Re: [PATCH v6 32/33] crypto: delegate permission functions to JobDriver .pre_run

2022-01-28 Thread Emanuele Giuseppe Esposito
On 26/01/2022 17:10, Hanna Reitz wrote: > On 21.01.22 18:05, Emanuele Giuseppe Esposito wrote: >> block_crypto_amend_options_generic_luks uses the block layer >> permission API, therefore it should be called with the BQL held. >> >> However, the same function is being called by two BlockDriver >

Re: [PATCH v6 32/33] crypto: delegate permission functions to JobDriver .pre_run

2022-01-26 Thread Hanna Reitz
On 21.01.22 18:05, Emanuele Giuseppe Esposito wrote: block_crypto_amend_options_generic_luks uses the block layer permission API, therefore it should be called with the BQL held. However, the same function is being called by two BlockDriver callbacks: bdrv_amend_options (under BQL) and bdrv_co_a

Re: [PATCH v6 32/33] crypto: delegate permission functions to JobDriver .pre_run

2022-01-24 Thread Paolo Bonzini
On 1/21/22 18:05, Emanuele Giuseppe Esposito wrote: diff --git a/job.c b/job.c index 39bf511949..cf0dc9325a 100644 --- a/job.c +++ b/job.c @@ -967,11 +967,24 @@ static void coroutine_fn job_co_entry(void *opaque) aio_bh_schedule_oneshot(qemu_get_aio_context(), job_exit, job); } +static

[PATCH v6 32/33] crypto: delegate permission functions to JobDriver .pre_run

2022-01-21 Thread Emanuele Giuseppe Esposito
block_crypto_amend_options_generic_luks uses the block layer permission API, therefore it should be called with the BQL held. However, the same function is being called by two BlockDriver callbacks: bdrv_amend_options (under BQL) and bdrv_co_amend (I/O). The latter is I/O because it is invoked by