[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Qemu-block] [Qemu-devel] [PATCH 2/7] block: Add Error parameter to
From: |
Max Reitz |
Subject: |
Re: [Qemu-block] [Qemu-devel] [PATCH 2/7] block: Add Error parameter to bdrv_amend_options |
Date: |
Wed, 2 May 2018 20:08:59 +0200 |
User-agent: |
Mozilla/5.0 (X11; Linux x86_64; rv:52.0) Gecko/20100101 Thunderbird/52.7.0 |
On 2018-05-02 19:52, Eric Blake wrote:
> On 04/21/2018 11:54 AM, Max Reitz wrote:
>> Looking at the qcow2 code that is riddled with error_report() calls,
>> this is really how it should have been from the start.
>>
>> Signed-off-by: Max Reitz <address@hidden>
>> ---
>
>> +++ b/block/qcow2.c
>> @@ -4049,7 +4049,8 @@ static int qcow2_load_vmstate(BlockDriverState
>> *bs, QEMUIOVector *qiov,
>> * have to be removed.
>> */
>> static int qcow2_downgrade(BlockDriverState *bs, int target_version,
>> - BlockDriverAmendStatusCB *status_cb, void
>> *cb_opaque)
>> + BlockDriverAmendStatusCB *status_cb, void
>> *cb_opaque,
>> + Error **errp)
>> {
>> BDRVQcow2State *s = bs->opaque;
>> int current_version = s->qcow_version;
>> @@ -4058,13 +4059,17 @@ static int qcow2_downgrade(BlockDriverState
>> *bs, int target_version,
>> if (target_version == current_version) {
>> return 0;
>> } else if (target_version > current_version) {
>> + error_setg(errp, "Cannot downgrade an image from version %i
>> to %i",
>
> %i is unusual compared to %d, but as they do the same, I won't make you
> change it, if we keep it. That said, this function is static; and the
> lone caller qcow2_amend_options is already doing sanity checks, right?
> (Yes - it does qemu_opt_get(opts, BLOCK_OPT_COMPAT_LEVEL), and then
> validates that the user's information is in range, which already filters
> out everything but v2 and v3; then _this_ function is called exactly for
> 3 downto 2). So this should be an assert() instead of an error message.
OK, I'll change it to an assertion.
>> + current_version, target_version);
>> return -EINVAL;
>> } else if (target_version != 2) {
>> + error_setg(errp, "Cannot downgrade an image to version %i
>> (only "
>> + "target version 2 is supported)", target_version);
>
> And again.
>
>> @@ -4072,6 +4077,8 @@ static int qcow2_downgrade(BlockDriverState *bs,
>> int target_version,
>> if (s->incompatible_features & QCOW2_INCOMPAT_DIRTY) {
>> ret = qcow2_mark_clean(bs);
>> if (ret < 0) {
>> + error_setg(errp, "Failed to make the image clean: %s",
>> + strerror(-ret));
>
> error_setg_errno(), please.
Right, forgot we had that...
>> @@ -4094,6 +4103,8 @@ static int qcow2_downgrade(BlockDriverState *bs,
>> int target_version,
>> ret = qcow2_expand_zero_clusters(bs, status_cb, cb_opaque);
>> if (ret < 0) {
>> + error_setg(errp, "Failed to turn zero into data clusters: %s",
>> + strerror(-ret));
>
> and again
>
>> return ret;
>> }
>> @@ -4101,6 +4112,8 @@ static int qcow2_downgrade(BlockDriverState
>> *bs, int target_version,
>> ret = qcow2_update_header(bs);
>> if (ret < 0) {
>> s->qcow_version = current_version;
>> + error_setg(errp, "Failed to update the image header: %s",
>> + strerror(-ret));
>
> etc. I'll quit pointing it out, but there are more.
>
>> @@ -4293,18 +4311,17 @@ static int
>> qcow2_amend_options(BlockDriverState *bs, QemuOpts *opts,
>> int refcount_order = ctz32(refcount_bits);
>> if (new_version < 3 && refcount_bits != 16) {
>> - error_report("Different refcount widths than 16 bits
>> require "
>> - "compatibility level 1.1 or above (use
>> compat=1.1 or "
>> - "greater)");
>> + error_setg(errp, "Different refcount widths than 16 bits
>> require "
>> + "compatibility level 1.1 or above (use
>> compat=1.1 or "
>> + "greater)");
>
> Not your fault, but that reads poorly. Better would be:
>
> Refcount widths other than 16 bits require compatibility level 1.1 or above
Sure it's my fault, I wrote this in the first place. :-)
Max
>> +++ b/qemu-img.c
>> @@ -3730,10 +3730,10 @@ static int img_amend(int argc, char **argv)
>> /* In case the driver does not call amend_status_cb() */
>> qemu_progress_print(0.f, 0);
>> - ret = bdrv_amend_options(bs, opts, &amend_status_cb, NULL);
>> + ret = bdrv_amend_options(bs, opts, &amend_status_cb, NULL, &err);
>> qemu_progress_print(100.f, 0);
>> if (ret < 0) {
>> - error_report("Error while amending options: %s",
>> strerror(-ret));
>> + error_report_err(err);
>
> Definitely nicer, but does change various outputs; glad you fixed
> iotests to match.
>
> Overall a nice patch, looking forward to v2.
>
signature.asc
Description: OpenPGP digital signature