I haven't traced down the entire flow for mem_align, so perhaps I should investigate more. My concern was when the user sets mem_align to something other than the system's page size or the user's block size. Regards, Jeff -----Original Message----- From: Sitsofe Wheeler [mailto:sitsofe@xxxxxxxxx] Sent: Tuesday, September 5, 2017 12:02 PM To: Jeff Furlong <jeff.furlong@xxxxxxx> Cc: fio@xxxxxxxxxxxxxxx Subject: Re: fio patch question on filesetup: align layout buffer On 5 September 2017 at 19:57, Jeff Furlong <jeff.furlong@xxxxxxx> wrote: > Right. A check of memalign or page size can help. But the patch is no longer compatible with the latest git, given the recent revert. However, my concern for mem_align still remains. How does the current git use the mem_align parameter in fill_io_buffer(td, b, bs, bs)? Or is it completley handled elsewhere, namely init_io_u() and allocate_io_mem()? When the I/O is normal and uncached won't it all be taken care of in the kernel? -- Sitsofe | http://sucs.org/~sits/ ��.n��������+%������w��{.n�������^n�r������&��z�ޗ�zf���h���~����������_��+v���)ߣ�
