject: Re: [PATCH v2] elf.5: Add description for ELF extension Date: Sat, 19 Jun 2010 15:53:00 +0200 > Cc: pasky@xxxxxxx, linux-man@xxxxxxxxxxxxxxx > Date: Sat, 19 Jun 2010 15:53:00 +0200 > Reply-To: mtk.manpages@xxxxxxxxx > > Hello Daisuke, Petr > > On Thu, May 13, 2010 at 7:55 AM, Daisuke HATAYAMA > <d.hatayama@xxxxxxxxxxxxxx> wrote: >> Hi. Petr. >> >> Thank you for your comments. >> >> I reflected your comments to the previous patch. >> Could you check it? >> >> # What a strange interface, the other two hold the max value while >> # e_shnum holds zero in case of overflow. >> >> I considered one answer to the question based on the difference of >> compatibility policy for each object format. >> >> For core object format, it is better for ordinary tools to understand >> as many program headers as possible, even if it is incomplete. On the >> other hand, for relocatable object format, it is meaningless if it is >> incomplete. This is why e_phnum is set to PN_XNUM(0xffff) and e_shnum >> is set to 0, respectively: in the latter case, any ordinary linker >> cannot recognize the relocatable object correctly. >> >> But I don't know why e_shstrndx is set to 0xffff. We could choose it >> from the reserved range of section header table indices, i.e. 0 and -- To unsubscribe from this list: send the line "unsubscribe linux-man" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
