Re: [go-nuts] systemstack stop traceback

2019-07-31 Thread jin wang
Thank you very much. 在 2019年7月31日星期三 UTC+8下午9:49:04,Ian Lance Taylor写道: > > On Wed, Jul 31, 2019 at 4:49 AM jin wang > wrote: > > > > so mstart just put on stack for call traceback , but actually will not > return at mstart? > > Yes. As you can see, the systemstack function changes the stack

[go-nuts] Re: Windows syscall.Open() change to fix os.Rename() & .Remove()

2019-07-31 Thread Liam
Microsoft recommends changing syscall.Open() for GOOS=windows to fix this. Pls reply if you know of existing apps that rely on it. This code fails with a "sharing violation" on Windows. That behavior is undocumented in package "os". path := "rename-after-open" fd, err := os.OpenFile(path, os.O_

Re: [go-nuts] How to implement a file download service in GRPC

2019-07-31 Thread Edward Muller
https://github.com/grpc-ecosystem/grpc-gateway > On Jul 30, 2019, at 4:12 PM, Rampradeep Nalluri > wrote: > > It is easy to pipe a file reader to http.ResponseWriter object using > io.copy() when I implement a large file download service over

Re: [go-nuts] How to implement a file download service in GRPC

2019-07-31 Thread Burak Serdar
On Wed, Jul 31, 2019 at 8:25 AM Rampradeep Nalluri wrote: > > If we do that clients like browsers or curl can not use the download API > directly. True. You would need an API front-end to expose the grpc stream. > > > > > On Tue, Jul 30, 2019, 5:41 PM Burak Serdar wrote: >> >> On Tue, Jul 30,

Re: [go-nuts] How to implement a file download service in GRPC

2019-07-31 Thread Rampradeep Nalluri
If we do that clients like browsers or curl can not use the download API directly. On Tue, Jul 30, 2019, 5:41 PM Burak Serdar wrote: > On Tue, Jul 30, 2019 at 5:49 PM Rampradeep Nalluri > wrote: > > > > It is easy to pipe a file reader to http.ResponseWriter object using > io.copy() when I i

Re: [go-nuts] systemstack stop traceback

2019-07-31 Thread Ian Lance Taylor
On Wed, Jul 31, 2019 at 4:49 AM jin wang wrote: > > so mstart just put on stack for call traceback , but actually will not return > at mstart? Yes. As you can see, the systemstack function changes the stack pointer back to its original value before it returns to its caller. It does not return o

Re: [go-nuts] systemstack stop traceback

2019-07-31 Thread jin wang
so mstart just put on stack for call traceback , but actually will not return at mstart? 在 2019年7月31日星期三 UTC+8下午12:08:51,Ian Lance Taylor写道: > > On Tue, Jul 30, 2019 at 8:58 PM > > wrote: > > > > I read go runtime code,but I'm poor in assembly language, > > i can't understand the following co

Re: [go-nuts] systemstack stop traceback

2019-07-31 Thread jin wang
在 2019年7月31日星期三 UTC+8下午12:08:51,Ian Lance Taylor写道: > > On Tue, Jul 30, 2019 at 8:58 PM > > wrote: > > > > I read go runtime code,but I'm poor in assembly language, > > i can't understand the following code > > // switch to g0 > > MOVQDX, g(CX) > > MOVQ(g_sc

[go-nuts] Re: [wasm] net/http crashes application

2019-07-31 Thread Patrick Gaubatz
Hi Jihoon, On Wednesday, July 31, 2019 at 5:58:54 AM UTC+2, Jihoon Chung wrote: > > > Changing doFetch() call in goFetch to "go doFetch()" should work. > You're right, this actually fixes the crashes. Thank you so much! Patrick -- You received this message because you are subscribed to the Go