Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]
Groups > linux.kernel > #1586999 > unrolled thread
| Started by | Mathias Nyman <mathias.nyman@linux.intel.com> |
|---|---|
| First post | 2017-02-23 17:30 +0100 |
| Last post | 2017-02-23 17:30 +0100 |
| Articles | 1 — 1 participant |
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.
Re: [RFC PATCH] xhci: Use Cached ring during endpoint ring allocation Mathias Nyman <mathias.nyman@linux.intel.com> - 2017-02-23 17:30 +0100
| From | Mathias Nyman <mathias.nyman@linux.intel.com> |
|---|---|
| Date | 2017-02-23 17:30 +0100 |
| Subject | Re: [RFC PATCH] xhci: Use Cached ring during endpoint ring allocation |
| Message-ID | <te4rh-3IU-35@gated-at.bofh.it> |
On 22.02.2017 17:05, Anurag Kumar Vulisha wrote: > Currently during endpoint initialization, a new endpoint ring is alloacte > using xhci_ring_alloc(), if this function fails to allocate ring a cached > ring(if available) is assigned to endpoint ring. > This patch modifies the code that during endpoint initialization, if cached > ring is available it is assigned to the endpoint ring. If cached rings are > not available then xhci_ring_alloc() is called to allocate a new ring. > Doing so will avoid unncessary memory allocations if cached ring is already > available for use. This also fixes endpoint "Ring expansion failed" error > which occurs due to insufficient memory during ring expansion. > > Signed-off-by: Anurag Kumar Vulisha <anuragku@xilinx.com> Thanks for bringing this to my attention. Your approach certainly makes more sense than the current way. I need to check history why this type of ring cache was created in the first place. It's possible that the whole ring cache is not really useful anymore -Mathias
Back to top | Article view | linux.kernel
csiph-web