On Tue, Nov 01, 2005 at 12:55:31AM -0500, Dmitry Torokhov wrote: > Now you are leaking memory if something else fails... FOr example when > chip is not present. Good point. I guess the original comment is because the final dmasound_init() can fail but we'll still have all sorts of memory, irq's and io that aren't cleaned up. So your previous patch probably introduces the least problems. -i
Attachment:
signature.asc
Description: Digital signature
- Follow-Ups:
- Re: [PATCH] Convert dmasound_awacs to dynamic input_dev allocation
- From: Dmitry Torokhov <[email protected]>
- Re: [PATCH] Convert dmasound_awacs to dynamic input_dev allocation
- References:
- [PATCH] Convert dmasound_awacs to dynamic input_dev allocation
- From: Ian Wienand <[email protected]>
- Re: [PATCH] Convert dmasound_awacs to dynamic input_dev allocation
- From: Dmitry Torokhov <[email protected]>
- Re: [PATCH] Convert dmasound_awacs to dynamic input_dev allocation
- From: Ian Wienand <[email protected]>
- Re: [PATCH] Convert dmasound_awacs to dynamic input_dev allocation
- From: Dmitry Torokhov <[email protected]>
- [PATCH] Convert dmasound_awacs to dynamic input_dev allocation
- Prev by Date: Re: [PATCH] Convert dmasound_awacs to dynamic input_dev allocation
- Next by Date: [PATCH] Simpler signal-exit concurrency handling
- Previous by thread: Re: [PATCH] Convert dmasound_awacs to dynamic input_dev allocation
- Next by thread: Re: [PATCH] Convert dmasound_awacs to dynamic input_dev allocation
- Index(es):
 |