Re: [PATCH V1] mtd : nand : denali :- No need of devm functions

From	Boris Brezillon <boris.brezillon@free-electrons.com>
Newsgroups	linux.kernel
Subject	Re: [PATCH V1] mtd : nand : denali :- No need of devm functions
Date	2017-01-02 10:00 +0100
Message-ID	<sV6Df-4iu-23@gated-at.bofh.it> (permalink)
References	<sLMaL-1Ll-77@gated-at.bofh.it> <sTMGB-14H-5@gated-at.bofh.it> <sV5Hb-3CL-5@gated-at.bofh.it>
Organization	linux.* mail to news gateway

Show all headers | View raw

On Mon, 2 Jan 2017 13:26:01 +0530
Arvind Yadav <arvind.yadav.cs@gmail.com> wrote:

> yes, if Memory is live out side function. Then devm_kzalloc()
> approach has the benefit of simplifying the different error paths.
> 
> Here, Memory is alive with in function. we are going to free allocate memory
> then why we need devm api. In this case Devm will first add this entry to
> list and immediately it will remove from list. In this case, It's just a 
> overhead
> for devm api.

Yes, it adds a small overhead, but ITOH, it simplifies the code (see
the kfree() calls you added in different error paths with your
approach). Sometime a small runtime overhead (especially when the code
is executed once at probe time) is acceptable if it improves
readability.

Back to linux.kernel | Previous | Next — Previous in thread | Find similar | Unroll thread

Thread

Re: [PATCH V1] mtd : nand : denali :- No need of devm functions Boris Brezillon <boris.brezillon@free-electrons.com> - 2016-12-29 18:30 +0100
  Re: [PATCH V1] mtd : nand : denali :- No need of devm functions Arvind Yadav <arvind.yadav.cs@gmail.com> - 2017-01-02 09:00 +0100
    Re: [PATCH V1] mtd : nand : denali :- No need of devm functions Boris Brezillon <boris.brezillon@free-electrons.com> - 2017-01-02 10:00 +0100

csiph-web