[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Qemu-devel] [PATCH 5/7] coroutine: rewrite pool to avoid mutex
From: |
Kevin Wolf |
Subject: |
Re: [Qemu-devel] [PATCH 5/7] coroutine: rewrite pool to avoid mutex |
Date: |
Fri, 28 Nov 2014 17:40:23 +0100 |
User-agent: |
Mutt/1.5.21 (2010-09-15) |
Am 28.11.2014 um 15:12 hat Paolo Bonzini geschrieben:
> I still believe we will end with some kind of coroutine bypass scheme
> (even coroutines _do_ allocate an AIOCB, so calling bdrv_aio_readv
> directly can help), but hey it cannot hurt to optimize hot code.
Not sure if speculations about the future belong into commit messages,
but while it may turn out that a bypass is required in the end (I hope
it doesn't), the part about AIOCBs is wrong if you really consistently
use coroutines all the way down from the device to the block driver.
I think Peter picked up all of my patches to actually handle requests
this way (i.e. virtio-blk already creates the coroutine).
Kevin
- [Qemu-devel] [PATCH 0/7] coroutine: optimizations, Paolo Bonzini, 2014/11/28
- [Qemu-devel] [PATCH 1/7] coroutine-ucontext: use __thread, Paolo Bonzini, 2014/11/28
- [Qemu-devel] [PATCH 2/7] qemu-thread: add per-thread atexit functions, Paolo Bonzini, 2014/11/28
- [Qemu-devel] [PATCH 3/7] test-coroutine: avoid overflow on 32-bit systems, Paolo Bonzini, 2014/11/28
- [Qemu-devel] [PATCH 4/7] QSLIST: add lock-free operations, Paolo Bonzini, 2014/11/28
- [Qemu-devel] [PATCH 5/7] coroutine: rewrite pool to avoid mutex, Paolo Bonzini, 2014/11/28
- Re: [Qemu-devel] [PATCH 5/7] coroutine: rewrite pool to avoid mutex,
Kevin Wolf <=
[Qemu-devel] [PATCH 6/7] coroutine: drop qemu_coroutine_adjust_pool_size, Paolo Bonzini, 2014/11/28
[Qemu-devel] [PATCH 7/7] coroutine: try harder not to delete coroutines, Paolo Bonzini, 2014/11/28