qemu-devel
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Qemu-devel] [PATCH v2 3/3] ahci-test: test atapi read_cd with bcl,


From: Kevin Wolf
Subject: Re: [Qemu-devel] [PATCH v2 3/3] ahci-test: test atapi read_cd with bcl, nb_sectors = 0
Date: Wed, 2 Nov 2016 14:33:05 +0100
User-agent: Mutt/1.5.21 (2010-09-15)

Am 01.11.2016 um 04:16 hat John Snow geschrieben:
> Commit 9ef2e93f introduced the concept of tagging ATAPI commands as
> NONDATA, but this introduced a regression for certain commands better
> described as CONDDATA. read_cd is such a command that both requires
> a non-zero BCL if a transfer size is set, but is perfectly content to
> accept a zero BCL if the transfer size is 0.
> 
> This test adds a regression test for the case where BCL and nb_sectors
> are both 0.
> 
> Flesh out the CDROM tests by:
> 
> (1) Allowing the test to specify a BCL
> (2) Allowing the buffer comparison test to compare a 0-size buffer
> (3) Fix the BCL specification in libqos (It is LE, not BE)
> (4) Add a nice human-readable message for future SCSI command additions
> 
> Signed-off-by: John Snow <address@hidden>

> diff --git a/tests/libqos/ahci.c b/tests/libqos/ahci.c
> index 5180d65..15fa888 100644
> --- a/tests/libqos/ahci.c
> +++ b/tests/libqos/ahci.c
> @@ -864,16 +865,12 @@ AHCICommand *ahci_command_create(uint8_t command_name)
>      return cmd;
>  }
>  
> -AHCICommand *ahci_atapi_command_create(uint8_t scsi_cmd)
> +AHCICommand *ahci_atapi_command_create(uint8_t scsi_cmd, uint16_t bcl)
>  {
>      AHCICommand *cmd = ahci_command_create(CMD_PACKET);
>      cmd->atapi_cmd = g_malloc0(16);
>      cmd->atapi_cmd[0] = scsi_cmd;
> -    /* ATAPI needs a PIO transfer chunk size set inside of the LBA registers.
> -     * The block/sector size is a natural default. */
> -    cmd->fis.lba_lo[1] = ATAPI_SECTOR_SIZE >> 8 & 0xFF;
> -    cmd->fis.lba_lo[2] = ATAPI_SECTOR_SIZE & 0xFF;
> -
> +    stw_le_p(&cmd->fis.lba_lo[1], bcl);
>      return cmd;
>  }

If I'm not mistaken, you're changing the endianness here, which seems
to be a silent bug fix.

For some reason the test passes both ways. Does the actual value even
matter with AHCI, as long as it's non-zero? Do we end up with the same
result with BCL=0x0200 and BCL=0x0002, just that we split it into some
more iterations for the latter (or deeper recursion, actually)?

Kevin



reply via email to

[Prev in Thread] Current Thread [Next in Thread]