> Correct, so if the app client sets an allocation length
> of 3, then at your discretion, you can either leave the
> code doing what it does now, or return those 3 bytes.
> IOW leave it alone, improve it but don't make it worse.
Ack, got the new spec and looks like the check < 4 is the correct on
g
> Subject: RE: [PATCH] scsi_debug: rework resp_report_luns
>
> > On 15-02-24 04:37 PM, Tomas Winkler wrote:
> > > 1. Fix the error check: the alloc length should be > 16
> > > and not > 4
> >
> > You are proposing to make a marginally incorrect test
On 15-02-25 02:54 AM, Winkler, Tomas wrote:
On 15-02-24 04:37 PM, Tomas Winkler wrote:
1. Fix the error check: the alloc length should be > 16
and not > 4
You are proposing to make a marginally incorrect test
completely incorrect!
Quoting from the spec:
The great thing about standards is t
> On 15-02-24 04:37 PM, Tomas Winkler wrote:
> > 1. Fix the error check: the alloc length should be > 16
> > and not > 4
>
> You are proposing to make a marginally incorrect test
> completely incorrect!
Quoting from the spec:
The ALLOCATION LENGTH field is defined in 2.2.6. The allocation length
On 15-02-24 04:37 PM, Tomas Winkler wrote:
1. Fix the error check: the alloc length should be > 16
and not > 4
You are proposing to make a marginally incorrect test
completely incorrect!
The governing rule for almost all "allocation length" fields
in SCSI commands (that return data-in) is:
"A
1. Fix the error check: the alloc length should be > 16
and not > 4
2. Remove duplicated boundary checks which simplify
the fill-in loop
3. Use more of scsi generic API
Signed-off-by: Tomas Winkler
---
drivers/scsi/scsi_debug.c | 119 +-
1 file changed
6 matches
Mail list logo
