Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > linux.kernel > #1722975 > unrolled thread

RE: [[PATCH v1] 15/37] [CIFS] SMBD: Post a SMBD data transfer message with data payload

Back to article view | Back to linux.kernel

This discussion starts older than the indexed window; earlier articles aren't shown. The article labeled Started by below is the oldest one visible, not the original post.

Contents

  RE: [[PATCH v1] 15/37] [CIFS] SMBD: Post a SMBD data transfer message  with data payload Long Li <longli@microsoft.com> - 2017-08-30 04:20 +0200
    Re: [[PATCH v1] 15/37] [CIFS] SMBD: Post a SMBD data transfer  message with data payload Christoph Hellwig <hch@infradead.org> - 2017-08-30 11:00 +0200
      RE: [[PATCH v1] 15/37] [CIFS] SMBD: Post a SMBD data transfer message  with data payload Long Li <longli@microsoft.com> - 2017-08-30 20:20 +0200

#1722975 — RE: [[PATCH v1] 15/37] [CIFS] SMBD: Post a SMBD data transfer message with data payload

> -----Original Message-----
> From: Christoph Hellwig [mailto:hch@infradead.org]
> Sent: Sunday, August 13, 2017 3:24 AM
> To: Long Li <longli@microsoft.com>
> Cc: Steve French <sfrench@samba.org>; linux-cifs@vger.kernel.org; samba-
> technical@lists.samba.org; linux-kernel@vger.kernel.org; Long Li
> <longli@microsoft.com>
> Subject: Re: [[PATCH v1] 15/37] [CIFS] SMBD: Post a SMBD data transfer
> message with data payload
> 
> You can always get the struct page for kernel allocations using virt_to_page
> (or vmalloc_to_page, but this code would not handle the vmalloc case either),
> so I don't think you need this helper and can always use the one added in the
> previous patch.

I partially addressed this issue in the V3 patch. Most of the duplicate code on sending path is merged.

The difficulty with translating the buffer to pages is that: I don't know how many pages will be translated, and how many struct page I need to allocate in advance to hold them. I try to avoid memory allocation in the I/O path as much as possible. So I keep two functions of sending data: one for buffer and one for pages.

[toc] | [next] | [standalone]

#1723131 — Re: [[PATCH v1] 15/37] [CIFS] SMBD: Post a SMBD data transfer message with data payload

On Wed, Aug 30, 2017 at 02:17:56AM +0000, Long Li wrote:
> I partially addressed this issue in the V3 patch. Most of the duplicate
> code on sending path is merged.
> 
> The difficulty with translating the buffer to pages is that: I don't
> know how many pages will be translated, and how many struct page I need
> to allocate in advance to hold them. I try to avoid memory allocation
> in the I/O path as much as possible. So I keep two functions of 
> sending data: one for buffer and one for pages.

You do: you'll always need speace for  (len + PAGE_SIZE - 1) >> PAGE_SIZE
pages.

That being said: what callers even send you buffers?  In general we
should aim to work with pages for all allocations that aren't tiny.

[toc] | [prev] | [next] | [standalone]

#1723574

> -----Original Message-----
> From: Christoph Hellwig [mailto:hch@infradead.org]
> Sent: Wednesday, August 30, 2017 1:52 AM
> To: Long Li <longli@microsoft.com>
> Cc: Christoph Hellwig <hch@infradead.org>; Steve French
> <sfrench@samba.org>; linux-cifs@vger.kernel.org; samba-
> technical@lists.samba.org; linux-kernel@vger.kernel.org
> Subject: Re: [[PATCH v1] 15/37] [CIFS] SMBD: Post a SMBD data transfer
> message with data payload
> 
> On Wed, Aug 30, 2017 at 02:17:56AM +0000, Long Li wrote:
> > I partially addressed this issue in the V3 patch. Most of the
> > duplicate code on sending path is merged.
> >
> > The difficulty with translating the buffer to pages is that: I don't
> > know how many pages will be translated, and how many struct page I
> > need to allocate in advance to hold them. I try to avoid memory
> > allocation in the I/O path as much as possible. So I keep two
> > functions of sending data: one for buffer and one for pages.
> 
> You do: you'll always need speace for  (len + PAGE_SIZE - 1) >> PAGE_SIZE
> pages.
> 
> That being said: what callers even send you buffers?  In general we should
> aim to work with pages for all allocations that aren't tiny.

I'll look through the code allocating the buffers, it probably can fit into one page. Will fix this.

[toc] | [prev] | [standalone]

Back to top | Article view | linux.kernel

csiph-web