Re: [FFmpeg-devel] [PATCH] HTTP: optimize forward seek performance

Attached is a git format-patch file 0001-HTTP-improve-performance-by-reducing-forward-seeks.patch Description: Binary data > On Jan 26, 2017, at 6:52 PM, Michael Niedermayer wrote: > > On Thu, Jan 26, 2017 at 10:44:33AM -0600, Joel Cunningham wrote: >> Michael, >> >>

Re: [FFmpeg-devel] [PATCH] HTTP: optimize forward seek performance

On Thu, Jan 26, 2017 at 10:44:33AM -0600, Joel Cunningham wrote: > Michael, > > Thanks for the review and testing! New patch included, see comments below > > > > > this segfaults with many fuzzed files > > backtrace looks like this: > > > > #0 0x7fffb368 in ?? () > > #1

Re: [FFmpeg-devel] [PATCH] HTTP: optimize forward seek performance

Michael, Thanks for the review and testing! New patch included, see comments below > > this segfaults with many fuzzed files > backtrace looks like this: > > #0 0x7fffb368 in ?? () > #1 0x005f9280 in avio_seek (s=0x7fffb278, offset=31, whence=0) > at

Re: [FFmpeg-devel] [PATCH] HTTP: optimize forward seek performance

On Fri, Jan 13, 2017 at 11:36:41AM -0600, Joel Cunningham wrote: > > > On Jan 12, 2017, at 10:59 AM, Joel Cunningham > > wrote: > > > > Nicolas, > > > > I’ve found existing “read-ahead logic” in avio_seek to do what I’ve > > implemented in http_stream_forward(). This

Re: [FFmpeg-devel] [PATCH] HTTP: optimize forward seek performance

Ping! Anyone have a chance to review my latest patch which leverages the read-ahead/discard logic in avio_seek? Thanks, Joel > On Jan 13, 2017, at 11:36 AM, Joel Cunningham wrote: > >> >> On Jan 12, 2017, at 10:59 AM, Joel Cunningham wrote:

Re: [FFmpeg-devel] [PATCH] HTTP: optimize forward seek performance

> On Jan 12, 2017, at 10:59 AM, Joel Cunningham wrote: > > Nicolas, > > I’ve found existing “read-ahead logic” in avio_seek to do what I’ve > implemented in http_stream_forward(). This is controlled by > SHORT_SEEK_THRESHOLD, currently set to 4KB. I proto-typed

Re: [FFmpeg-devel] [PATCH] HTTP: optimize forward seek performance

On Thu, Jan 12, 2017 at 10:59:56AM -0600, Joel Cunningham wrote: > Nicolas, > > I’ve found existing “read-ahead logic” in avio_seek to do what I’ve > implemented in http_stream_forward(). This is controlled by > SHORT_SEEK_THRESHOLD, currently set to 4KB. I proto-typed increasing this to >

Re: [FFmpeg-devel] [PATCH] HTTP: optimize forward seek performance

> On Jan 12, 2017, at 11:04 AM, Steinar H. Gunderson > wrote: > > On Thu, Jan 12, 2017 at 04:59:33PM +, Andy Furniss wrote: >> I've seen plenty of "legal" shrinks looking at tcpdumps - usually the >> app is throttling it's read speed. > > You're not really

Re: [FFmpeg-devel] [PATCH] HTTP: optimize forward seek performance

Steinar H. Gunderson wrote: On Thu, Jan 12, 2017 at 04:59:33PM +, Andy Furniss wrote: I've seen plenty of "legal" shrinks looking at tcpdumps - usually the app is throttling it's read speed. You're not really allowed to shrink by more than you've received, though, are you? Typically the

Re: [FFmpeg-devel] [PATCH] HTTP: optimize forward seek performance

On Thu, Jan 12, 2017 at 04:59:33PM +, Andy Furniss wrote: > I've seen plenty of "legal" shrinks looking at tcpdumps - usually the > app is throttling it's read speed. You're not really allowed to shrink by more than you've received, though, are you? Typically the buffer going down is just

Re: [FFmpeg-devel] [PATCH] HTTP: optimize forward seek performance

Nicolas, I’ve found existing “read-ahead logic” in avio_seek to do what I’ve implemented in http_stream_forward(). This is controlled by SHORT_SEEK_THRESHOLD, currently set to 4KB. I proto-typed increasing this to the 256KB (matches the initial TCP window in my test setup) and saw the same

Re: [FFmpeg-devel] [PATCH] HTTP: optimize forward seek performance

Joel Cunningham wrote: According TCP RFCs, a receiver is not allowed to decrease the window, Not sure that's true. It's certainly possible to do it illegally, on a linux box tcp will log a message something like heresy shrank window. That IME is where some middle box is "interfearing".

Re: [FFmpeg-devel] [PATCH] HTTP: optimize forward seek performance

> On Jan 12, 2017, at 10:08 AM, Andy Furniss wrote: > > Joel Cunningham wrote: >> Hi, >> >> I’ve been working on optimizing HTTP forward seek performance for ffmpeg and >> would like to contribute this patch into mainline ffmpeg. Please see the >> below patch for an

Re: [FFmpeg-devel] [PATCH] HTTP: optimize forward seek performance

Joel Cunningham wrote: Hi, I’ve been working on optimizing HTTP forward seek performance for ffmpeg and would like to contribute this patch into mainline ffmpeg. Please see the below patch for an explanation of the issue and proposed fix. I have provided evidence of the current performance

Re: [FFmpeg-devel] [PATCH] HTTP: optimize forward seek performance

> On Jan 12, 2017, at 6:47 AM, Ronald S. Bultje wrote: > > Hi Joel, > > On Wed, Jan 11, 2017 at 6:01 PM, Joel Cunningham > wrote: > >> Hi, >> >> I’ve been working on optimizing HTTP forward seek performance for ffmpeg >> and would like to

Re: [FFmpeg-devel] [PATCH] HTTP: optimize forward seek performance

Le duodi 22 nivôse, an CCXXV, Joel Cunningham a écrit : > This commit optimizes HTTP forward seeks by advancing the stream on > the current connection when the seek amount is within the current > TCP window rather than closing the connection and opening a new one. > This improves performance

Re: [FFmpeg-devel] [PATCH] HTTP: optimize forward seek performance

Hi Joel, On Wed, Jan 11, 2017 at 6:01 PM, Joel Cunningham wrote: > Hi, > > I’ve been working on optimizing HTTP forward seek performance for ffmpeg > and would like to contribute this patch into mainline ffmpeg. Please see > the below patch for an explanation of the

[FFmpeg-devel] [PATCH] HTTP: optimize forward seek performance

Hi, I’ve been working on optimizing HTTP forward seek performance for ffmpeg and would like to contribute this patch into mainline ffmpeg. Please see the below patch for an explanation of the issue and proposed fix. I have provided evidence of the current performance issue and my sample MP4